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: Waldeen who wrote (5924)	7/31/1998 11:31:00 AM
From: JC1X	Respond to of 16960

> 5. I counted on the company realizing that they had to produce > the best 2D and 3D board when they released Banshee. This > means I expected Banshee to blow away a single Voodoo2. How > can they not produce a board with better performance than a > Voodoo2? Let's go through the obvious reasons, As is obvious (now that the specs are out), 3Dfx "intended" that Banshee be slower than the Voodoo2. And I think the main reason for that is market positioning. They probably wanted a fast GENERAL market chip. not a gamer solution. The next reason is probably cost. The extra unit TexelFx2 chip would have meant EXTRA memory, board space, components.. (though, on the memory side, they have changed the architecture thus, I believe, eliminating the possibility of SLI - at least for now - with a standard Voodoo2) It would have been nice if we could just pop a second Texelfx2 chip into an open slot onto a Banshee board and get multi-texturing free but then, that would be about the only benefit. (the TNT ability to run the chips parallel doubling overall performance would have been nice, but TDFX doesn't seem to have the ability . . yet.) >Eventually, as people seem to be talking here, it is a given >that Banshee followed by Rampage etc. will eventually >lead to TDFX producing only combo 2D/3D cards? Why >give up the 3D add-on niche this company created? 3D add >on cards can be replaced more often as the market is more I think you sorta answered yourself in the few paragraphs before ie. the silicon is cheap. Why not have just dump the 2D functions in? J.

To: Waldeen who wrote (5924)	7/31/1998 2:37:00 PM
From: Sun Tzu	Read Replies (2) \| Respond to of 16960

Waldeen you make some very good points. Here is my take on it. With regards to your question on how much does a bit of extra silicon cost, it can potentially cost a lot. I have not done any semicon design for a while but in general you silicon cost (which is inversely proportional to yield) increases linearly for a while as the die size increases, and then it takes off like a rocket. Without knowing the current die size and today's processing parameters I can't give you more details. For what it is worth though, I have a nagging feeling that your final asessment (i.e not wishing to kill the V2 cash cow) was more the reason for the current VB design than silicon yields. You have done a good job of pointing out the down side of that decision also. Not allowing for VB SLI was a stupid decision IMO (unless they were really really yield constrained, which I doubt). A possible justification for that may be that they may wish to target the chip more to OEMs (ie on the motherboard) than to retail, in which case there would be no chance to convert a single VB to SLI. Even so, I think that was a poor choice. As for branding and other issues, most of it can be (and should have been) answered by proper publicity. 3Dfx looks to me like a restaurant with great a Chef and good food but really poor ambiance. They (the management) and their customers (us) would have been much better off if the restaurant would hire a nice decorator, play some music, put some candles on tables and would advertise a bit. I am not talking about advertising to the gamers, I'm talking about presenting themselves well to Wall St. The fact that the management comes from SGI may have something to do with this lack of talent (SGI never learned how to market itself to the street either). How can we expect the technophobes in the street to differentiate 3Dfx from a swampland in FLA when its management has done next to nothing to differentiate the company for them? They should be utilizing every opportunity to shout "3D is the way to go and we are the leading 3D chip maker". There is no shortage of research on this topic and they should hier some 3rd party firms to give them the favorable research that they need (this is not the same as lying. The third party firm would simply state the positive facts on 3Dfx and the 3D market). If they don't know how, they can look at some of their competetors and try to imitate them. ATI would be a good sample; even when they have nothing to say, they still find some way of making sure that the street will remember them and see them in a very positive light. Sigh! Timbur I loved your comment to Chip! Not that I expect a 50% revenue shortfall, but that it is good to realize there is more than one way for predictions to come true. Pat You asked for an explanation on why selling what goes down, so here it comes: We all know that to make money in stock market we should buy low and sell high. This as we all know is a lot easier said than done. In order for you to buy low and sell high in the classic meaning of the term, your point of view has to be drastically different than the vast majority of the market participants (i.e you have to be in the top 3% stock pickers in the market or know something about the company that others don't). Many people, professional and retail, find it much easier and more profitable to buy high and sell even higher. The answer to why should we sell something when it goes down has to do with which of the above two camps you believe in. In the same vain, many people are told to average down, but another way to make money is to average up and close the position when the market moves against you. Other reasons for closing the position when it goes down are, respect for the openion of the majority and acknowledging that you may be wrong, believing that the stock will still go lower (no sense in holding onto something if you know that it will not be going up anytime soon, finding better prospects in other investments, claiming the loss for tax purposes, and adjusting your asset allocation mix. Best Regards Sun Tzu

To: Waldeen who wrote (5924)	8/1/1998 2:35:00 PM
From: Marketeer	Respond to of 16960

Waldeen, Way back before Banshee was announced publicly, I had speculated that 3Dfx needed to make Banshee SLI-capable with Voodoo2. This would have allowed them to penetrate the PC OEM market without outright cannibalizing the existing sales. I guess it was technically unfeasible at this time. Moving forward, to break the graphics chip market convention of "my next chip is better than your current chip", they must offer such a solution. The easiest way to beat competition is to change the rules on them. 3Dfx, are you listening? Waiting to go long again...