[nikhsub1]

Guys, Intels Thermal Analysis Tool for C2D's generates more heat than any other program I've ever seen. It gets my core temps about 10 - 13C hotter than P95 or FAH. Because of this, this will be my new load tool for my informal WB testing. Check this out, and see how fast the temps go up and down. Anyone know of anything that can get a C2D hotter than TAT? You can get TAT here: http://anonforums.com/builds/TAT.zip

This is FAH
20:34:33: Processor 1 : DIGITAL TMP[ 51°C ]
20:34:35: Processor 1 : DIGITAL TMP[ 52°C ]
20:34:41: Processor 0 : DIGITAL TMP[ 50°C ]
20:34:43: Processor 0 : DIGITAL TMP[ 51°C ]
20:34:43: Processor 1 : DIGITAL TMP[ 51°C ]
20:34:45: Processor 1 : DIGITAL TMP[ 52°C ]
20:34:47: Processor 0 : DIGITAL TMP[ 50°C ]
20:34:49: Processor 0 : DIGITAL TMP[ 51°C ]
20:34:51: Processor 0 : DIGITAL TMP[ 50°C ]
20:34:53: Processor 0 : DIGITAL TMP[ 51°C ]
20:35:05: Processor 0 : DIGITAL TMP[ 50°C ]
20:35:05: Processor 1 : DIGITAL TMP[ 51°C ]
20:35:07: Processor 0 : DIGITAL TMP[ 51°C ]
20:35:07: Processor 1 : DIGITAL TMP[ 52°C ]
20:35:13: Processor 0 : DIGITAL TMP[ 50°C ]
20:35:15: Processor 0 : DIGITAL TMP[ 47°C ]
20:35:17: Processor 0 : DIGITAL TMP[ 51°C ]
20:35:21: Processor 0 : DIGITAL TMP[ 50°C ]
20:35:21: Processor 1 : DIGITAL TMP[ 50°C ]
20:35:23: Processor 0 : DIGITAL TMP[ 51°C ]
20:35:23: Processor 1 : DIGITAL TMP[ 52°C ]
20:35:25: Processor 0 : DIGITAL TMP[ 50°C ]
20:35:27: Processor 0 : DIGITAL TMP[ 51°C ]
20:35:31: Workload power level: 100%
20:35:32: ** Workload Started (Processor 0) **
20:35:32: Workload power level: 100%
20:35:32: ** Workload Started (Processor 1) **
This is TAT load
20:35:33: Processor 0 : DIGITAL TMP[ 59°C ]
20:35:33: Processor 1 : DIGITAL TMP[ 63°C ]
20:35:35: Processor 0 : DIGITAL TMP[ 60°C ]
20:35:41: Processor 1 : DIGITAL TMP[ 64°C ]
20:35:43: Processor 1 : DIGITAL TMP[ 63°C ]
20:35:47: Processor 1 : DIGITAL TMP[ 64°C ]
20:35:49: Processor 1 : DIGITAL TMP[ 63°C ]
20:35:51: Processor 1 : DIGITAL TMP[ 64°C ]
20:35:53: Processor 1 : DIGITAL TMP[ 63°C ]
20:35:59: Processor 1 : DIGITAL TMP[ 64°C ]
20:36:03: Processor 0 : DIGITAL TMP[ 55°C ]
20:36:03: Processor 1 : DIGITAL TMP[ 59°C ]
20:36:05: Processor 0 : DIGITAL TMP[ 60°C ]
20:36:05: Processor 1 : DIGITAL TMP[ 64°C ]
20:36:09: Processor 1 : DIGITAL TMP[ 63°C ]
20:36:11: Processor 1 : DIGITAL TMP[ 64°C ]
20:36:13: Processor 1 : DIGITAL TMP[ 58°C ]
20:36:15: Processor 1 : DIGITAL TMP[ 64°C ]
20:36:23: Processor 1 : DIGITAL TMP[ 63°C ]
20:36:25: Processor 1 : DIGITAL TMP[ 64°C ]
20:36:33: Processor 1 : DIGITAL TMP[ 63°C ]
20:36:35: Processor 1 : DIGITAL TMP[ 64°C ]
20:36:38: ** Workload Stopped (Processor 0) **
20:36:39: ** Workload Stopped (Processor 1) **
TAT load off, FAH load
20:36:39: Processor 0 : DIGITAL TMP[ 53°C ]
20:36:39: Processor 1 : DIGITAL TMP[ 55°C ]
20:36:41: Processor 0 : DIGITAL TMP[ 52°C ]
20:36:41: Processor 1 : DIGITAL TMP[ 52°C ]
20:36:43: Processor 0 : DIGITAL TMP[ 51°C ]
20:36:43: Processor 1 : DIGITAL TMP[ 53°C ]
20:36:45: Processor 1 : DIGITAL TMP[ 52°C ]

