PDA

View Full Version : Canon 7D, meet the 4K MegaTrumpets res chart



Pages : [1] 2

Barry_Green
10-09-2009, 04:39 PM
Today I introduced the Canon 7D to the greatest res chart in the world, the DSC Labs MegaTrumpet 12-SW. This is a chart that was designed because conventional video camera res charts weren't good enough to deal with new cams like the Red One. And, in testing my HMC40 against my old res chart, yeah, that little cam bested what the chart could do -- so, I got myself the greatest res chart in existence.

When I saw how the Canon performed on this chart, in 720p mode, I couldn't believe my eyes, so I went and got the HMC40 (the only other 1920x1080 camera I have right now; the GH1 is out on loan) and put it in 720 mode and shot it too, and then split-screened the results here.

Check it out:
http://dvxuser.com/barry/7D-HMC40-720-Mid-Trumpets.jpg


And, for good measure, I of course also shot 1080/24 on both.
http://dvxuser.com/barry/7D-HMC40-1080-Mid-Trumpets.jpg

Wow. As "The Contintental" would say, wowie wow wow wow.

The little HMC40 blows the doors off the 7D when it comes to sharpness. In 720p it's embarrassing; the Canon puts up a much better fight in 1080, but it's still subject to some bizarre color aliasing and vertical aliasing as far out as 500 lines! I mean, keep in mind that this chart is perfectly black and white, there's no gray, no color, no anything, but look at the bizarre detail extinction on the horizontal trumpets at about 800 lines, and the color moire happening on the 45-degree angles.

By my readings of these charts, vertically, the Canon doesn't even register on the chart in 720p mode. The chart starts at 500 lines, and it's already screwed up with bizarre color patterns right where the chart starts. The HMC40, on the other hand, delivers pretty clean up to 700 lines. Horzontally, the Canon looks to be delivering about 550 lines before detail extinction, and the HMC40 starts aliasing about 550 lines before complete detail extinction at about 720. And what's with the weird aliased color-moire pattern in the central zone?

In 1080 mode, the HMC40 is clean as a whisper in vertical resolution, easily hitting 900 lines and complete detail extinction at 1000. Horizontally there's some alias pattern but still pretty clean res down to about 900 lines. And look at the 45-degree-angle trumpets -- crystal clean and perfect down to about 1100 lines, where the detail becomes completely extinguised. The Canon, on the other hand, what the... ??? Look at the straight horizontal trumpet, measuring vertical res. There's nothing good about it. You can see the aliasing wants to take over starting early, and the aliases actually cross over the lines at about 700. On the vertical trumpet (measuring horizontal resolution) it's a much cleaner story, but still low-res. It aliases (crossing lines) as early as 700 lines, with complete detail extinguishing at about 780. The HMC40 doesn't hit the detail-extinguish point until about 1200 lines!

It's weird all the way through. The HMC40's 45-degree is clean to 1100 lines, the 7D is heavily aliased at 600 lines and has completely failed by 800. Look at the 7D's 292-degree trumpet (the one above the horizontal trumpet) -- it's actually rendering those just-off-horizontal lines as vertical lines! I have never seen anything like that, ever. Biiiiiizzzaaarrrreee.

In another thread, Glenn Przyborski said "I don't want to get 7D members on my butt, but others at our edit facility who are used to screening high-end HD, thought the 7D 1080p footage had the sharpness and image quality of 720P, blown up to 1080." I think I'd have to agree.

Now, res charts don't mean everything, they're only one way to evaluate an image, but -- yeah.

I'll put the GH1 through the same tests when it arrives. I frankly don't expect it to perform any better than the 7D, I mean -- DSLR to DSLR, same basic technology in each, I expect similar results.

Stephen Mick
10-09-2009, 04:46 PM
Thanks for putting this together, Barry.

Can you tell us more about how you had the cameras set up? Lens on the 7D? Aperture and focal length? Mode settings?

I haven't shot any res charts like this, but my experience with the 7D was much the same. Take it out of the box. Point it at the same thing as another camera. Be disappointed at the softness of the image.

Then I played around with some of the in-camera settings, and the results were a definite increase in "perceived" sharpness (again, to my eye, with no res charts).

ydgmdlu
10-09-2009, 04:47 PM
The 5D Mark II captured decently sharp video. In our testing, the camera measured 700 line widths per picture height (lw/ph) horizontal and 650 lw/ph vertical. This is very close to the results we measured on both the Canon T1i and Panasonic GH1. The Canon HF S100 had the best sharpness of this bunch, coming in with 800 lw/ph horizontal and 650 lw/ph vertical. When testing the vertical sharpness on the 5D Mark II we noticed its image had similar aliasing and blur to the Canon HF S100. All this data is taken from testing the cameras and camcorder in their Full HD (1920 x 1080) modes.
So according to Camcorderinfo.com, the DSLRs can do only about 650-750 lines. That squares with the "720p blown up to 1080p" idea.

5D Mark II sharpness review (http://www.digitalcamerainfo.com/content/Canon-EOS-5D-Mark-II-Digital-Camera-Review-19948/Video-Motion-amp;-Sharpness.htm)
GH1 sharpness review (http://www.digitalcamerainfo.com/content/Panasonic-Lumix-GH1-Digital-Camera-Review-20751/Video-Motion-amp;-Sharpness.htm)

William_Robinette
10-09-2009, 04:50 PM
Good thing we don't shoot res charts on a regular basis!

Crazy stuff.

Matthew Bennett
10-09-2009, 04:50 PM
Thanks for the tests! How expensive are these high-end charts by the way?

NoxNoctus
10-09-2009, 04:55 PM
Good thing we don't shoot res charts on a regular basis!

Crazy stuff.

Aw shucks.... :(

I guess I'll just stick with the strong DOF control and low light capabilities, which the 7D beats that little HV in spades.


Yes, it's still appreciated and interesting to see

commanderspike
10-09-2009, 05:13 PM
I'll put the GH1 through the same tests when it arrives. I frankly don't expect it to perform any better than the 7D, I mean -- DSLR to DSLR, same basic technology in each, I expect similar results.

Not even my eyes can view that chart without failing :2vrolijk_08:

The GH1 will do better for moire and aliasing but will be a little softer in res. Just my little bet.

Michael Olsen
10-09-2009, 05:43 PM
On one hand, I'm glad I haven't been mistakenwhen I said 7D has some pretty wacked out chroma and luma aliasing.

On the other hand, I wish I had been.

Barry, what software settings were used for this? Have you tried shots with the contrast/sharpness all the way down?

Thank you very much for taking the time to post this up and take the time to really look into it.

Eddy Robinson
10-09-2009, 05:50 PM
My god...it's full of moire!!

http://t0.gstatic.com/images?q=tbn:IAskK8YX5SEmFM:http://forgetomori.com/wp-content/uploads/2008/12/bowmandsa2.jpg

I wonder how much this has to do with the sensors, vs the way the sensor is being read out for video? If you have the time, Barry, maybe you could trigger the 7d via USB and get a shot of the chart and a regular still picture of the chart - which I predict will be perfectly crisp (duh, Captain Obvious). Scaling the video up to match the resolution of the still image might yield clues about what kind of pixel binning or skipping Canon are using.

I still very much like the footage I'm seeing out of this camera under most conditions, particularly night footage, but this is certainly an unwanted annoyance.

ASG
10-09-2009, 05:53 PM
Well I'm not shooting res charts but I can't get an image out of this camera that I'm happy with, in any settings. It is aliasing hell. I thought a lot of the footage posted online looked great, and when you magnify the image looks amazing on the camera, then when you get it on the computer.... well not so much. I think it might be ok for half scale web video, and shallow dof shots, but deep dof on the 10mm seems like a non-starter.

hermmermferm
10-09-2009, 06:12 PM
Great tests Barry, but I'm confused.. Wouldn't the lenses play a major factor as well?

Zim
10-09-2009, 06:16 PM
That's what I was wondering too. are you using the kit lens?


Great tests Barry, but I'm confused.. Wouldn't the lenses play a major factor as well?

PaPa
10-09-2009, 06:21 PM
Well I'm not shooting res charts but I can't get an image out of this camera that I'm happy with, in any settings. It is aliasing hell. I thought a lot of the footage posted online looked great, and when you magnify the image looks amazing on the camera, then when you get it on the computer.... well not so much. I think it might be ok for half scale web video, and shallow dof shots, but deep dof on the 10mm seems like a non-starter.

Turn your sharpness down to 2 or less. Not getting Aliasing on my end.

Duke M.
10-09-2009, 06:28 PM
Barry, could you post the ISO, etc. In an ISO test by Herme there were definatly some ISO levels that had lots more grain than others. 200 seemed about the best with 320 not far behind. Also, sharpness on/off, etc.

I would read that in the 800-900 range on most of those, which on my test of the HV20 on tape was about the same.

I had to laugh at the moire patterns though because you didn't label which one was on the left or right. That nasty star moire pattern on the left is almost duplicated on the top and bottom on the right.

Thanks for the test. Will you be trying again with different settings since some people seem to be getting different results, like sharpness of to tone down moire.

Ian-T
10-09-2009, 07:02 PM
It's got to be the lens that would make the difference.

ASG
10-09-2009, 07:02 PM
Turn your sharpness down to 2 or less. Not getting Aliasing on my end.

Even sharpness down to 0 it's everywhere. On the camera LCD I magnify to the max to set the focus, the image looks awesome. Then on the computer it looks really soft and riddled with aliasing. I'm pretty disappointed I have to say. Maybe I'm doing something wrong...

NoxNoctus
10-09-2009, 07:03 PM
how are you viewing the files? quicktime?`

commanderspike
10-09-2009, 07:07 PM
Even sharpness down to 0 it's everywhere. On the camera LCD I magnify to the max to set the focus, the image looks awesome. Then on the computer it looks really soft and riddled with aliasing. I'm pretty disappointed I have to say. Maybe I'm doing something wrong...

The camera doesn't scale when magnifying the image during live view to check focus. So no aliasing is visible. It's only when the downscaled 1080p footage is written to the memory card that the artefacts appear.

Actually you can see it on live view when not magnified, but not as much as it's downscaling it to 640x480 which obliterates most fine detail anyway.

The 7D's screen is 640x480. It's max magnification in live view is 1:1 sensor native (I'm guessing).

So am I right in thinking the middle magnification is 1920x1080? If so why doesn't the 7D show moire in live view at this setting? Could there be a way to enable sensor windowing and avoid moire altogether, despite with a dramatic crop factor?

ASG
10-09-2009, 07:07 PM
Quicktime 7.4.5, After Efftects CS3, Vegas 7.0, they all look the same. I AE I set the colour space to sRGB to get the detial back in the blacks.

Duke M.
10-09-2009, 07:10 PM
Down rez-ing will ofter create moire patterns that aren't there in the original footage. If you're monitor isn't full resolution, or its not full screen that can cause it. Also it will only be at certain frequencies of repetition.

Barry_Green
10-09-2009, 07:26 PM
Thanks for the tests! How expensive are these high-end charts by the way?
About $800.

Barry_Green
10-09-2009, 07:29 PM
Thanks for putting this together, Barry.

Can you tell us more about how you had the cameras set up?
On the HMC40, standard default F1 scene file setting, and then I walked through the detail levels from -7 to 0 to +7. Iris was f/2.0, and in fact, there may have been 6dB of gain in there too...

On the 7D, 18-135 lens, set at about 80mm. F/5.6. I changed the ISO to 400 to get equivalent brightness. Then I did the same thing, setting sharpness to 0, then to 3, then to 7.

The charts you see here are from the midpoint in sharpness on both, so 0 on the HMC40, and 3 on the 7D.


Then I played around with some of the in-camera settings, and the results were a definite increase in "perceived" sharpness (again, to my eye, with no res charts).
I'd love to know what you changed; I'm only at the beginning of digging into the 7D and it can't even really be a priority until next week, which is okay because by then the GH1 should be here too, for some side-by-sides.

plasmasmp
10-09-2009, 07:30 PM
totally expected, but glad you did the test with the proper chart. Sharpness has to go to 0, and you have to do a chroma blur on the footage depending on how bad the moire is in each scene. Is it the greatest cam in the world? Nope, but I still like the image over a ex series and other full raster cams.

Barry_Green
10-09-2009, 07:30 PM
Good thing we don't shoot res charts on a regular basis!
True, but this same stuff will happen in the real world. At least we're starting to learn what to look out for. Picket fences, venetian blinds, etc... the moire that Barlow Elton got in the 720p mode was simply awful and it's something that we're gonna have to learn how to anticipate and avoid.

The stuff that shows up on charts will show up in the real world too.

Barry_Green
10-09-2009, 07:31 PM
Barry, what software settings were used for this? Have you tried shots with the contrast/sharpness all the way down?
Software settings in the cameras? As outlined above, stock all the way except for taking the sharpness through three stages.

Barry_Green
10-09-2009, 07:37 PM
I wonder how much this has to do with the sensors, vs the way the sensor is being read out for video?
Well, two things -- the sensor itself is capable of brilliance, as exhibited by the stills. But you can't have both fantastic stills and fantastic video -- one or the other HAS to give.

So -- the sensor is capable of full-frame genius, but not at the video frame rates. That's why they employ pixel-binning and downscaling. But a further problem is the optical low-pass filter -- that filter can only be tuned for video, or for stills, but not both. Well, for clarification, that filter can be tuned for an 18-megapixel frame, or a 2-megapixel frame, but not both.

So, the problem could be with the binning, or it could be with the OLPF; I tend to think it's a combination of both. You can't just put a stronger OLPF in there, because the way the binning works (or, at least the way I suspect it works) it NEEDS to receive that ultra high resolution image, because each destination pixel is made up of six spread-out tiny pixels, and if you put a stronger OLPF (basically a blurring filter) in place to drop the transmitted detail down to 2mpix, then the individual pixel detail that the six-pixel bin needs, won't have what it needs.

Basically, it comes down to this: when Canon or Panasonic (or, some day, Nikon or Sony) decide to make a video-first camera using these sensors, and they spring for the proper hardware and they decide to forget/screw the stills capability, then the video these cameras produce should be astonishing!


If you have the time, Barry, maybe you could trigger the 7d via USB and get a shot of the chart and a regular still picture of the chart - which I predict will be perfectly crisp (duh, Captain Obvious). Scaling the video up to match the resolution of the still image might yield clues about what kind of pixel binning or skipping Canon are using.
Dang -- I meant to do that. Actually what I had in mind was that I wanted to take a raw still of it from both cameras, because the 7D is a still camera that shoots video, and the HMC40 is a video camera that also takes 10.6-megapixel stills. So I meant to take stills of the chart. Oh well, that'll go back on my to-do list.

plasmasmp
10-09-2009, 07:37 PM
Moire is all over 720p mode no matter what you do. You have no choice but to blur the chroma as much as possible.

Barry_Green
10-09-2009, 07:39 PM
Great tests Barry, but I'm confused.. Wouldn't the lenses play a major factor as well?
The lens can play a factor, but shouldn't be playing much of a factor because the lens is designed to handle 18-megapixel stills. Now, if I was using some horrible $10 lens, I'd agree with you. But that lens is designed to be way better than what the video mode can handle... And the stills don't have the aliasing hell all over them! Which means the culprit is, as described above, the way the hardware is processing the image off the sensor.

But I can try again with my ZF 50mm and see what it delivers.

Barry_Green
10-09-2009, 07:39 PM
That's what I was wondering too. are you using the kit lens?
The 18-135 kit lens, yes.

Barry_Green
10-09-2009, 07:50 PM
Turn your sharpness down to 2 or less. Not getting Aliasing on my end.
Sharpness makes no difference. It exaggerates the contrast so that you can see black and white as black and white instead of shades of gray, but all the same aliasing is still there.

Here's the 7D with sharpness set to maximum, and with it set to minimum.

http://dvxuser.com/barry/7D-720-Min-Max-Trumpets.jpg

Barry_Green
10-09-2009, 07:53 PM
It's got to be the lens that would make the difference.
Doubtful. It's a $1700 camera. It costs less than anything but pocket cameras. It's got limited hardware. Canon said so, in the 5D. It's the hardware, it's not optimized for video.

It's not the lens, because (as I should hopefully show soon) if you use that exact same lens and shoot the chart as a still, it's going to be incredible. I'm assuming, of course, but I did that with the GH1 and the GH1 in still mode was phenomenal as compared to video mode.

It isn't the lens. But, like I said, I can reshoot with the ZF, which is the same optics Zeiss uses in the Compact Primes series of cine lenses.

hermmermferm
10-09-2009, 07:53 PM
Amazing information, thanks for the follow-ups! I agree with avoiding the fences and grates, although sometimes impossible. At least the human face isn't a rez chart.

Barry_Green
10-09-2009, 07:57 PM
Well, that's where these cameras excel. Interviews, etc., put some shallow DOF behind the subject and it'll be fantastic -- but I still wouldn't shoot 720p mode on the 7D now. I mean, it's barely standard-def. The 1080 mode is obviously much better.

