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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Investor A who wrote (28478)	2/15/1998 12:42:00 AM
From: Ali Chen	Respond to of 1572356

Fuchi, <The technology loser is best at marketing. Look what their latest heavy duty marketing: CHEATING!> The question about 3D Winbench98 is not so simple. If you believe (and work hard in this direction) that the Microsoft Direct3D API is the future platform for all games, the 3D-Wb98 is a viable benchmark. It uses reasonable distribution of triange sizes in object tessellations, many additional effects for better image quality, etc. If a card is designed with explicit implementation of D3D API calls, it will apparently have a serious advantage on pure D3D applications/games. However, many game developers use their own 3D-geometry engines, and use some simplifications in vertices transformations, texture mappings, etc. This gave them better game dynamics at the expense of image realism. You probably would not call the wall textures in Quake as very realistic, don't you? For example, in many games monsters are just flat sprites (Doom), or very simplified objects. There are many corners for performance trade-offs. Just to get some impression about the variety of approaches to 3D rendering, you may want to visit cg.cs.tu-berlin.de This site contains references to 395 (!) 3D-engines. If the game engine is very different from the Direct3D (data structures, data locality, spectrum of tessellation, color representation, lightening model, filtration, special effects like fog, etc.), the card may be not so good in this particular game. Therefore, it would be a strong oversimplification to compare 3D-cards on a few specific games but skip applications with higher level of 3D-realizm. This diletant Tom does not know what he is doing. On the other side, Intel is well aware of the problem and has a special open(?) program for 3D-card validation. You probably cannot design a chip that will be optimal for all 395 3D-engines, and you have to bet on some API standard. I guess Intel bet on Microsoft D3D. Nothing unusual. Speaking about more serious approach, there is the Graphics Performance Characterization Group (GPC), now a part of SPEC, specbench.org There must be some benchmarks, but I am not very familiar with them. Best regards, Ali