Why the downer on the 1D C?

[cowpunk52]

3) You can save the "at least". If Canon thought the sensor was up for more, they would have said it.
8 bits limit you to exactly 256 stops in all shapes and forms. 10 stops is what I like to have without sacrificing precision in mids. Read more carefully.

You can easily prove me (and the numbers) wrong. Considering that around 8 stops is what most cameras will chop and encode in 8 bits in order to deliver pretty pictures, you can underexpose a scene 4 stops and show me a pretty picture after bringing the image up and fixing contrast in post. THAT will mean the camera handles 12 stops "extremely well". Alternatively, just shoot a step chart and show me good tonal precision in the low stops.

I put in the "at least" because the dynamic range of the C300 may go fractionally above 12 stops to 12.5. I haven't shot a step chart because, well, I don't have time for that sort of thing, I'm too busy actually working. But I take seriously the evaluations of other professionals, and I'll quote from Alan Roberts' test report on the C300 here, my own emphasis added:

"During the capture process, the very lowest exposure capture produced an image which was barely distinguishable from the background on two LCD monitors, had disappeared on a third, and was only just visible on a Sony 32” Grade 1 CRT HD monitor. It was also only just distinguishable from the background on a waveform monitor. Since this exposure was 10 stops below that for peak white, we can state that the exposure range is between 9 and 10 stops using the normal gamma curve, but can be extended by a further 1.75 stops by using the recommended knee settings. Therefore the range is about 9.5 stops with normal gamma-correction, extending to 11.25 stops with optimised knee.

If, as Canon claim, the camera’s sweet spot is at ISO850/+2.5 where there is headroom of 800% (3 stops), this would stretch the available exposure range to 12.5 stops, but this is available only by using the Canon Log gamma-correction curve."

source: http://blog.creativevideo.co.uk/wp-content/uploads/2012/01/canon_c300_alan_roberts.pdf

 

[cowpunk52]

Judge for yourself. To my eye, I see the GH2 image resolving more detail than the C-300 did, especially when you factor in that the C-300 image was shot under contrasty sunlight ( high contrast gives the impression of more resolution than is actually there ) where the GH2 image was shot under flat overcast sky.

Canon C-300 : 400 Percent Blow-Up

Panasonic GH2 : 400 Percent Blow-Up

I really don't want to turn this thread into a C300 vs GH2 resolution debate, because I know how high of a pedestal people put the GH2 in the face of all evidence. So I'll just put this here (from PB's site and his own frame grabs):

C300: http://philipbloom.net/wp-content/uploads/2011/12/AA0002_01.jpg
GH2: http://philipbloom.net/wp-content/uploads/2011/12/Clip-3.jpg

Open them both up in different tabs and view them at full size. Pay attention to the signs and written text (which are not affected by hard angle light since they're two dimensional). Not even accounting for the significant frame advantage of the GH2 due to crop factor, the C300 outresolves it. If you don't want to believe it, then you're just kidding yourself. I'm not saying that the GH2 isn't exceedingly good at resolution, but the claim was that it outresolves the C300, which it doesn't - just like it doesn't outresolve a XF300, as is claimed despite contrary proof on 'another' forum.

 

[cowpunk52]

These discussions remind me of a few memorable quotes:

Lies, Damn Lies and Statistics! - Mark Twain
......


"He uses Statistics as a drunken man uses lamp posts... more for support than for illumination" - Andrew Lang

Ha ha ha - it's the life-blood of internet forums. Few people care about real-world performance here, just numbers and charts and spec sheets. I'm just glad that the actual production world is completely opposite!

 

[TheDingo]

I really don't want to turn this thread into a C300 vs GH2 resolution debate

I really look forward to watching the new Zacuto "Revenge Of The Great Camera Shootout 2012" which should give us a better idea of how well the current crop of cameras compare to each other. ( including the hacked GH2 vs C-300 ) I expect the GH-2 will score near the bottom of the bunch overall ( a $600 camera vs. a $16,000 camera isn't exactly an even match ), but maybe not when it comes to image resolution. The list of cameras tested is as follows:

