ProRes RAW for UMP

The place for questions about shooting with Blackmagic Cameras.
  • Author
  • Message
Offline
User avatar

rick.lang

  • Posts: 8942
  • Joined: Wed Aug 22, 2012 5:41 pm
  • Location: Victoria BC Canada

ProRes RAW for UMP

PostSun Apr 08, 2018 8:58 pm

Andrew Kolakowski wrote:Bandwidth savings depend on what you compare it to. Average rate for ProRes RAW HQ is 15% above ProRes HQ (and about the same situation for normal mode). No saving here.
ProRes RAW HQ should save about 20% bandwidth compared to ProRes444 though...


You’re right there are two ways to look at the chart in the ProRes raw White Paper. With datarates for ProRes raw HQ positioned between ProRes 444 (12bit) and ProRes HQ (10bit), I was interpreting ProRes raw HQ as comparable in quality to ProRes 444. After all ProRes has to carry three colour values per pixel while ProRes raw has only one value per photosite. And BMD raw has always been 12bit log to record or unpacked 16bit linear in Resolve.

But Apple doesn’t actually say what it’s most comparable to. And there’s nothing to say that there will never be a ProRes raw XQ in the future, with datarates positioned between ProRes 444 XQ and ProRes 444.

Come to think of it, ProRes raw doesn’t even need to be 12bit Log, it could be 16bit linear. The White Paper seems flexible on that. Lots of unknowns to explain this week.





Sent from my iPad using Tapatalk
Rick Lang
Offline
User avatar

Cary Knoop

  • Posts: 688
  • Joined: Sun Mar 12, 2017 6:35 pm
  • Location: Newark, CA USA

Re: ProRes RAW for UMP

PostSun Apr 08, 2018 9:19 pm

rick.lang wrote:With datarates for ProRes raw HQ positioned between ProRes 444 (12bit) and ProRes HQ (10bit), I was interpreting ProRes raw HQ as comparable in quality to ProRes 444. After all ProRes has to carry three colour values per pixel while ProRes raw has only one value per photosite.

True, but for each 2x2 square the 4 RGB triplets carry less information than the 4 raw color values.
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostSun Apr 08, 2018 10:29 pm

rick.lang wrote:
Andrew Kolakowski wrote:Bandwidth savings depend on what you compare it to. Average rate for ProRes RAW HQ is 15% above ProRes HQ (and about the same situation for normal mode). No saving here.
ProRes RAW HQ should save about 20% bandwidth compared to ProRes444 though...


But Apple doesn’t actually say what it’s most comparable to. And there’s nothing to say that there will never be a ProRes raw XQ in the future, with datarates positioned between ProRes 444 XQ and ProRes 444.

Come to think of it, ProRes raw doesn’t even need to be 12bit Log, it could be 16bit linear. The White Paper seems flexible on that. Lots of unknowns to explain this week.

Sent from my iPad using Tapatalk


It's 12bit most likely because it's based on old engine which is 12bit max. In order to have 16bit support they would have to create new version.
Based on video with Atomos guy looks like ProRes RAW is targeted at non high-end market (which doesn't mean that it can't end up there) where 12bit is most likely enough. Nothing stops it to be extended to 14 or 16bits and it may happen if eg. ARRI or Sony will decide that they want to have ProRes RAW (they can push on Apple to have higher precision).

There can be even higher quality mode, but for target audience defined by ATOMOS guy there is really no need. At highest rate in HQ mode it pushes ratio close to 2:1 which is crazy high.
3:1 is already high quality used by RED or Sony F65- lower ratio is just waste of bandwidth/space for tiny quality improvements which most likely never going to be noticed in final video.
If you take into account that ProRes XQ is around 1:4.5 which already offers really high quality then you can predict that 3:1 should be easily good enough. Below 3:1 we are in range of mathematically lossless encoders.
Offline
User avatar

Robert Niessner

  • Posts: 1595
  • Joined: Thu Feb 21, 2013 9:51 am
  • Location: Graz, Austria

Re: ProRes RAW for UMP

PostMon Apr 09, 2018 12:31 am

I don't see why you should waste bandwidth storing into 16bit linear raw. Arri and BMD are using 12bit log raw:
viewtopic.php?f=2&t=35325#p208485

