Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

RED Color - A look at REDWideGamutRGB/Log3G10

sam karr said:
Agree, so far it adds some of what I always tried to dial down.
Click to expand...

It doesn't add anything.

RWG and Log3G10 bring larger range for luminance and saturation, in regards to which the recorded data is defined.

What you are seeing here is a) general effect of particular light (and mixture) meeting a particular camera base response properties (seen in all examples) + b) the effect of image transformations done to the base image, defined through various gamma and color spaces, and the initial testing results in the early stage are normal.

Reserve your final judgement for later.
These are very complex matters which take time.
 
  • Thread starter
  • Moderator
  • #43
Michael Lindsay said:
I think this is heading in the right direction... (it is easy to dial back colour from skin tones but adding always looks a bit synthetic) ...

Where can I get Phils transform back to Rec709?
Click to expand...

You're on the right path there.

These are Graeme's transforms and they are still in progress. I imagine very shortly everybody will have access to all of this once it's tightened up.

It's been interesting exploring it for sure.

For the comments regarding "enhanced chromatic aberrations", not exactly but..... If you have more saturation and the highlight roll-off is differently placed due to tonal remapping via a new gamma curve it will look different and perhaps more saturated. It's in the shot from the lens though.

Additionally, something a few people are looking closely at are images like this:

phfx_REDcolorTests_RWGLog3G10Rec709_32.jpg


Take a look at the blue LEDs on the DJ's board. This is one of my stress test files I keep handy and you'll notice you actually are getting the LED detail and the super saturation of the high end blue is much, much, much nicer. I have a few more oddball images like this I might post up that are under terrible lighting.

Also, do take a look at say the last few dozen films that have been shot and released in theaters. Specifically what they look like and the grading direction. I'll get more into a perspective on this later.

Also #2, remember you can take the color in any direction with REDCODE RAW. Color Science is always evolving. It will be interesting to see when this is finished up and how people will use it.
 
Another stress test image (courtesy of the other RGW thread):

I submitted a badly behaved R3D frame to RED some months ago. When viewed in linear light, it is clear that nothing is clipping. But when converted to DC2/RG4, multiple bizarre color artifacts emerged. Architectural lights (meaning colored glass that let light pass through, not lights aimed at architectural elements) that were blue started fringing purple in the hot spots. Changing the color temperature from warm to cold reduced that, but then architectural lights that were purple would fringe blue. And the colors were way, way too saturated.

With RedWideGamutRGB, the colors stay in the box, but they are very much desaturated from the real-life vibrant colors of the diamond lights. If you know what light looks like when it goes through cobalt-blue glass, that is the color blue we are aiming for. It's blue. It never turns purple. In any case, I think RedWideGamutRGB will be a very helpful starting place for me to build a LUT that ultimately handles my (architectural) glass.

Here is the R3D file, along with some JPG stills showing how things change across temperature (with DC2) compared to RedWideGamutRGB:

https://www.dropbox.com/sh/id1lhjnss1ot8uw/AAAZh3JrW_lGmQrhSw4Awb_6a?dl=0
Click to expand...

With Goran's LUT, RGW saves the day. Looking forward to what RED offers as the "official" solution. But clearly RGW is on the right path for dealing with super-saturated sources, which is a godsend for those of us who work in groovy environments.
 
I don't know, Phil...
On my Dell 27" monitor, REDgamma4 looks more natural.
 
Michael Lindsay said:
I think this is heading in the right direction... (it is easy to dial back colour from skin tones but adding always looks a bit synthetic) ...

Where can I get Phils transform back to Rec709?

thanks Michael
Click to expand...

People have been demanding better skin tones, its what everyone (including me) wants. Never in a million years would I want to add saturation to get to these skin tones. If I want the sky bluer, I will add saturation to the sky.. and no-one will be the wiser. What we need is less steps, not more, to get to a pleasing image.
 
If you look at something long enough, you can convince yourself it looks good. I do this all the time while editing photos and videos. Then I go to sleep, wake up and look back at it like "wtf was I thinking."
We're all providing some fresh eyes. It's subjective, but I wouldn't touch it right now in it's current state. At least it's early and we can make suggestions before it's good to go... part of the benefit of this community.
 
