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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: kash johal who wrote (59740)	5/27/1999 12:23:00 AM
From: kapkan4u	Respond to of 1572383

<Kash - re: The screwed up heatsinks on the voltage regulators are more worrying to me right now.> I wonder myself. Maybe Scumbria or someone else can explain. Kap.

To: kash johal who wrote (59740)	5/27/1999 1:32:00 AM
From: Scumbria	Read Replies (2) \| Respond to of 1572383

Kash, Could you explain why in some detail where they are clueless. CPU's are designed with all kinds of fail safe configuration registers, to allow the chip to run even if parts of the logic are not fully functional. Some common examples of this are L1 cache off, pipelining off, serialization of the pipes, branch prediction off, etc. Any one of these settings will cause dramatic performance degradation. In order for benchmarks to have any meaning, they have to be done in a controlled environment. For the "Foolish Squid" to release unofficial benchmarks with no clue about the BIOS settings, or a blessing from the manufacturer, is nothing short of stupid. Even a minor logic error may motivate the manufacturer to turn off a performance feature, to achieve stability. All that the Foolish Squid has accomplished is to prove that they can run a benchmark. The argument that they have a late model piece of silicon is worthless. One bad gate can make a CPU unstable. Scumbria