And about 14 bit log raw:
viewtopic.php?f=2&t=35325#p208493
Robert Niessner
LAUFBILDkommission
Graz / Austria
---------------------------------------------------------------------
Blackmagic Camera Blog (German):
http://laufbildkommission.wordpress.com

Read the blog in English via Google Translate:
http://tinyurl.com/pjf6a3m
Offline
User avatar

Cary Knoop

  • Posts: 688
  • Joined: Sun Mar 12, 2017 6:35 pm
  • Location: Newark, CA USA

Re: ProRes RAW for UMP

PostMon Apr 09, 2018 2:34 am

Andrew Kolakowski wrote:It's 12bit most likely because it's based on old engine which is 12bit max.

If they patched ProRes to carry RAW data I suppose they could use up to three 12 bit values to capture one single photosite. All they need in addition is some metadata on the sensor, ranges and gamma, but that is global and frame independent.
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostMon Apr 09, 2018 8:51 am

Robert Niessner wrote:I don't see why you should waste bandwidth storing into 16bit linear raw. Arri and BMD are using 12bit log raw:
viewtopic.php?f=2&t=35325#p208485

And about 14 bit log raw:
viewtopic.php?f=2&t=35325#p208493


If your camera doesn't use precise enough pipe than there is no point at all. LOG also helps, but in the future 12bit may become not enough to store newer sensors data. 12bit sounds like good enough for now, but in the same time it may become not enough quite quickly.
Last edited by Andrew Kolakowski on Mon Apr 09, 2018 8:56 am, edited 1 time in total.
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostMon Apr 09, 2018 8:54 am

Cary Knoop wrote:
Andrew Kolakowski wrote:It's 12bit most likely because it's based on old engine which is 12bit max.

If they patched ProRes to carry RAW data I suppose they could use up to three 12 bit values to capture one single photosite. All they need in addition is some metadata on the sensor, ranges and gamma, but that is global and frame independent.


Hmmm...you could do such a trick. Not sure if it's good idea though. Codec core could become bit messy.
Offline
User avatar

Robert Niessner

  • Posts: 1595
  • Joined: Thu Feb 21, 2013 9:51 am
  • Location: Graz, Austria

Re: ProRes RAW for UMP

PostMon Apr 09, 2018 12:33 pm

Andrew Kolakowski wrote:
Robert Niessner wrote:I don't see why you should waste bandwidth storing into 16bit linear raw. Arri and BMD are using 12bit log raw:
viewtopic.php?f=2&t=35325#p208485

And about 14 bit log raw:
viewtopic.php?f=2&t=35325#p208493


If your camera doesn't use precise enough pipe than there is no point at all. LOG also helps, but in the future 12bit may become not enough to store newer sensors data. 12bit sounds like good enough for now, but in the same time it may become not enough quite quickly.


I'd like to quote from someone working at ARRI directly:
Chase Hagen wrote:To answer Mr. PetePolyakov, I personally believe that in order to achieve the most benefit from 14-bit RAW files that the sensor would have to have an exceeding large number of output paths and the A/D converter would ideally have 16-bit precision (16-bit A/D) to convert the data to a 14-bit logarithmic data (14-bit linear contains a similar amount of data/information as 12-bit logarithmic). The processing power of the A/D converter would most likely come at an extreme price cost as the only 16-bit A/D converter equipped camera is the Sony F65 to my knowledge (RED does not provide this information). Also 14-bit RAW data would need considerable storage bandwidth and space compared to 12-bit especially at higher frame rates and would come at quite a cost.


