Your preferred 32bit Full Range Pixel Format Workflow / Pipeline ?

Comments

Wolfgang S. wrote on 1/31/2023, 2:09 PM

Yes, maybe I'd like to use 32FR's linear 1.0 gamma (I would, actually), but there's no way I'm aware of to go from 8-bit FR to 32FR (1.0) without having to regrade my clips.

Wolfgang, am I wrong about that?

Please, have a look to this picture here again. As far as I see, there is no need to regrade all your clips. But you have to apply the settings as suggested here. But please test it by yourself!

 

Desktop: PC AMD 3960X, 24x3,8 Mhz * GTX 3080 Ti * Blackmagic Extreme 4K 12G * QNAP Max8 10 Gb Lan * Blackmagic Pocket 6K/6K Pro, EVA1, FS7

Laptop: ProArt Studiobook 16 OLED (ProArt Studiobook 16 OLED (i9 12900H with i-GPU Iris XE, 32 GB Ram. Geforce RTX 3070 TI 8GB) with internal HDR preview on the laptop monitor. Blackmagic Ultrastudio 4K mini

HDR monitor: ProArt Monitor PA32 UCG, Atomos Sumo

RogerS wrote on 1/31/2023, 5:35 PM

Do your grading in 32-bit full and then return to 8-bit full for playback?

Custom PC (2022) Intel i5-13600K with UHD 770 iGPU with latest driver, MSI z690 Tomahawk motherboard, 64GB Corsair DDR5 5200 ram, NVIDIA 2080 Super (8GB) with latest studio driver, 2TB Hynix P41 SSD, Windows 11 Pro 64 bit

Dell XPS 15 laptop (2017) 32GB ram, NVIDIA 1050 (4GB) with latest studio driver, Intel i7-7700HQ with Intel 630 iGPU (latest available driver), dual internal SSD (1TB; 1TB), Windows 10 64 bit

VEGAS Pro 19.651
VEGAS Pro 20.411
VEGAS Pro 21.208

Try the
VEGAS 4K "sample project" benchmark: https://forms.gle/ypyrrbUghEiaf2aC7
VEGAS Pro 20 "Ad" benchmark: https://forms.gle/eErJTR87K2bbJc4Q7

ALO wrote on 1/31/2023, 6:01 PM

Please, have a look to this picture here again. As far as I see, there is no need to regrade all your clips. But you have to apply the settings as suggested here. But please test it by yourself!

I think the issue is not the levels changing, it's what happens when other fx get involved. The CC (Secondary) is a good plugin to test because you can apply a global color offset using the chrominance wheel (my preferred way to color correct).

When I do that, switching between the two modes with your recommended settings, unfortunately I do see a color shift which would necessitate regrading everything on the timeline.

ALO wrote on 1/31/2023, 6:04 PM

Do your grading in 32-bit full and then return to 8-bit full for playback?

I do wish we had a better option, although aside from the nicer crossfades, I'm not sure the 1.0 gamma in 32FR makes much difference in my workflow. Maybe it's easier to control the density across different brightness zones with the Color Curves fx?

RogerS wrote on 1/31/2023, 6:08 PM

Does working in the color grading panel function better? If not, I'd grade in 32-bit mode as that is what will be in effect at the render stage. Or go between 8-bit video to 32-bit video levels if the tools you use work better there?

Seems the issue with CC secondary could be a bug or just that it hasn't kept up with the program's development.

Custom PC (2022) Intel i5-13600K with UHD 770 iGPU with latest driver, MSI z690 Tomahawk motherboard, 64GB Corsair DDR5 5200 ram, NVIDIA 2080 Super (8GB) with latest studio driver, 2TB Hynix P41 SSD, Windows 11 Pro 64 bit

Dell XPS 15 laptop (2017) 32GB ram, NVIDIA 1050 (4GB) with latest studio driver, Intel i7-7700HQ with Intel 630 iGPU (latest available driver), dual internal SSD (1TB; 1TB), Windows 10 64 bit

VEGAS Pro 19.651
VEGAS Pro 20.411
VEGAS Pro 21.208

Try the
VEGAS 4K "sample project" benchmark: https://forms.gle/ypyrrbUghEiaf2aC7
VEGAS Pro 20 "Ad" benchmark: https://forms.gle/eErJTR87K2bbJc4Q7

set wrote on 1/31/2023, 10:25 PM

Trying experimenting between Compositing Gamma 1.000 vs 2.222 (in 32bit FR, with view transform off), and just apply gain or Log. exposure - there's a different result.

