AF-101 'on the set': some pics

[Gary Senda]

Here are some pics taken on the set of the short movie we have shooted in Italy to test the AF-101 on the set. It was not a quality test: we have used the prototype. We have only tested the functionality of the camera on the set and it is fantastic.

Bye bye HDSLR

Gary

[Spartacus]

Wil you tell us a little more about how you got your hands on a prototype and what your project is about?
Or are you just showing of here and try to make us hate you?:evil:

[vanvideo]

Oh my god! This camera makes people's faces break up into giant pixels!

[Gary Senda]

Wil you tell us a little more about how you got your hands on a prototype and what your project is about?
Or are you just showing of here and try to make us hate you?:evil:

101 has been presented in Italy during a special event after IBC, so the protoype was in Italy for a while (36 hours9).
Someone (not me, but I was on the set) has shooted a short film and we have showed it the day of presentation. Panasonic gave us the permit to make this, but they asked us to show the short film only at the presentation (no web diffusion: this was mandatory and I can understand).

The camera is not finished yet and a lot of functions were grayed, so we have not tested the image quality but the functionality of the camera on the set. The most of the film was shooted by night (as you can see there were cameracar shooting also) and in internal location, with a few light (led panels or existing lights).

- the camera is sensitive (better if you have fast lenses)
- the video noise is very low (even at +12bB)
- it is very easy to use handhelded and you don't need those terrible HD-SLR rigs
- EVF and monitor are very good (better than on HPX370)
- the results with mechanical photo lenses are stunning, expecially if you use fast lenses
- we have tried in every mode to evocate aliasing and moire, but we were not able

This is a camera for shooting in HD-SLR style but using a true videocamera and saving a lot of time and money also (no separate audio, no sync in post productions, no complicated rigs, no external ND, no mattebox mandatory, no loupe and, first of all, no aliasing and no moirè). But you have scene files, gamma, master pedestal, timecode, uncompressed audio (already in sync) and all the other features of a videocamera that increase your productivity.

The dynamic range needs to be a little increased: for the moment 5d and 7d are a little better, in my opinion. But we have shooted in HD norm mode and it was not possible to change the gamma, so perhaps you can have a better dynamic range from the prototype camera.

I apologize for my broken english.

If you have further question you can ask, I will try to answer

Gary

[Isaac_Brody]

Thanks for the review Gary, definitely answers some questions and concerns I had.

[Everts]

Thanks for sharing !

How was the jellow / skew on the car shots ?

Is that a Tamron or Sigma lens in the third photo, 101 hand2.jpg ?

Were you able to work with just the gain , why not ISO ?

Thanks

Everts

[AdrianF]

Thanks Gary, interesting insight to what we might be able to expect from this camera.

[Gary Senda]

Thanks for sharing !

How was the jellow / skew on the car shots ?

Is that a Tamron or Sigma lens in the third photo, 101 hand2.jpg ?

Were you able to work with just the gain , why not ISO ?


- No visible yello or skew in cameracar shootings (we used wide angle lens)

- The lens in the hand held pic is a nikon mount ZEISS (I have tested cheap russian lenses also with similar results)

- ISO was blocked on 400 so the only way to increase senstivity was the gain dial

As you can see, pics were taken without flash and we were shooting with the same light (with better results, beacuse GH1 kit lens is very dark and on AF-101 we were using fast lenses)

[Homunculus]

Wow red flag: Dynamic Range is better on a Canon 5d 7d?

Canons test at about 8 stops give or take dynamic range. Barry tested AF100 at 10.5 stops. Either something is terribly wrong with the camera you have or something else fishy is going on as there's no way the Canon DSLR's should have better dynamic range than AF100

[Gabrobot]

Wow red flag: Dynamic Range is better on a Canon 5d 7d?

Canons test at about 8 stops give or take dynamic range. Barry tested AF100 at 10.5 stops. Either something is terribly wrong with the camera you have or something else fishy is going on as there's no way the Canon DSLR's should have better dynamic range than AF100

They weren't shooting in a cine gamma mode though which you kinda have to do to really get a good sense of dynamic range on a panasonic camera.

[Homunculus]

They weren't shooting in a cine gamma mode though which you kinda have to do to really get a good sense of dynamic range on a panasonic camera.
Just curious how do you know they weren't shooting in a certain mode as I don't see him mention that anywhere? Or were you present at the shoot?

[Mike McNeese]

Wow red flag: Dynamic Range is better on a Canon 5d 7d?

Concerns me quite a bit, too. But, I'm confident that a finished product will give us better results than this 70% prototype.

[Lyris]

From what I saw of menu videos on a German site, there will be a lot of gamma curve options on the finished unit, so I'd imagine that would perhaps alleviate (or help alleviate) any issues there.

[Gary Senda]

Wow red flag: Dynamic Range is better on a Canon 5d 7d?

1. gamma node was not available on our camera copy
2. DSR stretching was not available on our camera copy

I've written.: this was not a quality test and, at the moment, quality tests are not possible with a limited camera. This was just an attempt to use the camera on the set to test the functionality and it is much, much, much better than on HD-SRL

[Gary Senda]

From what I saw of menu videos on a German site, there will be a lot of gamma curve options on the finished unit, so I'd imagine that would perhaps alleviate (or help alleviate) any issues there.

Yes, but we were not allowed to use them because it seems the software is not yet debugged and the camera could stop to work. There was a panasonic fellow with us and the only thing we were allowed to change in the menu was 'shoot without lens'.

[Gary Senda]

Furthermore: I've written 'a little better' that is quite different from 'better'. Perhaps less than 1 stop of difference, but we have not measured it and the most of shoots were taken nightime.

So, if we measure it, perhaps I'm wrong and anyway the grayscale charter was very satisfactory

[Homunculus]

1. gamma node was not available on our camera copy
2. DSR stretching was not available on our camera copy

I've written.: this was not a quality test and, at the moment, quality tests are not possible with a limited camera. This was just an attempt to use the camera on the set to test the functionality and it is much, much, much better than on HD-SRL


I feel that no special mode ESPECIALLY DR stretching should be NEEDED for this camera for it just to have equal DR with a 1400$ DSLR. This is a 6000 Prosumer Broadcast Division product. It should not need a special fancy "mode" (which raises a lot of noise by the way) like DRS just to have equal DR with a 1400$ camera.
I know it's not finished yet so we'll see how the final product is but I'm just saying, those are worrying things if with DRS turned off (which should only be used sparingly anyway) it doesn't have higher DR than a 1400$ DSLR.
If that ends up being the case then count me in on the GH2 bandwagon and forget this over priced thing.

[Barry_Green]

On the stouffer backlit chart the 7D measured 8.3 stops. On a DSC backlit chart the AF100 measured 10.5 stops.

There is no need to worry.

[Gary Senda]

it doesn't have higher DR than a 1400$ DSLR.


Deninitively, my eyes nightime are not a measurement instruments!

I've clearly written: test was on functionality and not on image. It is not possible to have a definitive idea of a prototype camera, expecially in our conditions and without measuring parameters.

[mcgeedigital]

I feel that no special mode ESPECIALLY DR stretching should be NEEDED for this camera for it just to have equal DR with a 1400$ DSLR. This is a 6000 Prosumer Broadcast Division product. It should not need a special fancy "mode" (which raises a lot of noise by the way) like DRS just to have equal DR with a 1400$ camera.
I know it's not finished yet so we'll see how the final product is but I'm just saying, those are worrying things if with DRS turned off (which should only be used sparingly anyway) it doesn't have higher DR than a 1400$ DSLR.
If that ends up being the case then count me in on the GH2 bandwagon and forget this over priced thing.

