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: Greg S. who wrote (7459)	9/21/1998 4:10:00 PM
From: Waldeen	Respond to of 16960

I want to add my two cents to this Intel discussion: Greg said, So I suppose that Sun, Apple, SGI, and IBM "can't win"? It's not all as simple as that; 3Dfx and Intel target different markets. Intel can kick tail in its market because of its immense marketing and distribution power. 3Dfx can smite the competition in its market because of the quality, speed, value, and support of its products. Show me a game enthusiast who buys an i740 and I'll show you a bumbling idiot. Different strengths cater to different markets - just like 3Dfx doesn't have the PR or the price to crack the OEM market wide open, Intel doesn't have the quality to enter the niche markets dominated by 3Dfx. Agreed. And not only that, Intel has no intention of dominating the graphics acceleration business. They just wanted to drive the prices down enough to sell more processors. They'll drive down the margins of a segment for a while, and when they realize it's not profitable, they drift back to their processor business. I've just been around too long and seen this before, especially with Intel. Case study: 3Com I am also a COMS holder, (and was long Intel from 1992 until a few months back) and last year Intel entered the NIC segment, COMS shareholders were all up in arms. Intel has entered the networking segment, and network cards will now be integrated onto the motherboard, was the conventional wisdom. We're doomed. Sound familiar? Well, what has happened. Sure there are more network cards on motherboards, but last week the rumor was Intel was at 3Com to talk about putting 3Com's networking on the motherboard (along with takeover rumors, Intel wanting "Intel Inside" the Palm Pilot etc. etc.) Since Intel entered the NIC market 3Com has reduced their NIC costs competitively where even Intel can't compete on price, and outside that area Intel has not kept pace. Routers have become switches, modems increased to V.90, networking is moving to ATM, voice over IP, and cable modems. In none of these areas is Intel dominating 3Com. Nor do they care. They pushed the prices down in NICs where it mattered to their market. The networking business is moving too fast to put all these things on the motherboard, as with the graphics accelerator business. Intel cares about pushing graphics accelerators costs down in the OEM market since that affects retail cost of their final product. If you're strictly an OEM supplier and not innovating, I would be somewhat worried. But even then, Intel doesn't seem to stay focused on products outside their core processor business once they have driven the prices down. Let's argue from another point of view: cost. When Banshee came out I was ticked they left the second TMU off. And this board thrashed me fairly good because that little bit of extra silicon real-estate can cost increasingly more as size moves up. Now imagine placing a Banshee size silicon area in with a Pentium II and tell me how that is cost effective to Intel's yields. Sure, Intel and AMD etc. are going to integrate some graphics technologies into the processor. Only if 3Dfx quits innovating do we care if they do that with our "old" technology. In fact, that helps 3Dfx concentrate on the high-end business. Besides, Simon has some real good points about geometry acceleration being the next key, and it is way too early for that to be motherboard bound. Does anyone know clearly how to do it? A lot more innovation left in this field before Graphics can truly simulate reality. At that point, Intel can put the whole thing on the Motherboard, and I'll be old and gray. Who knows, Intel might not even be around then.... Waldeen P.S. And if you are still worried, go over to the ASND board and post "Aren't you worried that Intel will integrate networking on the motherboard and put 3Com out of business?" Don't really recommend this though ;-)

To: Greg S. who wrote (7459)	9/22/1998 4:28:00 PM
From: timbur	Read Replies (1) \| Respond to of 16960

Is Brian Hook a "bumbling idiot"?!? Greg said: Show me a game enthusiast who buys an i740 and I'll show you a bumbling idiot. Brian Hook said in an editorial on June 30th, 1998 at voodooextreme.com : Graphics Accelerators: We live in the here and now. I was first amused about the Nth generation hardware accelerators coming around the bend, but at this point I'm getting a little tired of it. Much like the "Real Soon Now" Quake/Quake 2 killers, we have the "Real Soon Now" Voodoo/Voodoo2 killers. Why is it that whenever we talk about graphics accelerators, we don't talk about stuff we can actually buy RIGHT NOW? Fact: the best 2D/3D general purpose graphics accelerators out today are the NVidia Riva128 (which isn't very full featured), the Rendition V2200 (hugely underrated and a solid part), and the Intel i740 ($109 for a board with between Voodoo and Voodoo2 performance and 8MB?!). You can NOT buy a PVRSG, Banshee, Matrox G200 (that I've seen), or Riva TNT today. These boards are all "just about to ship". yawn Wake me when I can buy one. Fact: If I were to buy a new machine today, I'd get it with the Intel i740 part (I'd choose it over the Rendition because the i740 supports AGP). I'll grant you that it's almost 3 months since Brian wrote that. Still, if it was good enough to be in his system 3 months ago, I hope it's still a decent purchase! Tim the Troller