Temp/tint vs offset

Get answers to your questions about color grading, editing and finishing with DaVinci Resolve.
  • Author
  • Message
Offline

TCP786

  • Posts: 456
  • Joined: Thu Sep 16, 2021 7:05 am
  • Real Name: Cody Predum

Temp/tint vs offset

PostSun Sep 25, 2022 8:53 pm

Am I correct in assuming that the temp and tint sliders are equivalent to moving the offset wheel? i.e. can the same changes be made by either, but temp/tint exist to make sure you're moving exactly on the axis between the two colors you're working on? Also, what are the units for temp and tint? I'm assuming temp is +/- degrees Kelvin, but why are the numbers for tint changes orders or magnitude smaller than temp?
Offline

rNeil H

  • Posts: 553
  • Joined: Tue Jun 26, 2018 9:43 pm
  • Real Name: R. Neil Haugen

Re: Temp/tint vs offset

PostSun Sep 25, 2022 9:23 pm

It's been a while since I tested the Resolve WB tool. But I think it's as I recall.

WB controls Iike Temp and Tint actually are normally more like the Gain control, or using the white-point end of an RGB Curves tool.

Temp typically is essentially playing Red white point against Blue, one goes up as the other goes down.

Tint typically is Red and Blue ganged together and moved versus Green.

But this is normally moving the end point so affecting the upper mids/highlights far more than shadows.

Offset moves the entire scale, including to the black point.

If I'm wrong I'll be corrected of course! But that's what I recall.

Sent from my SM-S908U using Tapatalk
Offline

TCP786

  • Posts: 456
  • Joined: Thu Sep 16, 2021 7:05 am
  • Real Name: Cody Predum

Re: Temp/tint vs offset

PostSun Sep 25, 2022 9:41 pm

Very interesting. How can you test them? I can't think of a way to do it that doesn't involve an overwhelming amount of trial and error. Especially since all the tools I want to test use different units and I don't know what most of them are or how they relate to each other.
Offline

rNeil H

  • Posts: 553
  • Joined: Tue Jun 26, 2018 9:43 pm
  • Real Name: R. Neil Haugen

Re: Temp/tint vs offset

PostSun Sep 25, 2022 11:33 pm

Watching the scopes. Perhaps with a grayscale ramp as the image.

Say, using the Parade RGB, for example. Watching the movement of the trace on the scopes tells you what the control is doing.

An Offset channel change will clearly move an entire channel up or down.

While moving the right end point of a Curves tool will move the trace on the right proportionally more than the left.

So, watch the scopes while moving the Temp control. What changes?

Now move the Tint control ... what changes?

Sent from my SM-S908U using Tapatalk
Offline

TCP786

  • Posts: 456
  • Joined: Thu Sep 16, 2021 7:05 am
  • Real Name: Cody Predum

Re: Temp/tint vs offset

PostSun Sep 25, 2022 11:56 pm

Very interesting. It definitely looks like the temp slider behaves more like the gain control than the offset control.
Offline

Hendrik Proosa

  • Posts: 3015
  • Joined: Wed Aug 22, 2012 6:53 am
  • Location: Estonia

Re: Temp/tint vs offset

PostMon Sep 26, 2022 5:57 am

Ofcourse it does, white balancing is based on scaling the rgb components so that they get balanced. Scaling == gain. Whether it is actually correct depends on the state of image data. It must be linearized for this to work properly. On logarithmic data offset would be correct method and gain would be wrong.
I do stuff.
Offline

TCP786

  • Posts: 456
  • Joined: Thu Sep 16, 2021 7:05 am
  • Real Name: Cody Predum

Re: Temp/tint vs offset

PostMon Sep 26, 2022 6:07 am

Hendrik Proosa wrote:Whether it is actually correct depends on the state of image data. It must be linearized for this to work properly. On logarithmic data offset would be correct method and gain would be wrong.
Are you saying that the temp and tint controls behave incorrectly when used on a log image? Isn't that what the "use color space aware grading tools" checkbox is for?
Offline

Hendrik Proosa

  • Posts: 3015
  • Joined: Wed Aug 22, 2012 6:53 am
  • Location: Estonia

Re: Temp/tint vs offset

PostMon Sep 26, 2022 10:09 am

Not saying that. If colorspace aware tools work as they should, working space shouldn’t matter. What I’m saying is that for any color operator that mimics some kind of physical property or models a specific behavior, that operator must be applied to image data in correct state.
I do stuff.

Return to DaVinci Resolve

Who is online

Users browsing this forum: Bing [Bot], Google [Bot], Mark Sterne and 151 guests