Vs EX1 and stills

[morgan_moore]

People

I dont like my EX1 for many non technical reasons

I dont like my 7d which I swapped for a 5d

Information (at pre order time) pointed that the 7d had comparable real world visual appeal as the 5d but with PAL frame rates,

Information at that time also pointed that PAL was not going to happen ever on a 5d

To swap was therefore a sensible choice

I now believe this choice to be very far from sensible because the 7d, IMO has little visual appeal

Others reading the DSLR BOARD may in be in this boat of considering swapping a 5d for a 7d

I hope my test images help them with making their choice

Thanks for the interest

SMM

 

[Barry_Green]

im not that interested in the EX1 because it is big heavy and has a visually to me unapealing format

it does not take my nikkor prime set

you know perfectly well why I like a bigger chip

Well, then, I'd say that it's not that you prefer the Canons, you prefer the promise of what the Canons could be, right? Because if you want the ultimate in resolved detail, the Canons can't do it as well as the EX1 does.

A shallow focus can be less stressful on a codec, I guess; I know it can in terms of stills compression of which i have far more expertise, because it is simpler to 'describe' a simpler image - it has less information to communicate

This is correct. But none of the cameras being discussed have codec trouble, the 5D and 7D use a lot of bandwidth to overcome it. Well, at least, I haven't tested for codec trouble but I haven't noticed any in these two.

My understanding of moiree - as it is known in the stills game - alaising - is that it is an effect that happens when the frequency of the projected subject, comes close to the frequency of the pixel pitch

Aliasing is not moire. Moire is one example of how aliasing exhibits itself, and it is one of the most objectionable ways that aliasing exhibits itself. But aliasing happens in many ways, and moire is just one of the ways it shows itself.

For more on aliasing, check out the article in the articles section.

this is combatted with an AA filter that is deigned to blur the image before it occurs

Yes, but the problem with the HDSLRs is that they have such low ability to resolve true detail, if you used an appropriate AA filter that blocked out all
aliased detail, you'd end up with an image that's less than 720p.

I have, as stated, above owned two stills cameras that have no AA filter, I can tell you that my Kodak SRLn of 2004 or something still knocks any sub 20mp AA filter camera to bits for sharpness

Perceived sharpness, yes, but perhaps not actual resolved detail.

I know little about the 5d and 7d AA filters, but considering there is siginicant pixel pitch difference between the cameras one can assume they are very different in design, both aimed at optimising stills performance and may or may not have a different influence on video performance

They are unquestionably optimized for still performance, and they have pretty much *no* effect on video performance. The 5D and 7D (and GH1 and K-X and any other DSLR) are pretty much operating as if they had no AA filter when in video mode.

The fact that the 5d fails the zone tests at a point where pixel pitch closes to test chart is no suprise possiblly pointing to a situation where it is actually sharper when recording an image out of phase with the pixel pitch - in the same way as the Hassy and the SLRn are very sharp then fail dismally

Did you look at the zone plate? The 5D fails at about 500 lines. Your EX1 can easily handle in excess of 800 lines. That's why I'm telling you, regardless of any romanticizing of the 5D, that the EX1 is unquestionably going to be sharper and clearer in resolving detail.

You see (at the bottom of the page) that the filterless camera suffers worse on the test chart, but is clearly sharper on real world subjects the rope and the dollar bill showing that test charts to not always reflect real world sharpness

Test charts always reflect resolving power. Always. That's inherently what they do. What you're showing is that aliasing adds to the "perceived" sharpness. But it is not actually making the image sharper. You proved it for yourself, using the bar codes. And your 5D will perform the same.

Aliasing lets spurious detail through, that can indeed affect a person's perception of how sharp it is. But it's not accurate detail, and it moves and moires and confuses and muddles up the image.

There's a reason that camera manufacturers design AA filters specifically to remove that spurious alias detail!

 

[Barry_Green]

Didn't know that it was possible to aliase an image , I thought that aliasing had to do with rescaling a high resolution image to half or a third of its orginal size .

