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: view who wrote (9432)	12/8/1998 4:48:00 PM
From: Scott Garee	Respond to of 16960

Creative sells finished boards to Gateway, so most guesses are that TDFX is selling to Creative for ~$32 or so. I don't think TDFX is having yield problems with Banshee. We would have heard something by now. Plus, 1.2M parts is a lot. Certainly on the high side of anyone's expectations. If they were having yeild problems they would have had to forecast nearly 2 million to end up selling 1.2M. I've been told the typical fallout from fab to component shipment is 40-50%, though this is not an authoritive figure and is for different component types (but similar feature size.) Yield is affected by the process as well as the design. The reason TDFX spent all that money on design software was to get better initial silicon (which they apparently did with V3, since the first parts worked.) Yield can be due to failed gates (design feature size/locality issues) or thermal breakdown. My take on the TnT is that most yield problems were heat related. The die shrink will help this, but will still hold up the clock ramp. Thermal analysis should have given them the opportunity to refine the layout to reduce hot spots and thus improve yield. Unless the problems are extremely serious you don't respin. Respins cost a lot of money. Even design flaws are worked around as long as possible until enough are queued up (or you hit show stoppers) to justify a respin. I doubt we'll see the SIII part until Q3 at the earliest, if then. .18 is going to be very expensive compared to .25 (which appears to be next year's sweet spot.) I'm impressed that they are bold enough to try it, but it says a lot about how much trouble they're in. They announced they've "completed a design" 6 months early. Well, you've heard the saying "it looked good on paper." If they couldn't make a .25 part that could beat everyone else's .35 part I doubt their .18 part is going to save them. I doubt they're going to have the money to proceed, even if they use UMC's fabs. Ugh, got to ramlbing there, but I'm too tired to proofread, I hope it makes sense.