 

While learning this, I also try to relearn again the Sony camera Picture Profile settings, and found one good enough:

And there's an explanation in this video of why there's gamma 1.0 (linear) or 2.222 (as for me who don't know about the reason behind that (seriously!))

Last changed by set on 2/1/2023, 3:21 AM, changed a total of 4 times.

Setiawan Kartawidjaja
Bandung, West Java, Indonesia (UTC+7 Time Area)

Personal FB | Personal IG | Personal YT Channel
Chungs Video FB | Chungs Video IG | Chungs Video YT Channel
Personal Portfolios YouTube Playlist
Pond5 page: My Stock Footage of Bandung city

 

System 5-2021:
Processor: Intel(R) Core(TM) i7-10700 CPU @ 2.90GHz   2.90 GHz
Video Card1: Intel UHD Graphics 630 (Driver 31.0.101.2127 (Feb 1 2024 Release date))
Video Card2: NVIDIA GeForce RTX 3060 Ti 8GB GDDR6 (Driver Version 551.23 Studio Driver (Jan 24 2024 Release Date))
RAM: 32.0 GB
OS: Windows 10 Pro Version 22H2 OS Build 19045.3693
Drive OS: SSD 240GB
Drive Working: NVMe 1TB
Drive Storage: 4TB+2TB

 

System 2-2018:
ASUS ROG Strix Hero II GL504GM Gaming Laptop
Processor: Intel(R) Core(TM) i7 8750H CPU @2.20GHz 2.21 GHz
Video Card 1: Intel(R) UHD Graphics 630 (Driver 31.0.101.2111)
Video Card 2: NVIDIA GeForce GTX 1060 6GB GDDR5 VRAM (Driver Version 537.58)
RAM: 16GB
OS: Win11 Home 64-bit Version 22H2 OS Build 22621.2428
Storage: M.2 NVMe PCIe 256GB SSD & 2.5" 5400rpm 1TB SSHD

 

* I don't work for VEGAS Creative Software Team. I'm just Voluntary Moderator in this forum.

Wolfgang S. wrote on 2/1/2023, 12:41 AM

I think the issue is not the levels changing, it's what happens when other fx get involved. The CC (Secondary) is a good plugin to test because you can apply a global color offset using the chrominance wheel (my preferred way to color correct).

When I do that, switching between the two modes with your recommended settings, unfortunately I do see a color shift which would necessitate regrading everything on the timeline.

Level changes can be an major issue if you mix up full range and legal range, either in the footage or the project settings.

But the point is, that without any other effects the two modes show the same gamma and gamut.

If older filters like the CC secondary work in the 32bit project settings in the same way as in the 8bit mode - well, they should. But those filters were developed for the 8bit world, and if you use them in a 32bit pipeline is always dangerours. Unfortunately that are risks that the user takes, if he changes project settings as you do.

Have you checked the behaviour in the color grading panel?

I would either stay in the 8bit mode, or stay in the 32bit mode. I am aware that one major driver to switch between both modes could be performance issues. Even high end pcs will help only partly.

Desktop: PC AMD 3960X, 24x3,8 Mhz * GTX 3080 Ti * Blackmagic Extreme 4K 12G * QNAP Max8 10 Gb Lan * Blackmagic Pocket 6K/6K Pro, EVA1, FS7

Laptop: ProArt Studiobook 16 OLED (ProArt Studiobook 16 OLED (i9 12900H with i-GPU Iris XE, 32 GB Ram. Geforce RTX 3070 TI 8GB) with internal HDR preview on the laptop monitor. Blackmagic Ultrastudio 4K mini

HDR monitor: ProArt Monitor PA32 UCG, Atomos Sumo

ALO wrote on 2/1/2023, 9:46 AM

Have you checked the behaviour in the color grading panel?

I would either stay in the 8bit mode, or stay in the 32bit mode. I am aware that one major driver to switch between both modes could be performance issues. Even high end pcs will help only partly.

Well, that was a good suggestion, but unfortunately, yes, the same thing happens in the CG panel. I'm testing using the Offset wheel in the GG panel, and I see exactly the same behavior.

