To: Mark Carson who wrote (5420 ) 7/15/1998 10:37:00 PM From: c-horse Respond to of 16960
regarding the quickterm design tools PS: On the other hand, if written off over the next million chips, at $200,000, this Quickterm adds $0.20 to the cost of each and every chip. How long can this[...] Saving money on masks, etc etc is the real issue. When you have a 'virtual chip' [...] The HUGE reason for doing this is time! If you're selling chip X in the future at a clip of 1.5KK chips per quarter (once fully ramped), then any reduction in time to market allows you to sell an extra 30,000 chips (6KK/200days) PER DAY in reduced development time. Forget about the money you save not masking, etching and testing a bad design - you save probably 3 weeks in a development cycle that is likely to have 2 or 3 iterations between concept and prototype tooling. The reason you save time is that the quickterm equipment lets you "test drive" the new chip without going off and building a new chip. So, being realistic, what will happen is combination of improved quality (because additional iterations will get inserted into the design loop!) in a shorter time. Probably saving 10 business days out of a given cycle, but getting better designs than we otherwise would get (I say we as a shareholder, not an employee) which of course is a good thing. Lets be conservative, and say we save 5 business days. That's 150K chips more sold in the first quarter after releasing a new chip (if you cut 5 days off the beginning of the curve, the whole curve shifts left, and you measure the benefit in the fat part of the curve). I'll let the advanced people with 3rd grade math figure out the rest, I'm only crunching numbers at the 2nd grade level. C-Horse