[rc5] Overclocking (wasRe: [rc5] Another approach)

Skip Huffman SHuffman at Atl.Carreker.Com
Mon Jul 7 11:42:55 EDT 1997


On Mon, 7 Jul 1997 06:44:39 +0300 (EET DST), Tero Kilkanen wrote:

>It would be an enormous job to test every
>individual chip.

I am fairly sure that testing every chip is exactly what is done.  There is a significant failure rate, ie chips that simply do not work, so all the chips have to be tested to make sure that they fuction.  I believe that at the same time they are marking the "best speed" of the chip.  

If I remember correctly the first runs of Pentiums had a 75% failure rate, which was the primary reason for the short supplies and consequent high prices.

I also remember that chips that fail to perform at a marketable clock speed may be used in integrated applications at a much slower clock rate, perhaps less than 10Mz.

All of the above is coming out of my meat memory so I may have forgotten a lot.



Skip Huffman
Carreker-Antinori
Atlanta Office
Quality Group

----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list