Pages: [1] 2   Go Down

Author Topic: Srgb to linear to logC to rec709  (Read 5562 times)

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Srgb to linear to logC to rec709
« on: November 06, 2017, 03:23:11 AM »

This probably silly topic perfectly reflects my current ignorance when it comes to
Understand color science in the pipeline.

The question I ask is if it would makes sense to linearize an srgb material of a consummer camera then apply an Arri logC lut then ArrilogC to rec709 lut.

What I have in mind is this:
- 1 scene shooted with an Alexa in logC
- the same scene shooted from a different angle with a fuji with a look applied from capture in camera.
2 different animals for the same shot.
I'd like the Fuji footage to behave as if it had been shooted by the Arri logged.
So when I apply a logC-to-rec709 to both I got consistency.

In short, the idea is first to cancel in post what the Fuji in-camera did.
Then simulate the curve of the Alexa as if...

Does it make sense?
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #1 on: November 06, 2017, 12:28:33 PM »

This probably silly topic perfectly reflects my current ignorance when it comes to
Understand color science in the pipeline.

The question I ask is if it would makes sense to linearize an srgb material of a consummer camera then apply an Arri logC lut then ArrilogC to rec709 lut.

What I have in mind is this:
- 1 scene shooted with an Alexa in logC
- the same scene shooted from a different angle with a fuji with a look applied from capture in camera.
2 different animals for the same shot.
I'd like the Fuji footage to behave as if it had been shooted by the Arri logged.
So when I apply a logC-to-rec709 to both I got consistency.

In short, the idea is first to cancel in post what the Fuji in-camera did.
Then simulate the curve of the Alexa as if...

Does it make sense?

This makes sense, but you face a challenge.  Arri LogC is not just Log.  The Arri LogC - REC709 transform also applies a color matrix correction for camera original Arri LogC material.  Your Fuji camera REC709 (no camera shoots in sRGB that I know of) is already fully corrected, so converting it to Log will not make it match camera original Alexa footage color wise.  And so, when you apply the Arri LogC - REC709 LUT to the Fuji Log footage, it will have a matrix applied that will not look correct.

Here's what you might do:

Generally, when I grade Alexa LogC, I put the Arri LogC - REC709 LUT in the middle of the node tree.  All corrections done before the LUT are in Log space, corrections after the LUT are traditional Lift Gamma Gain.

For your Fuji material, you can apply a REC709 - Log (not LogC) node as your first node.  Corrections after that node will be in Log space and work the same as corrections (before the LUT) in the Arri clips.  After your Log corrections add a Log (not Arri LogC) to REC709 LUT.  And in nodes after this LUT, you will be again grading in Lift Gamma Gain.

So, the node tree and corrections for each camera will be different to get you to the same place in the end, which is REC709.  Make sense?

The other alternative, is to grade using ACES or Resolve color management.  In this case, the Arri material will be transformed firstly to Log space for grading and only transformed to REC709 at the end of the grade.  (no Lift Gamma Gain grading in this workflow at all).  And your Fuji, if not already tagged, will come in as REC709, transformed into a Log space for correction, and then again transformed to REC709 at the end.  But this will have the advantage of applying the matrix only to the Arri LogC camera originals and not the Fuji originals.  I have not tried ACES yet with mixed cameras, and it's possible that you'll end up grading the Fuji originals in Lift Gamma Gain.  You'll need to experiment with this to find out for sure.
Logged
Bruce Alan Greene
www.brucealangreene.com

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #2 on: November 06, 2017, 01:35:53 PM »

This makes sense, but you face a challenge.  Arri LogC is not just Log.  The Arri LogC - REC709 transform also applies a color matrix correction for camera original Arri LogC material.  Your Fuji camera REC709 (no camera shoots in sRGB that I know of) is already fully corrected, so converting it to Log will not make it match camera original Alexa footage color wise.  And so, when you apply the Arri LogC - REC709 LUT to the Fuji Log footage, it will have a matrix applied that will not look correct.

Here's what you might do:

Generally, when I grade Alexa LogC, I put the Arri LogC - REC709 LUT in the middle of the node tree.  All corrections done before the LUT are in Log space, corrections after the LUT are traditional Lift Gamma Gain.