And you are basing this on an unfinished prototype.

Way to jump the gun.

[Joe Walker]

Hey Gary (or Barry for that matter),

You've said this camera really benefits from fast lenses, if most of my lenses are say around T3 (f2.8) at there widest apertures, is that a decent starting point or should I be looking for faster glass around T1.3 or T2? I know its highly speculative right now since this camera is still technically a prototype but any insight is helpful. Thanks.

[Gary Senda]

Hey Gary (or Barry for that matter),

You've said this camera really benefits from fast lenses, if most of my lenses are say around T3 (f2.8) at there widest apertures, is that a decent starting point or should I be looking for faster glass around T1.3 or T2? I know its highly speculative right now since this camera is still technically a prototype but any insight is helpful. Thanks.

I think it now can be only speculative. F2.8 is a good starting point. Kit lens is 4-5.8 and if you use a 135mm f2.8 you have than 2 stop in that focal range.

F 2.8 is still good for low light, but if you are interested in extreme lowlights shootings and night shootings f2 (or less) lenses are required.

We should also consider that video noise is minimal, so we can use gain without a significative image degradation

[bwhitz]

It should not need a special fancy "mode" (which raises a lot of noise by the way) like DRS just to have equal DR with a 1400$ camera.



Maybe it will... who knows. Either way... just because something's more expensive, doesn't mean it will be better.

Case and point... Arri Alexa.

[Dino]

Barry measured 10.5 stops--that's impressive. Our HPX2700 in Film-Rec 600%/AVC-Intra 100 has 11 stops of latitude. When using HD gamma with the 2700, sensitivity is better, but latitude is compromised, same with our HPX170 and HDX900.

[xmephestox]

from what i saw of it at the hd expo in nyc, it looks like it's shaping up very well. I think that 30% more they are working on is definitely there though, but nothing unachievable but the best and brightest at panny. I felt like the blue channel still needed a lot of work, it suffered a lot under tungsten light, a common thing in digital sensors. Set at 400 iso (the lowest the camera to go, a bunch were greyed out) I still noticed (what i consider a unreasonable amount of noise) in the subtle shadow areas. If they can get their blue channel to get clean at, at least 400-800 iso, this camera is gonna be pretty cool.

[Homunculus]

On the stouffer backlit chart the 7D measured 8.3 stops. On a DSC backlit chart the AF100 measured 10.5 stops.

There is no need to worry.
yay barry to save the day. i feel better already ;-)

[...]

camera body only a mother could love, but no moire is going to be sweet.

can you tell me what camera mount that is?

[Bandido]

It is good to see it can be handheld operated with a nanoflash and a rod system.

[Gary Senda]

camera body only a mother could love, but no moire is going to be sweet.

can you tell me what camera mount that is?

do you mean the rod support? I think it is a Chrosziel

[Dan_Kanes]

On the stouffer backlit chart the 7D measured 8.3 stops. On a DSC backlit chart the AF100 measured 10.5 stops.

There is no need to worry.

Hey Barry,

I'm gonna be a pain in the butt here and ask what the 7D settings were?

Auto lighting optimizer off?
What iso?
Was the reading done from a QuickTime? A professional monitor?
If it was a QuickTime, what was the gamma?

Was D+ off?

What picture profile?

I don't want to defend Canon - just curious about all the many variables that comprise such a
Test.

-Dan

[aalleexx]

What kind of mechanical lenses, examples?

[ryanwieber]

As we seem to already have folks getting hands-on experience (and apparently testing), exploring the traditionally problematic aspects of DSLRs... the thing I am wondering most about doesn't seem to be talked about hardly at all. I'm extremely curious how the rolling shutter is working out. Not just in a "it seemed good!" sense, but in a "it looks better/same/worse than X camera" or something. I understand it's a difficult thing to quantize, but as the technical details are as of yet unreleased, hands-on comparison seems the only way to get a gauge of this. Has anybody studies that aspect yet?

Edit: My apologies, I hadn't read the NYC thread yet, where Barry said:

"In terms of skew? Blows away the HPX300 in 1080/24p skew performance, and about matches it in 60p. It seemed much more on par with the HPX370 (and, accordingly, the EX1/EX3 and Red One). They are all pretty much state of the art for skew, and the AF100 seemed to be about the same."

Which pleases me greatly.

[hunter richards]

Supposedly its on par with EX3/RED/Hpx370- you'll find that in many posts.

[Shipsides]

I saw the same thing on a DSC labs chart at Panasonic.. and this isn't the finished sensor.



On the stouffer backlit chart the 7D measured 8.3 stops. On a DSC backlit chart the AF100 measured 10.5 stops.

There is no need to worry.

[Barry_Green]

I'm gonna be a pain in the butt here and ask what the 7D settings were?
There's a whole thread on it, I think the title was something like "Canon 7D: meet the Stouffer 42-step chart" or something. All was explained there.

[bwhitz]

There's a whole thread on it, I think the title was something like "Canon 7D: meet the Stouffer 42-step chart" or something. All was explained there.

Eyes lie, tests lie... That test said 7 stops Zacuto said 11... Who knows what to believe.

Do people honestly believe the 7D is only hitting 7 stops because of this test? That's absurd. If the 7D only had 7-8 stops it would look like a flip video camera. No offense to anyone, but I'll believe my eyes over all these stupid tests.

[Barry_Green]

And I will believe documented tests with calibrated test equipment, over subjective opinion. To each his own.

The 7D is capable of 8.3 stops of dynamic range. And Art Adams got the exact same results out of the 5D. And DP Review got the exact same results out of both.

[Gary Senda]

What kind of mechanical lenses, examples?

In the short film were used:

- Zeiss primes with Nikon mount
- Nikon

Personallly I've tested my 2 sets of old lenses:

- Canon FD silver nose
- Zeiss Jena, Jupiter, Industar and other russian M42 lenses
- Samyang 85mm

Using these lenses on AF-101 was very difficult if not impossible to distinguish between expensive Zeiss primes and old DDR lenses. There was a big mechanic difference, but no visible difference on a plasma full HD monitor. Zeiss Jena 20mm, 35mm, Pancolar 50mm e Jupiter 85mm were incredibly sharp, even wide open (lenses are good and vignetting helps).

Regards

Gary

[acoelho1]

Just be happy that people don't watch test charts :)

[Dan_Kanes]

How do you like the Samyang 85mm? Both Optically and mechnically?

I think if Samyang decided to make a set of Cinema/Video Optimized Lenses in a full range of lengths from 16 to 100 they would really be able to sell A TON. Imagine the dampening of a cinema lens, geared Focus and Iris, with a univeral back the like the CP2's, universal 80mm OD fronts with 77mm ID filter threads and under $1500/lens - boom - you have Red Pro Prime / CP2 killers.

How did chromatic aberation manifest itself among the different lenses on the sensor?

Thanks!

[Gary Senda]

Samyang 85mm is a good lens for video and it is a 'must have', in my opinion: it is cheap, fast, sharp and the mechanic is not bad and it is usable with follow focus. Furthermore it is very good for photo. A 35mm f1.4 has been announced at photokina last week and a 14mm f2.8 is already available. If you add the Panasonic 20 mm f1.7 pancake, an 50mm f1.4 lens like the Canon FD and a 135mm f2 lens you will have a set of fast prime lenses at an affordable price.

