Elmer,
There is much confusion here. Intel isn't playing any tricks. Scumbria was confusing tjC with tcase. Intel was specing P4 at 72 deg C case temp while AMD is now specing Athlon at 95C die temp.
Actually Intel is specifying the processor case temperature, not case as in chasis temperature. Actually, the only thing that Intel says about the chasis temperature is that Intel recommends that the chasis temperature does not exceed 40 deg C, and if it exceeds 45 deg C, Thermal Monitor becomes active.
P4 has the Thermal monitor to slow down the CPU to prevent frying of the CPU when the CPU is under heavy load. What it means is that Intel may sell you an 1.5 GHz processor, but you may or may not be able to ever get full performace out of it, because when you do, the processor may overheat and slow down to 750 MHz.
It seems that Intel no longer reports the die temperature, only the processor case temperature, which seems like just another attempt at obfuscation.
Anyway, the more you read about the P4, the weaker it looks. Now add the voltage increase just to get the next speed grade, which should have been a routine, but it is proving to be a challenge, the only conclusion I derive is that Intel is again in position of weakness. P4 was supposed to save Intel, but now the new mantra is "wait until Northwood".
Joe
PS: Here are the documents in question:
ftp://download.intel.com/design/Pentium4/datashts/24919802.pdf
developer.intel.com
Here is what EricRR found: developer.intel.com
and his comment about it: Message 15365490 |
