Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

RedWideGamutRGB: discuss!

Sounds like to use this RWG/3G10 in the meantime it's best to avoid any LUT and just do basic primary and secondary grading?
 
Graeme has already stated that a proper transform has not been released yet. Attaching a Resolve film look LUT is not achieving a proper or even sensible color pipeline. It is throwing things against the wall to see if something sort of works. I understand people's impatience, but any LUT is designed for a certain set of expected input values. LUTs are not "generalized" and somehow able to magically yield great results regardless of the input color space, the input gamma curve, or the input exposure. They are specific. The new Red color space needs LUT's that are designed for it, just as Graeme has pointed out (even though people here don't seem to be listening.......).

Totally understand everything involved here, I was just trying to reverse engineer some of the nice looks that David was getting based on the assumption that he was using some modified standard film emulation luts. Sorry to have offended you...
 
No offense, and this is all new so we're all learning. I did some work with David to ensure his experimentation would be valid, and as we go down this route we'll all be able to reap the image benefits of the new log curve and colour space.

I realize this is a touch frustrating that the first part is in place (colour space and log curve) but that the transforms are not. Given that we can get delays in SDK implementation, it's important to put these elements out first while the subsequent parts are being worked upon, but of course, even the announcement of the new colour space breeds intense anticipation!!

Graeme
 
We will all try and be a little bit more patient hahaha!

Thank you so much for all the work you've done, and continue to do! Can't wait for this to be fully implemented!!!!
 
No offense, and this is all new so we're all learning. I did some work with David to ensure his experimentation would be valid, and as we go down this route we'll all be able to reap the image benefits of the new log curve and colour space.

I realize this is a touch frustrating that the first part is in place (colour space and log curve) but that the transforms are not. Given that we can get delays in SDK implementation, it's important to put these elements out first while the subsequent parts are being worked upon, but of course, even the announcement of the new colour space breeds intense anticipation!!

Graeme

Graeme,

can u share the XYZ values of RWG ?

Thanks.

- M
 
No offense, and this is all new so we're all learning. I did some work with David to ensure his experimentation would be valid, and as we go down this route we'll all be able to reap the image benefits of the new log curve and colour space.

I realize this is a touch frustrating that the first part is in place (colour space and log curve) but that the transforms are not. Given that we can get delays in SDK implementation, it's important to put these elements out first while the subsequent parts are being worked upon, but of course, even the announcement of the new colour space breeds intense anticipation!!

Graeme
Graeme, just trying to envision what this will be when done.

Will we be able to take footage from different RED cameras that doesn't necessarily match visually while being shot in RWG/3G10, and apply the same lut to the footage of each and get a match?
 
Graeme, just trying to envision what this will be when done.

Will we be able to take footage from different RED cameras that doesn't necessarily match visually while being shot in RWG/3G10, and apply the same lut to the footage of each and get a match?

nothing is shot in RWG, RWG is an output color space from the raw data... u can change that at any point to DC2 etc
 
I coulda sworn I saw this in the previous RCXbuild, but on 41 I can no longer find it listed (and yes, I know about the preferences>general>show advanced toggle.) Anyone else have this happen?
 
You need beta42 REDCine-X.

Graeme
 
You're killing me Graeme... :lol: I'm mid-project, so I'll play when it's in a release build I suppose.

Does it open up similar potential advantages (to a lesser degree) with MX footage?
 
So yes, I've tested with original M, MX, Dragon, and Helium footage. All works great.

Graeme
 
REDWideGamut and Log3G10 are coming to future camera firmware.

Graeme

refers to camera feed preview, like u currently have with RLF or RG4 etc and/or possibly Prores capture

Oh, thought it would be like outputting to Red log film.

So I guess my question now becomes, if you apply a lut to two differently shot sets of footage being graded from RED Wide Gamut color space, will one of your luts, Graeme, make the footage look much the same?
 
Last edited:
Ahoy.

I'm on the road for a while, but I am experimenting a bunch with REDWideGamutRGB/Log3G10 + the transform LUTs Graeme has been developing. So here's what I'm going to do until wrap on this sucker.

In each hotel I'll shoot my noggin with either the 6K or 8K VV Weapon w/ Dragon sensor and occasionally with a color chart. These are straight from camera with just white balance. What this should provide is a moderate baseline for skintones under daylight and mixed lighting conditions along the way. I might get tanner in the process and do forgive me for the hat hair or occasional goofy face. Various OLPFs as I'll be experimenting as much as possible.

Here's the first 4 hotels:

phfx_REDcolorTestsOnTheRoad_RWGLog3G10_hotel01.jpg


phfx_REDcolorTestsOnTheRoad_RWGLog3G10_hotel02.jpg


phfx_REDcolorTestsOnTheRoad_RWGLog3G10_hotel03.jpg


phfx_REDcolorTestsOnTheRoad_RWGLog3G10_hotel04.jpg



Flying to the next one tomorrow.
 
Last edited:
Plus one to this.

You can not just slap any LUT on this combination. I believe Graeme is developing a series of LUT's :)

Mike is spot on. A LUT is one part of an image pipeline. With RWG and Log 3G10 combo as the starting point it seems to allow for an improved debayer of the R3D.

David



Graeme has already stated that a proper transform has not been released yet. Attaching a Resolve film look LUT is not achieving a proper or even sensible color pipeline. It is throwing things against the wall to see if something sort of works. I understand people's impatience, but any LUT is designed for a certain set of expected input values. LUTs are not "generalized" and somehow able to magically yield great results regardless of the input color space, the input gamma curve, or the input exposure. They are specific. The new Red color space needs LUT's that are designed for it, just as Graeme has pointed out (even though people here don't seem to be listening.......).
 
When will the conversion LUTS be available in the wild? I'm guessing there are going to be different LUTs to go to different outputs (e.g. One for 709, P3, 2020, etc?)

Also, Phil, if time allows, could you post direct colour/gamma 709s for comparison (to see the improved accuracy/enhancement)?
 
So yes, I've tested with original M, MX, Dragon, and Helium footage. All works great.

Graeme

Kind of answers my question but I think I may need to reveal more of my thinking.

That is, I'll be shooting with a CF Weapon as my main camera and a RAVEN as secondary. The RAVEN will be at a different angle for the same scene and it is likely I'll be just exposing to the right on both cameras to save time as I will shoot some scenes sans crew.

What I'm hoping to learn is whether RWG will allow a like lut to be applied to both sets of footage and will visually match in all respects other than resolution.
 
Back
Top