Also a good read on the topic of log vs linear is David Newman essay (the developer of Cinemform):
http://cineform.blogspot.co.at/2007/09/ ... inear.html
Robert Niessner
LAUFBILDkommission
Graz / Austria
---------------------------------------------------------------------
Blackmagic Camera Blog (German):
http://laufbildkommission.wordpress.com

Read the blog in English via Google Translate:
http://tinyurl.com/pjf6a3m
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostMon Apr 09, 2018 12:48 pm

Yes, this is why most companies use LOG, but at some point 12bit LOG will start to be not enough.
There are some hints that ProRes RAW can store higher precision as well- not sure if it's already there or just a statement that it can (after e.g. new profile announcement).
Offline
User avatar

Rakesh Malik

  • Posts: 2182
  • Joined: Fri Dec 07, 2012 1:01 am
  • Location: Tacoma, WA

Re: ProRes RAW for UMP

PostMon Apr 09, 2018 2:21 pm

Red and Sony (in CineAlta) capture 16-bit linear, and generally not a lot of people regard either as particularly affordable. :)

The data rates are also the reason that both Red and Sony have put so much effort into developing compressed raw formats; without those the data rates would be untenable even for big productions, especially with 6K and 8K resolution.
Rakesh Malik
Cinematographer, photographer, adventurer, martial artist
http://WinterLight.studio
System:
HP Spectre x360 Kaby Lake-R w/ 1080Ti eGPU in RazerCore
Offline

Bruce Schultz

  • Posts: 69
  • Joined: Thu Aug 23, 2012 2:48 am

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 12:52 am

At the Atomos display center at NAB Las Vegas today (Monday April 9, 2018), I had a long talk with the fellow displaying the FCP-X implementation of Prores Raw and he was unable to demonstrate the ability to adjust WB, ISO, or anything prior to de-bayering which is done via a preset Lut or Log generated based on the original camera it was recorded on - Sony, Arri, Canon, Panasonic, BMagic were all listed.

Maybe someone has seen something different, but it looked to me like de-bayering to log or a custom lut based on the camera make was how it was initially processed to view the footage and work on an in at least this initial version of FCP-X today.
Offline
User avatar

rick.lang

  • Posts: 8942
  • Joined: Wed Aug 22, 2012 5:41 pm
  • Location: Victoria BC Canada

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 1:12 am

That’s in keeping with the early reviews. The camera profile is used almost like an ACES IDT. But no adjustment to those aspects that make raw so attractive. Here’s hoping Resolve does a better job of handling ProRes raw adjustments, at least when the ProRes raw is generated by a BMD camera. Thanks for that feedback.


Sent from my iPhone using Tapatalk
Rick Lang
Offline

Bruce Schultz

  • Posts: 69
  • Joined: Thu Aug 23, 2012 2:48 am

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 1:54 am

rick.lang wrote:That’s in keeping with the early reviews. The camera profile is used almost like an ACES IDT. But no adjustment to those aspects that make raw so attractive.


Yes, that's what I thought too - analogous to an ACES IDT implementation. Remembering that both FCP-X and Prores Raw are both Apple products, I'm not sure other vendors can fix or change this in their products if Apple can't - but we can hope . . .
Offline
User avatar

Jamie LeJeune

  • Posts: 915
  • Joined: Tue Feb 26, 2013 4:33 am
  • Location: San Francisco

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 8:16 am

After a long and informative day at NAB that started at the BMD Booth, included presentations by Apple + Atomos, and ended with a long chat over drinks with an Atomos engineer, here’s what I learned about ProResRAW:
1. Currently, Atomos devices are the only way to record ProRes RAW.
2. FCPX 10.4.1 is the only app that can currently play ProResRAW. The Resolve 15 beta does not support it (yet), despite reports saying otherwise. And there is no support yet in Adobe or Avid either, but that’s not a surprise.
3. FCPX does not allow you to change the WB and ISO metadata before the debayer. This has been 100% confirmed. ProResRAW currently does not offer the level of pre-debayer metadata control that you get with cDNG or .r3d or Arriraw.
4. Different cameras are sending out quite different raw signals. Panasonic’s raw signal over SDI is 12 bit log. Sony’s raw signal over SDI is 12 bit linear. Canon’s raw signal over SDI is 10 bit log. So, ProResRAW files recorded from Panasonic cameras offer the most code values in the crucially important midtones.
5. A 12 bit ProRes4444 log file and a ProResRAW HQ file from a 12 bit source offer equivalent levels of color quality and dynamic range in post.
www.cinedocs.com
http://www.imdb.com/name/nm4601572/
Offline

Vess Stoytchev

  • Posts: 36
  • Joined: Mon Nov 20, 2017 7:10 am

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 10:41 am

thanks for the info, Jamie LeJeune. :)
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 12:42 pm

Jamie LeJeune wrote:3. FCPX does not allow you to change the WB and ISO metadata before the debayer. This has been 100% confirmed. ProResRAW currently does not offer the level of pre-debayer metadata control that you get with cDNG or .r3d or Arriraw.


