the fujitsu pinolite board was a result of rendition's very lame drivers at the time. apparently, some new driver guy at rendition rewrote the GL driver to outperform the pinolite board. that helped kill the project.
IIRC, DX didn't have TNL at the time, and DX programs wouldn't have...
Re: good, bad and ugly
funny. he DID switch to the NV drivers because they are ** significantly ** more stable than the ATI variety. JC wouldn't say that, because he's a diplomat.
but it's obvious. he switches to an FX, despite it's obvious flaws, for the drivers and to execise new...
good, bad and ugly
you don't need to ask around here, since you will get all the usual ATI/NV employee's chiming in about their own products.
it's clear that NV makes better software all around (GL and DX), and the only proof you need regarding GL drivers is to ask Carmack.
he uses an FX...
sorry. don't give a fry about ATI either.
just that facts. NV FX is complete marketing HYPE.
to ATI's credit, the 9700P has been available at Fry's and at other retailers for 6+ months.
but the NV fan-butts still chime away on these boards about the virtues of the non-existant FX...
all i said was that you have to take into account all the processing that's going on in a renderer. modern out-of-order, superscalar CPU's like (P6 and beyond) have lots of execution resources -- multiple ALUs, AGUs, FPUs, vector FPUs (SSE). And, in comparing it to an r300 or FX class processor...
we're really oversimplifying this -- oh well
We also can't ignore all the other processing done by the GPU that must be explicitly done in a CPU, i.e. LOD, Texture Filtering, Alpha Blend, Texture Blend, Fog, ...
These are significant numbers of adders/multipliers (integer mainly). So our...
loops/branches
YOu are right and I made a mistake in a previous post. Branches out of a sub-pass are more tricky to handle. You could push an address into the F-buffer and resume on an address. Ideally, you want all branching and looping to stay in a sub-pass.
Data-dependent branching can...
stuff
It's not necessary to write out (to the F-buffer) ALL the register states between passes. Only the registers that need to live between passes.
Also, it's not necessary to involve shader code in the F-buffer. It's all data.
If you construct a code DAG from a shader, some portions...
confusion here
Stepper Retical's (typically) top out at 23x23 . There is miscommunication here. There is a physical limit to the retical sizes.
I'd be shocked to know to know it was 25 mm^2...
Besides this is really easy to determine if you are willing to blow a die. I'm sure NV has...