But what I am really waiting for is a cine style zoom at an affordable price: I hope that someone will make one

[Everts]

How do you like the Samyang 85mm? Both Optically and mechnically?

I think if Samyang decided to make a set of Cinema/Video Optimized Lenses in a full range of lengths from 16 to 100 they would really be able to sell A TON. Imagine the dampening of a cinema lens, geared Focus and Iris, with a univeral back the like the CP2's, universal 80mm OD fronts with 77mm ID filter threads and under $1500/lens - boom - you have Red Pro Prime / CP2 killers.

How did chromatic aberation manifest itself among the different lenses on the sensor?

Thanks!


Samyang 85mm is a good lens for video and it is a 'must have', in my opinion: it is cheap, fast, sharp and the mechanic is not bad and it is usable with follow focus. Furthermore it is very good for photo. A 35mm f1.4 has been announced at photokina last week and a 14mm f2.8 is already available. If you add the Panasonic 20 mm f1.7 pancake, an 50mm f1.4 lens like the Canon FD and a 135mm f2 lens you will have a set of fast prime lenses at an affordable price.

But what I am really waiting for is a cine style zoom at an affordable price: I hope that someone will make one

Go Samyang .....really !:thumbsup:

[Spartacus]

But what I am really waiting for is a cine style zoom at an affordable price: I hope that someone will make one

The only one who has the experience to pull this off would be Canon I guess.
But they seem to just like what´s going on right now: videoguys buying into pro-photo cameras and L-glass.
If the AF101 comes out this year, Canon will show a prototype new XL camera at NAB, and will deliver around August 2011- maybe along with a cinestyle manual lens we all could use...

[LoganMackay]

Maybe panny can adapt the olympus fast zooms to a more video focused version.

[Homunculus]

Eyes lie, tests lie... That test said 7 stops Zacuto said 11... Who knows what to believe.

Do people honestly believe the 7D is only hitting 7 stops because of this test? That's absurd. If the 7D only had 7-8 stops it would look like a flip video camera. No offense to anyone, but I'll believe my eyes over all these stupid tests.


Dude the 7d nor any other DSLR hits anywhere close to 10 or even 11 stops. quit dreaming the Red One barely does 11 stops you think 7d has the same dynamic range as the Red? you'll believe your eyes? come on man get real 7d shoots in STILLS mode at 9 to 10 stops, how do you figure it can have greater dynamic range in compressed video mode than in raw stills? LOL!

[mitsos68]

Samyang 85mm is a good lens for video and it is a 'must have', in my opinion: it is cheap, fast, sharp and the mechanic is not bad and it is usable with follow focus. Furthermore it is very good for photo. A 35mm f1.4 has been announced at photokina last week and a 14mm f2.8 is already available. If you add the Panasonic 20 mm f1.7 pancake, an 50mm f1.4 lens like the Canon FD and a 135mm f2 lens you will have a set of fast prime lenses at an affordable price.

But what I am really waiting for is a cine style zoom at an affordable price: I hope that someone will make one

http://www.samyang.pl/article,8,news?PHPSESSID=8d64bf08981f15f571dcf87d0 aa2790a

[New Direction]

I'll let the videos do the talking for me.

[hyphygreek]

i'll let the videos do the talking for me.

+1

[Rick Burnett]

I have the Bower 85mm F1.4 which people believe is the same as the Samyang. I've been pretty happy with it. One thing is for sure, I felt the focus throw was MUCH better than any Canon or Sigma I have. It was on par with some of my Nikon lenses.

Wish the 85mm Sigma would come out already. I want to compare it to the 30mm.

[bwhitz]

Dude the 7d nor any other DSLR hits anywhere close to 10 or even 11 stops. quit dreaming the Red One barely does 11 stops you think 7d has the same dynamic range as the Red? you'll believe your eyes? come on man get real 7d shoots in STILLS mode at 9 to 10 stops, how do you figure it can have greater dynamic range in compressed video mode than in raw stills? LOL!

Are you serious? The reason DSLRs look so much like film isn't just the shallow DOF... it's something called contrast ratio. I don't know why people have such a hard time beliveing something like the 7D can hit 11 stops. The sensors that go into these cams are designed from the ground up (by engineers that are better than Red's) to look like film and have similar contrast, color, exposure range... ect. Exposing for a shot in stills mode is exactly the same in video mode. This is easily observable by setting up a shot in stills mode and then switching to video... The image is the same. The camera doesn't automatically lose the ability to see light just because of the mode it's in. The sensors "sees" X to Y amount of light regardless. Of couse, however, the Advantage of Raw is that you can brig out hidden areas of the image and fix exposure mistake... This is 'latitude' and yes compression does ruin most of that. Compression also destroys allot of the mid-tones of the image... but it doesn't effect the overall contrast of the image. I guess maybe, I should rephrase my statement to something like "the 7D has 11 stops of 'contrast range'" Since 'Dynamic' implies a full spectrum of mid-tones in between the stops, which obviously is crippled by compression.

http://www.vimeo.com/13762496

I haven't already, take a look at this video. This defiantly shows off how good the DSLRs can look in the right hands with proper exposure. If the 7D really only had 7 stops of dynamic range, the image would not look anything like it does in that video. If you can't see how similar the contrast range is to film and Red in that piece, then you're eyes just aren't suited for cinematography... or any visual art for that matter.

Red does a good job of fooling people into believing their cameras are made with some sort of magic. Really, they are just DSLR sensors strapped to a bigger processor and a really clever compression scheme. Their cameras are great for sure... but if anything, their overall sensor technology is probably behind Canon and Nikon still. They're ahead in digital cinema right now, but it's only because they hit a target that nobody else saw... Now that everyone else knows "what's up" the next few years are going to be pretty interesting...

[Rick Burnett]

Are you serious? The reason DSLRs look so much like film isn't just the shallow DOF... it's something called contrast ratio. I don't know why people have such a hard time beliveing something like the 7D can hit 11 stops. The sensors that go into these cams are designed from the ground up (by engineers that are better than Red's) to look like film and have similar contrast, color, exposure range... ect. Exposing for a shot in stills mode is exactly the same in video mode. This is easily observable by setting up a shot in stills mode and then switching to video... The image is the same. The camera doesn't automatically lose the ability to see light just because of the mode it's in. The sensors "sees" X to Y amount of light regardless. Of couse, however, the Advantage of Raw is that you can brig out hidden areas of the image and fix exposure mistake... This is 'latitude' and yes compression does ruin most of that. Compression also destroys allot of the mid-tones of the image... but it doesn't effect the overall contrast of the image. I guess maybe, I should rephrase my statement to something like "the 7D has 11 stops of 'contrast range'" Since 'Dynamic' implies a full spectrum of mid-tones in between the stops, which obviously is crippled by compression.

http://www.vimeo.com/13762496

I haven't already, take a look at this video. This defiantly shows off how good the DSLRs can look in the right hands with proper exposure. If the 7D really only had 7 stops of dynamic range, the image would not look anything like it does in that video. If you can't see how similar the contrast range is to film and Red in that piece, then you're eyes just aren't suited for cinematography... or any visual art for that matter.