1. Sony F65 -- Sony Representatives
2. ARRI Alexa -- Rodney Charters, ASC
3. RED Epic -- Ryan Walters
4. Sony FS100 -- Den Lennie & Mick Jones
5. Sony F3 w/SLOG -- Nancy Schreiber, ASC
6. Canon C300 -- Polly Morgan
7. Canon 7D with Technicolor settings -- Michael Negrin, ASC
8. Panasonic GH2 (hacked and non-hacked) -- Jonny Zeller & Colt Seman

Zacuto : Revenge Of The Great Camera Shootout 2012

 

[cpc]

I put in the "at least" because the dynamic range of the C300 may go fractionally above 12 stops to 12.5. I haven't shot a step chart because, well, I don't have time for that sort of thing, I'm too busy actually working. But I take seriously the evaluations of other professionals, and I'll quote from Alan Roberts' test report on the C300 here, my own emphasis added:

"During the capture process, the very lowest exposure capture produced an image which was barely distinguishable from the background on two LCD monitors, had disappeared on a third, and was only just visible on a Sony 32” Grade 1 CRT HD monitor. It was also only just distinguishable from the background on a waveform monitor. Since this exposure was 10 stops below that for peak white, we can state that the exposure range is between 9 and 10 stops using the normal gamma curve, but can be extended by a further 1.75 stops by using the recommended knee settings. Therefore the range is about 9.5 stops with normal gamma-correction, extending to 11.25 stops with optimised knee.

If, as Canon claim, the camera’s sweet spot is at ISO850/+2.5 where there is headroom of 800% (3 stops), this would stretch the available exposure range to 12.5 stops, but this is available only by using the Canon Log gamma-correction curve."

source: http://blog.creativevideo.co.uk/wp-content/uploads/2012/01/canon_c300_alan_roberts.pdf

Nice link. Thanks!
It's a bit unclear, but it seems Alan might be misinterpreting reference white for white clip in Canon specs which would add a bit on top of 12 stops. At least that's what you can assume from the way he measures the range (white card and iris change + nds; this is not amongst the more accurate ways to measure, but that's another story).
The 3 stops headroom is over reference white (90% white) as described in Canon's white paper. At the mentioned sweet spot of the camera, ref white to noise floor is 54 db (9 stops). And a total range of 72 db (12 stops) with the headroom (numbers cited by Canon).
That's not particularly interesting though. At least not as interesting as precision in the blacks.

These discussions remind me of a few memorable quotes:

Lies, Damn Lies and Statistics! - Mark Twain
......


"He uses Statistics as a drunken man uses lamp posts... more for support than for illumination" - Andrew Lang

I have some of this kung-fu under my belt too:

"Someone told me that each equation I included in my book would halve the sales." - Stephen Hawking, "A brief history of time"
"A poet is always timely. A great philosopher is an urgent need. There is no rush for Newton." - Tom Stoppard, "Arcadia"

Ha ha ha - it's the life-blood of internet forums. Few people care about real-world performance here, just numbers and charts and spec sheets. I'm just glad that the actual production world is completely opposite!

Aah, gotta love living in the black & white world of exclusive opposites, eh?

 

[squig]

Ha ha ha - it's the life-blood of internet forums. Few people care about real-world performance here, just numbers and charts and spec sheets. I'm just glad that the actual production world is completely opposite!

Don't you know anything? :shocked: The most popular camera is the best camera, end of story. Therefore the 1DC is clearly a piece of junk. :huh:

 

[squig]

I really look forward to watching the new Zacuto "Revenge Of The Great Camera Shootout 2012" which should give us a better idea of how well the current crop of cameras compare to each other. ( including the hacked GH2 vs C-300 ) I expect the GH-2 will score near the bottom of the bunch overall ( a $600 camera vs. a $16,000 camera isn't exactly an even match ), but maybe not when it comes to image resolution. The list of cameras tested is as follows:

1. Sony F65 -- Sony Representatives
2. ARRI Alexa -- Rodney Charters, ASC
3. RED Epic -- Ryan Walters
4. Sony FS100 -- Den Lennie & Mick Jones
5. Sony F3 w/SLOG -- Nancy Schreiber, ASC
6. Canon C300 -- Polly Morgan
7. Canon 7D with Technicolor settings -- Michael Negrin, ASC
8. Panasonic GH2 (hacked and non-hacked) -- Jonny Zeller & Colt Seman

Zacuto : Revenge Of The Great Camera Shootout 2012

Damn no MKIII....I was really looking forward to munching on popcorn and watching the MKIII blitz the competition.....it being the most popular camera (the camera you own being the most popular camera) and therefore the best.....ahh well GH2 winner! :grin:

 

[Emanuel]

I haven't shot a step chart because, well, I don't have time for that sort of thing, I'm too busy actually working.