So the good news is there's nothing wrong with Vegas' Color Corrector (Secondary) plugin (I think).

The bad news is there's no way to edit in an 8-bit mode and then render out in 32-bit FR unless you're willing to regrade whatever effects you've added.

I guess creating an intermediate in 32FR for playback might help? I'm guessing virtually no hardware is going to be capable of smooth 4K playback in 32FR once you start adding fx.

-------------------

on the subject of levels: I generally prefer to manage levels myself, since that's how Vegas used to work. For a long time I was not happy that Vegas forced me to do that, but I've come to appreciate that I learned a lot because of it. Vegas Tough Love!

Howard-Vigorita wrote on 2/1/2023, 9:47 AM
If older filters like the CC secondary work in the 32bit project settings in the same way as in the 8bit mode - well, they should. But those filters were developed for the 8bit world, and if you use them in a 32bit pipeline is always dangerours.

I use Color Corrector (Secondary) all the time and it works spectacularly well for me in all my projects regardless of media. Most commonly to tweak specific colors. Like to juice the costume of a dancer. Or to do area-specific brightness/contrast adjustments under irregular stage lighting. Or to knockout burned-out elements of a scene, for instance when bright outside light shines through a window in an otherwise well-exposed indoor shot.

I only use CC Secondary when I need to generate a dynamic adjustment mask. That's a mask defined by brightness-, saturation-, and/or color-ranges. The mask will then work on moving video without key-frames, motion tracking, or AI. With very high performance.

Although these unique dynamic masks can do their own internal adjusting, that's just for starters. They can also be stacked with other FX to limit what they operate on. Like Brightness/Contrast or Style Transfer. Or to create dynamic knockouts, like Chroma Keyer on steroids. I've found CC Secondary to be one of the most flexible and powerful tools in Vegas. After I took the time to figure out how the masking works and dream-up ways to use it.

Illusion wrote on 2/1/2023, 10:04 AM

I only do the rough cut in 8b FR. Once done, I switch to 32b FR, Gamma 2.222, View transform to off. My PC is able to playback 32b FR without issues for the projects I am doing.

My delivery format is HEVC in full range. I no longer mess with Limited range.

  • ASUS ROG Strix X570-E
  • Ryzen 9 3900x 12-core
  • 128GB RAM (4x32GB)
  • Nvidia RTX 2070 8GDDR6
  • 1TB WD Black NVMe M.2 for OS/Prog
  • 1TB WD Black NVMe M.2 for Media
  • 1TB/2TB/4TB SATA3 SSDs for projects/media
  • 4TB WD Red drive for local cache
  • 10TB EXOS Enterprise, 14TB Toshiba drive for local cache
  • 32in 4K main monitor, 24in 1080 second
  • Win 11 Pro
  • 28TB NAS for long term archive storage
  • Sony a6000
  • Sony A7C
  • GoPro Hero 8
  • GoPro Hero 11
  • Sony BRAVIA XR X90J 75" 4K HDR10
  • Nvidia Shield TV Pro media player (GigE wired)
  • JBL Bar 5.1

 

Wolfgang S. wrote on 2/1/2023, 10:06 AM

Well, that was a good suggestion, but unfortunately, yes, the same thing happens in the CG panel. I'm testing using the Offset wheel in the GG panel, and I see exactly the same behavior.

If you say that there is a difference between the 32bit mode and the 8bit mode if you add filters, then this makes the swap between the modes tough - maybe impossible. The CC secondary is powerfull - right. But that may not help for the question discussed here.

If I grade my log footage to HDR in Vegas, I work in the 32bit mode only. I decided to invest in a powerfull machine, and with a reduced preview you can work even in the 32bit floating point mode. One possible solution.

Desktop: PC AMD 3960X, 24x3,8 Mhz * GTX 3080 Ti * Blackmagic Extreme 4K 12G * QNAP Max8 10 Gb Lan * Blackmagic Pocket 6K/6K Pro, EVA1, FS7

Laptop: ProArt Studiobook 16 OLED (ProArt Studiobook 16 OLED (i9 12900H with i-GPU Iris XE, 32 GB Ram. Geforce RTX 3070 TI 8GB) with internal HDR preview on the laptop monitor. Blackmagic Ultrastudio 4K mini

HDR monitor: ProArt Monitor PA32 UCG, Atomos Sumo

ALO wrote on 2/1/2023, 11:32 AM

 

And there's an explanation in this video of why there's gamma 1.0 (linear) or 2.222 (as for me who don't know about the reason behind that (seriously!))