But now, y'all know what I'm interested to try -- what'll happen if I put the Letus Ultimate on the HMC40, and compare that against the 7D's image? I mean, I know it's totally impractical, putting a $5,000 lens on a $2,000 camera, but... if the HMC40 delivers a so much crisper and cleaner image, and the Letus handles the DOF... how will they compare?

hermmermferm
10-09-2009, 08:09 PM
Well, that's where these cameras excel. Interviews, etc., put some shallow DOF behind the subject and it'll be fantastic -- but I still wouldn't shoot 720p mode on the 7D now. I mean, it's barely standard-def. The 1080 mode is obviously much better.

But now, y'all know what I'm interested to try -- what'll happen if I put the Letus Ultimate on the HMC40, and compare that against the 7D's image? I mean, I know it's totally impractical, putting a $5,000 lens on a $2,000 camera, but... if the HMC40 delivers a so much crisper and cleaner image, and the Letus handles the DOF... how will they compare?

I've already jumped the Brevis ship, so I might not like your results..

Barry_Green
10-09-2009, 08:13 PM
Well, keep in mind that even with the Brevis, you'd still be talking about a $3500 package (HMC40+Brevis) vs. a $2000 7D package, so even if the HMC40/Brevis spanks the 7D, they wouldn't be in the same price class at all.

Not to mention all the other aspects of image quality, such as grain/noise performance, low light sensitivity, dynamic range, all that other good stuff. That would all have to be compared!

Kellar42
10-09-2009, 08:21 PM
Very curious to see the GH1 tests.

Also, doesn't the 5D have one of the crispest images out there for less than $10,000 anyways? Does this mean that aside from the smaller sensor, there is something significantly different/off about the way the 7D handles the image from the 5D?

Lucian
10-09-2009, 08:21 PM
Well, that's where these cameras excel. Interviews,

The 7d's killer application.

Barry_Green
10-09-2009, 08:31 PM
Also, doesn't the 5D have one of the crispest images out there for less than $10,000 anyways?
Nowhere near. Not within a hundred miles; an EX1 or HPX300 or even an HMC40 will show a lot more sharpness than the 5D will. The zone plate that Alan Roberts shot for the 5D, when evaluating it for the BBC, shows the results; same kind of thing as the 7D has, but it was a zone plate and I don't have a zone plate chart so I couldn't see if they were handling identically.

You can see the 5D's report and zone plate extraction here: http://thebrownings.name/WHP034/pdf/WHP034-ADD39_Canon_5D_DSLR.pdf

I would bet that they are though. I thought the 7D would perform better since it's a year newer; without shooting the exact same chart I can't say whether it's performing better or not. But I'm pretty confident in saying that it's definitely not performing any worse than a 5D Mk II.

Barry_Green
10-09-2009, 08:31 PM
The 7d's killer application.
Especially at its price point.

NoxNoctus
10-09-2009, 08:31 PM
Well, that's where these cameras excel. Interviews, etc., put some shallow DOF behind the subject and it'll be fantastic



Speaking of....

50% res
http://captured-essence.com/photos/675586928_jjHPK-L.png



100% crop

http://captured-essence.com/photos/675586730_eWuCB-L.png

http://captured-essence.com/photos/675586804_U25Te-L.png


Shot on the 50mm f/1.8.


I guess the "No Striped Shirts" rule extends to "No shirts with anything resembling a shirt pattern" :-p

And god help you if they sweat.



I'm still floored with the cam, hopefully something can be dun bout dis though. It's not far from perfect for what I need it for

Barry_Green
10-09-2009, 08:35 PM
Well, that's the thing -- know what it can do, and what it can't do, and then decide if it works for you. Finely detailed patterns in sharp focus are going to be problematic. And in 720p mode, they might be very problematic. So let's know that, going in, so we have a better chance of getting great results.

All these cameras will fail at something. You cannot buy perfection for $1700, not when something that actually does approach "video perfection", like the HPX3700, costs $60,000. Something's gotta give, to get the price so low. So let's find out what the achilles heel is, and then we're prepared to work around it, embrace it, or forget it and wait for the next thing to come along...

Isaac_Brody
10-09-2009, 09:07 PM
So let's find out what the achilles heel is, and then we're prepared to work around it, embrace it, or forget it and wait for the next thing to come along...

Do you really think no one knows the achilles heels of each of these cameras at this point? I mean it's only hashed out in every other thread. I'm beginning to think it's a safety blanket and excuse so people don't actually make anything. It's the excuse not to make a film because it's got X problem or doesn't make toast.

Know what's crazy? I notice aliasing and moire in television shows on tv. No one else I know notices or cares about this stuff, but I actually notice it in big primetime shows all the time. That's definitely reminded me that we are a very small minority (thankfully) when it comes to noticing these issues.

hermmermferm
10-09-2009, 09:15 PM
Hahahahahaha, I've seen A LOT worse than moire on TV. Even QC'ed some of it recently!

sblfilms
10-09-2009, 09:29 PM
On the TV front, it is almost comical to me that the people producing big budget shows plan to show them in a format with image quality no where near what we are dealing with even on a cam like the 7D.

Broadcast TV is assy, and cable and satellite tend to be even worse. Hell, I think about all the DVDs I've watched over the years with terrible aliasing in high freq details.

I tend to agree that the "issues" with these cameras tend to be the crutch people use to not pick something and shoot with it. EVERYTHING has issues, learn them and work around them.

Mark Harris
10-09-2009, 10:01 PM
Know what's crazy? I notice aliasing and moire in television shows on tv. No one else I know notices or cares about this stuff, but I actually notice it in big primetime shows all the time. That's definitely reminded me that we are a very small minority (thankfully) when it comes to noticing these issues.

Funny I have been re-watching The Wire with my GF and noticed moire in several shots.

But I really agree with barry on this. Know what it's good for, and what it's not.

However, I guess I have to cut those focus charts out of my next film...:( I had the most amazing love scene planned, right in front of a focus chart...damn.

Rakesh Jacob
10-09-2009, 10:26 PM
However, I guess I have to cut those focus charts out of my next film...:( I had the most amazing love scene planned, right in front of a focus chart...damn.

Good the aliasing/moire would be too distracting LOL

Jean Dantes
10-09-2009, 11:27 PM
The results of this test are still rather disturbing...

Jean Dantes
10-09-2009, 11:32 PM
Barry, if you have time, would it be ok if you posted a live-action frame of anything (cat, flower, person) from the 7D @ 720P and the HMC40 @ 720P, and then again at 1080P for both. Being a novice to all of this stuff, I'd really like to see how sharp they are in reference to one another in a real life situation.

ydgmdlu
10-09-2009, 11:55 PM
Do you really think no one knows the achilles heels of each of these cameras at this point? I mean it's only hashed out in every other thread. I'm beginning to think it's a safety blanket and excuse so people don't actually make anything. It's the excuse not to make a film because it's got X problem or doesn't make toast.

Know what's crazy? I notice aliasing and moire in television shows on tv. No one else I know notices or cares about this stuff, but I actually notice it in big primetime shows all the time. That's definitely reminded me that we are a very small minority (thankfully) when it comes to noticing these issues.
Bravo, thank you!

squig
10-10-2009, 01:13 AM
Funny thing is I've never heard of anyone complaining about the MKII not being sharp enough, quite the contrary.

Uwe Lansing
10-10-2009, 03:09 AM
Today I introduced the Canon 7D to the greatest res chart in the world............

Interesting stuff Barry. Well done. But what exactly is the purpose of this test? I mean, are there any news?

Moire? It were made some very good sample clips like these ones:
http://www.vimeo.com/6549968
http://www.vimeo.com/6550033
So yes, 720p has a lot more moire + aliasing. We already knew that.

Sharpness? This has a lot to do with the lens. The kit lens 18-135 is quite soft + a bit unsharp at the wide-angle, especially at the widest aperture. - So, did i overlook something important? Please let me know... (i galopped pretty fast through the thread)...

ChipG
10-10-2009, 05:17 AM
Wow, this test shows how much people are 'snake charmed' by the shallow depth of field look. I'm blown away by the resolution performance of the little hmc40 but I'd still take the 7D over it anyday.

mhood
10-10-2009, 05:42 AM
Know what's crazy? I notice aliasing and moire in television shows on tv.

I was watching Larry King Live on CNN last night. It's unbelievable what that Christmas light map behind them does on a wide shot...and they keep taking the wide shots.

KeithAndrews.TV
10-10-2009, 06:03 AM
Well it's no surprise that you get better results with a video camera shooting video versus a stills camera shooting video. Obviously if you turned the table and compared the HMC40 as a stills camera, it would pale in comparison to the 7D.

The reason the 5D Mark II, D90, GH1 and now 7D look so nice and sharp to us all is because we are use to retro fitting a third party adapter to our video cameras and shooting through ground glass, which in my experiences has always resulted in a softer image.

I've been using the 7D for over a week now, and am very happy with the results. Haven't noticed much moire, or aliasing, no rolling shutter during camera movements, and resolution looks great. I'll be using the camera next week on two commercial shoots which I am very excited about. I wouldn't go back to a dof adapter/camera combo for anything now that I have the 7D. Just not worth the cost, size, or setup time for me.

Michael Olsen
10-10-2009, 06:08 AM
Barry, thanks again for posting these up and taking the time to shoot them, not to mention taking the financial responsibility of purchasing the chart. I really appreciate it.


Interesting stuff Barry. Well done. But what exactly is the purpose of this test?

It's a benchmark performance evaluation - think of it as a controlled experimental study. By shooting charts in a controlled environment, we can learn a lot about how specific cameras perform and what kind of images we can get from them. It also allows a point-by-point aspect of comparison that allows one to match one cameras performance against another. It's point-by-point, thought, so this chart alone shouldn't make anyone scream:

"Oh my God, I've been betrayed! The 7D is worse than a cheap camcorder! What were you all thinking?!"

It can help people who are wondering:

Do I keep my EX-1/3 or sell it for a 7D kit? Will the image quality be the same? Worse? Better?

So really, its a bit like putting a car on a dynamometer - you test the capabilities of the machine in an experimental environment to see what it is capable of under rather arbitrary conditions (unless you've met some trumpet charts or zone plate charts walking around looking for acting jobs...). The information gained from these charts should (and does, from what I've seen) apply to real world scenarios and can also act as a guide to people just learning to shoot (like me!).

For instance, after looking at the 0° and 180° lines, I know that high detailed areas that are horizontal could be a problem - I need to watch out for that when I'm shooting. This has been seen in plenty of videos, from moire on roofs to bricks and power lines. So I know it exists in the lab and in the field - its a real problem that isn't caused by user error or because someone wasn't using a tripod. I can't just ignore it and blame it on someone else's ignorance or stupidity. I can't say

"Well, when it is in the hands of a REAL DP we won't see these problems - they really know how to shoot!"
This issue is going to be there, no matter what.



Sharpness? This has a lot to do with the lens. The kit lens 18-135 is quite soft + a bit unsharp at the wide-angle, especially at the widest aperture. - So, did i overlook something important? Please let me know... (i galopped pretty fast through the thread)...

The problem with sharpness in relation to the lens in this case is that the lens Barry is using can probably do an okay job of resolving at least 4.5K - at least twice that of what we are seeing here. We know, because its a still camera that shoots ~5.3K images! If the lens wasn't sharp enough to resolve a full 1080p, the 5.3K RAW stills that come out of the camera would look absolutely terrible. Also, the chart is rather smart in that it increases in detail at it grows closer to the center. This takes lens factors out again, as most lenses lose sharpness and gain artifacts towards the edges, but are sharpest and clearest in the center.

The problem here is what the camera is doing to take a 5.3K image and turn it into a 1080p image. Like Barry said, some of this could come from an OLPF poorly designed for video. I tend to believe thats a least part of it since I've heard OLPF and optical engineers talk about it in relation to this camera and...well, they know a lot more than I do, that's for sure. The other problem is the actual downscaling and pixel binning procedure used. It is obviously rather artifact-inducing. And compression may play a part as well.

This is a major factor as to why I believe that DSLRs are a stepping stone towards large CMOS video cameras. There are hardware and processing issues involved that, except for the very highest end (say, 5K capture ala RED EPIC) require the camera to produce either great photographs or great footage. So until photographers scream at the top of their lungs for better codecs, better downscaling, and video-centric OLPFs, sensors, and chips, DSLRs will remain weighted towards photography. At least, that's what I think.

Right now, I just want to get my camera!

mhood
10-10-2009, 06:18 AM
Folks have been softening video to make it more "film-like" for a very long time now, haven't they? And how can the 5D be so widely acclaimed and look so very good to so very many people yet perform much the same as the 7D on the resolution chart?

At any rate..The Grinch has stolen Christmas from me even before Amazon shipped it. I hate you Barry!!!!! And your stoopid chart!!!!!! (running away sobbing like a little girl)

:-)

andrew00
10-10-2009, 07:13 AM
Aye interesting tests, although I think ultimately we'll put aside the tests and go with the results - what we see and what impressions we feel.

I like the 7D because it's tiny and shoots awesome footage. It may not be perfect, but to get the same organic look I need to bring along my ex1+letus, which isn't small or lightweight!

Therefore whilst the 7D isn't perfect, it certainly has it's many usages and will be an excellent tool in the right hands.

Barry_Green
10-10-2009, 07:24 AM
Do you really think no one knows the achilles heels of each of these cameras at this point?
Well, when I see folks saying "should I sell my EX1 for a 7D", then yeah, I think people don't really have any idea what the true differences are. That's why I'm out here digging.

Right now we have folks saying "I don't get aliasing", even in this thread. And that's impossible -- the 7D is absolutely aliasing all over the place. So the difference must be that PaPa isn't shooting things where the artifact rears its head. But if he were to shoot that, it'd be all over his shot.

Then, the next thing we get is "maybe it's just a bad batch. Maybe they fixed it. Maybe they'll fix it with firmware." No. It's not a bad batch, every single one of them is going to perform the same (unless we're talking about the 1-in-1000 physically broken issue). If you underexpose a GH1 by 3 stops, you're going to get streaking in the image, and it doesn't matter if you "burn it in" or not. If you shoot fine detail on a 7D, you're going to get splotchy ruined color.

Does it matter? If you're shooting head-shot interviews, probably not. If you're shooting a wedding, and the bride's face is covered in a lacy veil, and the 7D turns that veil into a purple orange blotch -- I'd say yes it matters a LOT. That guy's gonna get fired, and probably have to refund the job, and get badmouthed.

I'm looking for the stuff that happens, and pointing out that it will happen on every single camera, every time the situation is encountered. So we have to think this through and know what these issues really mean. We've got to get our heads out of the clouds and get off the hype and get down to what these products ACTUALLY deliver.

Greenscreen? I'd be very curious to see how the 7D handles greenscreen. And no, I don't mean a bald guy. I mean a redheaded woman with flyaway hairs against a green background. Or a blonde. If the 7D pukes on fine detail and turns it into a blob of color, that's a nightmare for greenscreen work. And I think every potential user would want to know about it.


I mean it's only hashed out in every other thread. I'm beginning to think it's a safety blanket and excuse so people don't actually make anything. It's the excuse not to make a film because it's got X problem or doesn't make toast.
Well, some people may feel that way, and if that's the case, whatever. What I'm concerned with is that people are making ostentatious plans around these products, and they simply cannot deliver the moon on a platter. There are definite failings, and it's far more important to know first, foremost, up-front, what those situations are. I'm concerned that our membership are making business decisions and committing themselves to a course of action without understanding that there are factors where these products simply will not perform well.

I'm the first to say that the price-performance ratio of these things is off the hizzle. But that doesn't mean the performance is bulletproof, and it doesn't mean that people are able to shoot freely anything and everything. If anything, getting the low-priced product puts *more* responsibility on the shooter's shoulders, to make sure that their low-price product doesn't puke on them when they can least afford it.

Barry_Green
10-10-2009, 07:27 AM
Barry, if you have time, would it be ok if you posted a live-action frame of anything (cat, flower, person) from the 7D @ 720P and the HMC40 @ 720P, and then again at 1080P for both. Being a novice to all of this stuff, I'd really like to see how sharp they are in reference to one another in a real life situation.
That's on the schedule, yes -- but not a closeup of a cat or anything like that. Buildings, downtown, wide shots. That's where it's gonna matter. If the results of the res test mean anything, it means that the 7D (and presumably all DSLRs) simply can't handle fine detail in a wide shot. That's where the stark differences should come in.

Closeups, we already know the 7D excels at those. Of course, so did DV cameras; it's much harder to tell the difference between HD and SD on closeups, than it is on wide-angle shots.

Barry_Green
10-10-2009, 07:30 AM
Funny thing is I've never heard of anyone complaining about the MKII not being sharp enough, quite the contrary.
Again, we may be pointing at the real culprit then. If the "sharpness" that's making it through is false detail and color-fringed detail, then that could explain a lot of why people might be complaining about such things.

"sharpness" does not equal "resolution". There's a lot more to the perception of sharpness than simple resolution. You can have a very, very, very high resolution image that looks quite soft -- just suck all the contrast out of it. Shoot a 4,000 DPI faint-gray circle on a white sheet of paper, and you'll have a very high resolution, very "soft" image. Shoot a 4,000 DPI pitch-black circle on a white sheet of paper and you'll have a very high resolution (in fact, identically high resolution) very "sharp" image. Contrast matters as much or more in our perception of sharpness than does resolution.