Check it out here.
http://www.dvxuser.com/V6/showthread.php?t=187503

 

[Barry_Green]

but given some more random subject matter, rope, dollar bills, the Filterless camera Beats the camera with a filter

But -- and please, follow along with me on this -- ONLY BECAUSE THEY ARE NOT PROPERLY RECREATING THE IMAGE PUT IN FRONT OF THEM!

Please, read the aliasing article at http://www.dvxuser.com/V6/showthread.php?t=187503. Look at the shots of the res chart, I think it will make everything become a lot more clear (pun intended).

Just because a camera moirees it does not mean it cannot resolve (out side of that moiree environment) - check out the images at the bottom of the page

Yes, it does -- inherently, that's exactly what it means! The presence of moire means that you have exceeded the camera's ability to resolve detail! The moire is false detail, fake, invented, and not real.

(Those who question the relevance of my test image; off angles, circles, cloth, should therefore possibly reconsider its value as potentialy a more relevant test than a test chart)

Again, ONLY if you don't care that it's not represented accurately!

 

[Barry_Green]

Im saying the 7d is not as good as either of them and that the terrible zone plate test result for the 5d may be somewhat misleading

Unfortunately I must point out the inherent impossibility in this statement. The zone plate is the hard cold truth. What's misleading is the scenes where the failings don't show up. The DSLRs use "cheats" to create their detail, and by and large, they get away with it. The zone plate doesn't let them cheat, it removes the "smoke and mirrors" and shows what the hard, stone cold reality is.

My opinion is that the 5d (when not filming wires or brides) and the EX1 meet my personal expectation of detail while the 7d does not

I do believe that that might be a romanticized opinion, remembering it more fondly than it actually performed. I think if you put the 5D and the 7D together, as Kholi has done, you'll find that they perform pretty much identically in video mode.

 

[Kholi]

I do believe that that might be a romanticized opinion, remembering it more fondly than it actually performed. I think if you put the 5D and the 7D together, as Kholi has done, you'll find that they perform pretty much identically in video mode.

Been saying this since the 7D got into the hands of users... They are pretty much identical and, if anything, the 7D has the edge on image quality thanks to updated compression and technology. And even THAT needs to be scientifically measured, something I have no~ time to do. LoL

I dunno how Barry finds the time or Patience. ;

 

[morgan_moore]

Thanks for your thoughts,

Barry I still dont see you acknowledging the D200 test of 'why charts can fail'

http://www.dvxuser.com/V6/showthread.php?t=188981

I have strated testing aginst the 5d

Another reasons that charts may 'fail' is that in the real world the 5d has less DOF and can use its bit ber frame to represent sharp parts of the image better (this is just a theory)

S

 

[astigmatic]

does the 7d resolve more detail in lower lighting situations than the ex1? I'm thinking once the ex1 ups the gain, resolution goes out the window. Where in that same lighting scenario, the 5d keeps up a very clean image, thus resolving more detail. This is perhaps why people in real world situations think these cams are sharper than the ex1.

 

[morgan_moore]

I started this whole thread using the EX1 as a kind of 'reference of expectation' for the 7d

I think DSLR sharpness perception is mainly driven by narrow DOF popping the subject off the background

I think the EX1 would be noisier in low light but this is not the reason for the sharpness perception

BTW

Testing against the 5d I have found them to be similar in ability to resolve the 5 having a weaker filter look, meaning that it is a little sharper but more alaised, the 5d is less noisy

I think that using the 7 requires costly primes in a way that the 5 does not to be made to pull off the 'sharp' illusion

I think that poor wideangle perception of images is actually the lack of DOF compared to a small chip camera

currently Im thinking I will stick with them both and despose of the 7 when the firmware becomes available

at the time of purchases word on the street was that no firmware would become available for the 5, this was an outrageous lack of comms from canon that pushed me to the 7

S

 

[ASG]

The resolving power of the 7 or 5 for that matter is so low I'm not convinced the quality of the glass makes a significant difference to the apparent sharpness of the video image. You only have to compare a still with video grab to prove it - even cheap lenses like sigmas produce very good quality sharp stills, not so in video mode. So it's not the lens.