I believe the video you linked to is describing capture gamma. The gamma in the Vegas project properties field is compositing gamma, which is not quite the same thing:

https://www.provideocoalition.com/gamma_intro/

ALO wrote on 2/1/2023, 11:33 AM
If older filters like the CC secondary work in the 32bit project settings in the same way as in the 8bit mode - well, they should. But those filters were developed for the 8bit world, and if you use them in a 32bit pipeline is always dangerours.

I use Color Corrector (Secondary) all the time and it works spectacularly well for me in all my projects regardless of media. Most commonly to tweak specific colors. Like to juice the costume of a dancer. Or to do area-specific brightness/contrast adjustments under irregular stage lighting. Or to knockout burned-out elements of a scene, for instance when bright outside light shines through a window in an otherwise well-exposed indoor shot.

I only use CC Secondary when I need to generate a dynamic adjustment mask. That's a mask defined by brightness-, saturation-, and/or color-ranges. The mask will then work on moving video without key-frames, motion tracking, or AI. With very high performance.

Although these unique dynamic masks can do their own internal adjusting, that's just for starters. They can also be stacked with other FX to limit what they operate on. Like Brightness/Contrast or Style Transfer. Or to create dynamic knockouts, like Chroma Keyer on steroids. I've found CC Secondary to be one of the most flexible and powerful tools in Vegas. After I took the time to figure out how the masking works and dream-up ways to use it.

agree -- CC (Secondary) is one of Vegas' most useful plugins!

ALO wrote on 2/1/2023, 11:35 AM

I only do the rough cut in 8b FR. Once done, I switch to 32b FR, Gamma 2.222, View transform to off. My PC is able to playback 32b FR without issues for the projects I am doing.

My delivery format is HEVC in full range. I no longer mess with Limited range.

So my question is, if you use 32-FR with view transform off and gamma set to video, is there any difference between that and 32-VL plus a computer-to-studio fx for output rendering?

Edit -- actually I see you don't even need the levels correction -- you want to output to FR, which 32-video levels will do for you automatically (provided you conform your footage as you edit on the timeline).

And wait a minute -- doesn't 32-FR output video levels to HEVC?

Howard-Vigorita wrote on 2/1/2023, 12:02 PM

I only do the rough cut in 8b FR. Once done, I switch to 32b FR, Gamma 2.222, View transform to off. My PC is able to playback 32b FR without issues for the projects I am doing.

My delivery format is HEVC in full range. I no longer mess with Limited range.

@Illusion Thanks for that. I just tested the template default limited-range render and a customized template full-range render from a 32b FR project with your settings. They both look great next to the preview of the same project switched back to 8b FR. Nice to know I can deliver either with ease this way.

Illusion wrote on 2/1/2023, 12:47 PM

I only do the rough cut in 8b FR. Once done, I switch to 32b FR, Gamma 2.222, View transform to off. My PC is able to playback 32b FR without issues for the projects I am doing.

My delivery format is HEVC in full range. I no longer mess with Limited range.

So my question is, if you use 32-FR with view transform off and gamma set to video, is there any difference between that and 32-VL plus a computer-to-studio fx for output rendering?

Edit -- actually I see you don't even need the levels correction -- you want to output to FR, which 32-video levels will do for you automatically (provided you conform your footage as you edit on the timeline).

And wait a minute -- doesn't 32-FR output video levels to HEVC?

With 32b FR timeline, you simply have to choose Color Space at render time (as @Howard-Vigorita pointed out just above) . I never mess with semw or computer to studio fx at all. Simple streamlined workflow.

Edit: I don't need to conform anything to 32FR, Vegas does it for me. This is my GoPro footage, notice the Color Space, Vegas sets that based on the file metadata.

And my A7C which is limited, and properly marked in metadata. Vegas expands it automatically to FR.

Any stills will be set to Full range as well.

I never use Video Levels. For me, 0 is black and 255 is white, period.

 

My delivery files. I use Voukoder HEVC.