For your Fuji material, you can apply a REC709 - Log (not LogC) node as your first node.  Corrections after that node will be in Log space and work the same as corrections (before the LUT) in the Arri clips.  After your Log corrections add a Log (not Arri LogC) to REC709 LUT.  And in nodes after this LUT, you will be again grading in Lift Gamma Gain.

So, the node tree and corrections for each camera will be different to get you to the same place in the end, which is REC709.  Make sense?

The other alternative, is to grade using ACES or Resolve color management.  In this case, the Arri material will be transformed firstly to Log space for grading and only transformed to REC709 at the end of the grade.  (no Lift Gamma Gain grading in this workflow at all).  And your Fuji, if not already tagged, will come in as REC709, transformed into a Log space for correction, and then again transformed to REC709 at the end.  But this will have the advantage of applying the matrix only to the Arri LogC camera originals and not the Fuji originals.  I have not tried ACES yet with mixed cameras, and it's possible that you'll end up grading the Fuji originals in Lift Gamma Gain.  You'll need to experiment with this to find out for sure.
Thanks for your time Bruce. As always it is a pleasure and I learned a lot.
The first solution (the non ACES) you mentionned is the route for me so far because I understand the all process.

On the contrary, in ACES my skills are not giving me enough confidence but as you point, one has to jump on the water if to swim wants.
But remember that we had a thread on ACES here some time ago. It wasn't easy. At the same time I created a thread in the Red forum that became pinned because Graeme Nattress from Red team's gave o lot of infos. And to my surprise, not only me was lost in the middle of the tech mess but 99per cent of the generally knowledgeable Red users ?said what?. Included some colo artists do not fully understand the procedures. So my ego felt a little less silly  :)

Many thanks.

Ps: did I say ”generaly knowledgeable in the Red forum”?
Mmm...as RLF implementation seems often a source of confusions...
« Last Edit: November 06, 2017, 03:04:56 PM by fredjeang2 »
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #3 on: November 07, 2017, 12:36:45 PM »

Thanks for your time Bruce. As always it is a pleasure and I learned a lot.
The first solution (the non ACES) you mentionned is the route for me so far because I understand the all process.

On the contrary, in ACES my skills are not giving me enough confidence but as you point, one has to jump on the water if to swim wants.
But remember that we had a thread on ACES here some time ago. It wasn't easy. At the same time I created a thread in the Red forum that became pinned because Graeme Nattress from Red team's gave o lot of infos. And to my surprise, not only me was lost in the middle of the tech mess but 99per cent of the generally knowledgeable Red users ?said what?. Included some colo artists do not fully understand the procedures. So my ego felt a little less silly  :)

Many thanks.

Ps: did I say ”generaly knowledgeable in the Red forum”?
Mmm...as RLF implementation seems often a source of confusions...

I have only tested ACES with Arri LogC originals.  The ACEScct transform is very pleasant.  But it's limiting, you're always working in the transform's out put curves for contrast and roll off for highlights and shadows.  I think, with a bit of practice with the LOG controls, one can get to the final result.  But, for me, it's not always easy.  Most shots look great, but a few are hard to shoehorn into what I want.

I'm grading a feature now using the LUT approach I outlined in the earlier post.  I was actually going to try ACES on this one, but... I'm not handling the render myself, and we need a proper render for DCP.  For me, the safest bet was to work entirely in REC709, gamma 2.4, to avoid any confusion for the output to P3/DCI for DCP.  Sure, I'm working in a smaller color space, which is very similar to working in sRGB instead of Adobe RGB space.  But, really, will any ticket buyers be aware that in some shot with some intense colors could have had more detail in these intense colors?  I doubt it.  ACES is kind of like Photoshop's color management, but with the limitation that the output gamma curves are linked to the color space transform.  In Photoshop only the colorspace and basic gamma are used in the profiles.  Creative transform curves are done separately when you work on the image.  In ACES this transform is built in and I'm not a fan of this, but it makes the idea of a one grade, many outputs the most simple.  In Photoshop, you're going to make a whole new set of corrections for mat paper vs. glossy.  The idea behind ACES is that one shouldn't need to, but it maybe impossible to truly achieve this "single grade" idea.
Logged
Bruce Alan Greene
www.brucealangreene.com

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #4 on: November 28, 2017, 07:46:21 AM »


... (no Lift Gamma Gain grading in this workflow at all). 