Barry_Green
10-10-2009, 07:33 AM
Interesting stuff Barry. Well done. But what exactly is the purpose of this test? I mean, are there any news?
The news is that it tells us what to investigate further. Like I said, right now I'd be quite concerned about greenscreen performance, about things like a bridal veil or lace or other fine-detail fabrics, etc.

If there's "news", it's that the 7D in 720p mode is a standard-def camera, and in 1080 mode it's a 720p camera.


Sharpness? This has a lot to do with the lens. The kit lens 18-135 is quite soft + a bit unsharp at the wide-angle, especially at the widest aperture.
That's the first I've heard of that; all the reviews I've read have said that the 18-135 is fantastic.

But, if it's soft, that should *hide* the moire and the aliasing, not accentuate it.

But, again, I'll reshoot with the ZF and see if that changes anything.

alpi69
10-10-2009, 07:41 AM
thanks Barry. This test totally confirms that it is yet too early to solely rely on a DSLR if you make a living from filmprojects.
I like the HVX Panasonic mojo nd use it often when there is enough light. but the EX-1/3 dig into the market. there are projects were the sharper SONY look is just better. so for these jobs I use the EX cameras.
For other jobs I rent a HPX500.
and now there will be shots where the DSLRs dig into our videocameras toolbox (interviews, low DOF shots etc)

It is just a tool. And while you canīt drill a whole with a hammer (well you can smash a hole) you can hammer a nail into wood with it.
DSLRs are added to the toolbox and for some might replace other tools (e.g. the 35mm Adapter Brevis or LEtus etc), but not more.
The good news is: the DSLR is relatively inexpensive. If you had Canon before it is a no-brainer. If you had Nikon, look into their line-up of VDSLRs. If you are like me stuck with something else (in my cse Minolta, Sony lenses) you either swap to Canon/Nikon or wait for SONY. But in no way will the 7D be able to do everything for you in all situations and this reschart shows just a small portion of the limitations of these DSLRs. Still I get one haha

mcgeedigital
10-10-2009, 07:45 AM
Thanks for the test Barry.

squig
10-10-2009, 07:49 AM
The MKII appears to be quite a bit sharper at most ISO's http://www.vimeo.com/6902912
It's gonna be interesting to see how the 7D compares to the GH1.

Barry_Green
10-10-2009, 07:52 AM
You know what else this means? It means that when Canon does build a proper oriented-for-video camera off this chip, it's got the potential to have a HUGE image quality boost over what we're getting today. So -- if you like what you see from the 5D and the 7D now, y'ain't seen nothing yet!

squig
10-10-2009, 07:57 AM
Just imagine the potential of the 1D MKIV with 2 DIGIC V chips and no line skipping.

Zim
10-10-2009, 07:57 AM
Thanks Barry. People should know the limits and the faults of these cameras.

Barry_Green
10-10-2009, 07:58 AM
Just imagine the potential of the 1D MKIV with 2 DIGIC V chips and no line skipping.
That is EXACTLY what I mean. Tim Smith said that they recognize what we want, and he said "more expensive products will do it right", and I can only assume that's what he meant...

mhood
10-10-2009, 08:05 AM
Wouldn't shutter speed have a big effect on sharpness? I wonder if shooting at 1/500 or even 1/4000 would produce different numbers?

These folks ( http://www.digitalcamerainfo.com/content/How-we-test-digital-cameras-20069.htm#videosharpness ) describe a process for testing video sharpness with the camera in motion. They also mention a pronounced effect on sharpness in stills by increasing shutter speed.

On the test 7D footage of the dog catching the Frizbee, the version shot at 1/4000 is much sharper to my eyes than the version shot at 1/60. Of course, sharpness is only one piece of the puzzle and I can't get the stark raving sharpness of the slomo motocross footage out of my mind's eye.

booggerg2
10-10-2009, 08:08 AM
True, but this same stuff will happen in the real world. At least we're starting to learn what to look out for. Picket fences, venetian blinds, etc... the moire that Barlow Elton got in the 720p mode was simply awful and it's something that we're gonna have to learn how to anticipate and avoid.

The stuff that shows up on charts will show up in the real world too.

Okay how often in the real world do you find individual high contrasting details juxtaposed to approximately equivalent of 500 lines on your res chart? LOL :2vrolijk_08:

squig
10-10-2009, 08:09 AM
I think Matt's quote still applies with the 7D but to a lesser extent than with any previous DSLR.

Peter J. DeCrescenzo
10-10-2009, 08:10 AM
Thanks for the test, Barry!

I wonder if CapRock would let you borrow a set of their anti-moire filters for testing with the 7D & GH1, or if a lens rental house rents CapRock filters?

See:
http://www.caprockdev.com/antimoire.htm
http://www.cinema5d.com/viewtopic.php?f=14&t=3247&st=0&sk=t&sd=a&start=40

If the CapRock filters work properly at all, they might produce a better-looking (sharper with less artifacts) result than using a traditional optical diffusion filter, or blurring in post.

YMMV.

Barry_Green
10-10-2009, 08:17 AM
Wouldn't shutter speed have a big effect on sharpness? I wonder if shooting at 1/500 or even 1/4000 would produce different numbers?
In motion, perhaps, but not on stills.


On the test 7D footage of the dog catching the Frizbee, the version shot at 1/4000 is much sharper to my eyes than the version shot at 1/60. Of course, sharpness is only one piece of the puzzle and I can't get the stark raving sharpness of the slomo motocross footage out of my mind's eye.
Well, at 1/4000 you're going to remove all motion blur from the equation, so yes, each frame is going to be sharper, perhaps substantially so. But it also will have an "unnatural" look to the motion.

Barry_Green
10-10-2009, 08:18 AM
Okay how often in the real world do you find individual high contrasting details juxtaposed to approximately equivalent of 500 lines on your res chart? LOL :2vrolijk_08:
All the time, actually. Electricity wires, buildings with windows at a distance, certain fabrics like herringbone suits or knit shirts or tweed or denim, striped shirts on people at a slight distance, picket fences, shingled roofs, tiled roofs, everywhere.

Barry_Green
10-10-2009, 08:26 AM
That's a great idea, and I've filed a request. If they respond positively, I'll do a review and we'll see what effect they have.

animanus
10-10-2009, 08:26 AM
Hi Barry, quote from another forum:

"Other issues have often been attributed to using non-UDMA compact flash media. Canon's Tim Smith explicitly said back in July that they have seen this kind of an issue with the 5D in testing. Banding in high detail areas can be attributed to the codec being instructed to drop the rate to accommodate slower cards. The bitrate is variable in the camera, and it will respond accordingly if you are cheaping out on your media. Because I have exclusively used UDMA media, I've never seen such banding or any quality issue that I can't attribute to the binning problem."

Is it possible to check this?

Barry_Green
10-10-2009, 08:29 AM
They also mention a pronounced effect on sharpness in stills by increasing shutter speed.
Where is that mentioned? I searched the article for shutter speed, and the only thing I found that looked like what you were talking about was on the image stabilization test.

On a perfectly static shot, no movement in the shot or by the camera, shutter speed will have no effect on sharpness at all. But if the camera or subject is moving, then yes, a higher shutter speed will preserve a sharper image than a slower shutter speed would.

Tom Roper
10-10-2009, 08:29 AM
Thank you Barry, once and for all exposing the myth of 5DMkII/7D video. I own the 5DMkII and love it, but the aliasing was so profound I shot a slant edge chart and ran the result through imatest, and came up with the same conclusion, horizontal resolution under 600 lines, vertical was better. Aliasing is horrid, totally unacceptable.

But the larger point, is the unmerited hoopla about the video images. Of course it's wonderful for still photography.

mhood
10-10-2009, 08:30 AM
In motion, perhaps, but not on stills.

Halfway down the page at http://www.digitalcamerainfo.com/content/How-we-test-digital-cameras-20069.htm#videosharpness is a "Sample Sharpness & Shutter Speed Graph" that charts the effect of shutter speed on stills. True, the chart is a part of their IS testing but the line with IS off demonstrates the shutter speed to sharpness that I am wondering about.

I can't help but wonder if that is how the wonderful motocross slomo was done...720p 60fps 1/500 160. That footage seems to contradict everything these tests are showing us.

Barry_Green
10-10-2009, 08:30 AM
Hi Barry, quote from another forum:

"Other issues have often been attributed to using non-UDMA compact flash media. Canon's Tim Smith explicitly said back in July that they have seen this kind of an issue with the 5D in testing. Banding in high detail areas can be attributed to the codec being instructed to drop the rate to accommodate slower cards. The bitrate is variable in the camera, and it will respond accordingly if you are cheaping out on your media. Because I have exclusively used UDMA media, I've never seen such banding or any quality issue that I can't attribute to the binning problem."

Is it possible to check this?
Check what -- whether using cheap media compromises image quality? Canon's already saying that it happens, and their specific advice was to use only UDMA cards.

Craig Zarkos
10-10-2009, 08:32 AM
Barry, if you have time, would it be ok if you posted a live-action frame of anything (cat, flower, person) from the 7D @ 720P and the HMC40 @ 720P, and then again at 1080P for both. Being a novice to all of this stuff, I'd really like to see how sharp they are in reference to one another in a real life situation.
This would really be great.. for me too as a novice. All great and interesting info... My 7d is on the way . ( my first real camera) and I hope it will do what i need it to do!

Barry_Green
10-10-2009, 08:32 AM
Halfway down the page at http://www.digitalcamerainfo.com/content/How-we-test-digital-cameras-20069.htm#videosharpness is a "Sample Sharpness & Shutter Speed Graph" that charts the effect of shutter speed on stills.
Yeah, but that's not a static shot -- that's moving the camera. They're trying to intentionally induce motion blur (which lowers sharpness), and then they try with the image stabilization to see if it overcomes that.

Rory_B
10-10-2009, 08:40 AM
Intriguing test seņor Green. I've been doing what you suggested on the previous page for the last couple of days actually. I am shooting nothing but deep focus shots on the 7d with my 35mm zeiss to see how wide angle shots hold up(day and night nothing but 720). Hopefully I can have some stuff up for people to see in a day or 2.

However, my main reason for posting was to comment on your underrated Ebonics skills :D

mhood
10-10-2009, 08:43 AM
Yeah, but that's not a static shot -- that's moving the camera. They're trying to intentionally induce motion blur (which lowers sharpness), and then they try with the image stabilization to see if it overcomes that.

Shouldn't your tests do the same...induce motion? Video sharpness only counts in motion (or it's not really video is it?) and shutter speed gr8ly affects sharpness in motion. I'm grasping for something that explains the 720p 60fps motocross slomo piece...

Jason Ramsey
10-10-2009, 08:47 AM
if you move the camera or shoot subjects in motion, you are going to reduce sharpness.... using image stabilization and/or increasing shutter speed is only going to be beneficial in trying to recover some of that lost sharpness due to motion blur. However, increasing shutter speed may have other undesirable effects on the look of your footage depending on the type of project you are shooting. Weddings at uber-high shutter speeds aren't going to look all that good. action shots, sure... or shooting in 60p at high shutter speeds to get crisp slo-mo later... sure :)

So, if the cam is producing x rez completely stable and shooting non-moving subjects, it's going to produce less when in motion. or so i would imagine.

later,
Jason

animanus
10-10-2009, 08:48 AM
Check what -- whether using cheap media compromises image quality? Canon's already saying that it happens, and their specific advice was to use only UDMA cards.

what i meant to say was, please let us know what card you're using seeing as its known to be something that can cause degradation of image

Jason Ramsey
10-10-2009, 08:54 AM
I think that was Barry's not so clear way of saying he was using a UDMA card :)

Uwe Lansing
10-10-2009, 09:04 AM
...
That's the first I've heard of that; all the reviews I've read have said that the 18-135 is fantastic.
...

Just shortly (during soccer half-time break russia-germany). I rummaged the whole internet and didnīt find any reviews - only some user reports (and pics) in german + international groups. If you have a link, please post it.

Yes, the 18-135 is pretty sharp but not at the wide angle. Here at this site you can compare different canon lenses. I took as a candidate on the other side the 17-55/2.8: http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=678&Camera=474&Sample=0&FLI=2&API=0&LensComp=398&CameraComp=474&SampleComp=0&FLIComp=3&APIComp=1. Youīll notice how soft + unsharp it is at 35mm - just go with the cursor over the chart for comparison...

Stephen Mick
10-10-2009, 09:21 AM
Capice-

This has been exactly my experience with the 7D. I have that very same 17-55 f2.8 lens, and the images are (objectively, to my eye), as tack-sharp as I would expect. On a recent shoot our "A" camera was a 7D with the Canon 85mm F1.2L lens, and the images were spectacular. Sharp, beautiful bokeh and spot-on color rendering.

So my takeaway is this…the better the glass the sharper the images will be. Which makes me think about glass in general as it relates to many of the cameras we've been using (HVX/HPX/etc.). These cameras are more expensive than the 7D, but couldn't a significant part of that cost difference be in the glass? Companies like Panasonic, Sony and Canon have invested dollars and years developing optics for camcorders designed to resolve as much detail as possible. So, if our test is putting a low-end kit lens on the 7D, the result would likely be a lower-quality image, comparatively-speaking. All of this is to say I'm not surprised by Barry's results, especially given the wide variations in lens quality, even within the same lot of a given product.

It's certainly possible that, when we move from a world of fixed-lens shooting (where there are few, if any variables) to interchangeable-lens shooting (more variables), these things could affect not only sharpness, but also other issues (CA, diffraction, etc.). Maybe?

f64manray
10-10-2009, 09:23 AM
Damn, this really puts a crimp in my plans to produce The Trumpet Chart Wars, The Trumpet Chart Strikes Back and Return of the Trumpet Chart ..... and 20 years later Revenge of the Trumpet Chart.


Excellent thread and one of the reasons I come here. To see people put the time and energy into testing technical aspects that frankly I'm to lazy and busy to do. Very informative to know where the strengths and weaknesses are.


It seems like on the set in the real world these weaknesses can be completely mitigated by simply knocking the background out of focus with a shallow depth of field when encountering challenging subject matter. I just don't really see a downside here. It seems very easy to work around this issue. Plan your production accordingly and your back on track with your award winning Sundance film. :-)

It's a very small price to pay for what you get in terms of low light stellar image quality that even Red One from my reading can't keep up with.

Stephen Mick
10-10-2009, 09:30 AM
I think the take-away from Barry's testing is simple… Know what the camera can (and can't do) and shoot accordingly. Understand that, if you're going to rely on the 7D (or any DSLR) as your primary tool, there will be trade-offs.

But more than anything, there's simply no better way to understand any of these issues without going out and shooting footage on your own. Res charts, skew tests and low-light comparisons are all important things to do, and they add value to the community for sure, but someone always comes along and posts a piece that makes all of us step back and say, "you did that with the 7D?"

And chances are, they were the ones out shooting and testing while many of us, myself included, were debating and discussing.

mhood
10-10-2009, 09:32 AM
I think I understand, thanks. What still confuses me is that in these video sharpness tests: http://www.digitalcamerainfo.com/content/Canon-EOS-5D-Mark-II-Digital-Camera-Review-19948/Video-Motion-amp;-Sharpness.htm , the 5D scored lower than both the T1i and GH1 and all three vDSLRs scored lower than the HF S100.

BTW: This from the site: "We test sharpness with the camcorder in motion, rather than a static shot, for a simple reason. How often do your video clips contain no motion? Motion is inherent in the nature of video. It's not the method that manufacturers would prefer, but we think it makes sense."

I find it very confusing that the 5D is so widely acclaimed as producing beautiful sharp footage and scores below the GH1, T1i and S100. How big a factor does sharpness actually have in the overall esthetics of the footage? I have read many complaints about the stark sharpness of the "video look" and haven't read much blasting the 5D for its inherent softness. Do you think a higher shutter speed would have any effect on moire?

Barry_Green
10-10-2009, 10:04 AM
Shouldn't your tests do the same...induce motion? Video sharpness only counts in motion (or it's not really video is it?) and shutter speed gr8ly affects sharpness in motion. I'm grasping for something that explains the 720p 60fps motocross slomo piece...
In video we typically never touch the shutter speed. You set it at 1/48 for 24p, and 1/60 for 720/60p, and leave it alone. Changing to a short shutter speed produces a very unnatural staccato effect into the footage. You might like it for a special effect (think "28 Days Later" zombies, or "Gladiator" fight scenes) but for normal footage, messing with the shutter speed is just something that's not normally done.

Barry_Green
10-10-2009, 10:07 AM
what i meant to say was, please let us know what card you're using seeing as its known to be something that can cause degradation of image
UDMA Extreme III. And there's no possible way this card is introducing any sort of degradation -- I'm shooting very short clips, of a static subject -- even if the card totally puked, a long-GoP codec would SCREAM in perfection at a still shot.