Red does a good job of fooling people into believing their cameras are made with some sort of magic. Really, they are just DSLR sensors strapped to a bigger processor and a really clever compression scheme. Their cameras are great for sure... but if anything, their overall sensor technology is probably behind Canon and Nikon still. They're ahead in digital cinema right now, but it's only because they hit a target that nobody else saw... Now that everyone else knows "what's up" the next few years are going to be pretty interesting...

It's not a DSLR sensor, it's a CMOS sensor that uses a Bayer pattern. DSLRS use other types of sensors as well. They've never claimed it to be anything other than what it was. And I VERY MUCH disagree that their sensor technology is behind Canons. Canon has worse rolling shutter, horrible aliasing/moire worse than ANY other CMOS sensor that is current, overheating problems, and I am sure there are more being left out.

In fact, their Epic cameras that are being used in real productions right now during their testing phase are absolutely amazing. 5K up to what was it 120fps? Canon can't even do 1080p at 60fps at the moment. In addition, without their new HDR stuff they were getting 13.5 stops of DR just like the Alexa?

Clearly they've shown that their technology is beyond Canon's at the moment.

Please, show me some tests from a Canon CMOS sensor that show 13 stops at 5k resolution at say even 24fps?

[Barry_Green]

I don't know why people have such a hard time beliveing something like the 7D can hit 11 stops.
Well, because it doesn't. It does 8.3 stops. That's what DPReview tested it at, that's what I tested it at, and that's what Art Adams tested the 5D at.

[bwhitz]

Well, because it doesn't. It does 8.3 stops. That's what DPReview tested it at, that's what I tested it at, and that's what Art Adams tested the 5D at.

And I'm not saying you or any of those other guys are wrong... I do believe on your tests it hit only 8.3 stops. However, other tests have shown 11 stops. Now the only logical explanation is that the tests that measured higher DR were staged vindictively to fuel a conspiracy bent on selling overpriced aluminum camera accessories... Or that dynamic range measures different results on different tests. I think I'm going to go with that latter.

All I know is that the cameras I've shot with... XH-A1, HPX 3000, ex-1, ect... are all supposed to be around 7-10 stops and they don't look anything like a film image... even the HPX with it's 2/3" sensor. So either your tests are correct and the 7D only has 8 stops... meaning that the tests for the other cams were BS and they're really only around 4-6 stops. Or that the tests used that register 8 stops, for some reason, measure the DR dynamic range differently and give a lower result.

Since Zacuto's "Shoot out" was the only test to actually show the back-light tests for an image from a film camera (13 stops) with an image from a DSLR (9-11 stops) in the same time and place (control variable), then logically and scientifically, I'm going to believe those results over other tests where only one camera was used at one point and time and then compared on a scale to the results from other tests performed in different and varied conditions.

[William_Robinette]

However, other tests have shown 11 stops.

Can you pass on a link please?



Now the only logical explanation is that the tests that measured higher DR were staged vindictively to fuel a conspiracy bent on selling overpriced aluminum camera accessories... Or that dynamic range measures different results on different tests. I think I'm going to go with that latter.


...or that these other tests may have errors?

[bwhitz]

It's not a DSLR sensor, it's a CMOS sensor that uses a Bayer pattern. DSLRS use other types of sensors as well. They've never claimed it to be anything other than what it was. And I VERY MUCH disagree that their sensor technology is behind Canons.
Please, show me some tests from a Canon CMOS sensor that show 13 stops at 5k resolution at say even 24fps?

Umm... DSLR sensors ARE cmos sensors that use bayer patterns. Red just has a lower pixle count and therefor better DR (at the moment). And just because a company doesn't have a product that measures up to Red's doesn't mean that their technology is behind. I'm sure if Canon dropped everything and built a dedicated cinema cameras, it would blow away Alexa and Epic. Why? Canon is bigger. They are an R&D powerhouse with revenue in more areas than just cameras. But Canon is in the business of staying in business and dropping everything to build a 120fps 5K cinema camera isn't exactly profitable for them simply because it would not be a consumer product. That's like saying Microsoft couldn't have made the ipod because they were behind in technology. Nope... In fact Apple was behind in the tech curve, but saw a product/market that others didn't and took it... just like Red. Unfortunately the pro-camera industry doesn't work the same. People arn't exactly running out and buying Red cameras for Christmas gifts.... which is why, surprise, the Scarlet isn't out yet! Red's business model is highly flawed... for many reasons, but mostly because, you can't make just one super camera that will never be obsolete and expect to generate consistent review for R&D. Eventually the market will be saturated and no one will need a Red camera anymore. Let's face it, consumers are not in the market for shooting 5k 96fps cinema grade images and for every "Pro" there are probably about 1 million consumers...


Can you pass on a link please?



...or that these other tests may have errors?

http://www.zacuto.com/shootout - episode one

Even if there are errors in the Zacuto test, it still showed film stock measuring at 13 stops and a DSLR measuring at 11 stops at the same time and place (something with other tests have not done, and are therefor are invalid), which means that just by the laws of f-ing logic... DSLRs are just generally 2-3 stops below film And since film is acceptably rated at 13-14 stops... that puts the 7D in the ball park of 10-11.

[Rick Burnett]

Umm... DSLR sensors ARE cmos sensors that use bayer patterns. Red just has a lower pixle count and therefor better DR (at the moment). And just because a company doesn't have a product that measures up to Red's doesn't mean that their technology is behind. I'm sure if Canon dropped everything and built a dedicated cinema cameras, it would blow away Alexa and Epic. Why? Canon is bigger. They are an R&D powerhouse with revenue in more areas than just cameras. But Canon is in the business of staying in business and dropping everything to build a 120fps 5K cinema camera isn't exactly profitable for them simply because it would not be a consumer product. That's like saying Microsoft couldn't have made the ipod because they were behind in technology. Nope... In fact Apple was behind in the tech curve, but saw a product/market that others didn't and took it... just like Red. Unfortunately the pro-camera industry doesn't work the same. People arn't exactly running out and buying Red cameras for Christmas gifts.... which is why, surprise, the Scarlet isn't out yet! Red's business model is highly flawed... for many reasons, but mostly because, you can't make just one super camera that will never be obsolete and expect to generate consistent review for R&D. Eventually the market will be saturated and no one will need a Red camera anymore. Let's face it, consumers are not in the market for shooting 5k 96fps cinema grade images and for every "Pro" there are probably about 1 million consumers...

No, actually you are wrong, not all DSLR sensors are CMOS sensors with a Bayer pattern. The Leica S2 uses a Kodak CCD sensor. The Sigma SD15 uses a Fovean sensor which is VERY different than standard CMOS/Bayer pattern sensors.

There is SO MUCH MORE to DR than the size of the pixel. Why doesn't the 5D have a larger DR than the epic when it's pixel sizes are 8.2um and the Epic has a pixel size of 5.4um? Clearly, there is more to DR than just the size of the pixel. In fact, most of it comes down to the elimination of noise in the system as that is what takes away from the lower end.

Do you personally have inside information to Canon that you understand their R&D structure? Have you worked for companies that large and with their R&D departments and understand the principles of how they work? I have, with MANY huge companies equal to and larger than Canon, like Samsung for instance. These companies are NOT unified powerhouses, they are generally fragmented smaller businesses inside a larger umbrella business where the left hand constantly doesn't know what the right hand is doing. Didn't you hear that Canon recently merged groups together in their company? There is a reason for that because it is hard to align separate business entities.

Do you have personal insight into RED? You claim that their business model is flawed, yet I don't see them folding OR complaining about their business not working. Sure, they realigned themselves to not cater to the prosumer market, but that was an addition to their original plan which was the professional cinema market. How can you possibly make a judgement of their business model without an insight to their books? You can't.