Do you mean, using the log wheels?
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #5 on: November 28, 2017, 11:37:01 AM »

Do you mean, using the log wheels?

In Davinci Resolve one sets the wheels to "primaries", which means traditional lift gamma gain controls, or to "LOG" which means shadows, mid-tones, highlights.  Offset is usually also used as a LOG control, but it doesn't map to a wheel (using an tangent elements control surface).  In the Elements panel, there is a knob for Offest luma, and separate knobs for offset RG and B.  Or, you can use a mouse with the offset wheel in the GUI.  On the control surface there is also a knob for contrast and another for pivot.  The LOG controls are generally used when grading in ACES.  A proper control surface for LOG grading should ideally have 4 wheels instead of three, Shadows, mid-tones, highlights, offset.

You could use the primaries control for lift and gain instead of contrast and pivot.  Does just about the same thing in a different way.  The gamma control is not useful as a LOG control.
Logged
Bruce Alan Greene
www.brucealangreene.com

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #6 on: November 28, 2017, 12:49:40 PM »

In Davinci Resolve one sets the wheels to "primaries", which means traditional lift gamma gain controls, or to "LOG" which means shadows, mid-tones, highlights.  Offset is usually also used as a LOG control, but it doesn't map to a wheel (using an tangent elements control surface).  In the Elements panel, there is a knob for Offest luma, and separate knobs for offset RG and B.  Or, you can use a mouse with the offset wheel in the GUI.  On the control surface there is also a knob for contrast and another for pivot.  The LOG controls are generally used when grading in ACES.  A proper control surface for LOG grading should ideally have 4 wheels instead of three, Shadows, mid-tones, highlights, offset.

You could use the primaries control for lift and gain instead of contrast and pivot.  Does just about the same thing in a different way.  The gamma control is not useful as a LOG control.

Thanks Bruce.
So I was on track, which is nice.
Those days I made encouraging progress. Color is a long learning curve but
At the same time very rewarding.
Cheers.

PS: there is a grade in the white house (CP) that reminds me of Goldfinger. I will use your DP reel for training to match look in Aces, because there are great stuff in there.
« Last Edit: November 28, 2017, 03:33:33 PM by fredjeang2 »
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #7 on: November 28, 2017, 07:52:53 PM »

Thanks Bruce.
So I was on track, which is nice.
Those days I made encouraging progress. Color is a long learning curve but
At the same time very rewarding.
Cheers.

PS: there is a grade in the white house (CP) that reminds me of Goldfinger. I will use your DP reel for training to match look in Aces, because there are great stuff in there.

I don't think I understand your "PS" comments but thanks!  Grade in the white house?  Do you mean the "oval office" scene with Eric Roberts?  What does (CP) mean?  :)
Logged
Bruce Alan Greene
www.brucealangreene.com

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #8 on: November 29, 2017, 06:21:26 AM »

I don't think I understand your "PS" comments but thanks!  Grade in the white house?  Do you mean the "oval office" scene with Eric Roberts?  What does (CP) mean?  :)
CP? indeed means nothing ;D. I meant the MCU on sec 23 of the reel. Keyboard mistake.
I don't know you guys but when I write in this forum with my phone, I often got a sort of auto-correct (which is turned off in my phone settings) that messes with the writing all the time, and it's quite annoying; if I don't reread the text I got weired things happening and then when correcting it sort of paste and mix words that become meaningless so I have to erase the entire word and rewrite again. I don't know why but it only happens in LuLa and only using the phones, no matter which ones.

My PS comments just meant that I often watch movie looks I find interesting, trying to understand how (and why) they did and as a self-training course, putting myself in the situation that I had to match this look with my own footage. It is the best school I have found so far. How masters do? Copy them first and step by step progress to find one's own style.  So I grab stills, Resolve in one monitor and the still I want to emulate in the other as a reference look.