There's absolutely nothing wrong with the tests, there's nothing defective about my camera, there's no UFO's landing anywhere, there's no radiation or zombies attacking anything. This is what the 7D does, pure and plain and simple. Just like there's no magical formula or burn-in that makes the streaking go away in the GH1, there's nothing going on here that's any different than will happen in every 7D in existence.

mhood
10-10-2009, 10:09 AM
In video we typically never touch the shutter speed.

I couldn't find the shutter speed adjustment on my Ikegami. lol! It only became a consideration for me when I bought a D90 and found out that I couldn't control it but was told I needed to. ;-)

Yea...these vDSLRs are a different breed of cat for sure...

robmneilson
10-10-2009, 10:10 AM
I refuse to believe this Barry! The 7D is unimpeachable, UFOs are real, Elvis never did drugs, and this tape I listen to when I sleep will raise my IQ!

Barry_Green
10-10-2009, 10:13 AM
I think I understand, thanks. What still confuses me is that in these video sharpness tests: http://www.digitalcamerainfo.com/content/Canon-EOS-5D-Mark-II-Digital-Camera-Review-19948/Video-Motion-amp;-Sharpness.htm , the 5D scored lower than both the T1i and GH1 and all three vDSLRs scored lower than the HF S100.
That doesn't confuse me at all. It totally confirms what I'm finding (as will be seen shortly when I post a comparison shot against the GH1).

The 7D and the 5D ain't sharp.


I find it very confusing that the 5D is so widely acclaimed as producing beautiful sharp footage and scores below the GH1, T1i and S100.
Beautiful footage, and sharp footage, aren't the same thing. Many HDV cameras can best an HPX170 on a res chart, but the HPX170 is widely acclaimed as having a gorgeous image. Sharpness is only one factor in the overall equation of how pleasing an image looks.


How big a factor does sharpness actually have in the overall esthetics of the footage?
Depends on what you're shooting. If you're shooting something where you need to see the detail, like landscapes of thin-leafed pine trees, then sharpness means a lot. If you're shooting faces, not so much, and too much sharpness could actually have a bit of a detrimental effect.

It's not the softness of the 7D that has me scratching my head and trying to think of the possible consequences -- it's that it blurs to color. That seems like it could really bite you at some point. Sooner or later I'll try some various greenscreen tests, since that seems the most likely place where color contamination could cause a problem.


Do you think a higher shutter speed would have any effect on moire?
Yes, a higher shutter speed will cause more moire. Not what you wanted to hear, but it's what the truth is. If you're tilting up past some mini-blinds, using a regular shutter, the blinds are going to be blurred together, and no moire will happen. But if you use a super-short shutter speed that freezes the blinds with no blur, then you'll see all the detail, and that detail will cause moire.

mhood
10-10-2009, 10:16 AM
There's a tape that makes me smarter!?!?!??? Where can I pre-order one?

hermmermferm
10-10-2009, 10:19 AM
-- it's that it blurs to color. That seems like it could really bite you at some point.

like this: http://www.vimeo.com/6546960 ?

Barry_Green
10-10-2009, 10:28 AM
like this: http://www.vimeo.com/6546960 ?
Well, yeah, that'd do it. But I don't know that that was "disastrous" like the video title said.

I'm thinking more like something of, say, a field of beige reeds blowing in the wind. That'd probably turn to a big red/purple/green mush. That's the kind of thing I'm worried about.

Barry_Green
10-10-2009, 10:40 AM
Okay, I want to forever dispell the myth that it has anything to do with the lens. Take a look at this chart extraction:

http://dvxuser.com/barry/MegaTrumpets-Still-Cam.jpg

THAT is what the 7D and the lens can do. What I did there is shoot a still photo, not video. I then resized the still photo down to 1920x1080 in PhotoShop, using bicubic sharper. Then I pulled out the relevant chunk out of the chart for all to see.

It. Is. Not. The. Lens. Causing. The. Moire. And. Glitching.

What's going on is that the way the camera has to read the chip (through pixel binning) combined with the inappropriate anti-alias filter (designed for stills and not video), combined with the inadequate hardware for downconverting to HD, is resulting in all the moire and aliasing.

Which pretty much directly matches what Canon's Tim Smith was saying. The appropriate hardware to do good video is not in the 7D or 5D. It will come, he pretty much hinted at exactly that, but he said it'll be in a "more expensive" product. Which, to me, means probably 1D Mk IV or an XLH2.

But it ain't the lens. It's not pixies or gamma radiation, it's not UFOs or tin-foil hats, it's not leprechauns causing it. It's just what you get for the $1700 price. GH1 has "mud", 7D has "aliasing/moire".

In many ways it's pretty much the same issue -- the Canon can't hold on to fine detail because of its hardware, the GH1 can't hold it because of its codec, but either way you're gonna have nastiness happening on fine detail in motion.

Barry_Green
10-10-2009, 10:42 AM
So, another test -- because so far, when comparing video, the 7D is getting its butt kicked, resolution-wise, by the HMC40. Are things really one-sided? Oh, I don't think so...

See, here we have two interesting products, as they're both hybrids; the 7D is a $2,000 camera (including kit lens) that's primarily a still camera, but also shoots video. The HMC40, on the other hand, is a $2,000 video camera that also shoots 10.6-megapixel stills. Where does one put their $2,000? I guess it all comes down to what your priorities are, stills or video... because while both can do both, they certainly can't do both equally!

So, when comparing video, at least in terms of sharpness and resolved detail, the video camera is taking the still camera out to the woodshed. So -- what happens when we turn the tables and compare them in still-camera functionality? Oh, about what you'd expect...

http://dvxuser.com/barry/Stills-7D-vs-HMC40.jpg

7D on top, HMC40 on the bottom (like you needed me to tell you that?) The 7D's still performance is about six million times better than the HMC40's still performance.

Anyway, this just goes to show that as far as "convergence" goes, we aren't there yet. You can optimize a camera for still performance, or for video, but not both. Not yet.

Stephen Mick
10-10-2009, 10:43 AM
The aliasing is not my question with regards to the lens. The sharpness is.

Barry_Green
10-10-2009, 10:54 AM
The aliasing is not my question with regards to the lens. The sharpness is.
The sharper the lens, the more aliasing you'll get. The aliasing is coming about because the lens is so sharp. The softer the lens, the less aliasing you'd see.

Look at post 101 -- the lens is extremely sharp, way way way way sharper than the 1080 video. Any softness being introduced is *not* because of the lens!

Stephen Mick
10-10-2009, 10:58 AM
Of course that's true. My interest is in getting back to the original test and why the 7D seems so much less sharp. I'm far less concerned with aliasing, though I may be in the minority.

Barry_Green
10-10-2009, 11:00 AM
Of course that's true. My interest is in getting back to the original test and why the 7D seems so much less sharp. I'm far less concerned with aliasing, though I may be in the minority.
Well, it's because the 7D's binning/scaling just isn't very good. The 7D is not a sharp video camera. Period. These tests are confirming what others have found. It's not the lens, because the lens has gobs of sharpness coming out of its ears. It's the processing of the image that causes it to be so soft. It's better in 1080, but even in 1080 mode it's only about as sharp as a good 720p camera.

hermmermferm
10-10-2009, 11:01 AM
Sticky this thread already!

mhood
10-10-2009, 11:10 AM
Sticky it in the spot with the fewest tear stains...

Michael Olsen
10-10-2009, 11:12 AM
I'll third the sticky request!

And say, once more, thanks to Barry for taking all of this time to shoot, post, compare, and answer questions. Most folks just stick up a still and disappear.

I look forward to a GH-1 (and 5D??) comparison!

Barry_Green
10-10-2009, 11:15 AM
The GH1/7D is going to get a little bit of exposure today, in that I recall that I do have a shot I took of the ISO 12233 printed-out chart, so I can re-shoot that with the 7D and that'll be an opportunity at a direct comparison.

It won't be a great comparison because ... well, let's just say that a home-printed, taped-together PDF printout isn't exactly up to my standards of what a res chart should be, and that's basically why I got the MegaTrumpets 12! But -- it will still be a direct comparison and, as such, valid in that respect. Should be up within an hour or so.

Michael Olsen
10-10-2009, 11:17 AM
It's better than hand-drawn! :beer:

You'll have to see if Jannard can ship you out an M-X cam to poke at! That would be interesting. :)

Barry_Green
10-10-2009, 11:30 AM
From what I understand of the DSMC's, the still res is the same as the motion video res. Because of that, they can use the same AA filter and they won't be compromising sharpness of one for the other. But, you'll probably have relatively limited still image size (the 3K scarlet would have maximum 6-megapixel stills) depending on the native res of the sensor. A 6K Scarlet should deliver 20-megapixel stills.

Michael Olsen
10-10-2009, 11:32 AM
From what I understand of the DSMC's, the still res is the same as the motion video res. Because of that, they can use the same AA filter and they won't be compromising sharpness of one for the other. But, you'll probably have relatively limited still image size (the 3K scarlet would have maximum 6-megapixel stills) depending on the native res of the sensor. A 6K Scarlet should deliver 20-megapixel stills.

Yep, that seems to be the deal. End of October can't come quickly enough in that regard...

Barry_Green
10-10-2009, 11:59 AM
Okay, last for now. I found a GH1 720p version of the hand-printed 12233 chart, so we can put that up against the 7D's 720p and see how they compare. I'll do 1080 testing next week when my GH1 gets back.

http://dvxuser.com/barry/7D-GH1-12233-720p.jpg

This is a 2x blowup of a portion of the chart that shows, really, all you need to see. At least in 720p mode, the GH1 slays in terms of sharpness, primarily in vertical resolution. The Canon's got weird color aliasing happening as soon as 330 lines, it's crossed the aliasing boundary at about 430, and it's totally lost all detail at 530. That's no better than an HVX200.

The GH1, on the other hand, shows clean and pristine up to about 550, and loses all detail at 700. That's solid 720p performance.

In terms of horizontal resolution, the 7D does much better than it did vertically, and even manages to eke out higher res than the GH1; the GH1 fades out at about 570 whereas the 7D ekes out about 620 lines.

As for color/aliasing -- the GH1 shows only the faintest bit of color mush beyond 700 lines of vertical res, whereas it's incredibly exaggerated on the 7D, starting at about 400 lines and getting obscene at about 600.

Kholi
10-10-2009, 12:04 PM
Thanks for your hard work as usual, Barry. I wish I had the patience to test like this. But it's definitely what I found as soon as I turned on the 7D.

Need 720? Shoot with a real camera. Heck, and in some cases, 1080 as well. LoL.

HMC-40 sounds like a sweet camera to have... sigh.

squig
10-10-2009, 12:15 PM
Keep in mind though that the GH1 codec falls apart with movement

Barry_Green
10-10-2009, 12:18 PM
The reason the 5D Mark II, D90, GH1 and now 7D look so nice and sharp to us all is because we are use to retro fitting a third party adapter to our video cameras and shooting through ground glass, which in my experiences has always resulted in a softer image.

It seems like it should, but does it really? Well, I guess it depends on the adapter. But seeing as I have a pristine chart of the HMC40, what would shooting through the adapter do to the sharpness?

Answer: not much.

The HMC40 raw is the right half of this picture:

http://dvxuser.com/barry/7D-HMC40-1080-Mid-Trumpets.jpg




and here's the HMC40 through the Ultimate:
http://www.dvxuser.com/barry/HMC40-Ultimate-ZF-1080-Trumpets.jpg


They're very close. If there's any softening going on, it's not much at all.

Kholi
10-10-2009, 12:19 PM
Holy crap. Letus Ultimate and HMC-40!?!?!? Wow...

Barry_Green
10-10-2009, 12:20 PM
Keep in mind though that the GH1 codec falls apart with movement
And the 7D falls apart even without movement. The GH1 looks like it delivers a sharper image to the codec, which then can pooch it. The 7D has a better codec, but it delivers a softer/muddied/aliased image to the codec.

Six of one, half dozen of the other. Pick your poison.

Barry_Green
10-10-2009, 12:21 PM
Holy crap. Letus Ultimate and HMC-40!?!?!? Wow...
Yeah, it's kind of ridiculous to put a $5,000 adapter on front of a $2,000 camera, but what the hey, I've got 'em hanging around so why not?

I'd like to do some comparison shots of the 7D vs. the Ultimate/HMC40 to see how they compare, but I don't know what I'd shoot other than a cat... bah. :)

Kholi
10-10-2009, 12:25 PM
Yeah, it's kind of ridiculous to put a $5,000 adapter on front of a $2,000 camera, but what the hey, I've got 'em hanging around so why not?

I'd like to do some comparison shots of the 7D vs. the Ultimate/HMC40 to see how they compare, but I don't know what I'd shoot other than a cat... bah. :)


No, it isn't that ridiculous. I'm just surprised at how much it resolves in comparison to the 7D, man. =T

And I bet it's NOT that bulky at all considering the small size of the HMC-40...

XLR inputs, internal NDs, 720/60P that's USEABLE. No codec falling apart...

squig
10-10-2009, 12:25 PM
And the 7D falls apart even without movement. The GH1 looks like it delivers a sharper image to the codec, which then can pooch it. The 7D has a better codec, but it delivers a softer/muddied/aliased image to the codec.

Six of one, half dozen of the other. Pick your poison.

"All i need in a camera is out there, just not in the same body at the same time. :-) " mattsand

I'm sticking with the MKII for now. The 1D MKIV is rumoured for January release, fingers crossed they get it right.

morgan_moore
10-10-2009, 12:32 PM
NO mention of 1080

is this thread just knocking the 720version

my impression is that the 720 looks very toy like, internet only use

1080 is better ??

S

Kholi
10-10-2009, 12:34 PM
I wouldn't doubt that the HMC+Ultimate combo would run the 7D into the ground @ 1080, to be honest. But good question.

ChipG
10-10-2009, 12:35 PM
Barry, thanks for doing this! I know your busy but if you have extra time could you please take a still pic of the hmc40, 7D & GH1 all together. I'm hoping I just found a new fairly small beach / travel cam :)

Ben_B
10-10-2009, 12:36 PM
NO mention of 1080

is this thread just knocking the 720version

my impression is that the 720 looks very toy like, internet only use

1080 is better ??

S

Barry said in the first post that people feel like the 1080p on the 7D just looks like upresed 720p.

Kholi
10-10-2009, 12:36 PM
Man I'm with you, Chip. I've been thinking about the HMC-40 with a knitted brow for weeks straight; I now have an endless headache.

It sounds like a REALLY friggin' good camera. Might be talkin' to Andrew at Spec-Comm soon. ICK

morgan_moore
10-10-2009, 12:42 PM
Barry said in the first post that people feel like the 1080p on the 7D just looks like upresed 720p.

Im not sure about that, http://vimeo.com/6960394, I can clearly see the 720 looking pretty poor

It may look like upressed 720 from a more expensive capture device that has a better codec

BUt 1080 is clearly better in my eyes

of course because of a lower shutter on 1080 I could shoot at a lower ISO than 720

Just my impressions

Hoping to some some outdoor tommorrow at lowest ISO

S

ChipG
10-10-2009, 12:44 PM
Man I'm with you, Chip. I've been thinking about the HMC-40 with a knitted brow for weeks straight; I now have an endless headache.

It sounds like a REALLY friggin' good camera. Might be talkin' to Andrew at Spec-Comm soon. ICK

Kholi,

Reason for me why I like the GH1 & 7D so much is because of their small size, throw them in a beach bag type camera, if this hmc40 is fairly small I'm going to buy one because it has xlr inputs (optional). I need a good small high quality travel cam and I'm not that worried about low light performance so it could be a big winner for me!

Barry_Green
10-10-2009, 12:45 PM
NO mention of 1080
Not sure what you mean, there's plenty of pix and talk about 1080 in this thread.


my impression is that the 720 looks very toy like, internet only use
If that. It's bad.


1080 is better ??
It is better than 720, yes. It doesn't match what a 1080 video camera can do though. And again, we're only talking in terms of resolved detail in this thread. Color and sensitivity and latitude, that's all for other threads, and I'm sure the Canon will redeem itself there.

But as an HD camera, it doesn't have much "H" in its "HD"; 720p mode is about standard-def res, and 1080p mode is about 720p-def.

Barry_Green
10-10-2009, 12:46 PM
I wouldn't doubt that the HMC+Ultimate combo would run the 7D into the ground @ 1080, to be honest. But good question.
In sharpness, yes. But def. not in sensitivity, and I'd like to see how the latitude and color play out.

I think this is a really great opportunity for folks to finally, once and for all, end all and be all, learn that "resolution" isn't everything!

Barry_Green
10-10-2009, 12:47 PM
Barry, thanks for doing this! I know your busy but if you have extra time could you please take a still pic of the hmc40, 7D & GH1 all together. I'm hoping I just found a new fairly small beach / travel cam :)
Don't have the GH1 here at the moment. The 7D and HMC40 are about the same size, volumetrically, when using the stock lens. When using a smaller prime the 7D is smaller. But it's still a behemoth compared to the GH1, it's probably a good 50% to 100% larger than the GH1.

Kholi
10-10-2009, 12:47 PM
In sharpness, yes. But def. not in sensitivity, and I'd like to see how the latitude and color play out.