 

[Ian-T]

Ha ha…I got to laugh at all of these complaints about resolving power etc. I’ll say this…it does not matter. The only time it matters is if you are comparing cameras or trying to match footage with a higher resolving camera. But other than that no one is going to tell….”hmm…this cam doesn’t resolve well.” Fact is….they still look sharp. I mean….no one cared about resolving issues when the 5Dll first came out. It just looked darn good in…”what it can do.” LOL

From as many movies and movie trailers I have watched I can say that not all FILM have the same resolving power either. But you don’t hear people complaining about that. I never worked with film but I can imagine that not all film stocks are the same. Some go for this or that for its resolution, color. Grain, yada, yada, yada. They are all different. I look at all of these DSLRs as different kinds of film stocks having their own characteristics. Now it’s up to you to go and choose your poison. One will prefer the GH-1 for its (fill in the blank_____) while the other will choose the 7D/5Dll for their (fill in the blank______) while another will choose Kodak Film for its (fill in the blank______) etc. That list can keep going on and on. It’s just crazy to see folks getting up in arms and defending their choice of camera as if it were their offspring.

I had an HV20 for a few years….and most people (including myself) loved using its Cinemode feature. Now….we know that the HV20 resolves details better than a 5Dll….but when using Cinemode in the HV20…detail is not as good as when you are using it in its standard mode…yet…we all still loved to use Cinemode. It just looked better for post production (regardless…it did not look as good as these DSLRs…and yes…even the crappy codec ridden D90). Not all cameras resolve the same kind of detail…but that does in no way dictate which cam is better than which in regards to their overall image. Ergonomically one could make a good case against these DSLRs….but for me it’s all about the final picture and coming from an HV20 I’m used to cutting corners to get a good image (or an image that I like)..

The 7D does not look soft to me….neither does the 5Dll or GH-1. If you put them side by side with an EX-1 then my perception of sharpness is likely to change. Leave them separate and I’ll probably give the same answer every time which is….I love what I see from these DSLRs over what I see from the EX-1. If you put all these cameras together and somehow make them work seamlessly then I’ll say....”more power to you.” It’s no different from a professional trying to match up two very different types of film for a feature. If they can get away with it then more power to them. If they can’t…then it just might end up being distracting.

 

[SPZ]

I'm currently editing a video that has footage from an HVX200, HVX200+ Brevis, EX1 and just a bit of 7D. In sharpness, the 7D is about the same to what I get with the HVX without adapter. Since I-ve sucessfully been able to broadcast HVX 200 footage with and without 35mm adapter to International TV Stations, the 7D sharpness doesn't seem like much of a concern to me. I bought the 7D not as a replacement to my HVX, but as a complement.

Someone talked about the quailty of the glass not influencing the picture. Well, I can confirm that it does, and in a big way: color can change from a top lens to a "cheapo", Bokeh, even sharpness. For the 7D I have currently the 16-35 L and the EF 28-200 (no stabilization). The difference between them is very, very noticeable-to the point of me being seriously considering both the 24-70 and the very expensive 70-200, for latter use when Canon comes out with their proper Video camera, which should be soon (this is a small hunch from no source whatsoever)

 

[morgan_moore]

The resolving power of the 7 or 5 for that matter is so low I'm not convinced the quality of the glass makes a significant difference to the apparent sharpness of the video image. You only have to compare a still with video grab to prove it - even cheap lenses like sigmas produce very good quality sharp stills, not so in video mode. So it's not the lens.

Indeed but for the image to 'pop' you need narrow DOF which is easier to create with the 5 or with expensive cine primes at the wide end

There are not many 2 or 2.8 options that are cheap less than 20mm

especailly if you want the convenience of zooms

S

 

[dcloud]

I'm currently editing a video that has footage from an HVX200, HVX200+ Brevis, EX1 and just a bit of 7D. In sharpness, the 7D is about the same to what I get with the HVX without adapter. Since I-ve sucessfully been able to broadcast HVX 200 footage with and without 35mm adapter to International TV Stations, the 7D sharpness doesn't seem like much of a concern to me. I bought the 7D not as a replacement to my HVX, but as a complement.