General
Complete name                            : W:\2021\Ouest\New set\Partie 9_0009 vk.mp4
Format                                   : MPEG-4
Format profile                           : Base Media
Codec ID                                 : isom (isom/iso2/mp41)
File size                                : 10.7 GiB
Duration                                 : 21 min 20 s
Overall bit rate mode                    : Variable
Overall bit rate                         : 71.4 Mb/s
Writing application                      : Voukoder (VEGAS)

Video
ID                                       : 1
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main@L5@High
Codec ID                                 : hev1
Codec ID/Info                            : High Efficiency Video Coding
Duration                                 : 21 min 20 s
Bit rate                                 : 71.1 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 29.970 (30000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 8 bits
Scan type                                : Progressive
Bits/(Pixel*Frame)                       : 0.286
Stream size                              : 10.6 GiB (100%)
Color range                              : Full
Codec configuration box                  : hvcC

Audio
ID                                       : 2
Format                                   : AAC LC
Format/Info                              : Advanced Audio Codec Low Complexity
Codec ID                                 : mp4a-40-2
Duration                                 : 21 min 20 s
Source duration                          : 21 min 20 s
Bit rate mode                            : Variable
Nominal bit rate                         : 314 kb/s
Maximum bit rate                         : 320 kb/s
Channel(s)                               : 2 channels
Channel layout                           : L R
Sampling rate                            : 48.0 kHz
Frame rate                               : 46.875 FPS (1024 SPF)
Compression mode                         : Lossy
Source stream size                       : 48.0 MiB (0%)
Default                                  : Yes
Alternate group                          : 1

Voukoder does not set the Color Range metadata in the output file by default. You have to tell it to do that using set_params.

Last changed by Illusion on 2/1/2023, 1:01 PM, changed a total of 4 times.

  • ASUS ROG Strix X570-E
  • Ryzen 9 3900x 12-core
  • 128GB RAM (4x32GB)
  • Nvidia RTX 2070 8GDDR6
  • 1TB WD Black NVMe M.2 for OS/Prog
  • 1TB WD Black NVMe M.2 for Media
  • 1TB/2TB/4TB SATA3 SSDs for projects/media
  • 4TB WD Red drive for local cache
  • 10TB EXOS Enterprise, 14TB Toshiba drive for local cache
  • 32in 4K main monitor, 24in 1080 second
  • Win 11 Pro
  • 28TB NAS for long term archive storage
  • Sony a6000
  • Sony A7C
  • GoPro Hero 8
  • GoPro Hero 11
  • Sony BRAVIA XR X90J 75" 4K HDR10
  • Nvidia Shield TV Pro media player (GigE wired)
  • JBL Bar 5.1

 

Illusion wrote on 2/1/2023, 1:44 PM

Here is a quick demonstration of the benefit of 32b FR vs 8b FR.

This is the original footage with severe backlighting, in a 8b FR timeline. This is GoPro ProTune Flat Native. WB and Input LUT not applied yet.

Now using the CGP to trim the input levels to bring back the histogram to 0, still in 8b FR. Notice the banding in the histogram.

Switching to 32b FR, Gamma 2.222, View transform off. Banding is gone.

Now with input LUT applied (Leeming LUT Pro), CGP White Balance, and my personal touch to correct the severe backlighting, still 32b FR.

If I switch back to 8b FR, notice the banding in the histogram and the visible banding in the preview, mainly in the sky, which is not visible in 32b FR.

Last changed by Illusion on 2/1/2023, 2:28 PM, changed a total of 3 times.

  • ASUS ROG Strix X570-E
  • Ryzen 9 3900x 12-core
  • 128GB RAM (4x32GB)
  • Nvidia RTX 2070 8GDDR6
  • 1TB WD Black NVMe M.2 for OS/Prog
  • 1TB WD Black NVMe M.2 for Media
  • 1TB/2TB/4TB SATA3 SSDs for projects/media
  • 4TB WD Red drive for local cache
  • 10TB EXOS Enterprise, 14TB Toshiba drive for local cache
  • 32in 4K main monitor, 24in 1080 second
  • Win 11 Pro
  • 28TB NAS for long term archive storage
  • Sony a6000
  • Sony A7C
  • GoPro Hero 8
  • GoPro Hero 11
  • Sony BRAVIA XR X90J 75" 4K HDR10
  • Nvidia Shield TV Pro media player (GigE wired)
  • JBL Bar 5.1

 

ALO wrote on 2/1/2023, 4:18 PM

With 32b FR timeline, you simply have to choose Color Space at render time (as @Howard-Vigorita pointed out just above) . I never mess with semw or computer to studio fx at all. Simple streamlined workflow.

