Look, I can’t speak for other people, but here is a timeline about how it personally worked for me:
I saw JC’s .plan long before nvidia had released anything official. I was very excited about 3d texturemapping being hardware accelerated.
Later, I did see something from nvida claiming full OpenGL 1.2 support in hardware. This confirmed my belief that 3d texmap would be there.
Someone on this board later noticed that it says “sw” in nvida’s opengl extensions pdf, meaning that the gf3 would only support them in software.
Cass confirmed this, also on this board. At this point, I was sure they would NOT be supported.
After this, I saw a review of herc’s 3d prophet III that said that it supported dx8 volumetric textures and volumetric texture compression in hardware. Confusion about whether or not it was there was re-introduced in my mind.
At this point, I bought the card. I was assuming 3d texturemapping WASN’T supported, but hoping that it was, that’s all. I tested it and this feature is missing. The card is amazing for a huge variety of other reasons and I am not a bit upset with the purchase. I wouldn’t trade it for a Radeon.
I would, however, like to know what happened with that one feature. Why isn’t there an official explanation? Is MS forcing this to be dx8 only? Xbox only? Is it for quadros only? Ultra version of the card?
JC doesn’t work for Nvidia, but he may as well. They optimize their drivers around his code. They give him alpha versions of their cards to use. He has input on the future direction of hardware. Therefore, if he says a feature is there, I believe it. His words are certainly better than any from a “PR guy”, who would just spew out cool sounding words like quncunx and nfinite fx without even really knowing what they mean.