LOL One of best quotes ever read on dvxuser

 

[Emanuel]

Judge for yourself. To my eye, I see the GH2 image resolving more detail than the C-300 did, especially when you factor in that the C-300 image was shot under contrasty sunlight ( high contrast gives the impression of more resolution than is actually there ) where the GH2 image was shot under flat overcast sky.

Canon C-300 : 400 Percent Blow-Up

Panasonic GH2 : 400 Percent Blow-Up

I bet Scorsese should be addressed to some one of most interesting videoish cameras ever tested under the bang for the buck method. Who knows if he wouldn't re-think his support to C300 or re-shoot Hugo (shot on Alexa) again after that?! LOL ;-)

 

[Jim Martin]

Go Cowpunk!!!!!!!

Jim Martin
Filmtools.com

 

[Simon Shasha]

How do you guys think Canon will be in regards to offering improvements via firmware updates for the 1DC? Do you think they will constantly try and improve the power of the camera the way RED does via their firmware updates? Or do you think they will treat it more like the EOS stills line with firmware updates appearing only every so often?

I wonder if they could do 4K 48FPS @ 250Mbit? Or 4K 72FPS @ 166Mbit? I think this camera will sell a lot more if there is potential for "bonus" features via free firmware updates, like how it is with RED's line of cameras.

If the 1DC's street-price comes in at $10K or under, and there is potential for at least 48FPS @ 250Mbit, or something like that, I'd definitely be in.

 

[Amr Rahmy]

the red company releases there cameras unfinished without the proper firmware and then uses the members as guinea testers. they don't "add" or "improve" the capabilities of the camera through free firmware upgrades per say. as a matter of fact canon actually released firmware upgrades for there hdslrs on request from some cinematographers.

and that's not how firmware works, if the camera hardware is not built for something, it can't be added through firmware later. it's not magical.

 

[Simon Shasha]

the red company releases there cameras unfinished without the proper firmware and then uses the members as guinea testers. they don't "add" or "improve" the capabilities of the camera through free firmware upgrades per say. as a matter of fact canon actually released firmware upgrades for there hdslrs on request from some cinematographers.


and that's not how firmware works, if the camera hardware is not built for something, it can't be added through firmware later. it's not magical.

Well, that wasn't condescending...


Anyway, RED does add and improve their cameras through free firmware upgrades (the most recent being the REDcolor3 and REDgamma3). That's a well known fact, and something RED prides themselves on.


Virtually no camera is released "finished" with "proper" firmware. The 5DMK3 is about to get it's second firmware update in as many weeks (http://www.canonrumors.com/2012/05/new-5d-mark-iii-firmware-before-the-end-of-may-cr2/).


And yes, that's how firmware works. If you actually read what I wrote, I said "4K 48FPS @ 250Mbit? Or 4K 72FPS @ 166Mbit" - I specifically used half the bitrate used at 4K 24FPS for 48FPS, and a third of the bitrate for 72FPS - the dual Digic 5, if capable of processing 24FPS @ 500Mbit, theoretically could do 48FPS @ 250Mbit - the obstacle, from my experience with electrical-engineering, will most likely be with cooling the processor efficiently enough to stop the camera from crashing/shutting down/melting the processor. Firmware is low-level software - it serves the very purpose of controlling and manipulating the hardware (e.g. GH2 hack, Magic Lantern for Canon DSLRs, Canon's adding of 24P/25P to the 5DMK2). How do you think Nvidia and ATi/AMD get more frame-rates out of their video-cards with every generation of firmware they release? - They find a more efficient way, via low-level firmware, for their hardware to process data.


Neither of us have seen what is under the hood of the 1DC, but given the information about the specs of the bitrate and dual Digic 5 processor, I wouldn't necessarily consider it ridiculous to spectuale double the frame-rate at the sacrifice of half the bitrate (or half the resolution even - i.e. RED Scarlet X crops the sensor to deliver more frames).