[Patched]

does it also error check to see if your OC is stable? ^_^

[nikhsub1]

Originally Posted by Patched

does it also error check to see if your OC is stable? ^_^

No.

[Patched]

Too bad since it seems to be putting more load than even F@H can by itself, temps went up more than 10+C. People might find out their OC is not as stable as they thought or errors pop up quicker.

[roscal]

You won't find any software better than TAT to maximise heat dump because TAT is what we call a "power virus" especially coded to saturate a given architecture, here the C2D one. For P4, the soft was called P4PowerMax and there are coded by Intel with the exact knowledge of his architectures. TAT gives the bigger load we can pull off the chip (without checking because it lowers the load), softs like F@H are a joke for that part.

[nikhsub1]

roscal, is that what Intel bases TDP from? I'm just wondering if this consumes more when TAT is at load?

[roscal]

No, Intel TDP is defined as a typical "max" load under several typical softwares. We can say that TAT defines the max consumption because you can't go further with a given situation (I don't talk about o/c here) and Intel TDP is about 80 % of max power according to Intel. TAT consumption is over TDP.

It's the Intel definition. AMD has a different one with max power based on max current at stock voltage and max Tj IIRC. Both can't be compared directly, but Intel TDP is far more realistic and useful than AMD TDP because the chip never go at its very max load. TAT is a very special (and useless) case.

EDIT : Add a pic to understand Intel TDP :



Chip power is somewhere between Pidle and Ptdp for a typical use of a computer

[BillA]

Thanks for the lucid description Roscal

more junk here: http://www.xtremesystems.org/forums/...&postcount=236
when does "validate" not mean validate ?
a sink is independant of the source; air, water, phase change all do the same - according to their specific capability

NO mfgr "revalidates", only Swiftech (marketing) attempting to sell to the uninformed

clarification:
Scott, I am not panning your use of TAT
a comparison is simply a comparison, and the more extreme the conditions the greater the (apparent) differences
OK for testing; actual utility ??

[nikhsub1]

Originally Posted by BillA

Thanks for the lucid description Roscal

more junk here: http://www.xtremesystems.org/forums/...&postcount=236
when does "validate" not mean validate ?
a sink is independant of the source; air, water, phase change all do the same - according to their specific capability

NO mfgr "revalidates", only Swiftech (marketing) attempting to sell to the uninformed

clarification:
Scott, I am not panning your use of TAT
a comparison is simply a comparison, and the more extreme the conditions the greater the (apparent) differences
OK for testing; actual utility ??

Bill I really don't want to get into the TTV again... all I mean is that it is not the end all be all of WB testing. In any case, I think that for my menial testing with my limited tools that I will use TAT to generate load, I think the higher temps may give better resolution somehow. Interesting so far pitting the G5 against the Fuzion. Only have one mount of the G5 so far so my lips are still shut for now.

[BillA]

no no Scott, I do agree; do it
max stress = max resolution
I support all testers, no flames from here