I like the highlight rolloff, that's major
 
I do think patience is needed on this one. There's work happening behind the scenes to get this right and when it officially all comes together it will be a solid pipeline to process your RED images. Right now, there's a lot of experimentation and testing. Graeme has mentioned he is tweaking things. I would wait for the official rollout with the mathematical transforms in place to arrive at a final judgement. I think even at this point, it's clear where some of the improvement is but I think it's important to keep it all in perspective.
 
Jake Wilganowski said:
I like the highlight rolloff, that's major
Click to expand...

You should be able to get much smoother highlight roll-off over a wider range of ISO footage with the new Log3G10/RWG. We're certainly seeing that here.

Graeme
 
Akin A said:
If you look at something long enough, you can convince yourself it looks good. I do this all the time while editing photos and videos. Then I go to sleep, wake up and look back at it like "wtf was I thinking."
We're all providing some fresh eyes. It's subjective, but I wouldn't touch it right now in it's current state. At least it's early and we can make suggestions before it's good to go... part of the benefit of this community.
Click to expand...

What you're seeing here are Phil's experiments on how to take Log3G10/RWG out to a viewable image. Elements of this are indeed subjective, but other elements are objective. Phil's images do indeed help immensely when I get to read such interesting feedback as I have done on this thread, and that's one of the reasons why Phil has gone to all this bother.

Because there are elements of subjectivity, we also employ blind testing of images too, and a couple of rounds of that process provided immense amounts of feedback that I've been spending hours deep in the code to respond to.

Graeme
 
Graeme Nattress said:
What you're seeing here are Phil's experiments on how to take Log3G10/RWG out to a viewable image. Elements of this are indeed subjective, but other elements are objective. Phil's images do indeed help immensely when I get to read such interesting feedback as I have done on this thread, and that's one of the reasons why Phil has gone to all this bother.

Because there are elements of subjectivity, we also employ blind testing of images too, and a couple of rounds of that process provided immense amounts of feedback that I've been spending hours deep in the code to respond to.

Graeme
Click to expand...

Awesome.. exactly why these discussions are so important. I am just so tired hearing how Arri has such a better look, Years and years of hearing it now... And the sucky part is for the most part they are right.. I think sometimes we get so lost in the pixel peeping that we forget that we are artists first. An image may not 100% reflect what exists, but thats what cinema is all about... creating beautiful images, pleasing to the eyes... My personal hope is after you are done with your magic making, the Arri color bandwagon can finally come to an end... Anyways thanks for everything you are doing!
 
  • Thread starter
  • Moderator
  • #57
And here's what those images look like with the current transforms and method (which are still being worked on)

phfx_REDcolorTests_RWGLog3G10Rec709_49.jpg
 
Graeme Nattress said:
Because there are elements of subjectivity, we also employ blind testing of images too, and a couple of rounds of that process provided immense amounts of feedback that I've been spending hours deep in the code to respond to.
Click to expand...
And thanks for reading and responding to our feedback!
 
Looks to be about ~5% too yellow to me, based off nothing but the mild hyper-saturation of yellow tones... ...of course without seeing it in person, I've got no real clue as it could be tungsten spill... However, even in all the sunlight shots, the sunlight side looks more yellowy than normal (which is nice for highlight roll-off and reminiscent of some filmstocks, but not necessarily the most pleasing/accurate for the rest of the image.)

I definitely see more red/capillaries/rosiness and tonal differentiation to the skintones, but again I didn't see their actual skin, so it could just be subtle over saturation of those reds.

I think Kem has a good point; it'd be nice to have comparison Alexa (and maybe even a film stock) frames as reference... It's kind of ambiguous comparing just DC2/RG4 to 310/RWG; all I can say for sure is that they generally look subtly warmer and more saturated, have more muted/subdued extreme highlights (perceived as a softer roll-off), and there seems to be more tonal differences.
 
Thankyou Phil for the swathe of imagery.

DC2/RG4 looks likes its colours are 'bleached' (i.e. White in place of colour) compared to WCG.

The blue LEDs (WCG) looks phenomenal.

I am wondering of people colour perception changes for when looking at yellow stills vs moving image.

Is it possible that we think the WGC images are too yellow because we have then beside the whiter DC2 (i.e. Is our brain playing tricks on us and adjusting the yellow More than other colours?).

Perhaps relative colours within an image are more important than the absolute colours? Skin vs lipstick (even if there is a slight yellow look) in WCG looks more convincing than DC2/RG4 (to my eyes).

AJ
 
You must log in or register to reply here.
Back
Top