I've been in the CMOS design world for 15 years. I've worked on chips in almost every imaginable type of equipment. I have the experience and position to tell you that from my standpoint, RED is ahead of the game in terms of technology with CMOS design compared to Canon on the market. If Canon has something better, and they could, because we just don't know, they've done NOTHING to prove to me that their technology is getting ANY better. Each camera they release, including now the 60D, continues to have the same problems as the 5D, 7D and T2I. While companies like Panasonic and Nikon are actually releasing improved cameras.

And I hate to break it to you but there were plenty of MP3 players before the iPod came out. Diamond Rio? It doesn't matter how much money you have, it's about understanding what it is that people want. The iPod gave the user a pleasant experience. Even the Wii is VASTLY inferior technology wise to the Playstation 3 and the Xbox 360 (which I worked on as well), and yet it was an absolute hot item because of how it interacted with people.

Canon doesn't have to have better technology to beat others in the market, they just have to anticipate what it is people want in a camera. Would you buy a camera for 1/4 the price if it only did 60fps instead of 120fps? It's what is important. Canon doesn't seem to listen much to what people want, whereas both Panasonic and RED seem to bend over backwards to listen. That's why I think the AF100 exists.

RED is not trying to sell to consumers, so your point has no relevance. They've already stated they are making cameras for professionals. You don't know what their revenue targets are or what their bottom line is. Just because they might not bring in the same profit amount as Canon doesn't mean they don't have the same profit margin. But really, NONE of that matters to me. Therefore Canon is targeting the consumer. They don't have to make anything more expensive since that is not their target audience. RED does, because they've got a higher standard to hit. They have the motivation.

[mcgeedigital]

And I'm not saying you or any of those other guys are wrong... I do believe on your tests it hit only 8.3 stops. However, other tests have shown 11 stops.

Then those tests are flawed.

Period.

I have the 5d, 7d, and HPX-500 and have shot on RED as well as 16mm and 35mm film.

There is no way in hell that the 7d has 11 useable stops.

The 5d only has 10:

http://provideocoalition.com/index.php/aadams/story/canon_5d_how_much_dynamic_range_does_it_have_reall y/

So there is no way that the 7d, having a SMALLER sensor, has MORE.

8.5 is more realistic.

[bwhitz]

Then those tests are flawed.

Period.



You're not in any position to make that call... Nobody really is as long as there are different tests showing different results.

We can all agree that film easily hits 13-14 stops right? So if the Zacuto test was accurate in that area... how can you say that when they measure the 7D at 11 stops that there all the sudden wrong? There has to be some truth to the findings if the control variable (film stock) measured correctly. If they found that film stock had 18 stops or something like that and then said the 7D had 11 stops... that would probably mean the test was crap.

You can't just say "Oh the one that is the lowest is correct... " That just makes you sound like you're pissed off and jealous.

BTW, the reason that the 7D might be showing better DR that the 5D is because of the gapless micro-lensing. Supposedly, they waste less light by focusing it into the photodiods...

[Rick Burnett]

Because the Zacuto test was about photo mode, which uses the entire sensor and doesn't throw out any pixel information.

It's not the same image between shooting full resolution and shooting in video. In video, a much smaller subset is used that includes pixel binning and some suspect line skipping. Whatever they are doing to reduce the data, it is reducing the DR between both modes.

[Steve Kahn]

Because the Zacuto test was about photo mode,.
amazing...

[acoelho1]

Who cares... you can get a great image from all of the above. The HDDSLR has proven to be an essential tool in the tool box. Leave it at that move on.

[dcloud]

Then those tests are flawed.

Period.

I have the 5d, 7d, and HPX-500 and have shot on RED as well as 16mm and 35mm film.

There is no way in hell that the 7d has 11 useable stops.

The 5d only has 10:

http://provideocoalition.com/index.php/aadams/story/canon_5d_how_much_dynamic_range_does_it_have_reall y/

So there is no way that the 7d, having a SMALLER sensor, has MORE.

8.5 is more realistic.

That 5d test is hardly ten. 9 isnt even usable at that level.

[mcgeedigital]

That 5d test is hardly ten. 9 isnt even usable at that level.

I agree.

[bwhitz]

Because the Zacuto test was about photo mode, which uses the entire sensor and doesn't throw out any pixel information.


uhhh... I think you need to watch it again. They never used photo mode until one part at the very end where they showed what a RAW video image could look like. Everything else was video mode.

[Danielvilliers]

I am not going into who is best, who will be best, etc, etc... But as a photographer I would like to bring some reality here. I am not talking about video dynamic range of the Dslr here but on the photo side. It is a fact now that all the recent sensor of the last two to three years give at least 11.5 to 12 stop of dynamic range. The highest being the Nikon D3x rated at about 13.7 stop at DXO mark. Result varies between reviewers site like Dpreview but by acceptable margins. These result are in RAW so as to give a little sense of perspective towards the red sensor which itself is in raw. So in pure sensor performance the red is very very good, but not some sort of alien technology trumping the other manufactures by 2 to 3 stop of dynamic range. You cannot use the result of the jpeg engine to say the Canon or Nikon show 8 - 9 stop, again I am talking pure sensor raw still performance.

Now in video I don' t know, the only site that I see which has a collection of test from the canon 5d to the Arri Alexa and Red (One, Mx) is provideocoalition.com and it puts the Canon 5D quite close to the Redone, about a stop difference only but his test is a bit confusing in some place. The individual conclusion of his original red mx and the comparison of the red Mx and Alexa test below being an example. I don' t know if he has changed methodology or not. The last thing is the the 7d and 5d share about the same dynamic range in RAW still mode. Smaller pixel does not mean lower dynamic range necessarily as illustrated by the Nikon D3x 24 mega pixel compared to the Nikon d3 12 mega pixel having about a stop more dynamic Range. Iso is the same thing, technology advances until someday we will reach some physical limit. But until now we can see the latest crop of 16 mega pixel sensor giving a stop better Iso quality than the last 12 mega pixel ones.

http://provideocoalition.com/index.php/freshdv/story/epic_vs_alexa_dynamic_range/

[f64manray]

Just to confuse things further, ASC DP Shane Hurlbut says he gets 12 stops out of the 5DII with his camera/post settings and choice of lens. No one has more experience with the 5DII than he does.

Of course Red has announced that they are getting 18 stops with their HDRX system that will be incorporated in all their new cameras.

[Rick Burnett]

uhhh... I think you need to watch it again. They never used photo mode until one part at the very end where they showed what a RAW video image could look like. Everything else was video mode.

Ahhh, you are correct. I had this confused with another test I had looked at. Thanks for pointing that out. Wish they had a good fast forward feature, had to watch the whole thing again! :)

Would be nice for them to share some waveform views of the results like some of the tests did.

[bwhitz]

So in pure sensor performance the red is very very good, but not some sort of alien technology trumping the other manufactures by 2 to 3 stop of dynamic range.

http://provideocoalition.com/index.php/freshdv/story/epic_vs_alexa_dynamic_range/

Thank you... that was the whole point I was trying to make. Red is an awesome company doing allot of innovative stuff... but in terms of pure sensor design... it's not something that Nikon or Canon aren't capable of or haven't seen before. In fact, I'm not 100 percent sure, but I could swear I remember hearing that the original mysterium sensor was actually manufactured by Canon or Nikon.