FCP X doesn't allow for such a adjustments, but if I have RAW data all what I have to do is:
- decode ProRes RAW to uncompressed RAW
- now I have uncompressed RAW data and I can do whatever I want with it including all adjustments
- send adjusted RAW to debayer process

Is there a problem with my thinking? Maybe ProRes RAW doesn't allow to be decode to RAW uncompressed, but strictly to already debayered version (RED seems to have such a lock but for different reasons). If this is the case than it's bad news.
Offline
User avatar

Cary Knoop

  • Posts: 688
  • Joined: Sun Mar 12, 2017 6:35 pm
  • Location: Newark, CA USA

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 2:11 pm

Andrew Kolakowski wrote:... if I have RAW data all what I have to do is:
- decode ProRes RAW to uncompressed RAW
- now I have uncompressed RAW data and I can do whatever I want with it including all adjustments
- send adjusted RAW to debayer process


Right.
Offline
User avatar

Dmitry Shijan

  • Posts: 400
  • Joined: Wed Sep 17, 2014 7:15 pm
  • Location: UA

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 2:48 pm

See about ProRes RAW future support In this video at 7:56
All my custom made accessories for BMMCC/BMMSC now available here https://lavky.com/radioproektor/
Offline
User avatar

Jamie LeJeune

  • Posts: 915
  • Joined: Tue Feb 26, 2013 4:33 am
  • Location: San Francisco

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 3:36 pm

Andrew Kolakowski wrote:
Jamie LeJeune wrote:3. FCPX does not allow you to change the WB and ISO metadata before the debayer. This has been 100% confirmed. ProResRAW currently does not offer the level of pre-debayer metadata control that you get with cDNG or .r3d or Arriraw.


FCP X doesn't allow for such a adjustments, but if I have RAW data all what I have to do is:
- decode ProRes RAW to uncompressed RAW
- now I have uncompressed RAW data and I can do whatever I want with it including all adjustments
- send adjusted RAW to debayer process

Is there a problem with my thinking? Maybe ProRes RAW doesn't allow to be decode to RAW uncompressed, but strictly to already debayered version (RED seems to have such a lock but for different reasons). If this is the case than it's bad news.


No one could tell me whether it’s a limitation of the codec or instead that Apple just decided in FCPX it was better to make it simple and limit the interface. Someday it will be supported by other applications and we’ll get to see whether they offer more control.
www.cinedocs.com
http://www.imdb.com/name/nm4601572/
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 4:49 pm

It's probably just how Apple implemented it in FCP X, but there is no 100% guarantee. I hope you have access to RAW uncompressed data after decoding, so developers can introduce adjustment on RAW.
Offline
User avatar

rick.lang

  • Posts: 8942
  • Joined: Wed Aug 22, 2012 5:41 pm
  • Location: Victoria BC Canada

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 4:57 pm

My money is on the latter. Resolve 15 will eventually do it right. There’s a reason for Resolve to distinguish their ProRes raw support since they are essentially a colour grading app at the core whereas FCP X is an editing app at the core.

DJI is supporting ProRes raw as well as Atomos. Funny the engineer didn’t mention that. I think it’s the Zenmuse 7 that will record ProRes raw.

Thanks for the update from NAB 2018!