I think this is a really great opportunity for folks to finally, once and for all, end all and be all, learn that "resolution" isn't everything!

Resolution is what I meant. Or resolved detail. I'm sure everything else it would fall short ASIDE from being a REAL camera.

Barry_Green
10-10-2009, 12:47 PM
Barry said in the first post that people feel like the 1080p on the 7D just looks like upresed 720p.
Well, to be exact, I quoted someone else who said that. That was not my assessment at the time, although it's becoming my assessment...

Barry_Green
10-10-2009, 12:48 PM
Man I'm with you, Chip. I've been thinking about the HMC-40 with a knitted brow for weeks straight; I now have an endless headache.

It sounds like a REALLY friggin' good camera. Might be talkin' to Andrew at Spec-Comm soon. ICK
The big major knock against the HMC40 is sensitivity. It's about 50 ISO. It's dark. Other than that, it's really cute and pretty darn good for the price.

Barry_Green
10-10-2009, 12:50 PM
Im not sure about that, http://vimeo.com/6960394, I can clearly see the 720 looking pretty poor

It may look like upressed 720 from a more expensive capture device that has a better codec

BUt 1080 is clearly better in my eyes
Okay, for clarity -- I don't think anyone is asserting that the 7D's 1080 is made from upressing its 720 -- that's obviously and clearly not the case whatsoever.

The 7D's 720p is bad. The 7D's 1080p looks like pretty good 720p, in terms of sharpness. That's what we're saying. It doesn't resolve the full detail that a native 1080p video camera can do, but it does pretty well as compared to a 720p video camera (meaning, 7D in 1080, video camera in 720).

Barry_Green
10-10-2009, 12:51 PM
Resolution is what I meant. Or resolved detail. I'm sure everything else it would fall short ASIDE from being a REAL camera.
That is my guess as well. But now I'm not taking *anyone's* word for it, hype or no hype, messiah or no messiah, holy grail or not, they're gonna have to prove themselves On. The. Court. Nobody's getting a free pass anymore.

ChipG
10-10-2009, 12:52 PM
The big major knock against the HMC40 is sensitivity. It's about 50 ISO. It's dark. Other than that, it's really cute and pretty darn good for the price.

Sounds like a great little outdoor daylight cam!

morgan_moore
10-10-2009, 12:53 PM
Sorry I re read some of the thread including the initial post including spotting the 1080 version !

The 1080 is 720 quote I would not be suprised by

I dont know too much about higher end video (never having payed with any of the $50k models) but am sure there is a reason why for example Sony sell cameras that cost 10X what the EX1 costs and the EX1 too

to expect greatness from any sub $10k device and low rate codec would surely be a little much to ask

In terms of the 1080 pics on the first page the other cam looks softer but does not moiree, this I have seen many times over the years having owned two filter free stills cams, the Kodak SLRn and the Sinar 22mp, both sharp as razors and moiree hell

The filter is obviously designed for stills 18mp frequencies

S

Kellar42
10-10-2009, 01:37 PM
I'm pleased at the way the GH1 is coming through all of this, and I can't wait till Barry gets his back to see some of these scientific tests applied to the 1080 mode. I'm not an expert yet, but I haven't had much ruined by mud, and it's starting to look the GH1's mud issues might not be as bad as the 7D aliasing/moire issues.

Either, way, this is all making me appreciate the fact that I still have my HFS100, which back in April, seemed like a good bump over the Hv30. As near as I can tell, the HFS series is essentially identical to the HMC40, with a less professional form factor and no XLR inputs, but less than half the price. And for crisp video, with lots of controls (if a bit of work to get at) you can't beat this baby.

Either way, we're living in interesting times. For a kid messing with different projects in a lower quality broadcast market, the GH1, HFS100, a decent tripod and a steadicam Merlin offer great looking, filmic video, low light performance, cinematic shallow dof and relative ease of use for less than $5,000. Suddenly I'm a major power in a region where the news is shot on DVXs. Now if I can only learn how to use everything...

Sumfun
10-10-2009, 01:51 PM
Thanks for doing the tests for us Barry. I was thinking about getting the 7D, but now I'll probably wait for the next generation, be it 1D mk4, or something from another company.

It seems like Canon can have a winner if they can throw in enough processing power to do the compression right. That is: proper scaling without skipping lines, upgraded codec (high profile 4:2:2 would be nice), and better heat management. Hopefully that will be coming soon.

Another option would be to do like Red and record raw images that can be compressed externally by software. This is nice for filmmakers, but for people who only want 1080 or lower resolution, it requires too much post-processing and storage.

dadoboy
10-10-2009, 02:06 PM
Thanks for the test Barry! Charts do have a purpose.

ChipG
10-10-2009, 02:08 PM
Hmm, wondering what the resolution is through the hdmi output. We know it's lower but just not sure how much lower.

powervideo
10-10-2009, 03:01 PM
Thanks for doing the tests for us Barry. I was thinking about getting the 7D, but now I'll probably wait for the next generation, be it 1D mk4, or something from another company.

From my point of view, I see Barry's somewhat depressing assessment as more a challenge than a capitulation. Rather than wait for the next big thing, I'm going full steam ahead on the 7D and learning to master the idiosyncrasies, strengths and weaknesses of the 7D.

If a superior 1D IV comes out in a few months then great. I'll flick the 7D off and go that way. At least I'll have all the support equipment and experience. No matter what the charts say, just look at so much of the great work that is out there; Bloom, Vienna, Stillmotion, etc etc. You more you know your camera, the more you can work to great work out of it.


Peter

NoxNoctus
10-10-2009, 03:07 PM
What I'm wondering, and I have no idea how it would work, if theres such a way to interject that "Binning" or whatever code with something more effective.

Almost on that note, what's the likelihood of it being "cured" with Firmware?

Michael Olsen
10-10-2009, 03:08 PM
From my point of view, I see Barry's somewhat depressing assessment as more a challenge than a capitulation. Rather than wait for the next big thing, I'm going full steam ahead on the 7D and learning to master the idiosyncrasies, strengths and weaknesses of the 7D.

If a superior 1D IV comes out in a few months then great. I'll flick the 7D off and go that way. At least I'll have all the support equipment and experience. No matter what the charts say, just look at so much of the great work that is out there; Bloom, Vienna, Stillmotion, etc etc. You more you know your camera, the more you can work to great work out of it.


Peter

I absolutely agree, Peter. The 7D won't give you the quality of an F35, DVX3700, or a RED ONE, but it can make pretty pictures. Hopefully, I won't have any time to spend on here...I'll be too busy shooting!

Michael Olsen
10-10-2009, 03:09 PM
What I'm wondering, and I have no idea how it would work, if theres such a way to interject that "Binning" or whatever code with something more effective.

Almost on that note, what's the likelihood of it being "cured" with Firmware?

Virtually none. Canon has already stated that they understand the problems, know how to fix them, and will make them readily available shortly. At a higher price point.

As for Magic Lantern, I'd wager at least half of the problem is hardware.

powervideo
10-10-2009, 03:14 PM
I'm looking at the 7D as a short term solution. My sights are on the 1D IV, but I'm not going to wait around and not shoot on this format over our summer.

Barry_Green
10-10-2009, 03:18 PM
If a superior 1D IV comes out in a few months then great. I'll flick the 7D off and go that way. At least I'll have all the support equipment and experience.
Exactly. These cams are so cheap, and ebay will pick up your castoffs, so there's no reason not to take the plunge with whatever you need at the current moment, and upgrade when you need to upgrade!

Barry_Green
10-10-2009, 03:19 PM
What I'm wondering, and I have no idea how it would work, if theres such a way to interject that "Binning" or whatever code with something more effective.
Binning is the process of combining multiple pixels on the sensor. It's implemented in hardware.


Almost on that note, what's the likelihood of it being "cured" with Firmware?
None whatsoever.

mhood
10-10-2009, 03:41 PM
With an eye on having to unload the 7D even faster than my D90 was obsoleted, I guess it's FF glass only? Anyone think Canon will get there on APS-C? I'm sure not buying any EF-S lenses. I'm also getting buyer's remorse before Amazon has even shipped my kit. It seems the GH1 might have been a much better choice for me...

Barry_Green
10-10-2009, 03:57 PM
It seems the GH1 might have been a much better choice for me... Why do you say that? They both have their issues. In fact, ANY camera in this price range, is going to have issues! It's guaranteed because of the price bracket, the price bracket dictates the kind of hardware they'll build into it.

Someone sent me a chart of the Pentax K-X -- it looked like a cross between the GH1 and 7D's chart. More res than the 7D, but with the same funky color aliasing.

So pick your poison. If you like the 7D's looks and you like the larger frame, then that's the one to get. If you like the GH1's functionality, that's the one to get.

Just don't go thinking that any of them are anywhere near a match for a $6,000 or $8,000 professional camcorder, because they're not. And don't go thinking that any one of them is dramatically or substantially better than another one. They may be in some ways, but will fall short in other ways.

Don't buy hype. Look at what the product actually does, and look at what your needs are, and then match the two together -- and don't care or even think about what "the other guy" bought -- who cares if everyone's going with the 7D, if it doesn't do what you want? Who cares if everyone's gaga over the Pentax K-X, if it doesn't do what you want?

That's all I'm trying to do -- find out what these products actually DO, and what they don't do.

Now, if you're saying that the GH1 is what you wanted, but you felt that the 7D was obviously "better" for some reason so you went with it for that, then -- sure, it may be time to re-evaluate. But if you like the 7D for what it is, why change to something else?

Ben_B
10-10-2009, 04:13 PM
Now, if you're saying that the GH1 is what you wanted, but you felt that the 7D was obviously "better" for some reason so you went with it for that, then -- sure, it may be time to re-evaluate. But if you like the 7D for what it is, why change to something else?

Well put. I went with the GH1 over the 5D because it was the camera I wanted more for a variety of reasons...now that the 7D is out the same holds true, I like the GH1 more...and it turns out the image is probably better in many situations or ways (but not in others.)

I think the reason he said this was because so many people touted the 7D as being the "better" camera in all IQ regards, without anyone realizing that this simply isn't the case. They both have strengths and weaknesses, and the only people really disappointed here should be people that expected the 7D to simply be the better camera...that's not how things work.

My GH1 is "better" than my friends HVX-200 + Letus extreme setup for certain things, portability, low-light, and even image quality in certain situations. His setup has clear advantages over the GH1 in other areas, like codec performance, sound, the other things that come along with being a dedicated video camera, etc.

No one camera is ever going to be simply better...

Michael Olsen
10-10-2009, 04:21 PM
Interesting comment from Graeme Natress @ REDUSER:


Also, what you can't see directly is that all these artifacts eat at the available data rate to record the image, and the more bits spent on aliases and artifacts, the less you have for your image data. And that problem travels with you all through to broadcast where again you're using a lossy motion adaptive codec.

Never thought about that, but it's important to understand!

Barry_Green
10-10-2009, 04:37 PM
It is indeed. And as far as "better" goes -- every manufacturer has the same basic budget, they're all trying to hit that $1500-$2000 mark. They will spend their dollars on the hardware they think best gets them what they are trying to do. Someone pointed out that in the pro video world, Panasonic would put out fantastic recording mechanisms and weak chips. Whereas Sony would put out fantastic chips and weak codecs, etc. Every manufacturer is going to deliver what they can, to hit a certain price point, and their priorities will vary.

Anyone who expects one of these products to be heads and shoulders better than another in all ways, is just asking for their expectations to be dashed.

What they may find, however, is that one is better than the other in all the ways that they care about -- and where it's worse, that may be on things that they don't care about. When someone is able to make that determination, that's when they're a happy camper.

If you want it all, you can have it all, but not at a $1700 price point, you're going to have to pony up the big bucks to have it all.

At the $1700 price point you can still get some amazing stuff. Yes, the 7D disappoints me in its resolved detail, but ... think about it -- if its image was only as sharp as an HVX (and in 720, it's not even as sharp as an HVX, in 1080 it's sharper) but even if it was only as sharp as an HVX... for $1700 you're getting the sharpness of an HVX, plus the shallow DOF of a $1500 adapter, plus more low light sensitivity -- it's still a heck of an accomplishment! At about the same cost as an adapter alone! (oh, and you also get a very, very, very good still camera too).

mhood
10-10-2009, 04:40 PM
I think the reason he said this was because so many people touted the 7D as being the "better" camera in all IQ regards

That's pretty much it. And now, after having bought 2 Canon EF USM Primes and a Canon Tamron in anticipation of the 7D, my bed is pretty much made for me here. I do think I'll regret not being able to use the many SDHC cards I have, record a talking head for as long as the head wants to talk, never have to worry about overheating and $500.

Michael Olsen
10-10-2009, 05:05 PM
The thing that is nice about these cameras as that, even though they aren't a godsend, they aren't useless either. The cost, even with some equipment, is relatively inexpensive and should be easy to recoup. Lenses, media, and audio equipment could all be used on future platforms.

Lucian
10-10-2009, 05:22 PM
Dunno guys, don't you think you're over valuing rez chart tests?

In many ways this camera will take down an EX3 (shallow focus shooting, low light, portability, lens options, media price).

boulder
10-10-2009, 05:29 PM
Arrrgh...wish I would have known this before getting my wife on board and turning my car down the 7D road. :(
I guess if something seems too good to be true...
Another lesson learned.

ChipG
10-10-2009, 05:44 PM
Come on man, if your having second thoughts just go watch Phillp Blooms little dublin footage again, learn the cam and it looks awesome!

Barry_Green
10-10-2009, 05:57 PM
In many ways this camera will take down an EX3 (shallow focus shooting, low light, portability, lens options, media price).
And in many, many, many ways, the EX3 will take down the 7D. Professional audio, screen viewing angles, variable frame rates, sharpness, customizability, control over the image, and -- well, it's an EX3, so you have many of the same lens options; the Adaptimax adapters let you use Nikon glass and probably Canon glass, so you probably have more lens options with the EX3. Plus no overheating issues, and the ability to roll continuously for hours of footage, power zoom, autofocus while recording video, the ability to actually plug in headphones and monitor your audio, zebras, and so many other things...

There is never any clear-cut answer as to what is empirically "better".

The charts are showing that the 7D does not deliver sharpness comparable to a conventional video camera, and that it has serious artifacting in its image, which will likely show up under fine-detail conditions. That's what the charts are showing. If that's what you intend to be shooting, then these charts are invaluable. If that's not what you intend to be shooting, if you're getting it for an "interview" cam for talking heads with shallow DOF, then it's great for that.

boulder
10-10-2009, 06:05 PM
Question, did no one do chart tests like these with the 5d?

Jim Klatt
10-10-2009, 06:06 PM
If you are having second thoughts...watch the footage on an HDTV. Your jaw will drop.

Ben_B
10-10-2009, 06:13 PM
Good call.

Looking at moire in chart tests doesn't make a damn bit of difference if you go shoot something amazing.

If your film is kick ass and has a bit of moire but a good story, good acting, and overall a great aesthetic, we're all going to love it. Besides it's not like we haven't been dealing with moire for years...it was really awful on most SD stuff IMO.

Tracey Lee
10-10-2009, 06:14 PM
Hey folks...go out and shoot something spectacular with the combination of all facets of film making; don't let a res chart test stop you.

boulder
10-10-2009, 06:15 PM
*and shoot really, really close up.

Getting it anyways as the Tamron lens on it's way and wife is insistant on getting an SLR now that I talked her into it.

Jason Ramsey
10-10-2009, 06:16 PM
it's just information... use it to arm yourself. no need to freak out... I know some people thought this was just the holy grail of cameras and all.... but, reality sets in eventuallly. it has limitations just like any other piece of equipment.

Ben_B
10-10-2009, 06:22 PM
Exactly.

Sorry to all the people who bought it expecting it to be the holy grail of cameras...everyone knows the 1d mk iv will be the holy grail...duh. :P

boulder
10-10-2009, 06:26 PM
Oh no no no, you know it will be the Scarlet, how dare ye!

At my next shoot:
Ok people, I need to get in a bit closer…
honey, I’m wondering if you could touch up some of those pores on your nose some…it’s just that I’m doing this movie in EXTREME close-ups only, I’m kind of avant-garde that way…

Kellar42
10-10-2009, 06:27 PM
This whole thing is changing the way I think about resolution and images. 6 months ago I didn't know that 1080p isn't always actually 1920 x 1080p and I'm still a little confused by the fact that it sounds like much of the footage we see on HDTV or blu-ray probably isn't exactly 'resolving' at full res.

My gut instinct is that EX sharpness and above is what I want out of an HD camera, and I didn't think I'd join the 'softer is more filmlike' camp, but I've got to say, what I've seen people do with a MKII and the GH1 (and maybe the 7D, we'll see) is pretty awesome looking. It seems to have a certain quality that is lacking in other video footage I've seen and I believe it's more than the shallow DOF (which I'm not a HUGE fan of, though I think it's useful), so I'm going to continue to try and get the best I can on these things.

I might also be wrong, but these DSLRS may be the best low-budget ways for a learner to learn lots of film type workflows. The paint settings, ISOs, dual system sound, manual focusing, it feels like a positive learning curve for things to come, although I wouldn't know for sure. There is more going on when I walk out the door with my GH1 than with another consumer cam.

