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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: wbmw who wrote (239123)	8/21/2007 7:00:37 AM
From: tecate78732	Respond to of 275872

Damn fine post! I appreciate the work you put into it. -tecate-

To: wbmw who wrote (239123)	8/21/2007 3:08:24 PM
From: Petz	Read Replies (1) \| Respond to of 275872

re: me: In Intel's case, for the quad core X5300 series 120w (145w max) Xeons, Intel states categorically that a heatsink designed to allow Tcase to reach 70C when dissipating 120w WILL result in "noticeable performance loss due to increased TCC activation [thermal throttling]." They state this is true even if the actual Tcase temp doesn't come close to 70C! you: <In this statement, you are only referring to the Thermal Profile B solution, which according to the spec is a reduced specification meant for 1U form factors. It's meant to give the designers a means of cooling the X-series Xeon in volumetrically constrained spaces, where otherwise, the choice would be to drop down to E-series parts.> You are just obfuscating here. I am setting out to prove that all Clovertowns NEED to have a HSF that keeps the temperature at 63 and below or they will suffer "noticeable performance loss" -- Intel's words, not mine. The name Intel gives to the 70C solution is irrelevant. The spec also notes that servers will maintain the 63C degree Tcase of Thermal Profile A due to TCC activation, so you can extrapolate that the Thermal Profile B heat sink is really only dissipating 90W ...yea, and they will have "noticeable performance loss" According to this spec, Intel's Thermal Profile B Psi(ca) guideline is 0.249 C/W. ... This makes Xeon actually slightly easier to cool than Opteron. ...but we've established that Thermal Profile B results in noticeable performance degradation. Don't even go there! No AMD solutions have any performance degradation whatsoever. Only the standard X-series Thermal Profile A Psi(ca) is 0.190 C/W. This makes Xeon actually slightly easier to cool than Opteron. I did my own calculations of required HSF performance in °C/W because AMD and Intel might have different assumptions about the temp inside the server case. Turns out I was right -- Intel uses 40°C and AMD uses 38°C. There's no magic here, but its clear that keeping Tc below 63 to avoid Intel thermal throttling is a lot more difficult than keeping it below 67-72°C. So comparing AMD's SE series parts to Intel's X series parts, AMD is easier to cool. Actually, my claim is that Pete has invented a term that he calls TDPmax, but it doesn't relate to anything in the Intel specs. Yes, Intel's TDPmax of 145w for X-series Xeons is not the same as Pete's. Petz