Using this approach, I've been learning a lot on common mistakes rookies often do.

PS: Using the log Wheels in ACES I like much more the control of the wheels, although being different and at first needs to get used to, but it responds much better to fine adjustments and in the end coming back to the traditional workflow ends to appear unatural.
« Last Edit: November 29, 2017, 07:22:26 AM by fredjeang2 »
Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1905
    • Pieter Kers
Re: Srgb to linear to logC to rec709
« Reply #9 on: November 29, 2017, 07:20:54 AM »

...no camera shoots in sRGB that I know of...

sorry to interrupt and bring down the level of this topic:)
- i am new to video - and always am puzzled with the different video formats ..

my question is this; if i play my ready (nikon) video on my tv it looks OK
But if i play the edited version  ( Premiere or other) it often looks different - different contrast and colour...
What do i have to do to keep that the same after editing?  ( in photography that would be the assigned profile - sRGB or other)
regards Pieter kers
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #10 on: November 29, 2017, 12:47:43 PM »

sorry to interrupt and bring down the level of this topic:)
- i am new to video - and always am puzzled with the different video formats ..

my question is this; if i play my ready (nikon) video on my tv it looks OK
But if i play the edited version  ( Premiere or other) it often looks different - different contrast and colour...
What do i have to do to keep that the same after editing?  ( in photography that would be the assigned profile - sRGB or other)
regards Pieter kers

This topic is way too big to answer here.  Check out the forum at:  http://liftgammagain.com/forum/index.php

Good luck.  This is not a simple topic.
Logged
Bruce Alan Greene
www.brucealangreene.com

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #11 on: November 29, 2017, 01:09:59 PM »

PS: there is a grade in the white house (CP) that reminds me of Goldfinger. I will use your DP reel for training to match look in Aces, because there are great stuff in there.

Goldfinger LOL!  That is Eric Roberts in the Oval Office. It was shot REC709 (low contrast) at 720p.  The movie is called "First Dog", it's a children's film.  I believe it's streaming on Netflix now.
Logged
Bruce Alan Greene
www.brucealangreene.com

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #12 on: November 29, 2017, 02:00:00 PM »

Goldfinger LOL!  That is Eric Roberts in the Oval Office. It was shot REC709 (low contrast) at 720p.  The movie is called "First Dog", it's a children's film.  I believe it's streaming on Netflix now.
Lol! My 2cents of humour...not that much of a Bond but I like the grade indeed.
Currently working with Red material with Aces. It's cool because no IDT with Red, no RLF, no color science...nothing! All bypassed. Hyper minimalist zen no messy workflow.
Then when all this is constrained by the rec709 ODT, the entire colour palette ain"t there
Any longuer but output samples does not weired things and the look is indeed preserved.
But I find that (again...just like a mantra in the wonderful world of colour engineering)
Where most detail goes away is in the shades and luminance is involved.
Going from a wide space to a narrower one, at the expense of DR.
The look is correct but the rec709 renders appear slightly different with a simplify color palette.
Specialy visible when conforming for the web.
I'm going to test if the solution could be to lift a bit artificialy the lum
To sort of compensate the effect. (A bit the same story than with the DCP)
For what I'm seeing so far, there is not one solution for all delivery
Formats but one specific for each.



« Last Edit: November 30, 2017, 09:17:42 AM by fredjeang2 »
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #13 on: November 29, 2017, 07:46:32 PM »

Lol! My 2cents of humour...not that much of a Bond but I like the grade indeed.
Currently working with Red material with Aces. It's cool because no IDT with Red, no RLF, no color science...nothing! All bypassed. Hyper minimalist zen no messy workflow.
Then when all this is constrained by the rec709 ODT, the entire colour palette ain"t there
Any longuer but output samples does not weired things and the look is indeed preserved.
But I find that (again...just like a mantra in the wonderful world of colour engineering)
Where most detail goes away is in the shades and luminance is involved.
Going from a wide space to a narrower one, at the expense of DR.
The look is correct but the rec709 renders appear slightly different with a simplify color palette.
Specialy visible when conforming for the web.
I'm going to test if the solution could be to lift a bit artificialy the lum
To sort of compensate the effect. (A bit the same story than with the DCP)
For what I'm seeing so far, there is not one solution for all delivery
Formats but one specific for each.