I guess all I'm saying is that I love the charts, the science, and want to see and compare and learn all I can about them, but I wouldn't sell my new 7D based on this stuff if I had one!!! The night I got my GH1 the 7D was announced with the now somewhat disproved statement that it was going to annihilate the GH1 in IQ and all that. I didn't rush out and sell my camera, and send in a pre-order, I kept shooting with what I'd just got!

You 7D guys, I'd say, if you've got it, do the same! It's extremely cool that we have easy access to people like Barry giving us the skinny on how we get mud, moire, color shifting, banding, pattern noise, aliasing and soft images out of all our cameras, and how to avoid it, but I don't know that the knowledge that such things exist is an immediate deal-breaker.

(Five more posts appeared while I was writing this! Hot thread!)

f64manray
10-10-2009, 06:30 PM
And in many, many, many ways, the EX3 will take down the 7D. Professional audio, screen viewing angles, variable frame rates, sharpness, customizability, control over the image, and -- well, it's an EX3, so you have many of the same lens options; the Adaptimax adapters let you use Nikon glass and probably Canon glass, so you probably have more lens options with the EX3. Plus no overheating issues, and the ability to roll continuously for hours of footage, power zoom, autofocus while recording video, the ability to actually plug in headphones and monitor your audio, zebras, and so many other things...

There is never any clear-cut answer as to what is empirically "better".

The charts are showing that the 7D does not deliver sharpness comparable to a conventional video camera, and that it has serious artifacting in its image, which will likely show up under fine-detail conditions. That's what the charts are showing. If that's what you intend to be shooting, then these charts are invaluable. If that's not what you intend to be shooting, if you're getting it for an "interview" cam for talking heads with shallow DOF, then it's great for that.

Really? Only good for interview work? How about filming any narrative cinema pieces with people talking etc and controlling for problem subject matter with a shallow depth of field when encountered. These chart samples are great to push these cams to the edge and find their flaws, but lets not forget that DPs are using these cams in the industry and intercutting their footage with much more expensive cameras. Sorry, but just saying they're good for interview work is selling the 7D and 5DII (if your including the 5DII in that statement) a little short. These cams are being used already in the industry for far more than interview work.

The 7D would have kicked the DVX100's butt filming "November" The DVX was clearly struggling in some of those low lit scenes. Clean low light capability is meat and potatoes. Most of those other features you mentioned are dessert.

Michael Olsen
10-10-2009, 06:31 PM
This whole thing is changing the way I think about resolution and images. 6 months ago I didn't know that 1080p isn't always actually 1920 x 1080p and I'm still a little confused by the fact that it sounds like much of the footage we see on HDTV or blu-ray probably isn't exactly 'resolving' at full res.

My gut instinct is that EX sharpness and above is what I want out of an HD camera, and I didn't think I'd join the 'softer is more filmlike' camp, but I've got to say, what I've seen people do with a MKII and the GH1 (and maybe the 7D, we'll see) is pretty awesome looking. It seems to have a certain quality that is lacking in other video footage I've seen and I believe it's more than the shallow DOF (which I'm not a HUGE fan of, though I think it's useful), so I'm going to continue to try and get the best I can on these things.

I might also be wrong, but these DSLRS may be the best low-budget ways for a learner to learn lots of film type workflows. The paint settings, ISOs, dual system sound, manual focusing, it feels like a positive learning curve for things to come, although I wouldn't know for sure. There is more going on when I walk out the door with my GH1 than with another consumer cam.

I guess all I'm saying is that I love the charts, the science, and want to see and compare and learn all I can about them, but I wouldn't sell my new 7D based on this stuff if I had one!!! The night I got my GH1 the 7D was announced with the now somewhat disproved statement that it was going to annihilate the GH1 in IQ and all that. I didn't rush out and sell my camera, and send in a pre-order, I kept shooting with what I'd just got!

You 7D guys, I'd say, if you've got it, do the same! It's extremely cool that we have easy access to people like Barry giving us the skinny on how we get mud, moire, color shifting, banding, pattern noise, aliasing and soft images out of all our cameras, and how to avoid it, but I don't know that the knowledge that such things exist is an immediate deal-breaker.

(Five more posts appeared while I was writing this! Hot thread!)

Absolutely agreed! I can't wait to get mine and I think this is a really good attitude to have. There is a lot to learn here, not only about the specific camera, but the direction of digital cinematography in general. :thumbsup:

Jason Ramsey
10-10-2009, 06:42 PM
Really? Only good for interview work?

I didn't see anywhere that he said it's ONLY good for that... I believe it is called an example....

Also... keep in mind... one mans dessert may be another's meat and potatoes... the opposite being true as well.

Important for you is not necessarily important for someone else... There's a reason why I haven't jumped on the vdslr bandwagon. Sure, great price, outstanding relative image quality... But, they are missing many of the features that I consider mandatory/standard parts of the feature set... for ME... and I recognize that it's for ME.

Later,
Jason

Finster
10-10-2009, 06:49 PM
My clients have been blown away by the footage I've shot with the 5D & 7D. Of course, they know nothing about this current DSLR video revolution. And they surely know nothing about res charts.

They're happy. I'm happy. Presently, in my little corner of the video world, Canon rules! :happy:

Still, thanks for the tests Barry. Very interesting results.

squig
10-10-2009, 06:51 PM
There's nothing about the 7D that's stopping anyone from making films and or making money, sure it has limitations but we have this place to figure out workarounds. My MKII is paying itself off on it's first job, a couple more jobs and I'll have the $$$ for the 1D MKIV.

boulder
10-10-2009, 07:02 PM
Yes! Away with you and your facts dear sir! Your charts confuse and cloud my mind with doubt...I will accept these new findings with a smile and go forth and make movies anyways, for we must not linger on negative issues that may have been hidden from us...besides they told us it was 1080 so I belive Canon.

mhood
10-10-2009, 07:07 PM
If you are having second thoughts...watch the footage on an HDTV. Your jaw will drop.

Exactly how do you go about watching your 7D footage on an HDTV? I want to be able to do that as soon as mine gets here.

Sumfun
10-10-2009, 07:09 PM
From my point of view, I see Barry's somewhat depressing assessment as more a challenge than a capitulation. Rather than wait for the next big thing, I'm going full steam ahead on the 7D and learning to master the idiosyncrasies, strengths and weaknesses of the 7D.

If a superior 1D IV comes out in a few months then great. I'll flick the 7D off and go that way. At least I'll have all the support equipment and experience. No matter what the charts say, just look at so much of the great work that is out there; Bloom, Vienna, Stillmotion, etc etc. You more you know your camera, the more you can work to great work out of it.


Peter

I don't currently have a job that requires a 35mm DOF, or I might go out and get a 7D tomorrow. I was just going to buy one to "play" with, so I don't mind waiting a little longer.

But in any case, Barry's tests show a weakness in the 7D. I'm sure you can make great videos with the 7D, but it's nice to know the strengths and weaknesses of your tools.

Tom Roper
10-10-2009, 07:12 PM
I can take my 5DMkII, and by using shallow DOF, diffusing filters or other techniques control the aliasing. But am I doing this because it's the look I'm trying to achieve, or because I'm reacting to cover up a flaw in the camera system? If the latter, then fixing the artifact problem may mean I don't achieve the look I'm after.

Useful DSLR footage requires the convergence of the technique(s) for controlling the artifacts and simultaneously getting the desired look. Those ends are not always congruent.

dcoughla
10-10-2009, 07:27 PM
Exactly how do you go about watching your 7D footage on an HDTV? I want to be able to do that as soon as mine gets here.

Easy way to do it would be with a PS3 hooked up to an HDTV...

Stephen Mick
10-10-2009, 07:28 PM
Mini-HDMI to HDMI cable works fine for me.

Barry_Green
10-10-2009, 07:46 PM
Really? Only good for interview work?
I didn't say "only" interview work, I gave that as an example of where the 7D would excel.

Frankly, I don't know what situations it's going to fail in, or what it will excel in. The chart data is to start gathering ideas, the next step is to put it into action and see what happens out there. But right now, I feel pretty comfortable saying that it'll do very well in a close-up shot with an out of focus background, so long as the subject isn't wearing a fine herringbone or other repeating-pattern fabric. You will have to watch out for stuff like that -- just like we always used to have to watch out for it on BetaSP too.

mhood
10-10-2009, 07:51 PM
Mini-HDMI to HDMI cable works fine for me.

So it will playback full resolution from the card? With audio?

Stephen Mick
10-10-2009, 07:54 PM
Indeed it will.

Stephen Mick
10-10-2009, 07:58 PM
To anyone gnashing their teeth or clenching their fists over these tests, remember, they're just tests. They should give you some ideas as to the strengths and weaknesses of any camera.

And in case you've forgotten out there…

http://vimeo.com/6487566

http://vimeo.com/6860546

mhood
10-10-2009, 08:01 PM
Now comes my dumbest of questions: Can you "print" a master in h.264 (or whatever) from Premiere CS4.1 to the CF card using a reader and play it to your HDTV using the 7D? Thanks for your patience...

Stephen Mick
10-10-2009, 08:03 PM
If you had a stick of gum, a hockey ticket and a paper clip, maybe.

Honestly, I have no idea. Maybe the Premiere forum folks would have a better idea.

Barry_Green
10-10-2009, 08:08 PM
No gnashing of teeth necessary -- what you're getting for $1700 is still phenomenal. There are limits, just like there are in all pieces of equipment. We're finding out what those limits are in a controlled environment, so that we are then prepared to work around or overcome the limits in a practical shooting situation. I have no doubt that shooting on a 7D or GH1 is going to be more work than shooting on something "better" would be, but that's just how it goes when you get something like this for the bargain basement price of $2,000.

Next week I'm going to get to dynamic range testing, and greenscreening, and compression quality testing, and noise/sensitivity stuff. We will then know what this cam can do, and what it can't -- and everyone will be better off for knowing what the real capabilities are.

Barry_Green
10-10-2009, 08:10 PM
I haven't tried exporting an h.264 mov out to the camera... it might work. There's not a big ol' folder structure, it's just stored right in the regular still-photo directory, so ... maybe. You might have to pick an export profile that doesn't support B-frames though, because it's possible that the 7D couldn't play back a B-frame file...

... yet another thing to test... :)

Lucian
10-10-2009, 08:35 PM
And in many, many, many ways, the EX3 will take down the 7D. Professional audio, screen viewing angles, variable frame rates, sharpness, customizability, control over the image, and -- well, it's an EX3, so you have many of the same lens options; the Adaptimax adapters let you use Nikon glass and probably Canon glass, so you probably have more lens options with the EX3. .

A basic EX3 box is over $8,000. EIGHT THOUSAND DOLLARS!
And no, the lens comparison doesn't work, 1/2" video lenses are prohibitively expensive and using 35mm lenses doesn't work well because of image cropping. Power zoom? I never use that. Autofocus on the EX3 is junk. 3rd party media cards for the EX3 are unreliable from my experience. Perhaps they work better with the EX1, but on the EX3 I've been burnt. For another few hundred you can get a 24 bit audio source for the 7d. Sure, EX3 is stronger overall, it's fantastic, but it's a lot more money and my point that the 7D whips it in select and critical areas remains (shallow focus, media, low light, COST).

Lucian
10-10-2009, 08:35 PM
Next week I'm going to get to dynamic range testing, and greenscreening, and compression quality testing, and noise/sensitivity stuff. We will then know what this cam can do, and what it can't -- and everyone will be better off for knowing what the real capabilities are.

Awesome! Can't wait.

Ben_B
10-10-2009, 09:34 PM
Seems like with the potential for overheating and 12 minute record limit interview work would not be something the 7D would be good at? Not really seeing how it would be good for this...am I missing something?

Seems like a great narrative filmmaking camera...I'm trying to get the TV station at my university to replace the current crop of Canon DSLRs they have with them (xsi to 40D.)

Isaac_Brody
10-10-2009, 10:17 PM
Seems like with the potential for overheating and 12 minute record limit interview work would not be something the 7D would be good at? Not really seeing how it would be good for this...am I missing something?

I've been shooting for a week with no overheating warnings. I think it's an overstated issue. A lot of people who get the warning continue to shoot with no problems, or they it stop for a minute and then start up again.

As for 12 minute limits, not an issue for interviews. Unless you're used to asking a question and letting a person ramble on for 12 minutes. There's always a moment to pause and interrupt and down time. Not too long ago people were shooting HVX with 4 gig P2 cards and that wasn't a problem.

Ben_B
10-10-2009, 11:08 PM
Hmmm...I suppose so.

Interviews I do usually end up being a bit more of a fluid exchange (wow pardon that unfortunate double entendre) wherein the person is talking quite a bit and I'm just prodding them along, but I can see how this would not always be the case.

That said I grew up around college professors so you can see why I would think 12 minutes wouldn't be enough time :D

sblfilms
10-10-2009, 11:10 PM
This thread is another depressing reminder of how finicky the indie filmmaker community is. The 7D isn't perfect and the same people who said "If only the GH1 didn't have THIS issue" that Canon included in the 7D, they would be content. So now it is on to "If only the 7D didn't have THIS issue" and Nikon will probably drop a fantastic D90 update that has its own niggling issues that will be the excuse of many on this board and other why they aren't shooting and making their great masterpiece.

I shot a fun little short with my filmmaking society this afternoon on two 7Ds and when we screened some of the footage for the admittedly non-pixel-peeping crew...they were blown away and reinvigorated for working together in a way I haven't seen since my best friend showed us the first footage from his DVX100. That makes all the issues of the 7D melt away for me because I don't make movies for pixel peepers.

ChipG
10-10-2009, 11:45 PM
My eyes & head hurts, I'm going to curl up and go to bed for a few days.

I wanna dream about girls again and not cameras

powervideo
10-10-2009, 11:56 PM
I don't make movies for pixel peepers.

:grin:

Kholi
10-11-2009, 01:04 AM
At the end of the day, DVXuser is more a TECHNICAL board than anything else. Because of Barry_Green and others like him we got a 7D with 1080/24 in the first place. You think they would have done it without the backlash that they received from the 5D only being 30P?

You think we would have gotten the HPX170 updates without Luis Caffesse' upgrade wishlist?

Letus Ultimate? Brought forth by pixel peepers.

We ALL know the mantra: it's about what's in front of the BLAH BLAH BLAH.

If you're upset about the technical side then this isn't the place to look. There's a footage section all the same.

What Barry showed here is completely relevant and lots of us are glad to see it since we couldn't or wouldn't have done it ourselves.

dadoboy
10-11-2009, 01:20 AM
Word up! Squeaky wheels... you always have to push Canon/RED/Panasonic, etc. to do better. It's not pixel peeping, it's plain out bitching!

"Don't underestimate the technical aspects of filmmaking." - William Friedkin (I'm sure Stanley Kubrick basically said the same thing).


At the end of the day, DVXuser is more a TECHNICAL board than anything else. Because of Barry_Green and others like him we got a 7D with 1080/24 in the first place. You think they would have done it without the backlash that they received from the 5D only being 30P?
.

filmmaker's gang
10-11-2009, 01:33 AM
Funny thing is I've never heard of anyone complaining about the MKII not being sharp enough, quite the contrary.post of the thread

can me say it or my post will be deleted asap?

filmmaker's gang
10-11-2009, 01:40 AM
Interesting stuff Barry. Well done. But what exactly is the purpose of this test? I mean, are there any news?

Moire? It were made some very good sample clips like these ones:
http://www.vimeo.com/6549968
http://www.vimeo.com/6550033
So yes, 720p has a lot more moire + aliasing. We already knew that.

Sharpness? This has a lot to do with the lens. The kit lens 18-135 is quite soft + a bit unsharp at the wide-angle, especially at the widest aperture. - So, did i overlook something important? Please let me know... (i galopped pretty fast through the thread)...

looooool don't push too much man ehehehehe :beer:



Funny thing is I've never heard of anyone complaining about the MKII not being sharp enough, quite the contrary.post of the thread

can me say it or my post will be deleted asap?

PappasArts
10-11-2009, 01:45 AM
I'm pleased at the way the GH1 is coming through all of this, .

Me too!

Like Barry's 720 GH1 vs 7D example shows. The GH1's 1080 is even better than it's 720. I had the option to get the Canon 5D, tested it, didn't like it for many reasons. I'm completely happy with the GH1; and the results are stunning. When "Jack Daniel Stanley" { http://www.dvxuser.com/V6/showthread.php?t=175852 } did his GH1 vs HPX170; and the GH1 was better in 720 then the 1080 of the HPX170, let alone how good the 1080 tests he did on the GH1 was over the HPX170, I was sold. Not a single regret. The GH1 is just a better well rounded design over all IMHO.



Michael Pappas
http://www.pbase.com/Arrfilms
Arrfilms@hotmail.com
http://www.PappasArts.com

powervideo
10-11-2009, 01:52 AM
My view is we can get too hung up on the numbers rather than leaving it to the subjective viewpoint of the end-user. I've just has a video exhibition, "WALK" (http://www.netsvictoria.org/walk/), finish touring throughout Australia for three years which was shot on a Sony HC-1 HDV camera and edited with PPro/Magic Bullet Swing-shift filter because I couldn't afford a decent DOF adaptor at the time. It played off an early Toshiba HD-DVD player and despite the up-rezzing and crappy HDV specs, it still looked great and all I got was accolades about the picture quality.

