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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Tony Viola who wrote (66164)	10/9/1998 7:59:00 PM
From: Jeff Fox	Read Replies (1) \| Respond to of 186894

Tony, re: Maverick wafers Thank you - very well said. Perhaps you should teach this stuff? :) My only comment is that I fully believe Intel's fab machine is on par with IBM. Jeff

To: Tony Viola who wrote (66164)	10/9/1998 8:34:00 PM
From: nihil	Read Replies (2) \| Respond to of 186894

Tony Re: Chip identification I was told that Intel did not serial number chips, so that it was impossible to identify the wafer or batch or fab date once it left the test stand or was sliced up. Do you know if IBM does serialize, as you suggest in the early part of your post (as a source of post-test failure data)?

To: Tony Viola who wrote (66164)	10/9/1998 10:53:00 PM
From: Elmer	Read Replies (2) \| Respond to of 186894

Re: "If, after it gets through some percentage of the wafer, the yield is less than a programmed amount, testing stops and the whole wafer is thrown away. They call these wafers 'Mavericks'." Almost every company now has SPC guidelines that set an upper as well as a lower limit on yields. Wafers aren't just thrown away if they yield outside these limits because it could be a tester or probecard problem and not necessarily the wafer itself. It would simply be flagged for disposition by an engineer at any company. Why throw away a wafer because the probecard has a worn probe? Re: "Most companies probably package and sell die from any wafer that has any good ones at all." This is an Intel thread. Shouldn't you be posting this on the AMD thread? <ggg> EP

To: Tony Viola who wrote (66164)	10/10/1998 11:01:00 PM
From: kash johal	Read Replies (2) \| Respond to of 186894

Yony, Oh great yield expert. Perhaps you could help us understand a little bit more. 1. You must be aware that yield is a function of die size. So wrt to recent chips like the HP 8500 chips. These yields should naturally be low due to large die size. Does this mean that HP's 8500's are doomed and will be highly unreliable. Should we be calling Platt and telling him about these major reliability threats. 2. Also Intel is putting massive cache memories on chip. So these chips will be much larger than chips with smaller caches. Are these new Intel chips a long term yield/reliability threat? 3. Intel is using redundancy to improve yields. Basically a defective portion of the chip is bypassed and a good portion muxed in it's place. But these defects are still there and they can grow and fester can they not. Could this cause long term reliability problems. 4. As you are probably aware yields are tuned over time. When new products come out yields are low and then gradually improve. Should we be designing in the latest Intel 450 Xeons as they are at high end of yield cure. How about all those memory chips as they move up the technology curve. Regards, Kash