By the way, if you have time, would you visit SLR Magic at booth C2663? They are showing the unrealeased SLR Magic APO 32mm prime and their four new Micro Prime high speed lenses for MFT mounts. I’d be very interested in your opinions. I haven’t been able to find any of the media doing a review of those products yet. Thanks again, Jamie.
Rick Lang
Offline
User avatar

Cary Knoop

  • Posts: 688
  • Joined: Sun Mar 12, 2017 6:35 pm
  • Location: Newark, CA USA

Re: ProRes RAW for UMP

PostTue Apr 10, 2018 11:51 pm

Jamie LeJeune wrote:No one could tell me whether it’s a limitation of the codec or instead that Apple just decided in FCPX it was better to make it simple and limit the interface.

If you can't change the white balance and ISO it is not RAW or otherwise I think it is pointless marketing.
Offline

Mihail Moskov

  • Posts: 37
  • Joined: Wed Jun 24, 2015 11:18 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 9:43 am

Andrew Kolakowski wrote:
Jamie LeJeune wrote:3. FCPX does not allow you to change the WB and ISO metadata before the debayer. This has been 100% confirmed. ProResRAW currently does not offer the level of pre-debayer metadata control that you get with cDNG or .r3d or Arriraw.


FCP X doesn't allow for such a adjustments, but if I have RAW data all what I have to do is:
- decode ProRes RAW to uncompressed RAW
- now I have uncompressed RAW data and I can do whatever I want with it including all adjustments
- send adjusted RAW to debayer process

Is there a problem with my thinking? Maybe ProRes RAW doesn't allow to be decode to RAW uncompressed, but strictly to already debayered version (RED seems to have such a lock but for different reasons). If this is the case than it's bad news.


It must be decompressed before debayering so there is definitely a stage where uncompressed raw is present (in memory).
decompress -> linearize -> debayer -> transform to a target gamut and gamma
is a typical pipeline


Cary Knoop wrote:If you can't change the white balance and ISO it is not RAW or otherwise I think it is pointless marketing.


The only thing "raw" implies in the video world is Bayer data. There are certain benefits in baking exposure adjustments (baking ISO) and white balance in the Bayer image, particularly at lower bit depths.
www.shutterangle.com
The science & magic of shooting moving pictures

www.slimraw.com
A fast CinemaDNG compressor
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 10:03 am

Mihail Moskov wrote:
It must be decompressed before debayering so there is definitely a stage where uncompressed raw is present (in memory).
decompress -> linearize -> debayer -> transform to a target gamut and gamma
is a typical pipeline


Yes, but for example RED didn't (doesn't?) allow you to access uncompressed RAW data in their SDK, which is the only way to read R3D due to encryption. I think (although have no idea for sure) that if Apple wanted they could lock access to uncompressed RAW data. We definitely don't want this to be true :)
Offline
User avatar

Cary Knoop

  • Posts: 688
  • Joined: Sun Mar 12, 2017 6:35 pm
  • Location: Newark, CA USA

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 2:23 pm

Mihail Moskov wrote:There are certain benefits in baking exposure adjustments (baking ISO) and white balance in the Bayer image, particularly at lower bit depths.

This does not make any sense to me, there is nothing to "bake in" if the data is raw.
Offline

roger.magnusson

  • Posts: 474
  • Joined: Wed Sep 23, 2015 4:58 pm
  • Location: Sweden

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 2:37 pm

Just tried a ProRes RAW HQ 4K DCI sample in FCPX and at least it's efficiently implemented. Even a low end 2014 MacBook Pro without dedicated GPU had no performance issues at all with multiple tracks (didn't try more than two).
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 2:51 pm

Where did you get it from?
Can you post small sample file?

update:

Found some here: https://mega.nz/#F!Nd0XxJzC!hlfXGYfIDwiw_UiwB3ZS-w
Offline

roger.magnusson

  • Posts: 474
  • Joined: Wed Sep 23, 2015 4:58 pm
  • Location: Sweden

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 2:56 pm

Yes, those are the ones I tested.
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 3:42 pm

Do they also open in QT X (or QT 7) or only FCP X?
Last edited by Andrew Kolakowski on Wed Apr 11, 2018 4:54 pm, edited 1 time in total.
Offline

Mihail Moskov

  • Posts: 37
  • Joined: Wed Jun 24, 2015 11:18 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 4:31 pm

Cary Knoop wrote:
Mihail Moskov wrote:There are certain benefits in baking exposure adjustments (baking ISO) and white balance in the Bayer image, particularly at lower bit depths.