I'd never disparage what Barry does. He's been a shining light for pushing small format HD video, but we all need to take a deep breath sometimes and look at it from our end-users's viewpoint and expectations.

squig
10-11-2009, 02:15 AM
I hope this doesn't turn into another "mud" thread. There's nothing wrong with it only resolving the equivalent of 720p, the D90 is 720p with a shi*e codec and that hasn't stopped people shooting a few features and a shi*load of music vids.

And thx for the heads up Barry it's important to know all the limitations this not being a big one.

Rakesh Jacob
10-11-2009, 02:18 AM
My 2cents:
7D, AWESOME! GH1, AWESOME! 5DmkII, AWESOME! D90... it was first... :)
Seriously, whatever you buy right now is going to do an amazing job. "Pixel Peeping" is an important part of the game, we wouldn't be at this level if it wasn't for it. And thanx to the pixel peepers, we are gonna get better and better gear. But seriously, don't be afraid to pick up any of these cameras, 3 or4 years ago people would have easily paid over $100k for the image quality and aesthetic of these DSLRs. It's a great time to be a filmmaker. Take it in stride, do your peeping, make your decision and be happy. I chose the 7D, the GH1 is an oh so close 2nd but what bothers me is different than what bothers the guys that are going GH1. And honestly when I get some extra cash I'm probably gonna pick up a couple of GH1s too, especially if they're on sale or refurbished by then.

plasmasmp
10-11-2009, 02:26 AM
I doubt this will turn into a "mud" thread. I think when it comes down to looking at an image produced by these cams, the dynamic range is more impressive to the viewer than just resolution. I really wanted to love the GH1, and certainly almost bought one, but the range and the way it processes colors just doesnt compare to the 5D/7D cameras. This may be completely fixed in the GH2, and I expect Panasonic to fix the issues before we see another update from Canon or Nikon. In that case, I'll happily buy a GH2 body and more lens adapters. :happy:

Jean Dantes
10-11-2009, 02:35 AM
My 2cents:
7D, AWESOME! GH1, AWESOME! 5DmkII, AWESOME! D90... it was first... :)
Seriously, whatever you buy right now is going to do an amazing job. "Pixel Peeping" is an important part of the game, we wouldn't be at this level if it wasn't for it. And thanx to the pixel peepers, we are gonna get better and better gear. But seriously, don't be afraid to pick up any of these cameras, 3 or4 years ago people would have easily paid over $100k for the image quality and aesthetic of these DSLRs. It's a great time to be a filmmaker. Take it in stride, do your peeping, make your decision and be happy. I chose the 7D, the GH1 is an oh so close 2nd but what bothers me is different than what bothers the guys that are going GH1. And honestly when I get some extra cash I'm probably gonna pick up a couple of GH1s too, especially if they're on sale or refurbished by then.

Amen!

filmmaker's gang
10-11-2009, 02:53 AM
seems there are other fine "post of the thread" ex.s after all..


Wow, this test shows how much people are 'snake charmed' by the shallow depth of field look. I'm blown away by the resolution performance of the little hmc40 but I'd still take the 7D over it anyday.


Just shortly (during soccer half-time break russia-germany). I rummaged the whole internet and didnīt find any reviews - only some user reports (and pics) in german + international groups. If you have a link, please post it.

Yes, the 18-135 is pretty sharp but not at the wide angle. Here at this site you can compare different canon lenses. I took as a candidate on the other side the 17-55/2.8: http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=678&Camera=474&Sample=0&FLI=2&API=0&LensComp=398&CameraComp=474&SampleComp=0&FLIComp=3&APIComp=1. Youīll notice how soft + unsharp it is at 35mm - just go with the cursor over the chart for comparison...


NO mention of 1080

is this thread just knocking the 720version

my impression is that the 720 looks very toy like, internet only use

1080 is better ??

S


Really? Only good for interview work? How about filming any narrative cinema pieces with people talking etc and controlling for problem subject matter with a shallow depth of field when encountered. These chart samples are great to push these cams to the edge and find their flaws, but lets not forget that DPs are using these cams in the industry and intercutting their footage with much more expensive cameras. Sorry, but just saying they're good for interview work is selling the 7D and 5DII (if your including the 5DII in that statement) a little short. These cams are being used already in the industry for far more than interview work.

The 7D would have kicked the DVX100's butt filming "November" The DVX was clearly struggling in some of those low lit scenes. Clean low light capability is meat and potatoes. Most of those other features you mentioned are dessert.


Hey folks...go out and shoot something spectacular with the combination of all facets of film making; don't let a res chart test stop you.


A basic EX3 box is over $8,000. EIGHT THOUSAND DOLLARS!
And no, the lens comparison doesn't work, 1/2" video lenses are prohibitively expensive and using 35mm lenses doesn't work well because of image cropping. Power zoom? I never use that. Autofocus on the EX3 is junk. 3rd party media cards for the EX3 are unreliable from my experience. Perhaps they work better with the EX1, but on the EX3 I've been burnt. For another few hundred you can get a 24 bit audio source for the 7d. Sure, EX3 is stronger overall, it's fantastic, but it's a lot more money and my point that the 7D whips it in select and critical areas remains (shallow focus, media, low light, COST).


This thread is another depressing reminder of how finicky the indie filmmaker community is. The 7D isn't perfect and the same people who said "If only the GH1 didn't have THIS issue" that Canon included in the 7D, they would be content. So now it is on to "If only the 7D didn't have THIS issue" and Nikon will probably drop a fantastic D90 update that has its own niggling issues that will be the excuse of many on this board and other why they aren't shooting and making their great masterpiece.

I shot a fun little short with my filmmaking society this afternoon on two 7Ds and when we screened some of the footage for the admittedly non-pixel-peeping crew...they were blown away and reinvigorated for working together in a way I haven't seen since my best friend showed us the first footage from his DVX100. That makes all the issues of the 7D melt away for me because I don't make movies for pixel peepers.

squig
10-11-2009, 03:33 AM
Just to throw a spanner in the works it looks like Nikon are announcing the D3s this week with 1080p 24p. I hope it has a crappy codec and low resolution and mud and overheats because I can't afford one.

filmmaker's gang
10-11-2009, 03:41 AM
ehehehehe me likes you anyway

Sttratos
10-11-2009, 04:26 AM
What puzzles me is how can a camera which performs so badly in a resolution chart test put out sharp, high quality images like these:

http://www.vimeo.com/6938509

http://www.vimeo.com/6860546

The answer is I don't care how. All I care is what I see in the final result. I have an EX1 for over a year now and this stuff looks as sharp and as high quality as anything out of an EX1. If the 7D is not putting out full 1080 and the EX1 is and the 7D looks as good and looks that sharp, Canon must have some voodoo going on under the hood and it is just more credit to the 7D. And the clips above were even shot in 720p, which is supposed to be closer to SD according to the test. Now try that with a DVX.

Jean Dantes
10-11-2009, 04:37 AM
What puzzles me is how can a camera which performs so badly in a resolution chart test put out sharp, high quality images like these:

http://www.vimeo.com/6938509

http://www.vimeo.com/6860546

The answer is I don't care how. All I care is what I see in the final result. I have an EX1 for over a year now and this stuff looks as sharp and as high quality as anything out of an EX1. If the 7D is not putting out full 1080 and the EX1 is and the 7D looks as good and looks that sharp, Canon must have some voodoo going on under the hood and it is just more credit to the 7D. And the clips above were even shot in 720p, which is supposed to be closer to SD according to the test. Now try that with a DVX.

That's pretty refreshing coming from someone that has owned an EX1 for over a year! :)

commanderspike
10-11-2009, 04:43 AM
It doesn't perform badly in terms of resolution. It's sharper than the HMC40. But because of that, and because it's hard to downscale a 18MP sensor to 1080p when you have an image processor designed for processing stills, not video, you get the moire and aliasing artefacts on fine patterns like those found in the res charts.

Only occasionally (and on some idiotic shirt designs worn by 'real film makers'!) does it occur in real life.

A guy at the BBC did the same test on the 5D and it too failed badly while the HF100, which we all know is a worse camera in almost every other regard, passed fine. This is understandable from the BBC's point of view because a lot of their set designs have wild patterns in them and they do a lot of interviews with politicians who wear awful shirts, but for us - whilst it does crop up occasionally and damage the image, it's nowhere near the level of concern for most people that Barry has whipped up here.

What this test DOES show is what we already know, that 7D uses same downscaling method as 5D and this won't change until 1DS Mark IV with Digic V comes out.

I believe the GH1 does better at the resolution chart, but that uses softer compression and a slightly more advanced method of downscaling (from a lower res sensor in turn), so it's bound to be a little better in this regard. But boy does it have some IQ problems of it's own!!

Michael Olsen
10-11-2009, 05:04 AM
Only occasionally (and on some idiotic shirt designs worn by 'real film makers'!) does it occur in real life.


I say this not to bash things, or to bring people down, but to inject reality into this sentiment. High details are not present only occasionally, nor especially in "idiotic shirts".

Here's what Barry has to say:

All the time, actually. Electricity wires, buildings with windows at a distance, certain fabrics like herringbone suits or knit shirts or tweed or denim, striped shirts on people at a slight distance, picket fences, shingled roofs, tiled roofs, everywhere.

And I agree. In fact, I'll go further. Even fabrics which are a solid color but have a distinct weave could prove a problem. Pique, a favorite fabric for Polo shirts, concerns me. "Waffle knit" cardigans and shirts concern me. Any kind of lace or tulle concerns me. Voile, a more expensive shirting fabric for summer dress, concerns me. Silk ties, especially those with physical ribs or knit patterns, concern me. The list goes on.

This is not to say you could stand in front of the camera wearing any or all of these and be just fine. It's conceivable. It's not to say that other cameras don't have similar problems - I believe people have occasional moire problems with the RED. And I see it on TV, DVD, and BR all the time.

What I'm saying is that people can't get away with the lasseiz-faire "It won't happen to me!" mentality and repeatedly receive the best results with this camera. Especially for interviews. When people dress up, the quality of their clothing increasing, usually using finer cloth and fabrics with more detail. I plan on asking any future interviewees to bring with them whatever they want to wear as well as a very plain ensemble - something free of patterns or weaves. Because what happens if that Congressman finally has time to give me the 10 minutes I need and he sits down in his new J. Press suit, Brooks Brothers shirt and his Drakes tie and every fabric on him shimmers with moire on my LCD. I'm screwed.

Estote Parati - Be Prepared

hermmermferm
10-11-2009, 05:27 AM
From IMDB, deep-focus films:

Black and White



Twilight of a Woman's Soul (http://en.wikipedia.org/w/index.php?title=Twilight_of_a_Woman%27s_Soul&action=edit&redlink=1) (1913)[8] (http://en.wikipedia.org/wiki/Deep_focus#cite_note-7)
Foolish Wives (http://en.wikipedia.org/wiki/Foolish_Wives) (1922)
Nosferatu (http://en.wikipedia.org/wiki/Nosferatu) (1922)
Greed (http://en.wikipedia.org/wiki/Greed_%28film%29) (1924)
Downhill (1927)
Working Girls (http://en.wikipedia.org/w/index.php?title=Working_Girls_%281931_film%29&action=edit&redlink=1) (1931)
Waltzes from Vienna (1934)
Mad Love (http://en.wikipedia.org/wiki/Mad_Love_%281935_film%29) (1935)
Dead End (http://en.wikipedia.org/wiki/Dead_End) (1937)
Grand Illusion (http://en.wikipedia.org/wiki/Grand_Illusion_%28film%29) (1937)
The Rules of the Game (http://en.wikipedia.org/wiki/The_Rules_of_the_Game) (1939)
The Long Voyage Home (http://en.wikipedia.org/wiki/The_Long_Voyage_Home) (1940)
Rebecca (http://en.wikipedia.org/wiki/Rebecca_%28film%29) (1940)
Citizen Kane (http://en.wikipedia.org/wiki/Citizen_Kane) (1941)
The Little Foxes (http://en.wikipedia.org/wiki/The_Little_Foxes_%28film%29) (1941)
The Magnificent Ambersons (http://en.wikipedia.org/wiki/The_Magnificent_Ambersons_%28film%29) (1942)
The Best Years of Our Lives (http://en.wikipedia.org/wiki/The_Best_Years_of_Our_Lives) (1946)
Ugetsu (http://en.wikipedia.org/wiki/Ugetsu) (1953)
Sansho the Bailiff (http://en.wikipedia.org/wiki/Sansho_the_Bailiff) (1954)
Seven Samurai (http://en.wikipedia.org/wiki/Seven_Samurai) (1954)
Mr. Arkadin (http://en.wikipedia.org/wiki/Mr._Arkadin) (1955)
Ashes and Diamonds (http://en.wikipedia.org/wiki/Ashes_and_Diamonds_%28film%29) (1958)
L'avventura (http://en.wikipedia.org/wiki/L%27avventura) (1960)
The Trial (http://en.wikipedia.org/wiki/The_Trial_%281962_film%29) (1962)
The Manchurian Candidate (http://en.wikipedia.org/wiki/The_Manchurian_Candidate_%281962_film%29) (1962)
Birdman of Alcatraz (http://en.wikipedia.org/wiki/Birdman_of_Alcatraz_%28film%29) (1962)
Seven Days in May (http://en.wikipedia.org/wiki/Seven_Days_in_May) (1964)
The Train (http://en.wikipedia.org/wiki/The_Train) (1964)
Chimes at Midnight (http://en.wikipedia.org/wiki/Chimes_at_Midnight) (1966)
Faces (http://en.wikipedia.org/wiki/Faces_%28film%29) (1968)
The Last Picture Show (http://en.wikipedia.org/wiki/The_Last_Picture_Show) (1971)
Paper Moon (http://en.wikipedia.org/wiki/Paper_Moon_%28film%29) (1974)
The Good German (http://en.wikipedia.org/wiki/The_Good_German) (2006)

[edit (http://en.wikipedia.org/w/index.php?title=Deep_focus&action=edit&section=7)] Color



The Bridge on the River Kwai (http://en.wikipedia.org/wiki/The_Bridge_on_the_River_Kwai) (1957)
For a Few Dollars More (http://en.wikipedia.org/wiki/For_a_Few_Dollars_More) (1965)
The Good, the Bad and the Ugly (http://en.wikipedia.org/wiki/The_Good,_the_Bad_and_the_Ugly) (1966)
The Offence (http://en.wikipedia.org/wiki/The_Offence) (1972)
Rumble Fish (http://en.wikipedia.org/wiki/Rumble_Fish) (1983)
52 Pick-Up (http://en.wikipedia.org/wiki/52_Pick-Up) (1986)
Heat (http://en.wikipedia.org/wiki/Heat_%281995_film%29) (1995)
Eyes Wide Shut (http://en.wikipedia.org/wiki/Eyes_Wide_Shut) (1999)
Peter Pan (http://en.wikipedia.org/wiki/Peter_Pan_%282003_film%29) (2003)
Harry Potter and the Prisoner of Azkaban (film) (http://en.wikipedia.org/wiki/Harry_Potter_and_the_Prisoner_of_Azkaban_%28film%2 9) (2004)
The New World (http://en.wikipedia.org/wiki/The_New_World_%28film%29) (2005)
Scrubs (TV series) (http://en.wikipedia.org/wiki/Scrubs_%28TV_series%29) (2000-2008)
Six Feet Under (http://en.wikipedia.org/wiki/Six_Feet_Under_%28TV_series%29) (2001-2005)
The Black Dahlia (http://en.wikipedia.org/wiki/The_Black_Dahlia_%28film%29) (2006)



It's pretty niche.. IMHO, deep focus and the 7D is asking for a sharp video look.

Uwe Lansing
10-11-2009, 05:56 AM
Haha... this thread is fun at its best :-). - Believe me, rather than all the "chart-babble" it matters more what you shoot and how it looks like. And the sharpness looks way good enough for every purpose. And if surprisingly not, it can be fixed with a bit of sharpening in post - absolutely no problem at all. But i promise you, there will be some postings in the near future like "How can i soften my 7d-footage and make it less sharp". There are a lot of similar posts out there concerning the 5DII. So, relax and enjoy your new 7d...

Rakesh Jacob
10-11-2009, 07:30 AM
What puzzles me is how can a camera which performs so badly in a resolution chart test put out sharp, high quality images like these:

http://www.vimeo.com/6938509

http://www.vimeo.com/6860546

The answer is I don't care how. All I care is what I see in the final result. I have an EX1 for over a year now and this stuff looks as sharp and as high quality as anything out of an EX1. If the 7D is not putting out full 1080 and the EX1 is and the 7D looks as good and looks that sharp, Canon must have some voodoo going on under the hood and it is just more credit to the 7D. And the clips above were even shot in 720p, which is supposed to be closer to SD according to the test. Now try that with a DVX.