I can't quite follow all of what you've posted, but there is an issue when outputting for the web from ACES in Resolve.

The issue is the gamma of "REC709"  What is the gamma of your display?  If your display is gamma 2.4 and you're grading to that, I think you'll find that the gamma on the web makes the images a little bit light and washed out looking.  Since your project is in ACES, there's no way to change the gamma.  What you'll need to do is render out your REC709 version to a very good codec or uncompressed format.  Then open a new project, import your rendered material and use the OFX to convert from gamma 2.4 to gamma 2.2 and render again.  Then upload to the web and see if that makes things better.  If you can't find the OFX plug-ins, you can just make a gamma adjustment node to make the image a little bit darker.  Make sure this new project is not ACES or color managed at all.  Let me know if that helps!
Logged
Bruce Alan Greene
www.brucealangreene.com

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #14 on: November 30, 2017, 02:57:11 AM »

I can't quite follow all of what you've posted, but there is an issue when outputting for the web from ACES in Resolve.

The issue is the gamma of "REC709"  What is the gamma of your display?  If your display is gamma 2.4 and you're grading to that, I think you'll find that the gamma on the web makes the images a little bit light and washed out looking.  Since your project is in ACES, there's no way to change the gamma.  What you'll need to do is render out your REC709 version to a very good codec or uncompressed format.  Then open a new project, import your rendered material and use the OFX to convert from gamma 2.4 to gamma 2.2 and render again.  Then upload to the web and see if that makes things better.  If you can't find the OFX plug-ins, you can just make a gamma adjustment node to make the image a little bit darker.  Make sure this new project is not ACES or color managed at all.  Let me know if that helps!
Gosh! This is exactly what I saw when outputting for the web. The shades are a bit washed out and the overall image is slightly lighter.
Thanks for the tip! Indeed.
I will try to post some images when got it because this thread could be usefull for others who'd like to dig into ACES.
Thanks to your advices on that vast topic.
« Last Edit: November 30, 2017, 09:19:49 AM by fredjeang2 »
Logged

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #15 on: November 30, 2017, 03:13:05 AM »

sorry to interrupt and bring down the level of this topic:)
- i am new to video - and always am puzzled with the different video formats ..

my question is this; if i play my ready (nikon) video on my tv it looks OK
But if i play the edited version  ( Premiere or other) it often looks different - different contrast and colour...
What do i have to do to keep that the same after editing?  ( in photography that would be the assigned profile - sRGB or other)
regards Pieter kers
Have you found help in LiftGammaGain forum? As Bruce said, this is a complex
And long topic to write. There are no easy answers.
You could try a dedicated topic here and maybe point towards more specific
Parts of the subject to have more chances of answers.
Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1905
    • Pieter Kers
Re: Srgb to linear to logC to rec709
« Reply #16 on: November 30, 2017, 06:04:23 AM »

Thank for the answers!
I will check it out and just was hoping it would be a simple answer... as it is in photography.
I am always puzzled by the many video formats.
I like to know the basic of things so if you have other sites with basic video-info i will look into it.
regards,

Pieter
« Last Edit: November 30, 2017, 06:07:24 AM by kers »
Logged

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #17 on: December 01, 2017, 10:03:51 AM »


I'm grading a feature now using the LUT approach I outlined in the earlier post.  I was actually going to try ACES on this one, but... I'm not handling the render myself, and we need a proper render for DCP.  For me, the safest bet was to work entirely in REC709, gamma 2.4, to avoid any confusion for the output to P3/DCI for DCP. 
This is one of the problems of Aces. In what collaborative workflow is concerned, all the people involved in the chain
Have to be used of Aces workflow otherwise there are unexpected problems.
For many, if not most, it remains oscur to date and still a work in progress before it becomes implemented
Everywhere and as natural as breathing.
I see gazillion tutos packed with mistakes that I don't do myself any longuer, from the very beginning of their workflows, done by people who are supposed to know; it does not bring any confidence on the credit one can assign to the online tutos on Aces. In the end, very few really know
And the rest are guessing.
Resolve IMO has embrassed it differently than Baselight and Assimilate at the beginning.
I really do not get the point of their DaVinci flavours.I beleive gamma 2.6. Some pointed that it was supposed to
Give the wheels a natural control/feel but from my experience they are almost unmanegable for being too sensitive.
And it breaks the all idea of Aces.
As far as I'm concerned, I see the potential but not become obssessed with it.
Color is complicated enough not to ad more complications on something that is designed to simplify/unify.
And as you point, in the end the audience will never ever realise the super wide color space and its wonders.
Just us that we are mentaly sick with our toys. ;D
« Last Edit: December 01, 2017, 10:10:40 AM by fredjeang2 »
Logged