This does not make any sense to me, there is nothing to "bake in" if the data is raw.


Well, raw isn't raw data. Arri raw isn't raw data. BM raw isn't raw data. Red raw isn't raw data.
Various processing steps go between digitized sensor values and what goes into a raw image. Processing steps that often make it into the "raw" image include tonal compression, denoising, hot pixel removal, various forms of massaging like applying illumination and black calibration, lossy compression, etc.

"Baking" ISO and WB into the Bayer data is something that, for example, Canon's been doing to their cinema raw since they offer raw. This basically improves quality (more efficient use of coding space = better effective density) and is easily reversed and revised in post.
www.shutterangle.com
The science & magic of shooting moving pictures

www.slimraw.com
A fast CinemaDNG compressor
Offline

roger.magnusson

  • Posts: 474
  • Joined: Wed Sep 23, 2015 4:58 pm
  • Location: Sweden

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 5:14 pm

Andrew Kolakowski wrote:Do they also open in QT X (or QT 7) or only FCP X?

I tested all three in High Sierra, only FCPX can play the files.
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 5:23 pm

Ok- thank you.
Offline

roger.magnusson

  • Posts: 474
  • Joined: Wed Sep 23, 2015 4:58 pm
  • Location: Sweden

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 5:27 pm

Motion works as well. But not Compressor, it's like they don't want you to be able to batch convert these files. ;)
Offline

Bruce Schultz

  • Posts: 69
  • Joined: Thu Aug 23, 2012 2:48 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 5:29 pm

I haven't found that a "baked" ISO/WB/Exposure from Canon, Sony, Panasonic RAW is easily reversed in post. Baked is baked, only Red R3D is truly fluid and changeable in all those parameters in my post experiences.
Offline

Mihail Moskov

  • Posts: 37
  • Joined: Wed Jun 24, 2015 11:18 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 5:58 pm

Bruce Schultz wrote:I haven't found that a "baked" ISO/WB/Exposure from Canon, Sony, Panasonic RAW is easily reversed in post. Baked is baked, only Red R3D is truly fluid and changeable in all those parameters in my post experiences.


"Baked ISO" means that the linear raw values are multiplied uniformly before tonal compression. Note that it is a much simpler operation than tonal compression itself (which all BM, Canon and Arri cameras do, for example). It is easily reversible -- you simply divide the linear values after tonal expansion (linearization). Same for WB -- it is very similar to ISO except that in native camera space it does per channel multiplication (still a much simpler operation than tonal compression). In contrast, tonal compression will lose precision in the highlights, which is irreversible. Yet, people will object to the benign operation, and won't care for the lossy operation. (And I am not saying that tonal compression is bad, on the contrary -- it is very useful. But it is lossy.)

In post, the Gain control in a color corrector will manipulate both "ISO"/exposure and WB when working on a linear image. You don't need specific controls to adjust both of these in post, as long as you have a linear image (which doesn't mean that a good raw processor can't provide explicit controls, it certainly can).
www.shutterangle.com
The science & magic of shooting moving pictures

www.slimraw.com
A fast CinemaDNG compressor
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostWed Apr 11, 2018 7:05 pm

We can expect more ProRes RAW flavours:

<array>
<string>aprx</string>
<string>aprh</string>
<string>aprn</string>
<string>aprs</string>
<string>apro</string>

Current files are aprh and aprn (HQ and Normal).
Offline
User avatar

Jamie LeJeune

  • Posts: 915
  • Joined: Tue Feb 26, 2013 4:33 am
  • Location: San Francisco

Re: ProRes RAW for UMP

PostThu Apr 12, 2018 8:40 pm

Mihail Moskov wrote:In post, the Gain control in a color corrector will manipulate both "ISO"/exposure and WB when working on a linear image. You don't need specific controls to adjust both of these in post, as long as you have a linear image (which doesn't mean that a good raw processor can't provide explicit controls, it certainly can).