Edit: I don't need to conform anything to 32FR, Vegas does it for me. This is my GoPro footage, notice the Color Space, Vegas sets that based on the file metadata.

Ah -- you're using Vokoder. That explains how you're getting full-range instead of video levels from your 32-FR renders.

Like you, I prefer to work on the Vegas timeline in 0-255 for obvious reasons, but I always have to deliver to video levels, so adding the output levels fx doesn't bother me (when needed).

With 8-bit footage, I prefer to conform it myself, because a lot of the time I like the extra headroom. As I am probably now shifting to 10-bit log, I will have to learn how to do everything all over again. :)

Howard-Vigorita wrote on 2/1/2023, 4:41 PM

To render full range with any of the Vegas-supplied templates, just go the Project tab and select it. Then give it a name and save it as a custom template if you want to reuse it again.

Illusion wrote on 2/1/2023, 5:01 PM

 

Ah -- you're using Vokoder. That explains how you're getting full-range instead of video levels from your 32-FR renders.

Like you, I prefer to work on the Vegas timeline in 0-255 for obvious reasons, but I always have to deliver to video levels, so adding the output levels fx doesn't bother me (when needed).

With 8-bit footage, I prefer to conform it myself, because a lot of the time I like the extra headroom. As I am probably now shifting to 10-bit log, I will have to learn how to do everything all over again. :)

@ALO

Using Voukoder has nothing to do with that. You can use any template and choose Full or Limited at render time, regardless of the timeline. Again, no need for output level fx.

Contrary to Voukoder however, Magix templates will properly mark the output file as Full or Limited depending on your choice in the project tab of the render dialog, see below

General
Complete name                            : F:\Renders\Untitled.mp4
Format                                   : MPEG-4
Format profile                           : Base Media / Version 2
Codec ID                                 : mp42 (isom/mp42)
File size                                : 333 KiB
Duration                                 : 10 s 777 ms
Overall bit rate mode                    : Variable
Overall bit rate                         : 253 kb/s
Encoded date                             : UTC 2023-02-01 23:03:44
Tagged date                              : UTC 2023-02-01 23:03:44

Video
ID                                       : 1
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main@L5@High
Codec ID                                 : hvc1
Codec ID/Info                            : High Efficiency Video Coding
Duration                                 : 10 s 777 ms
Bit rate                                 : 60.6 kb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 29.970 (30000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 8 bits
Bits/(Pixel*Frame)                       : 0.000
Stream size                              : 79.8 KiB (24%)
Language                                 : English
Encoded date                             : UTC 2023-02-01 23:03:45
Tagged date                              : UTC 2023-02-01 23:03:45
Color range                              : Full
Codec configuration box                  : hvcC

Audio
ID                                       : 2
Format                                   : AAC LC
Format/Info                              : Advanced Audio Codec Low Complexity
Codec ID                                 : mp4a-40-2
Duration                                 : 10 s 773 ms
Bit rate mode                            : Variable
Bit rate                                 : 192 kb/s
Channel(s)                               : 2 channels
Channel layout                           : L R
Sampling rate                            : 48.0 kHz
Frame rate                               : 46.875 FPS (1024 SPF)
Compression mode                         : Lossy
Stream size                              : 248 KiB (75%)
Language                                 : English
Encoded date                             : UTC 2023-02-01 23:03:45
Tagged date                              : UTC 2023-02-01 23:03:45

 

Last changed by Illusion on 2/1/2023, 5:05 PM, changed a total of 1 times.

  • ASUS ROG Strix X570-E
  • Ryzen 9 3900x 12-core
  • 128GB RAM (4x32GB)
  • Nvidia RTX 2070 8GDDR6
  • 1TB WD Black NVMe M.2 for OS/Prog
  • 1TB WD Black NVMe M.2 for Media
  • 1TB/2TB/4TB SATA3 SSDs for projects/media
  • 4TB WD Red drive for local cache
  • 10TB EXOS Enterprise, 14TB Toshiba drive for local cache
  • 32in 4K main monitor, 24in 1080 second
  • Win 11 Pro
  • 28TB NAS for long term archive storage
  • Sony a6000
  • Sony A7C
  • GoPro Hero 8
  • GoPro Hero 11
  • Sony BRAVIA XR X90J 75" 4K HDR10
  • Nvidia Shield TV Pro media player (GigE wired)
  • JBL Bar 5.1

 

ALO wrote on 2/1/2023, 7:40 PM

@Illusion thanks -- I didn't know that!