smthopr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 477
    • Bruce Alan Greene Cinematography
Re: Srgb to linear to logC to rec709
« Reply #18 on: December 01, 2017, 12:46:36 PM »

This is one of the problems of Aces. In what collaborative workflow is concerned, all the people involved in the chain
Have to be used of Aces workflow otherwise there are unexpected problems.
For many, if not most, it remains oscur to date and still a work in progress before it becomes implemented
Everywhere and as natural as breathing.
I see gazillion tutos packed with mistakes that I don't do myself any longuer, from the very beginning of their workflows, done by people who are supposed to know; it does not bring any confidence on the credit one can assign to the online tutos on Aces. In the end, very few really know
And the rest are guessing.
Resolve IMO has embrassed it differently than Baselight and Assimilate at the beginning.
I really do not get the point of their DaVinci flavours.I beleive gamma 2.6. Some pointed that it was supposed to
Give the wheels a natural control/feel but from my experience they are almost unmanegable for being too sensitive.
And it breaks the all idea of Aces.
As far as I'm concerned, I see the potential but not become obssessed with it.
Color is complicated enough not to ad more complications on something that is designed to simplify/unify.
And as you point, in the end the audience will never ever realise the super wide color space and its wonders.
Just us that we are mentaly sick with our toys. ;D

About wheel sensitivity:  Are you using a control surface?  Or just the mouse?

I you are using a control surface (the contraption with the balls and wheels), in Resolve preferences there are settings for the sensitivity of the wheels and that should solve your issue.  Mouse control of the GUI wheels in Resolve is often too coarse.

About gamma 2.6 that you mention, it just applies to DCI/P3 color space.  If you're not grading with a projector display, I don't think this is relevant anyway.  Using a control surface with Resolve and ACES is very usable.  But in ACES (or any Log grading), get very used to adjusting the "low range" and "high range" adjustments that determine what part of the curve these operations work on.
Logged
Bruce Alan Greene
www.brucealangreene.com

fredjeang2

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1376
Re: Srgb to linear to logC to rec709
« Reply #19 on: December 01, 2017, 12:59:17 PM »

About wheel sensitivity:  Are you using a control surface?  Or just the mouse?

I'm using the mouse currently because I'm not in my main studio but on the coast for awhile. With the mouse it's a bit painfull. I need to grab one in E-bay or Amazon.
I could change the sensitivity from the OS but it would affect the mouse behaviour out of Resolve.
On the CS panel, it doesn't really matter if I'd have to use a knob for the offset, as long as I keep control on S-M-H and avoid the mouse it's already a step.
A tangent
Or the BM


This Logitc G13 is also usefull

and there is a mapping profile to download for Resolve here: https://jonnyelwyn.co.uk/wp-content/uploads/2016/04/Will-Tejeda-G13-for-Resolve-12.zip

Yes, the key point with log wheels is the high/low range. I use them included in a non Aces workflow for fine tuning. balance highights etc...

I stick to gamma 2.4 in my monitors and color temp 6500, although I'm working in almost complete darkness most of the time.

Let me/us know if/when you finaly grade a feature film in Aces. That would be nice to have your comments on your findings and experience.

For people interested in ACES and currently bored like hell in a plane, I recommend this conference: https://www.youtube.com/watch?v=kUOo35Sd7rE
 
« Last Edit: December 01, 2017, 02:00:09 PM by fredjeang2 »
Logged
Pages: [1] 2   Go Up