YUCK!!! I can't believe I used to think that looked great. :)

Zim
10-11-2009, 08:09 AM
Hey you left off one!!


My 2cents:
7D, AWESOME! GH1, AWESOME! 5DmkII, AWESOME! D90... it was first... :)
Seriously, whatever you buy right now is going to do an amazing job. "Pixel Peeping" is an important part of the game, we wouldn't be at this level if it wasn't for it. And thanx to the pixel peepers, we are gonna get better and better gear. But seriously, don't be afraid to pick up any of these cameras, 3 or4 years ago people would have easily paid over $100k for the image quality and aesthetic of these DSLRs. It's a great time to be a filmmaker. Take it in stride, do your peeping, make your decision and be happy. I chose the 7D, the GH1 is an oh so close 2nd but what bothers me is different than what bothers the guys that are going GH1. And honestly when I get some extra cash I'm probably gonna pick up a couple of GH1s too, especially if they're on sale or refurbished by then.

Zim
10-11-2009, 08:18 AM
What would happen if you tried to put it on film and put it on the big screen? Can you make a film like this with a 7D or any HDSLR like a DVX100 did.

http://www.apple.com/trailers/independent/iraqinfragments/


What puzzles me is how can a camera which performs so badly in a resolution chart test put out sharp, high quality images like these:

http://www.vimeo.com/6938509

http://www.vimeo.com/6860546

The answer is I don't care how. All I care is what I see in the final result. I have an EX1 for over a year now and this stuff looks as sharp and as high quality as anything out of an EX1. If the 7D is not putting out full 1080 and the EX1 is and the 7D looks as good and looks that sharp, Canon must have some voodoo going on under the hood and it is just more credit to the 7D. And the clips above were even shot in 720p, which is supposed to be closer to SD according to the test. Now try that with a DVX.

NoxNoctus
10-11-2009, 08:24 AM
There's someone around here that took their 5D footage and had it transferred to Film. Said it looked incredible

Lucian
10-11-2009, 08:25 AM
I say this not to bash things, or to bring people down, but to inject reality into this sentiment. High details are not present only occasionally, nor especially in "idiotic shirts".

Here's what Barry has to say:


And I agree. In fact, I'll go further. Even fabrics which are a solid color but have a distinct weave could prove a problem. Pique, a favorite fabric for Polo shirts, concerns me. "Waffle knit" cardigans and shirts concern me. Any kind of lace or tulle concerns me. Voile, a more expensive shirting fabric for summer dress, concerns me. Silk ties, especially those with physical ribs or knit patterns, concern me. The list goes on.

This is not to say you could stand in front of the camera wearing any or all of these and be just fine. It's conceivable. It's not to say that other cameras don't have similar problems - I believe people have occasional moire problems with the RED. And I see it on TV, DVD, and BR all the time.



Estote Parati - Be Prepared

I used to own an $40,000 betacam. It had serious issues with any striped clothes, any fine detail on clothes, orange shirts, black shirts, ties with fine patterns and it had a host of other issues too. We worked around its limitations and never looked back. Shooting city scapes does seem to be a problem with 7d though.

Rakesh Jacob
10-11-2009, 08:26 AM
Hey you left off one!!

I left off a bunch, which one do you mean?

Tom Roper
10-11-2009, 08:30 AM
These are MTF50 resolution numbers from the Imatest software:

Canon 5DMkII:

1080/30p Horizontal - 623 LW/PH
1080/30p Vertical - 756 LW/PH

****************************

Canon XH-A1:

1080/60i Horizontal - 824 LW/PH
1080/60i Vertical - 675 LW/PH

1080/30F Horizontal - 824 LW/PH
1080/30F Vertical - 590 LW/PH

****************************

Canon HV-20

1080/60i Horizontal - 785 LW/PH
1080/60i Vertical - 705 LS/PH

****************************

Sony EX-1

1080/30p - Horizontal - 931 LW/PH
1080/30p - Vertical - 1109 LW/PH

Rakesh Jacob
10-11-2009, 08:35 AM
The EX-1 is a resolution having sum-a-ma-beech! Had the DSLR thing not happned, that's probably what I would have been rocking by now. But resoultion is only one parameter of the aesthetic I want in my image and appearently not as important as I used to think it was.
And almost as important to me is the guerilla aspect of the DSLR. Something about having everything I need in a small backpack that is very compelling to my sense of independent, minimalist filmmaking style.

Ian-T
10-11-2009, 08:42 AM
The EX-1 is a resolution having sum-a-ma-beech! Had the DSLR thing not happned, that's probably what I would have been rocking by now. But resoultion is only one parameter of the aesthetic I want in my image and appearently not as important as I used to think it was.
You know. I agre with you JC. I remember Mattias and I having this discussion over a year ago in another forum when he was contemplating getting the D90. I didn't think about color, dynamic range etc but was focusing more on resolution. I had the HV20, whos resolution was higher than this cam...but there was always "something" missing from the image that I could not put my fnger on. Even with the A1 and a majority of the EX-1 videos I've watched (though they are higher in resolution).

Rory_B
10-11-2009, 08:43 AM
Been playing around with some deep focus stuff i shot. I posted an ungraded video and the aliasing isn't horrendous in real world situations. This morning I've dabbled with some colour correction and throwing a diffusion filter of some sort on the footage and then bringing the contrast and sharpness back through a 2nd pass seems to help mitigate the effects a bit. I didn't find the aliasing horribly objectionable in the first place, but it's there. I've shown it to plenty of non pixel peepers and they don't even notice it in spots I would say it's blatant, so there's solace in that.

Jason Ramsey
10-11-2009, 08:46 AM
I don't understand why so many of you have to over-react over a simple test.... Either panic and freak out, or get all defensive.... what is the point? It's a test... It's valuable information. That is it's purpose. Used to inform yourself... Some of you behave as though the test shouldn't be posted unless it is good news..

it's just information, people... why are some folks getting so worked up over it like there is all kinds of drama about it? there isn't... shouldn't be anyways.

later,
Jason

squig
10-11-2009, 08:51 AM
Some of you behave as though the test shouldn't be posted unless it is good news..



I thought it was you guys that only wanted good news :Drogar-BigGrin(DBG)

Yeah Johnny I can personally carry everything I need to shoot and cut a film including lighting and I'm pretty weak.

Zim
10-11-2009, 08:52 AM
That is one of the things I was thinking about to with the HDSLRs is carrying a smaller load. Two cameras in a bag that can do both video and stills would be great. I would like to see a flip out LCD screen. I would also like to not use a shoulder rig but a mono pod or maybe a Flow-Pod would work. Which means I'm going to see what happens after Oct 14th.


The EX-1 is a resolution having sum-a-ma-beech! Had the DSLR thing not happned, that's probably what I would have been rocking by now. But resoultion is only one parameter of the aesthetic I want in my image and appearently not as important as I used to think it was.
And almost as important to me is the guerilla aspect of the DSLR. Something about having everything I need in a small backpack that is very compelling to my sense of independent, minimalist filmmaking style.

Jason Ramsey
10-11-2009, 08:52 AM
I thought it was you guys that only wanted good news

we want productive discussion.... not incessant negativity. :)

Rory_B
10-11-2009, 09:01 AM
That is one of the things I was thinking about to with the HDSLRs is carrying a smaller load. Two cameras in a bag that can do both video and stills would be great. I would like to see a flip out LCD screen. I would also like to not use a shoulder rig but a mono pod or maybe a Flow-Pod would work. Which means I'm going to see what happens after Oct 14th.

A Monopod is actually pretty sturdy for quite a few situations, provided you still don't mind having the handheld feel and look, because there's always going to be a bit of float.

Jason Ramsey
10-11-2009, 09:15 AM
Like Barry said... for 2 grand, it's still an unbelievable deal. Just good to figure out all the quirks and where the shortcomings will be so you aren't surprised on a shoot. Learn it now, arm yourself. That way you are in a much better position to get the best out of the camera and not get caught by something when you least expect it.

It's valuable information. Not something meant to work people up or freak them out or whatever.

J Davis
10-11-2009, 09:19 AM
When it comes to having everything you need in one backpack and guerilla style shooting the GH1 leaves the 7D dead in the water.

I say this as a GH1 owner who shot constantly with it for nearly 3 months and now I've been shooting with the 7D for almost 2 weeks.

The GH1 is a much lighter and more compact camera body. From memory I would say the 7D body weight is about double the body weight of the GH1. The other huge factor to consider with guerilla shooting is monitoring. Having an articulated LCD is a winner. Even if you are not doing low shots and the cam is shoulder height most of the time there is reflected sunlight to deal with. An articulated LCD can be angled even by 5 to 10 degrees and bingo you just increased your chances of a better run & gun shot tenfold.

My work is narrative with actors, tripods and controlled lighting conditions where laptop monitoring with the EOS live utility make the 7D worth it. My only run around town shooting is for family and friends occasions and these happen so regularly it is almost a distraction so lack of pulldown there is a plus.

But, anyone looking at doc work, guerilla, run & gun should go with a GH1, especially if their delivery is for the web. If its video not stills, I would advise this even if you were knee deep in Canon EF glass.

mhood
10-11-2009, 09:29 AM
But, anyone looking at doc work, guerilla, run & gun should go with a GH1, especially if their delivery is for the web. If its video not stills, I would advise this even if you were knee deep in Canon EF glass.

I guess it's safe to assume adapters are available (and affordable?) for the GH1 that allows the use of Canon EF USM lenses with AF and exposure?

J Davis
10-11-2009, 09:41 AM
Please note I said video not stills. If its video the 7D won't do video AF, the GH1 will do it with the kit lens but this is something I never explored.
My GH1 kit lens went on the shelf after the first day as its too slow and aperture changes with zoom.

Rakesh Jacob
10-11-2009, 09:41 AM
When it comes to having everything you need in one backpack and guerilla style shooting the GH1 leaves the 7D dead in the water.

I say this as a GH1 owner who shot constantly with it for nearly 3 months and now I've been shooting with the 7D for almost 2 weeks.

The GH1 is a much lighter and more compact camera body. From memory I would say the 7D body weight is about double the body weight of the GH1. The other huge factor to consider with guerilla shooting is monitoring. Having an articulated LCD is a winner. Even if you are not doing low shots and the cam is shoulder height most of the time there is reflected sunlight to deal with. An articulated LCD can be angled even by 5 to 10 degrees and bingo you just increased your chances of a better run & gun shot tenfold.
While I completely agree with you about the articulated LCD and low weight advantages of the GH1, saying that the 7D is "dead in the water" LOL That's VERY HYPERBOLIC.

But like I said I completely agree, in principle the GH1 is way more practical as a video camera than the 7D, my choice came down to 3 things: a) overall image quality, b) native 1080-24p, c) future cameras in the line up (DigicV based) make me feel confident about investing in the lenses.

Now the weight thing, the 7D is heavy for a noobie (like me) especially with a good (which=heavy) zoom lens. It will kill your right hand for, I would guess, about a month while you build up those muscles and get used to it. But I've been through WAAAAAY worse lugging an HVX with Redrock M2 on rails hand held all day, NO SHOULDER RIG, just grabbing that piece of crap any which way!

Stephen Mick
10-11-2009, 09:44 AM
Guys, let's not turn this into a 7D vs. GH1 thread. Barry's original post was about res chart testing, sharpness and aliasing.

If there are opinions on size, weight and usability, please start a new thread and discuss it there.

(I'm not a mod, but I play one on TV.)

Rakesh Jacob
10-11-2009, 09:46 AM
Guys, let's not turn this into a 7D vs. GH1 thread. Barry's original post was about res chart testing, sharpness and aliasing.

If there are opinions on size, weight and usability, please start a new thread and discuss it there.

(I'm not a mod, but I play one on TV.)

I iz sorry pretend TV mod :embarasse

J Davis
10-11-2009, 09:49 AM
I wrote my post in case anyone reading the previous page was inspired to buy a 7d for video DSLR doco, run&gun, all in one back pack work. My apologies for drifting off topic.

mhood
10-11-2009, 09:57 AM
...but I still don't know if exposure can be adjusted on my EF 50mm 1.4 USM while using a GF1 2 Canon adapter... please forgive me pretend mod person...the EF 50 is a *very* sharp lens you know? :-)

Stephen Mick
10-11-2009, 10:01 AM
AFAIK, absolutely no adapter on the GH1 will allow you to control anything on a Canon EF lens. It's fully manual, all the time, and unless it has an aperture ring (which that lens does not), you'll have to set it to the aperture you want on a Canon camera before putting it on the GH1.

I think.

mhood
10-11-2009, 10:02 AM
Then my particular bed continues to be "made". Thanks.

bwwd
10-11-2009, 10:57 AM
This thread should be locked and sticky as info only without comments.
Someone put some titles of deep focus films ,most of them from 40's-60's B-W,and there was one Harry Potter film which is suppose to be with deep focus too but its not,ive seen the trailer like a minute ago.So better check first after research cause its just not truth.
Id say lock the thread ,no comment will add more res to this cam and this thread can end up quite bad.

new2dvx
10-11-2009, 11:00 AM
Thanks Barry for doing these tests, my first time posting on the forums.

Technically speaking, I have some noob questions:

When doing green screen work what matters more, the bitrate or the resolution? I know Barry does not have his gh1 and but doesn't expect it to do any better than the 7d. But charts and numbers wise, I understand the bitrate of the 7d to be much higher than the gh1. Will this help in grading the 7d footage considerably better?

And ultimately, in layman terms, what have these tests proven? I see that many types of visual juxtaposing would cause aliasing problems with the 7d. Are these the issues or is there something more we can conclude from these tests?

Thanks!

sunburst
10-11-2009, 11:05 AM
this thread is depressing, and absurd at the same time.

It should be trimmed to technical info only,

and put in a SUB category away from the main posts.

Isaac_Brody
10-11-2009, 11:55 AM
This thread is just a reminder that these cameras aren't perfect, and it's best to know what the limitations are and make an informed choice for your project. I've been shooting for a week and the footage looks gorgeous. I'll be using this for gigs, I know what the limitations are, and it's not going to stop me from shooting. If you want to sell your cameras and get back on the fence for something better I'm sure someone else will buy your 7D.

A little advice though, if you spend too much time on here you will get the sense that these cameras are totally hopeless for any kind of shooting. Charge your batteries, go out and do some shooting. I am not selling mine because of res charts that show its weaknesses. This looks better than HVX to me, and remember that of all the people you'll show footage to, DVXuser is the smallest minority and most vocal critic when it comes to image quality. HVX footage was skewered because of noise, but when you showed it to the average viewer people didn't notice it. GH1 mud is noticed here but again the average viewer doesn't care.

If you want to make films for people with their heads pressed against their LCD's looking at the corner of your frame or the side of a building be my guest. If you want to spend all day shooting "tests" and comparing cameras and res charts and then abandon a camera after you can do that too. But move on, don't keep pissing in the pool here and telling people they can't make films with these cameras.

The 7D isn't perfect, the message has been made, loud and clear and over and over again. Still, the footage it puts out looks great. And if you just want to go back and forth and whine about a decision to keep one camera or the other perhaps creating a blog is a better use of your time. I am interested in how people are using it to create great projects.

Alan Bradley
10-11-2009, 11:59 AM
Well said Isaac.

OPHERBA
10-11-2009, 12:05 PM
My answer to anyone who spend too much time in front of res charts ext.

http://www.vimeo.com/7011771

Know the camera limitations, go out and enjoy shooting.


Thanks,
Johnnie



My aim was to abuse DOF and test 720/50p slowed down.

Master to Cineform AVI.
No color correction or other picture manipulation was done.

Lens used: Samyang 85mm f/1.4
Samyang 8mm f/3.5 (one shot only)

Filters: Format ND 6, 9

Zim
10-11-2009, 12:17 PM
Yes all the of the HDSLRs have limits, like I guess all cameras do and what is important is going out and using them. Even after the massive dump on Nikon's HDSLRs people continue to make some great stuff with them. Now that we know the Holy Grail of these cameras has not been made yet people can get on with making films and not kicking each other cameras down. Like you said Issac I'm back on the fence waiting on my brand to come around, but I'm really enjoying watching the videos people are making from all of these cameras. I think we are moving into a really good time to be making films. New cameras, great editing systems, HD online, talking to other filmmakers around the world on websites like DVXusers,,,,,it just keeps getting better.


T


The 7D isn't perfect, the message has been made, loud and clear and over and over again. Still, the footage it puts out looks great. And if you just want to go back and forth and whine about a decision to keep one camera or the other perhaps creating a blog is a better use of your time. I am interested in how people are using it to create great projects.

Zim
10-11-2009, 12:19 PM
Looks great. I really liked the leaves falling out of the trees,,,,,


My answer to anyone who spend too much time in front of res charts ext.

http://www.vimeo.com/7011771

Know the camera limitations, go out and enjoy shooting.


Thanks,
Johnnie



My aim was to abuse DOF and test 720/50p slowed down.

Master to Cineform AVI.
No color correction or other picture manipulation was done.

Lens used: Samyang 85mm f/1.4
Samyang 8mm f/3.5 (one shot only)

Filters: Format ND 6, 9