Someone talked about the quailty of the glass not influencing the picture. Well, I can confirm that it does, and in a big way: color can change from a top lens to a "cheapo", Bokeh, even sharpness. For the 7D I have currently the 16-35 L and the EF 28-200 (no stabilization). The difference between them is very, very noticeable-to the point of me being seriously considering both the 24-70 and the very expensive 70-200, for latter use when Canon comes out with their proper Video camera, which should be soon (this is a small hunch from no source whatsoever)

agreed. ive been scouting for lenses and its really hard to find great lenses without the high price

 

[Kholi]

Indeed but for the image to 'pop' you need narrow DOF which is easier to create with the 5 or with expensive cine primes at the wide end

There are not many 2 or 2.8 options that are cheap less than 20mm

especailly if you want the convenience of zooms

S

I don't understand what you mean at all. You mean to create a decent image you need to throw things out of focus? Trying to understand what you mean and how it's relevant to the quality of glass?

 

[Kholi]

The resolving power of the 7 or 5 for that matter is so low I'm not convinced the quality of the glass makes a significant difference to the apparent sharpness of the video image. You only have to compare a still with video grab to prove it - even cheap lenses like sigmas produce very good quality sharp stills, not so in video mode. So it's not the lens.

It only takes slapping on a cheap lens and then a Canon L or Zeiss prime and comparing the two images to see this.

I've got a cheap-ass 17-35/2.8~4 Tamron zoom and bet your money that I can tell the difference between this @ 35 F4 and my Zeiss Contax 35 @ F4.

If you want an even easier way to compare, just use the camera's zoom function. Focus on any object and zoom in, check what it looks like then switch the lens.

The glass absolutely, undoubtedly matters.

 

[Ian-T]

The resolving power of the 7 or 5 for that matter is so low I'm not convinced the quality of the glass makes a significant difference to the apparent sharpness of the video image. You only have to compare a still with video grab to prove it - even cheap lenses like sigmas produce very good quality sharp stills, not so in video mode. So it's not the lens.

Huh? Tell me something....what video camera (other than Red maybe) do you know of that can resolve its image just as good as its stills capability? What video camera do you use? Have you compared its still and video grab? i assure you unless you are coming out via HDMI or SDI whatever compression and color scheme it uses will muddy up the image as compared to the still capture of the very same image.

 

[mtan]

It think it all depends on what you are after. If you are primarily a filmmaker then the most absolutely high resolution detail might not be as important as other factors, so long as it is a nice clean image. The 7d can certainly produce a nice clean image and i think many of the 7d videos I have seen are amazing.

However if you are expecting a crisp high resolution image then you might be disappointed. I expected more when I bought the camera since it advertises as 1920 X 1080p. How would I expect the hv20 to have better resolution at 1440 X 1080i and being $1000 cheaper? I read a lot of these forums and never saw anything about the the hv20 having a better resolution image than the canon dslrs, otherwise i certainly never would have bought the 7d.

Thank goodness I still have an hv30 and will use that primarily for my purposes. I like the still capabilities of the 7d but it is too expensive for just that purpose.

Not sure what i am going to do, maybe I will sell the 7d whenever the next camera comes out and I will withhold buying any lenses or accessories for the 7d. I feel sort of misled by Canon with this purchase, just my personal feeling.

Not to take away from the positives of the camera though, it is great tool for professionals who can work miracles in post, I don't have the skills or patience to spend that much time on my footage.

 

[astigmatic]

Knowing that the ex1 is tact sharp... i bet one can pump the gain up to get close to the lowlight of the 7d, then run a de-noiser such has neatvideo, do a sharpen.. and end up with about the same resolution as the canon. Throw in a 35mm adapter and crush the blacks and I bet the picture quality would be pretty close.

 

[Kholi]

Knowing that the ex1 is tact sharp... i bet one can pump the gain up to get close to the lowlight of the 7d, then run a de-noiser such has neatvideo, do a sharpen.. and end up with about the same resolution as the canon. Throw in a 35mm adapter and crush the blacks and I bet the picture quality would be pretty close.

Well, there are other anomalies that come with the 35mm Adapter other than lowlight.

So while that's slightly true, it's a bit trickier.