AVS Forum Special Member
Join Date: Jan 2003
Location: Portland, Oregon, USA
Mentioned: 25 Post(s)
Tagged: 0 Thread(s)
Quoted: 233 Post(s)
First let me say that I am actually looking at whether this chip might make sense for Lumagen to consider. It does have some nice features.
However, I would like to point out that "3D color space conversion" and "6-axis color control" is not the way to do CMS.
ANY color conversion is "3D." This applies to YCrCr to/from RGB as much as YCbCr to xvYCC. This does not mean they have a 3D palette like the one we use in the Radiance, and in fact their claim of "6-axis-color-control" pretty much tells me they do not.
I believe the only correct way to do CMS is with a linear-gamma RGB color cube palette, and definitely not "6 axis color control." While the 6-axis color can help some, it is the wrong color space. We looked at 6-axis color for our CMS and immediately dropped it once we looked at the math. This has been confirmed to me by calibrators who have tried to use 6-axis-color-control systems to do CMS. They note that while some improvement can be made, the 6-axis approach fails in two regards:
- It is not the right way to adjust color. You need control of R, G, and B to adjust primaries and secondaries, but 6-axis-color-control gives you color and hue.
- It does not keep the primaries correct over intensity changes.
To illustate this, consider a ball and a cube. Walking a "straight line" on the ball is not the same as a "straight line" on the cube. While this is not a perfect analogy, the 6-axis-color-control is the ball and you want a cube (aka RGB).
Still the IDT chip looks interesting. I hope they have overcome some of the issues which caused us to switch from its predecessor "Realta" chip to the "VXP GF9450."