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 : IDTI - an IC Play on Growth Markets -- Ignore unavailable to you. Want to Upgrade?

To: BULL who wrote (6541)	2/12/1998 1:59:00 PM
From: flickerful	Respond to of 11555

i agree bull... so far it seems to be doing respectably well, but the strength at the close will be telling. consider the last few days has seen consistent, to the upside trades through close.

To: BULL who wrote (6541)	2/12/1998 3:01:00 PM
From: Rob S.	Read Replies (2) \| Respond to of 11555

Must be. OK, to make things simple let's start fresh: Rob S. has no credibility except for whatever each of you thinks. Read my posts or just pass over them from now on. No one forces K. or anyone else to click the mouse button on anyone else's posts. The internet is not a broadcast media - we have the power to surf where we want and read what we want. Personal decision. Enough already. Someone questioned my posts on yields. I posted comments made in Infrastructure web newsletter which everyone is welcome to verify for themselves and I encourage you to do so. Then I called up Brian and he confirmed that IDTI was getting 60% yields on the runs in the Oregon fab. Brian reported they have completed 39 wafer runs so far and the yield targets have been exceeded. Again, if K. or anyone else doesn't want to accept this statement, please do your own checking. Several weeks ago I suggested to Brian that IDTI might be able to achieve yields of 70% once they had gone down the experience learning curve. At that time he thought the number was probably too agressive. yesterday Brian said that 70%+ ws a reasonable target for the C6 at 0.25 um. I think this is fantastic. As I have been encouraging IDTI in past posts, they have reported (Lehman) discussions with fab partners. IDTI has at least three potential partners. The downside of building parts at an IBM, SGS, TI, Hatachi or other major semi company is that the wafer level cost will be about 50% higher than for IDTI to manufacture the product in-house. Cyrix, for instance reportedly pays over $3,000 per 8" wafer from IBM. IDTI's internal cost is closer to $2,100/8" wafer. With the high yielding, small die size C6, the raw die cost is only half of the finished packaged and tested cost. So it may still make very good sense to fill orders with outside manufacturers. The other advantage in doing a deal is that it will almost certainly be with one that has cros-licenses with Intel. IBM, SGS, TI, Hitachi, and a few other Asian semi houses have long standing Intel cross-licenses. This will provide IDTI with the ability to prevent any potential lawsuit from Intel. So what is so different about the situation with IDTI than it has been for Cyrix or AMD? Cyrix has a fab agreement with IBM that was renewed after the buy-out by NSM. Their M2 part is about twice the area of the IDTI C6 and uses five metal layers and more complex memory cells (according to IDTI). That has led to horrible yields - reportedly less than 40%. So, Cyrix pays $3000 per wafer that they can fit maybe 200 die onto and that will yield about 80 good raw die. So each die cost them about $37. That's more than 3X what it costs IDTI to build the C6. AMD reportedly is now getting under 35% yields on the K6. They probably have a lower incremental cost per wafer (but one heck of a sunk cost). What's so different about IDTI's situation, and why I have been stressing the issue about yields, is that the dynamics of having parts built at IBM, SGS (even with their less capable fabs) or others is that the parts can be built at the higher costs compared to internal production and still turn out to be very profitable while Cyrix/NSM is wrapping money around every part they have been selling. This fact is not being lost on the managers at IBM, Compaq, Toshiba, and elsewhere from what I've been told. K. and others. I've asked before and I'll ask again, please do some research, talk to whomever, check your knowledge banks and pour over these statements of opinion and details. If you have some contradicting, supporting, new, old, or whimsical information that adds or detracts (intelligently) from this, then please post it. These are just my findings and opinions. Until I'm told otherwise, the internet and phone are free for anyone to use and no one forces anyone to accept what I post.