Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : 3DFX -- Ignore unavailable to you. Want to Upgrade?

To: Obewon who wrote (13466)	6/24/1999 11:19:00 PM
From: Patrick Grinsell	Read Replies (5) \| Respond to of 16960

Important! Please read... I will be interviewing Scott Sellers tomorrow. Post any good questions and I'll see if I can work them in. Most of the 3dfx interviews have been gamer oriented. This one will be business oriented and I'll try to be comprehensive. Pat

To: Obewon who wrote (13466)	6/25/1999 4:33:00 PM
From: Greg S.	Read Replies (1) \| Respond to of 16960

Obewon - this has re-ignited my long-extinguished spark of enthusiasm for 3dfx. Rumor, maybe, but the trademark names lend it some credence. Of course, T-buffer could be for "Transformation" and M-buffer could be for "Matrix" and it would just be a new set of transform buffers. But let's examine the situation for a minute. We can assume that 3dfx is adding SOME kind of new feature or set of features to the chipset, based on repeated hints from company sources. They've just been tightlipped about what those features will BE. Two positives for LOD and two negatives for T&L: Transform and Lighting: a) Competitors also recognize the potential of T&L and have openly discussed its possible presence on future chipsets. 3dfx would be first to market, maybe, but the other competitors would play catch-up real fast. 3dfx would want to implement something no one else though to accelerate. b) The benefit of T&L at this point is dubious because games may not have been written to take full advantage of hardware T&L, and there are possible issues with not being efficient enough at offloading the CPU and hence not seeing enough performance gains. Releasing a T&L chip now might be wasteful because real gains wouldn't be seen until later down the road, when competitors' chips are out there. LOD: a) This is a feature that AFAIK -NO- vendor has hinted they will use in their next-gen product, but there is an obvious use for it (being able to send parametric curves to the hardware and have it tesselate it for you). Does this remind anyone of 3dfx's flagship feature, the V2 and its multitexturing? Something no one else had planned to do? b) Gains will be almost immediate - games are shipping now that use MRM and RTDAT, the only problem is getting those routines into hardware. Perhaps it will coincide with a new release of Glide? If 3dfx is the company we all hope it is (the company of two years ago) my opinion is they may in fact blow us all away with a superb implementation of LOD. The impact this will have on games? It will encourage much higher polygon counts and more detailed models as well as curved surfaces for both models and environments. Imagine for the first time encountering real-time 3d models of humans in a game that don't look like triangulated chunks? -G