I’m trying to understand better exactly how profiles are constructed and how they work. I’ve read Anatomy of a Profile and a few dozen other articles on the Color Wiki.
I have watched your Rendered Gamut video (which I enjoyed) and understood what you were showing. I also read Myth #26 and understood that, but neither answered a question I have. Steve touched on the “rendered gamuts” in that article, but didn’t go into what they actually are.
When I view a 3D graph of an RGB printer profile, it defaults to “Device Gamut”. I understand that this is the same as the Abs Colorimetric rendering.
What puzzles me is that if this is the device gamut, or the maximum this printer/paper/ink combination can do, then why do Perceptual, RelCol and Saturation 3D views of the profile display larger than the AbsCol rendering? Exactly what am I viewing when plotting those rendering intents?
RelCol looks like AbsCol, but the white point is raised to 100L* instead of the paper white (which is about 96L* in this profile. It also looks like RelCol provides extended Chroma compared to AbsCol, mainly on the lighter end of the tonal scale.
When I plot Perceptual, it looks like the L* range goes from 0 100 and I get even more Chroma than RelCol. The perceptual view completely encloses the AbsCol view by an almost equal percent all the way around.
These plotted “gamuts” are larger than what I get when I convert a wide gamut, ProPhoto RGB color list to the same profile using Perceptual or RelCol. Those points, when plotted, fit nicely into the Device Gamut (AbsCol) and take most of the points out to the gamut border (like in your rendering video).
I think if I understood exactly what these Perceptual, RelCol and Saturation 3D plots represent, it would help me fill in some gaps in my understanding.
I’ve never posted an image, so I hope this works. This screen shot from CTP shows what I am trying to understand. The solid image is the Device Gamut, or Abs Colorimetric rendering of the RGB printer profile. The wire frame is the same profile displayed with Perceptual Rendering Intent.
The rendered gamut described in the video is different than the intent you see in the ColorThink Pro interface.
The intent shown in CT Pro today sets the rendering intent used when CT Pro calculates the RGB/CMYK -> Lab transform. This transform is typically set to abs col for printer profiles and rel col for monitor / working space profiles (as you noticed). When set to abs/rel col it effectively shows the gamut of the device.
When set to perceptual or saturation it shows how the profile will convert device values to Lab, not a rendered gamut. This setting is useful for some who are evaluating part of a profile’s tables and shows how data is handed off to the second profile in a transformation but it is pretty technical and of limited use in workflow analysis.
Rendered gamut uses the Lab->device transformation that you would use when printing an image - and shows what each output intent does… I have no idea if this is clear…
Let me illustrate. A typical printing workflow would go like:
Adobe RGB -> Lab -> printer RGB
A rendered gamut plot will show the effect of the 2nd half of this transform.
A typical “soft proofing” transform would go like:
printer RGB -> Lab -> monitor RGB
When you select an intent in ColorThink Pro, you are affecting the first 1/2 of this transform.
We appreciate the value of the rendered gamut and plan to add it to a future version of ColorThink. In the meantime, we released the video / explanation to show people how to do it in ColorThink today.
Thanks for the response. I think I am beginning to grasp this. Please tell me if my layman’s paraphrase of what you wrote is correct or incorrect…
When graphing a 3D plot of an RGB printer profile, the grapher is always plotting the proofing direction (A2B, or RGB-> Lab). The default is Device Gamut, which is Abs Color, and this shows the device’s full gamut.
RelCol is used for sending this back to the display, and besides, it’s the only rendering intent understood by Adobe RGB, sRGB and other matrix profiles. Unlike Abs Col, RelCol leaves white set to L*=100. Consequently, will display brighter whites, no preview of the paper color, and the color gamut will be wider since the colors aren’t being diluted by the darker, colored substrate. I’m guessing this is like doing a soft proof in Photoshop using RC without BPC. If I click Paper White in the soft proof, I am seeing the AbsCol intent on my screen (same as what is displayed in the grapherif I had a perfect profile).
Also, Perceptual and Saturation have no real useful function in the A2B direction for monitor proofing, since my monitor and working profiles can’t use this data effectively, and they expect to receive RC values. So, is the Grapher showing you the Lab values being handed off to Photoshop?
If this explanation is incorrect, it is probably my misunderstanding of terminology. I’ll get it eventually.
Thanks for your answer. Hope I have it correct, or close to correct. I LOVE ColorThink Pro. What an awesome program and learning tool!