Wolfgang S. wrote on 2/6/2023, 8:20 AM

Coming back to this discussion, a more systematic test approach has shown following results - for the question, if we can switch between 8bit and 32bit settings (without transformation), and without color shift. So mainly for the purpose to render out the footage in 32bit instead of 8bit:

  • one can switch between "Legacy 8-bit (video levels) and "32-bit floating point (video levels)".
  • No transformation should be applied
  • No color shift seems to take place then
  • No clipping seemt to take place then
  • Media settings of the media properties can be full, limited or undefined.

So do not use the project settings 8bit full and 32-bit floating point full, since here the color shift takes place (even with disabled view transformation).

Please test this.

Desktop: PC AMD 3960X, 24x3,8 Mhz * GTX 3080 Ti * Blackmagic Extreme 4K 12G * QNAP Max8 10 Gb Lan * Blackmagic Pocket 6K/6K Pro, EVA1, FS7

Laptop: ProArt Studiobook 16 OLED (ProArt Studiobook 16 OLED (i9 12900H with i-GPU Iris XE, 32 GB Ram. Geforce RTX 3070 TI 8GB) with internal HDR preview on the laptop monitor. Blackmagic Ultrastudio 4K mini

HDR monitor: ProArt Monitor PA32 UCG, Atomos Sumo

Illusion wrote on 2/6/2023, 8:28 AM

Coming back to this discussion, a more systematic test approach has shown following results - for the question, if we can switch between 8bit and 32bit settings (without transformation), and without color shift. So mainly for the purpose to render out the footage in 32bit instead of 8bit:

  • one can switch between "Legacy 8-bit (video levels) and "32-bit floating point (video levels)".
  • No transformation should be applied
  • No color shift seems to take place then
  • No clipping seemt to take place then
  • Media settings of the media properties can be full, limited or undefined.

So do not use the project settings 8bit full and 32-bit floating point full, since here the color shift takes place (even with disabled view transformation).

Please test this.

There is no color shift if you use 2.222 compositing gamma in Full Range.

  • ASUS ROG Strix X570-E
  • Ryzen 9 3900x 12-core
  • 128GB RAM (4x32GB)
  • Nvidia RTX 2070 8GDDR6
  • 1TB WD Black NVMe M.2 for OS/Prog
  • 1TB WD Black NVMe M.2 for Media
  • 1TB/2TB/4TB SATA3 SSDs for projects/media
  • 4TB WD Red drive for local cache
  • 10TB EXOS Enterprise, 14TB Toshiba drive for local cache
  • 32in 4K main monitor, 24in 1080 second
  • Win 11 Pro
  • 28TB NAS for long term archive storage
  • Sony a6000
  • Sony A7C
  • GoPro Hero 8
  • GoPro Hero 11
  • Sony BRAVIA XR X90J 75" 4K HDR10
  • Nvidia Shield TV Pro media player (GigE wired)
  • JBL Bar 5.1

 

Wolfgang S. wrote on 2/6/2023, 9:08 AM

True, but this is the gamma of video anyway. And if you use the 32bit floating point (full range) with a gamma of 2,222 (video), then the media properties of the footage MUST be set to full. With undefined or limited, clipping in the luminance takes place.

So to avoid this riks, the recommendation is to stick to video levels.

Desktop: PC AMD 3960X, 24x3,8 Mhz * GTX 3080 Ti * Blackmagic Extreme 4K 12G * QNAP Max8 10 Gb Lan * Blackmagic Pocket 6K/6K Pro, EVA1, FS7

Laptop: ProArt Studiobook 16 OLED (ProArt Studiobook 16 OLED (i9 12900H with i-GPU Iris XE, 32 GB Ram. Geforce RTX 3070 TI 8GB) with internal HDR preview on the laptop monitor. Blackmagic Ultrastudio 4K mini

HDR monitor: ProArt Monitor PA32 UCG, Atomos Sumo