Oh and here's a direct quote from that article...

"... Also, reading a Dynamic Range chart is NOT an exact science. Because there is rarely a sudden clip in blacks and the highlights, you have to guesstimate at which point in the highlights or shadows the recorded image becomes unusable…and that is subject to personal bias and your own threshold of what is usable and what is not. Some people don’t mind a little extra noise in the shadows, others demand pristine near-blacks. It’s a very subjective thing interpreting these charts."

There you go... If you feel that the 7D is a 8 stop camera, light and shoot like it's an 8 stop camera. But if you like to tweak settings and use some decent glass, you're going to find that you can stretch the performance quite a bit.

[Rick Burnett]

And this article compares Alexa and Epic, NOT Epic and Canon or Epic and Nikon. Until I see proof that Canon has made such a sensor, they are behind the curve. Period. It really has NO RELEVANCE what they are capable of if they don't do it. You could say "Canon is capable of making a 20+ stop sensor" So what? If they don't make it, I can't use it, and so it's a pointless statement.

TSMC is one of the biggest fabs in the world yet that doesn't mean they can make a great microprocessor. Yet they fab them all the time. Because their part of the process has nothing to do with design, it's all a matter of chemical engineering at that point. There is a LOT more to engineering than just the process.

The Epic and Alexa are MUCH more expensive that the 7D, they should be better! But thinking you can get the same freedom in performance out of the 7D is just plain wrong. It's more than DR. It's rolling shutter, it's compression, it's moire/aliasing.

It's the same argument over and over. The point is, you can get great beautiful stuff on the 7D and even older cameras, that said, the conditions that you can get great images is greatly reduced. It's not about what the BEST thing you can produce on those cameras, it's about what option you have about what goes into the camera.

With a camera like the AF100, I will:

1. Be able to shoot scenes with a larger DR. I don't have to keep all the light in my scene within a smaller range if I want to protect maybe the light coming in a window that is a bit too much. Sure, I can bring this down or bring up the light, but in many cases this will save me extra time on set. I like the freedom of not having to fine tune EVERY scene because the DR is more limited IN MY OPINION. I say this because *I* make the determination of where usable DR is on my 7D and how much noise I have to bring up in the blacks.

2. I can pan my camera a little faster with tele lens, or, if I am moving, allow a little more movement in my motion without as huge a jello issue. This is a constant issue for more dynamic shots that I deal with all the time. I don't want to lock down every shot, and sometimes, I am shooting from a position or location where a stabilizer doesn't fit. There is IS but not all fast lenses have this.

3. It allows me to get a cleaner pull from greenscreen footage, especially with hair and such, with better compression. Macro blocking even with proper lighting on my 7D has made footage that is difficult to key. Lots of color contamination.

4. It allows me to shoot deeper focus without losing all my detail in the distance.

5. It allows me to shoot urban scenes without having distant brick buildings creating rainbow moire all over them. Or people wearing suits and having the suit moire dance all over the place. In addition, moire/aliasing causes huge problems with keying with fine hair detail and the color contamination Canon exhibits.

Sure, I can work around all these, I do every day, but these are ALL instances of where the sensor in my 7D by CANON is causing me to limit what I can shoot, or, increase the amount of work I need to do in my scene to push it into the cameras limitations. If I shoot everything shallow focus, check every fabric, lock the camera down, bring in the DR of the scene, sure, I'll get really nice shots.

It's not just tweaking the settings of the camera, or changing my glass, my glass looks great, it's the sensor that needs to be updated, and the system speed.

I want more freedom in my shooting.

[Rick Burnett]

And by the way, this is why I don't always trust my eyes:

http://www.scientificamerican.com/media/inline/blog/Image/CurveBallShapiroIC2009.swf

Because our eyes do some crazy things under certain conditions!

[ROne]

Thank you... that was the whole point I was trying to make. Red is an awesome company doing allot of innovative stuff... but in terms of pure sensor design... it's not something that Nikon or Canon aren't capable of or haven't seen before. In fact, I'm not 100 percent sure, but I could swear I remember hearing that the original mysterium sensor was actually manufactured by Canon or Nikon.

.

The thing is RED offers much more than just much higher quality than vDSLR - it offers Cinema Style integration - it was designed to ween people away from film in terms of modularity, flexibility and lens systems. It emulates the film workflow as much as possible but in the digital domain.

That's why vDSLR can't even hold a candle here.

The AF100 is at least moving in the right direction, vDSLR are good stills cameras with iffy video quality as a bolt-on that looks good on vimeo.

Are we seriously still comparing vDSLR to RED?

[bwhitz]

And this article compares Alexa and Epic, NOT Epic and Canon or Epic and Nikon. Until I see proof that Canon has made such a sensor, they are behind the curve. Period. It really has NO RELEVANCE what they are capable of if they don't do it. You could say "Canon is capable of making a 20+ stop sensor" So what? If they don't make it, I can't use it, and so it's a pointless statement.

I was mainly pointing to the article for the points it was making about dynamic range being tricky to measure and how allot of it is subjective...

BTW, here's some proof... http://www.technewsdaily.com/canon-giant-new-image-sensor-photography-cmos-1184/



1. Be able to shoot scenes with a larger DR. I don't have to keep all the light in my scene within a smaller range if I want to protect maybe the light coming in a window that is a bit too much. Sure, I can bring this down or bring up the light, but in many cases this will save me extra time on set. I like the freedom of not having to fine tune EVERY scene because the DR is more limited IN MY OPINION. I say this because *I* make the determination of where usable DR is on my 7D and how much noise I have to bring up in the blacks.


All signs seem to point to this... But we simply don't know this for a fact yet. I really hope it's better... I'm not a fan-boy of anything. I really could give a shi* what manufacturer makes what. I just give credit where credit is due and if the blacks on the AF-100 look anything like the GH1... I'm going to have to take a pass. Blacks are handles extremely well on the 7D and 5D, despite compression. I really hope the DR is better on the AF-100, shi* I hope it's hitting 14 stops and gives the Epic a run for it's money, but it is a smaller sensor than APS-C and if Panny isn't using any of that gap-less Micro-lensing that Canon is... we could very well see a lower overall dynamic range. Though, the mid-tones should definaly be better on the AF-100.

Overall, yea, I think it's going to be a better cam though. Especially with that 60fps over-cranking in 1080p... I NEED that. I've got money set aside right now, but I'm going to wait until there is some real results.


The thing is RED offers much more than just much higher quality than vDSLR - it offers Cinema Style integration - it was designed to ween people away from film in terms of modularity, flexibility and lens systems. It emulates the film workflow as much as possible but in the digital domain.

Are we seriously still comparing vDSLR to RED?

Yea, that's the problem. The Cinema style work-flow... how can I eloquently put this? ...sucks balls. Giant camera systems are a paint in the ass. That's the main reason the DSLRs are being compared to Red. They give you the "look", not necessarily the quality, that Red and film have in a tiny package that all the sudden drops the cost of production by 100 fold. You can throw them on a steady-cam with no need for a $10,000 vest and arm system... you can use lighter less-expensive cranes... cheaper car mounts... they go in tighter spaces... the list goes on. Overall the DSLRs ARE a much better film-making camera. They just give the lower quality image at the moment, but still... it's an great image. That's why the Epic is going to rock the shit out of Alexa and all other cameras for that matter. You get a better image... plus it's small and modular like a SLR.

