October 25, 2014, 12:23:53 AM

Author Topic: 5D Mark III & RAW Video, A Case Study  (Read 13072 times)

Drizzt321

  • 1D X
  • *******
  • Posts: 1674
    • View Profile
    • Aaron Baff Photography
Re: 5D Mark III & RAW Video, A Case Study
« Reply #15 on: May 30, 2013, 01:47:01 PM »
Ah, I see, thanks for finding it :)

Yea, sounds like Canon is doing some crazy tricks. It's funny, Canon engineers clearly could have made a better/simpler output pipeline I'm sure with more or less the hardware that's in there, but I'll bet much of it was marketing & feature segmentation decisions to avoid that. Yet ML still has managed to coerce the camera into giving us this crazy high quality output.
5D mark 2, 5D mark 3, EF 17-40mm f/4L,  EF 24-105mm f/4L IS USM, EF 135mm f/2L, EF 85mm f/1.8
Film Cameras: Mamiya RB67, RB-50, RB-180-C, RB-90-C, RB-50, Perkeo I folder, Mamiya Six Folder (Pre-WWII model)
http://www.aaronbaff.com

canon rumors FORUM

Re: 5D Mark III & RAW Video, A Case Study
« Reply #15 on: May 30, 2013, 01:47:01 PM »

ddashti

  • Rebel T5i
  • ****
  • Posts: 143
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #16 on: May 30, 2013, 02:50:30 PM »
The ML raw is significantly sharper! Those guys are brilliant.

LetTheRightLensIn

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 3937
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #17 on: May 30, 2013, 05:33:57 PM »
Hmm... I prefer the Canon one over the Magic Lantern one.

Give it a real try and you will NOT say that.

LetTheRightLensIn

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 3937
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #18 on: May 30, 2013, 05:37:42 PM »
The difference is amazing, especially at high ISO.

What is the H.264 doing to the image to destroy the contrast and brilliance of the color?

It is not so much the h.264 they use since even going HDMI clean out to Ninja 2 it still loses a lot of micro-contrast and color. Something they do in the image processing chain after it gets fed raw to liveview and before it hits the h.264 engine is knocking it down hard.

Drizzt321

  • 1D X
  • *******
  • Posts: 1674
    • View Profile
    • Aaron Baff Photography
Re: 5D Mark III & RAW Video, A Case Study
« Reply #19 on: May 30, 2013, 07:05:17 PM »
The difference is amazing, especially at high ISO.

What is the H.264 doing to the image to destroy the contrast and brilliance of the color?

It is not so much the h.264 they use since even going HDMI clean out to Ninja 2 it still loses a lot of micro-contrast and color. Something they do in the image processing chain after it gets fed raw to liveview and before it hits the h.264 engine is knocking it down hard.

If you look at what bchernicoff posted on the other page, it looks like they are doing some weird stuff before the LiveView/HDMI buffer. See http://www.eoshd.com/content/10250/canon-5d-2k-raw-feed-update-1920x720-possible-on-1000x-card.
5D mark 2, 5D mark 3, EF 17-40mm f/4L,  EF 24-105mm f/4L IS USM, EF 135mm f/2L, EF 85mm f/1.8
Film Cameras: Mamiya RB67, RB-50, RB-180-C, RB-90-C, RB-50, Perkeo I folder, Mamiya Six Folder (Pre-WWII model)
http://www.aaronbaff.com

syder

  • Rebel T5i
  • ****
  • Posts: 106
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #20 on: May 30, 2013, 08:20:13 PM »
Hmm... I prefer the Canon one over the Magic Lantern one.

Give it a real try and you will NOT say that.

It depends what you mean by a real try... On a typical day I might shoot 150 clips. Processing each one of those using the current methods for getting raw video out the 5dm3 would be 1) a massive pain in the ass 2) so time consuming that it isn't worthwhile for anyone actually making professional work.

At the moment it's amazing for hobbyists (or other people for whom time doesn't equal money) - but unusable for anyone who makes videos for a living.