Great info. Thank you.
According to the Atomos engineer that I spoke with, while Sony is sending out linear, Canon and Panasonic are actually sending log out the SDI. 10bit in the case of Canon, and 12bit for Panasonic.
www.cinedocs.com
http://www.imdb.com/name/nm4601572/
Offline
User avatar

Jamie LeJeune

  • Posts: 915
  • Joined: Tue Feb 26, 2013 4:33 am
  • Location: San Francisco

Re: ProRes RAW for UMP

PostThu Apr 12, 2018 8:51 pm

rick.lang wrote:By the way, if you have time, would you visit SLR Magic at booth C2663? They are showing the unrealeased SLR Magic APO 32mm prime and their four new Micro Prime high speed lenses for MFT mounts. I’d be very interested in your opinions. I haven’t been able to find any of the media doing a review of those products yet. Thanks again, Jamie.

Hey Rick, I was only at NAB for a day and a half before I had to head back to SF for a gig, so I never did make it to the SLR Magic booth. A bummer as I wanted to check out those lenses as well.
Although a video hasn't dropped yet, perhaps Newsshooter or Cinema5D got something that will be released once they have a chance to post it. I saw them all over the show floor capturing tons of stuff that I haven't seen them put up yet.
www.cinedocs.com
http://www.imdb.com/name/nm4601572/
Offline
User avatar

rick.lang

  • Posts: 8942
  • Joined: Wed Aug 22, 2012 5:41 pm
  • Location: Victoria BC Canada

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 12:14 am

Jamie, no problem and hope you had a blast at the show. Like you, I’ll check out videos possibly on the weekend. Based on previous NAB shows it can take up to a couple of weeks to post all the news since there’s so much to cover. The big guys get covered fast, especially BMD since they always have something interesting and sometimes a real game changer.


Sent from my iPhone using Tapatalk
Rick Lang
Offline

Wayne Steven

  • Posts: 1113
  • Joined: Thu Aug 01, 2013 3:58 am

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 1:16 am

So, what has happened with this raw on the micro studio. ProRes on micro studio, maybe new micro studio?
Offline

Mihail Moskov

  • Posts: 37
  • Joined: Wed Jun 24, 2015 11:18 am

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 11:19 am

Jamie LeJeune wrote:
Mihail Moskov wrote:In post, the Gain control in a color corrector will manipulate both "ISO"/exposure and WB when working on a linear image. You don't need specific controls to adjust both of these in post, as long as you have a linear image (which doesn't mean that a good raw processor can't provide explicit controls, it certainly can).

Great info. Thank you.
According to the Atomos engineer that I spoke with, while Sony is sending out linear, Canon and Panasonic are actually sending log out the SDI. 10bit in the case of Canon, and 12bit for Panasonic.


Yes, "log" stands for tonal compression. It is used for more efficient storage. The log image gets linearized during raw processing, before ending in the target gamut and gamma. It can always be linearized back with the appropriate gamma transform.
Panasonic EVA1 dng raw is actually 10-bit log. Or at least that's what Atomos records. Varicam LT raw can be both 12- or 10-bit.
www.shutterangle.com
The science & magic of shooting moving pictures

www.slimraw.com
A fast CinemaDNG compressor
Offline

spitandspite

  • Posts: 6
  • Joined: Fri Mar 23, 2018 5:32 pm
  • Real Name: Abel Salazar

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 12:59 pm

Wayne Steven wrote:So, what has happened with this raw on the micro studio. ProRes on micro studio, maybe new micro studio?


I just jumped into BMD w/ the micro and really hope they can make that option usable but w/ 4kpocket pretty sure updates or solutions are going to be slooooowwwww to non-existent...

what can u do...
Offline

Wayne Steven

  • Posts: 1113
  • Joined: Thu Aug 01, 2013 3:58 am

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 1:45 pm

You realise that the pocket was a pre-release, meaning they have a latter release schedule. So, there might be more cameras coming. The micro's are good target's. I hope there is a top end camera to fill in where the Ursa was. Would be great, Micro, Pocket, Mini, max.
Offline
User avatar

Frank Glencairn

  • Posts: 1624
  • Joined: Wed Aug 22, 2012 7:07 am
  • Location: Germany

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 3:42 pm

To me, this is all pretty bizarre.