Anyways...I guess I believe in "magic"...

 

[Amr Rahmy]

the hardware needed to increase frame rates goes beyond the processing power needed.

and even at that doubling the frame rate and half-ing the bit-rate is not a balanced processing equation if that. and happens in a different part of the process.

and not mentioning the additional requirements in sensor and hardware to be able to achieve better frame rates.

actually half-ing the bit-rate alone may require more processing power not to have the image brake apart.

you are talking about magic that you do not understand.

 

[Kholi]

Damn no MKIII....I was really looking forward to munching on popcorn and watching the MKIII blitz the competition.....it being the most popular camera (the camera you own being the most popular camera) and therefore the best.....ahh well GH2 winner! :grin:

#26

xD It very well may be the winner.

But on a more serious note: "it that shall not be named" isn't out-resolving C300 by any measure. The aliasing etc may make it appear to be in the same league, it is not. It still looks pretty good, but again, it's not doing that kind of work.

I agree with CP, and at the same time am glad that CPC (lots of PC) is dropping information that's very interesting to read. Real world results are the best results, but some understanding of how things work is great. Neither should be bemoaned. Well, to me, anyway.

Just look at the response to "it that shall not be named", there's a WHOLE mess of flaws in that image but it's really turning heads every where the footage pops up. The responses I've heard coming out of the Zacuto revenge screening are kind of surprising, I imagine that a lot of people will be surprised when it hits the net as well.

 

[Graemsay]

I agree with what Amr says.

The bottleneck on the EOS 1D C with respect to frame rates could be the bandwidth between the sensor and the Digic processor, the downsamping / de-Bayering of the raw image data, compression into the JPEG format, or writing it to the CF card.

If you wanted to go from 24 fps to 48 fps then you'd need to double the maximum data rate between the sensor and processor. Unfortunately there are no shortcuts here, as it's an issue of pumping the raw data through the camera. I'd expect the processing of the image to take around the same time, but it's hard to say about the compression step. I don't know the workings of the JPEG algorithm, but would guess that more calculations would be required to optimise an image for a lower quality setting. However, memory IO is expensive, and a smaller file would reduce this.

The only part of the chain where a lower quality JPEG would lead to a performance boost would be actually writing the file onto the CF card.

With regards to the bitrate, you've got your numbers confused Simon. The 500 Mb/sec number is the amount of data that it can output. That works out at around 2.6 MB per exposure. If you wanted to support 48 fps then the bitrate could remain constant, with the size of each frame falling to 1.3 MB.

I suspect that Canon can up the framerate to 25 fps, particularly as there have been mutterings about its absence being an oversight, but they might not go a lot further than that. A lot depends on whether they've overspecced the system, and it's possible that it could be pushed further, but whether or not it would be stable is a different matter.

 

[derek7]

Does 1DC have continuous AF in video mode like the new cheap models from Canon (650D/70D/7D MKII)? or is the 1DC already obsolete?

 

[Michael Vera]

Does 1DC have continuous AF in video mode like the new cheap models from Canon (650D/70D/7D MKII)? or is the 1DC already obsolete?

At $10,000+ I think many pro users of the body (I say "many" and not "all", so as not to be dogmatic), won't consider using a first generation "autofocus" feature, but will likely surround the camera with a rig and follow focus, etc.

That sensor based continuos AF is on the perfect bodies for the right demographics (parents wanting to film kids at the beach, etc), I cant see it being a highly desired feature by the demographic the 1D-C is aimed at...

 

[deltoidjohn]

Does 1DC have continuous AF in video mode like the new cheap models from Canon (650D/70D/7D MKII)? or is the 1DC already obsolete?

Most proper Cinema cameras and lenses do not have AF - DP's on real film sets need to be able to control their focus selectively, and this is the job of the focus puller.

Also, you're talking about cameras that have not even been announced yet (apart from the 650D). How do you know these cameras will even come to market with continuous AF?

 

[camarade]

4k in a high-end DSLR body + a 1DX to go with it? I'm game... If I had the money. Also, from my understanding this beast is actually decent in low light... don't know many other 4k cameras that are...