RGomezPhotos

  • Canon 6D
  • *****
  • Posts: 330
    • View Profile
    • Ricardo Gomez Photography
Re: 5D Mark III & RAW Video, A Case Study
« Reply #21 on: May 30, 2013, 08:38:05 PM »
Yes..  I was thinking the same thing....  This is great for hobbyist...  But pros really can't us it as effectively as 3rd party devices to control the camera won't work with it.
EOS 5D MKII & 50D, Zeiss 50mm f1.4
www.ricardogomezphotography.com

canon rumors FORUM

Re: 5D Mark III & RAW Video, A Case Study
« Reply #21 on: May 30, 2013, 08:38:05 PM »

cayenne

  • 1D X
  • *******
  • Posts: 1213
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #22 on: May 30, 2013, 08:43:27 PM »
Hmm... I prefer the Canon one over the Magic Lantern one.

Give it a real try and you will NOT say that.

It depends what you mean by a real try... On a typical day I might shoot 150 clips. Processing each one of those using the current methods for getting raw video out the 5dm3 would be 1) a massive pain in the ass 2) so time consuming that it isn't worthwhile for anyone actually making professional work.

At the moment it's amazing for hobbyists (or other people for whom time doesn't equal money) - but unusable for anyone who makes videos for a living.

How much time do you currently spend, on footage from you shoots in post? I'm talking total time from editing, to color correction to color grading....sound...etc?

Just curious, I mean, for most people it isnt' like they shoot, and BAM, have a finished product out the door in 1-2 hours later.

I usually take a good bit of time auditioning takes, sync'ing or dubbing sound, layering on effects, titles...hell, just figuring out the music for things takes time, etc.

I'm guessing when the ML guys get this stuff ready for the general public (remember, it ain't even release to the general public as alpha yet really), I'm guessing they'll have the tools to prep it for whatever entry into the usual workflow worked out pretty straightforward.

Sure, you're gonna need a bit of horsepower to do the early processing of that RAW footage, but you need that with any camera that is currently outputting RAW video...so, I shan't think its gonna be as much of a PITA as you seem to be describing it when this stuff from the ML boys is ready for prime time.

And for some of us, it isn't just turning and burning for the sake of getting something out the door. If it is another hour or so to get super high quality footage with an incredible amount of dynamic range allowing for major color correction/grading in post...then it is definitely worth it.

Quality over Quantity.....it strikes again what balance you are gonnna go for.

C

AAPhotog

  • Rebel T5i
  • ****
  • Posts: 119
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #23 on: May 30, 2013, 08:46:05 PM »
Hmm... I prefer the Canon one over the Magic Lantern one.

Give it a real try and you will NOT say that.

It depends what you mean by a real try... On a typical day I might shoot 150 clips. Processing each one of those using the current methods for getting raw video out the 5dm3 would be 1) a massive pain in the ass 2) so time consuming that it isn't worthwhile for anyone actually making professional work.

At the moment it's amazing for hobbyists (or other people for whom time doesn't equal money) - but unusable for anyone who makes videos for a living.
There are professionals who shoot RAW and make a living doing so. Whats the difference between the other cameras RAW and the 5d3's?

The ONE step of dragging your RAW clips to raw2dng???
ok.
iPhone 4s camera

dirtcastle

  • Canon 6D
  • *****
  • Posts: 391
    • View Profile
    • Eric Nord Flickr Page
Re: 5D Mark III & RAW Video, A Case Study
« Reply #24 on: May 30, 2013, 09:02:33 PM »
[smell of coffee]

Pros are already using this.

dilbert

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 3095
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #25 on: May 30, 2013, 09:50:20 PM »
Yea, sounds like Canon is doing some crazy tricks. It's funny, Canon engineers clearly could have made a better/simpler output pipeline I'm sure with more or less the hardware that's in there, but I'll bet much of it was marketing & feature segmentation decisions to avoid that. Yet ML still has managed to coerce the camera into giving us this crazy high quality output.

Or maybe Canon decided to just reuse whatever work they had done for the 5D2 in order to shorten the time to market for the 5D3 by cutting out extra software R&D?