It's 2018 now. Drive space never was so cheap before - I can by a 4TB drive for under 100 bucks now.

Drive speed never was so fast before, even a single halfway decent spinning drive, is fast enough to play 4k DNGs in real time.

We have a really small production company here, though we work with nothing but raw DNGs (plus some Arri raw and R3D) since 2012 without any problems or huge inventions. Even our oldest workstation can play, edit and grade DNGs in realtime. 5000 bucks buy you a workstation that totally flies with DNGs.

Raw DNG workflow with Resolve is super fast, super easy, super convenient. The fact that DNGs come in folders, instead of .mov containers doesn't make the slightest difference.

Though even huge companies and productions, with tons of budget shoot ProRes. I don't get it how ProRes is "easier" in any way.

And now we get ProRes Raw, which probably even has a similar data rate. So what's the big deal? Is everybody working with a laptop in Starbucks now?
http://frankglencairn.wordpress.com/

I told you so :-)
Offline
User avatar

Tommaso Alvisi

  • Posts: 118
  • Joined: Wed Aug 22, 2012 7:53 am
  • Location: ITALY

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 4:27 pm

Frank:

- convenience

- compatibility

- less cpu/gpu intensive supposedly
Offline
User avatar

Frank Glencairn

  • Posts: 1624
  • Joined: Wed Aug 22, 2012 7:07 am
  • Location: Germany

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 4:32 pm

How is it more convenient?

For now, compatibility is the worst, only FCPx can deal with ProRes Raw (that might change though).

With DNGs the CPU is totally bored, not much to do,
and even a 3 year old midrange gamer GPU runs in circles around DNGs.

So no, and no.
http://frankglencairn.wordpress.com/

I told you so :-)
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 5:21 pm

Frank Glencairn wrote:How is it more convenient?

For now, compatibility is the worst, only FCPx can deal with ProRes Raw (that might change though).

With DNGs the CPU is totally bored, not much to do,
and even a 3 year old midrange gamer GPU runs in circles around DNGs.

So no, and no.


Are you talking about compressed DNGs? I though they push CPU quite a bit.
There were many complains about Cinema DNG when it was released (lack of support, slow etc) and it's stil not all that shiny.
ProRes RAw definitely has much better potential. It's all down to Apple and manufactures now. I would ideally instead of those 2 see Cineform RAW to be most popular :)

Latest Scratch allows you with a 1 touch of a button decode just 1/2 or 1/4 of original resolution in Cineform which makes it working for 4K even on basic laptop. It's such a killer feature, specially for some cases.
Last edited by Andrew Kolakowski on Fri Apr 13, 2018 5:29 pm, edited 2 times in total.
Offline
User avatar

Frank Glencairn

  • Posts: 1624
  • Joined: Wed Aug 22, 2012 7:07 am
  • Location: Germany

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 5:23 pm

No, they don't, super easy on the CPU
http://frankglencairn.wordpress.com/

I told you so :-)
Offline

roger.magnusson

  • Posts: 474
  • Joined: Wed Sep 23, 2015 4:58 pm
  • Location: Sweden

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 5:30 pm

Depends on the CPU. :)

A six core CPU or less at about 2.4GHz will have trouble with 4K lossless compressed raw. Uncompressed raw should be quite easy but I think all BMD cameras have lossless compression now.
Offline

Andrew Kolakowski

  • Posts: 4315
  • Joined: Tue Sep 11, 2012 10:20 am

Re: ProRes RAW for UMP

PostFri Apr 13, 2018 5:57 pm

I don't get why those formats are so difficult to decode (I just blame lazy and poor design/implementation). There quite a few examples of lossless codecs created by single "guy" (student) which outperform DNG speed by miles. Lossless compression is very different to "normal" one and you can make codec crazy fast.
PreviousNext

Return to Cinematography

Who is online

Users browsing this forum: No registered users and 19 guests