[Rick Burnett]

I was mainly pointing to the article for the points it was making about dynamic range being tricky to measure and how allot of it is subjective...

BTW, here's some proof... http://www.technewsdaily.com/canon-giant-new-image-sensor-photography-cmos-1184/



All signs seem to point to this... But we simply don't know this for a fact yet. I really hope it's better... I'm not a fan-boy of anything. I really could give a shi* what manufacturer makes what. I just give credit where credit is due and if the blacks on the AF-100 look anything like the GH1... I'm going to have to take a pass. Blacks are handles extremely well on the 7D and 5D, despite compression. I really hope the DR is better on the AF-100, shi* I hope it's hitting 14 stops and gives the Epic a run for it's money, but it is a smaller sensor than APS-C and if Panny isn't using any of that gap-less Micro-lensing that Canon is... we could very well see a lower overall dynamic range. Though, the mid-tones should definaly be better on the AF-100.

Overall, yea, I think it's going to be a better cam though. Especially with that 60fps over-cranking in 1080p... I NEED that. I've got money set aside right now, but I'm going to wait until there is some real results.

I agree with you that DR is subjective. It all comes down to how much noise you are willing to accept. But unfortunately, these chart tests are the closest thing for us to try and relate one cameras response to another. I think they are very useful for someone when they have a camera they know, like me and my 7D, and they want to see what kind of relationship they have to each other. But, as I am sure you know and many point out, DR is one aspect.

Btw, I have seen that sensor from Canon. I wonder what glass they are going to use on that!!! :) Until I see some test footage and tests, I'll just leave it as "interesting" in my mind. Don't forget Canon's 4k hair dryer prototype either. Again, "Interesting" but useless to me at the moment. I'll be excited when they get rid of their rainbows in their moire. (And I'd DEFINTELY upgrade my 7D to a newer 7D if they did this and use it as a B camera NO QUESTION).

[mcgeedigital]

Just to confuse things further, ASC DP Shane Hurlbut says he gets 12 stops out of the 5DII with his camera/post settings and choice of lens. No one has more experience with the 5DII than he does.

Of course Red has announced that they are getting 18 stops with their HDRX system that will be incorporated in all their new cameras.

Yeah, he also says that Master Primes "suck balls", so excuse me if I don't believe what he says.

[Homunculus]

Just to confuse things further, ASC DP Shane Hurlbut says he gets 12 stops out of the 5DII with his camera/post settings and choice of lens. No one has more experience with the 5DII than he does.

Of course Red has announced that they are getting 18 stops with their HDRX system that will be incorporated in all their new cameras.

I dunno Hurlbut has also stated some very 'interesting' things, like particular lenses increase his dynamic range by like 2 stops or more. Sounds fishy to me, he seems a tad eccentric with his claims of the 5d. He also claims to love 5D compression and what it does to skin tones, saying the compression gives it a film like quality. (WTF??)

[Homunculus]

Yeah, he also says that Master Primes "suck balls", so excuse me if I don't believe what he says.


yeah here's another amazing remark of his. master primes 'suck' but his special canon stills lenses give him 2-3 more stops of DR and the crappy DSLR compression is amazing and gives a film like quality to skin tones. This guy is just amazing. I don't know what canon is paying him but it better be something huge :thumbsup:

[Veteran2010Tor]

At the end of the day THIS camera PWNZ0RZ everything...

http://www.popsci.com/technology/article/2010-09/ultra-high-def-bbc-broadcast-16-times-sharper-hdtv

The end.
:grin::grin::grin::grin:

[bwhitz]

He also claims to love 5D compression and what it does to skin tones, saying the compression gives it a film like quality. (WTF??)

It does act as an instant "digital make-up" hiding blemishes and details that something like a Red One would show... film also hides these blemishes too, due to it's softer-than-digital nature. He could have meant it in that regard... or... he's part of that dern global Canon conspiracy.

[Rick Burnett]

I still think this is the best video camera EVER built!

http://www.retrothing.com/2005/11/fisherprice_pxl.html

:P

(And the same style as the AF100 if you move the handle from the bottom to the top)

[GRENCH]

I still think this is the best video camera EVER built!

http://www.retrothing.com/2005/11/fisherprice_pxl.html

:P

(And the same style as the AF100 if you move the handle from the bottom to the top)

This camera gave you the truest B&W image. My cousin and I used to make classic videos with it back in the late 80's early 90's.

Wickedly Lo-Fi !

[Homunculus]

ultra high def 8k? That ain't squat. Red is already planning and/or developing like a 28k resolution camera that will blow this thing out of the tub

[ROne]

Yea, that's the problem. The Cinema style work-flow... how can I eloquently put this? ...sucks balls. Giant camera systems are a paint in the ass. That's the main reason the DSLRs are being compared to Red. They give you the "look", not necessarily the quality, that Red and film have in a tiny package that all the sudden drops the cost of production by 100 fold. You can throw them on a steady-cam with no need for a $10,000 vest and arm system... you can use lighter less-expensive cranes... cheaper car mounts... they go in tighter spaces... the list goes on. Overall the DSLRs ARE a much better film-making camera. They just give the lower quality image at the moment, but still... it's an great image. That's why the Epic is going to rock the shi* out of Alexa and all other cameras for that matter. You get a better image... plus it's small and modular like a SLR.

None of this is a problem for 'crewed' films done properly.

[heavyG]

None of this is a problem for 'crewed' films done properly.

Amen to that!
Even when shooting web video I hire in the same crew members that I use on Big Budget TV commercials, there is no substitute for qualified technicians that know how to get things done. (and done properly)

[f64manray]

I dunno Hurlbut has also stated some very 'interesting' things, like particular lenses increase his dynamic range by like 2 stops or more. Sounds fishy to me, he seems a tad eccentric with his claims of the 5d. He also claims to love 5D compression and what it does to skin tones, saying the compression gives it a film like quality. (WTF??)

I just want to see "Acts of Valor" on the big screen. If it looks incredible, I'll give him the benefit of the doubt.

[bwhitz]

None of this is a problem for 'crewed' films done properly.

Just because it's 'not a problem' doesn't mean it can't be improved. Just because you have a crew doesn't guarantee a quality production. This is all part of that "Pro" fallacy... where people think that if they have a 'pro' camera and a 'pro' crew than their result must therefore be 'pro' as well... even if it sucks. "Well we did it all like the pros! It must be pro!" In fact I think it's been observed over the last 20 years of indie cinema that the 'ones who made it' did it differently and definitely didn't have 'proper film crews'. Innovation is the only way for an indie to succeed... To take the route that a big production crew would never think of taking. They simple cannot match the scale and numbers of a real film production, so why try? If you think that just because you crew up and 'look' and act professional that you're gonig to get acceptable results then you're heading down the wrong road. People watching your film don't care if you had catering and assistants... just like they wouldn't care if you had a dildo strapped to your head while performing oral sex on a dog... as long as the result is interesting.

[Osslund]

All talk about DR numbers with regards to DSLR cameras made me think on how I handle video from a DSLR. It's pretty much spot on exposure and colours and often only slight adjustments to the image.

This is the way I shoot, to make it look like final result in camera rather than doing heavy grading afterwards. I don't think any DSLR with the poor codec is a good choice for grading. Get it right in camera and it looks great. A hassle to work with these cameras but the final result is rewarding. So it does look great from a 7D or 5D mkII but I doubt it can touch real RAW or decent codec 10-bit when it comes to number crunching DR.