At this point in time, a very small number of CF cards work with this feature. Had Canon of brought this out in the mainstream model, it is highly likely that they would have received a large number of complaints/returns because people would expect it to work with all CF cards.

Mantanuska

  • PowerShot G1 X II
  • ***
  • Posts: 48
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #26 on: May 30, 2013, 09:54:32 PM »
they look the same except for contrast... pretty hard to compare them that way . if they wanted to show a valid comparison they should have processed the RAW to match the h.264 in terms of contrast and saturation

dirtcastle

  • Canon 6D
  • *****
  • Posts: 391
    • View Profile
    • Eric Nord Flickr Page
Re: 5D Mark III & RAW Video, A Case Study
« Reply #27 on: May 30, 2013, 10:36:52 PM »
Yea, sounds like Canon is doing some crazy tricks. It's funny, Canon engineers clearly could have made a better/simpler output pipeline I'm sure with more or less the hardware that's in there, but I'll bet much of it was marketing & feature segmentation decisions to avoid that. Yet ML still has managed to coerce the camera into giving us this crazy high quality output.

Or maybe Canon decided to just reuse whatever work they had done for the 5D2 in order to shorten the time to market for the 5D3 by cutting out extra software R&D?

At this point in time, a very small number of CF cards work with this feature. Had Canon of brought this out in the mainstream model, it is highly likely that they would have received a large number of complaints/returns because people would expect it to work with all CF cards.

Business is all about marginal return. The extra effort and headaches probably wouldn't have translated into proportionally larger sales for the 5D3. The 5D3 is already a winning product without RAW video. And the directly competing DSLRs (D800, for example) don't yet have it, so there was no major competitive advantage to including it.

RAW video seems revolutionary now, but in 3-4 years it will be standard for high-end DSLRs. It will be relatively easy to implement for smaller resolutions. Already, the cards are more than capable of handling the lower resolutions. And eventually 1920x1080 will be supported.

canon rumors FORUM

Re: 5D Mark III & RAW Video, A Case Study
« Reply #27 on: May 30, 2013, 10:36:52 PM »

LetTheRightLensIn

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 3937
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #28 on: May 30, 2013, 10:38:48 PM »
Hmm... I prefer the Canon one over the Magic Lantern one.

Give it a real try and you will NOT say that.

It depends what you mean by a real try... On a typical day I might shoot 150 clips. Processing each one of those using the current methods for getting raw video out the 5dm3 would be 1) a massive pain in the ass 2) so time consuming that it isn't worthwhile for anyone actually making professional work.

At the moment it's amazing for hobbyists (or other people for whom time doesn't equal money) - but unusable for anyone who makes videos for a living.

Yeah but the other guy was talking about output quality not workflow.

If you are really small or hobby then it is workable or if you are really big it is workable. If you are shooting like crazy for money and just barely getting by and barely able to keep up with even current workflow to get enough money in then it's probably a pretty rough or probably impossible way to go though yeah. But if you are in quite in that situation it should be possible, especially if you get some big, fast RAID setups going and higher-end PC.

I mean it is more of a pain, especially since even quick previews are so now (but already some attempts at rawtodng that also make quick preview files are being tested at which point you'd then be able to quickly scan stuff you shoot without all that much more extra time being needed). In some ways the RAW is so good it can make color development a breeze. But with slow HDs and an old computer it would take ages and yeah be a rough go for some, too rough in many cases.
« Last Edit: May 30, 2013, 10:44:49 PM by LetTheRightLensIn »

LetTheRightLensIn

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 3937
    • View Profile
Re: 5D Mark III & RAW Video, A Case Study
« Reply #29 on: May 30, 2013, 10:39:59 PM »
Yes..  I was thinking the same thing....  This is great for hobbyist...  But pros really can't us it as effectively as 3rd party devices to control the camera won't work with it.

Does it really prevent anything? For now it prevents full scale external monitoring until they make it work with 1.2.1, but people did have that all this time anyway.

canon rumors FORUM

Re: 5D Mark III & RAW Video, A Case Study
« Reply #29 on: May 30, 2013, 10:39:59 PM »