[Thomas Kist]

"None of this is a problem for 'crewed' films done properly"
Please keep in mind that from a producers point of view Crew is the biggest chunk out of the budget.
(unless Harrison Ford stars in it)
So offering a camera solution with less crew gets their attention.
On the other hand, having 20+ crewmembers staring at an actor, chewing gum, while texting at the same time and discussing
overtime rates, makes some Directors think it would be nice to shoot a movie on a small intimate set with few people.
So it gets their attention too.
There is more to movie making then camera's and image quality, as a DP you always have to keep that in the back of your mind.

[ROne]

Just because it's 'not a problem' doesn't mean it can't be improved. Just because you have a crew doesn't guarantee a quality production. This is all part of that "Pro" fallacy... where people think that if they have a 'pro' camera and a 'pro' crew than their result must therefore be 'pro' as well... even if it sucks. "Well we did it all like the pros! It must be pro!" In fact I think it's been observed over the last 20 years of indie cinema that the 'ones who made it' did it differently and definitely didn't have 'proper film crews'. Innovation is the only way for an indie to succeed... To take the route that a big production crew would never think of taking. They simple cannot match the scale and numbers of a real film production, so why try? If you think that just because you crew up and 'look' and act professional that you're gonig to get acceptable results then you're heading down the wrong road. People watching your film don't care if you had catering and assistants... just like they wouldn't care if you had a dildo strapped to your head while performing oral sex on a dog... as long as the result is interesting.

I dunno about all of this, 20 years experience has told me (starting off in the real no money end of things) that there is little point trying to emulate the big-boys without actually doing what they do. I chose my path which was to dig us out of the bottom end, invest, improve and become more professional that way I can earn more put more back in and so forth. I see many film-makers who set there sights on the indie side of things and then become so obsessessed with doing things on the cheap that I can't help but notice their films reflect it.

I'm far far from the high-end but I've always moved up aiming high and keeping professional and charging likewise. That attitude has took me from Hi8 to RED One. It's a progression in my eyes. Again personal philosophy.

[morgan_moore]

A couple of thoughts

The AF has a small sensor compared to the 5d or 7, they would have to have better technology to equal the low light performance, it is likely that they dont

DR is one thing but IMO there is a second element to usable DR, and that is the characteristics of the 'roll to clipped' it no good having a wide DR that then just stops making an image look like it has been cut out with scissors

IMO my 5 and 7 clip in a more attractive way than my EX1, and you cant mesure 'attractive' on a chart

The canons have more usable DR than the EX1 IMO whatever the charts say (not to mention the superior glass on the canons see below)

---

Hurlburts wide DR..

So a cheap lens flares all the time, creating a fog of grey across the image and clipping information from the bottom, slip on that $50k glass and the problem is less = more DR

And I include L glass as 'cheap' at that level, Im talking about a very subtle effect not a visible fog unless the lighting is very 'contra'

Less flare gets you more DR in practice - that how his expensive glass 'gives more DR'

---

My guess is that the 1ds and 5 will have better lowlight than the AF100

My guess is that the AF100 will beat the canons in 90% of other situations , maybe 99%

My guess is that there is a risk that the AF100 will not clip in an 'attractive' manner

[Homunculus]

morgan moore. i thought hurlbut uses L glass, or at least used it for his last 3 minutes short or whatever it's called so when he talks about his lenses giving more DR I would assume he's referring to L glass

[Noel Evans]

ultra high def 8k? That ain't squat. Red is already planning and/or developing like a 28k resolution camera that will blow this thing out of the tub

Actually its a bit more than squat. 8k is shaping up to be the standard in around 20's. NHK Japan are also working to Ultra High Vision 8k. That means cameras the shoot 8k will be in high demand as it moves to become the broadcast standard (in about 10 -15 years).

[morgan_moore]

morgan moore. i thought hurlbut uses L glass, or at least used it for his last 3 minutes short or whatever it's called so when he talks about his lenses giving more DR I would assume he's referring to L glass

http://www.hurlbutvisuals.com/blog/ look at the home page image him and the small redrock rig

he uses l glass when shooting for canon !

[TheReverend]

I know this is a bit off-topic, but 8k broadcast standard in tens years? Bandwidth limits and compression schemes are nowhere near that capable and won't be for sometime. I mean you've got to consider how long HD has been around and it's still not even close to 100% broadcast implemented (in US). I mean 8k is a great number to throw around, but really it's not going to have any place for 15-25 years outside the cinema world. Displays capable of this kind of res are, well, basically non-existent other than multiscreen formats, but even then not close at all to 8k. Acquisition is easier to get up to speed than distribution.

HD is the public standard and it most likely will be for a long time. If anything, we'll probly see a 1080p120/3D standard sooner than anything else.

[Barry_Green]

The AF has a small sensor compared to the 5d or 7
Compared to the 5D, yes, all cameras have a small sensor.
Compared to the 7D, no. The sensors are nearly the same size, the effective crop factor is 1.18. Not exactly the same, but very close, and certainly close enough that it's unfair to say one has a "small sensor" compared to the other.

[Darren Levine]

Jeez this turned into a DR fest then rez fest. all i can think of reading through all this is why? Why debate what isn't in stone yet? and why oh why talk about 8k?....

But really, DSLR's don't have real curve profile abilities, the af100 will, i'll bet any of my cameras that the af100 will be less contrasty than the 5/7d. and does anyone here really need 8k? or 4k? or ANYTHING over 1080? if resolution is the most important thing to you, i'd never want to watch your films

It looks like DSLR fanatics are cradling their cameras and defending them against anything that could 'possibly' be better

[Rick Burnett]

Actually, yes, I'd want 3k or 4k if I could get it on something APS-C sized. The reason being, going higher means when I pull it down to 1080p, I will eliminate even more aliasing and actually get closer to 1080p resolution since with a bayer pattern, which most of these cameras use, you can only get something like between 70 and 80% resolution. I'd be okay with it even getting downsampled in camera to 1080p.

I worked with some Red footage and it was really nice for VFX work. Working with my 7D footage has been PAINFUL. I am of course getting better at knowing how to work around the 7D and make my life easier.

[Michael Olsen]

4K is coming, and coming fast, at least to theaters.

And, as was stated above, broadcasters have been laying the roadmaps for 8K broadcast.

And considering that some compression schemes can deliver almost flawless 4K motion with 5.1 surround sound at 15Mbps - that's Megabits...and they have been publicly demonstrated...

[morgan_moore]

Compared to the 5D, yes, all cameras have a small sensor.
Compared to the 7D, no. The sensors are nearly the same size, the effective crop factor is 1.18. Not exactly the same, but very close, and certainly close enough that it's unfair to say one has a "small sensor" compared to the other.

My quote was .. "My guess is that the 1ds and 5 will have better lowlight than the AF100"

No mention of the 7,

S

[Danielvilliers]

4K is coming, and coming fast, at least to theaters.

And, as was stated above, broadcasters have been laying the roadmaps for 8K broadcast.

And considering that some compression schemes can deliver almost flawless 4K motion with 5.1 surround sound at 15Mbps - that's Megabits...and they have been publicly demonstrated...

Are you sure, it has taken a while for theaters to get into digital 2k and many place are still film base. Now many are doing the transition to 3d, I don' t see the movement to 4k in any near future.

[dcloud]

if the af100 performs well at 1600 ISO, it would have similar low light performance to a 5d mk II