Alternative Athlon Coolers reviewed @ Toms Hardware

***Veridian3*** · Sep 18, 2003, 10:16 PM

"CPU cooling solution suppliers obviously agree. What else could explain the fact that many a manufacturer is presenting "new" cooling solutions for the "old" platform only days/weeks before the launch of the new platform? Totally unprecedented in the practice of presenting a cooler is the way Coolermaster has attached an NDA to any reporting on its latest model, the Jet 7. In other words: only after the embargo ends (as of September 17) can the Jet 7 be reported on. "

http://www.tomshardware.com/cpu/20030917/index.html

DriveEuro · Sep 18, 2003, 10:46 PM

As usual. One hell of a great review by Tom's

Matth · Sep 19, 2003, 05:50 PM

I little off-track on the specific thermal power though, I'd say.

The Barton core dissipates more power, but through an area that's larger = lower W/cm-sq.

What matters to the heatsink overall, is the higher TOTAL.
The lower value per area will mean lower loss across the thermal interface (be it pad or paste) and that is ALL.

It's possible it balances, and the improvement in transfer efficiency to the heatsink counters the increased power - just depends how well a GOOD thermal paste is performing in the first place, as to how much there is to gain there.

Vampyromaniac · Sep 20, 2003, 12:41 AM

Quote:

Originally posted by Matth
What matters to the heatsink overall, is the higher TOTAL.
The lower value per area will mean lower loss across the thermal interface (be it pad or paste) and that is ALL.

It's possible it balances, and the improvement in transfer efficiency to the heatsink counters the increased power - just depends how well a GOOD thermal paste is performing in the first place, as to how much there is to gain there.

I didn't understand that part, can you please explain?

I saw on THG's chart the Palomino is way below the others, but don't Palominos run hotter than Thoroughbreds?

Matth · Sep 20, 2003, 04:45 PM

He SAYS the Barton is easier to cool, due to it's lower watts per square centimetre (more power, but over a larger core)
"However, a lower specific thermal power means that an XP 3200+ is also easier to cool. Or, put another way, coolers that can keep the die temperature of an Athlon XP 2700+ under control will not have any problems with a 3200+ CPU."

The Palomino blows that out of the water, as it's much larger core gives it a much lower SPECIFIC thermal power (watts per square centimetre), but the total wattage made them real swines to cool.

As I see it, watts per squre centimetre are more demanding on the CPU to heatsink thermal interface (paste or one-time pad), while the total power is more demanding on the heatsink itself (fins/fan)

A small core with high specific figure, probably benefits more from a well finished copper contact area and the finest paste.

Not adept with the extrem maths, but I've used power transistors on heatsinks, where you also have several thermal resistance figures:
1a. Junction to ambient (if using without heatsink)
1b. Junction to case
2. Heatsink to ambient - and in an unfanned heatsink, less than 1 degree per watt is one serious hunk of extruded aluminium - a figure achieved by a CPU heatsink you can hold in one hand.
3. Case to heatsink - and with the relatively large contact area, usually considered neglegible other than for the mica insulator that often had to be used

To take the specic point on balance....

The Barton produces more heat than the Thoroughbred, therefore, the differential from the heatsink point of contact to ambient WILL be higher.
It transfers TO the heatsink over a greater area, so the differential over the core to heatsink contact will be lower.

The net result, is that the Heatsink itself will be hotter, but the core underneath might be cooler - but that depends hown much the interface area actually contributes - I'd guess it to be probably less difference than between different thermal pastes!

Sep 18, 2003, 10:16 PM	#1
*Veridian3* Administrator Join Date: Nov 2002 Location: Cloaked Posts: 5,028 Rep Power: 181	Alternative Athlon Coolers reviewed @ Toms Hardware "CPU cooling solution suppliers obviously agree. What else could explain the fact that many a manufacturer is presenting "new" cooling solutions for the "old" platform only days/weeks before the launch of the new platform? Totally unprecedented in the practice of presenting a cooler is the way Coolermaster has attached an NDA to any reporting on its latest model, the Jet 7. In other words: only after the embargo ends (as of September 17) can the Jet 7 be reported on. " http://www.tomshardware.com/cpu/20030917/index.html

Sep 18, 2003, 10:46 PM	#2
DriveEuro HardwareHeaven Extreme Member Join Date: Jul 2002 Location: Gurnee Illinois Posts: 4,677 Rep Power: 0	As usual. One hell of a great review by Tom's __________________ [color=orange]ASUS P4P800 Deluxe [color=black]-[/color] 2.4C --> 3300mhz [color=black]-[/color] Mushkin PC3200 at 220 5-2-2-2 (cpu limited) [color=black]-[/color] BBAti 9800Pro at 430/375 [/color] [color=black]3[/color][color=gray]D[/color]M[color=black]a[/color][color=gray]r[/color]k[color=yellow]01[/color] [color=black]3[/color][color=gray]D[/color]M[color=black]a[/color][color=gray]r[/color]k[color=yellow]03[/color] 16001200 12801024

Sep 20, 2003, 04:45 PM	#5
Matth Flash Banner Hater Join Date: Jun 2002 Location: UK Posts: 3,426 Rep Power: 93 System Specs	He SAYS the Barton is easier to cool, due to it's lower watts per square centimetre (more power, but over a larger core) "However, a lower specific thermal power means that an XP 3200+ is also easier to cool. Or, put another way, coolers that can keep the die temperature of an Athlon XP 2700+ under control will not have any problems with a 3200+ CPU." The Palomino blows that out of the water, as it's much larger core gives it a much lower SPECIFIC thermal power (watts per square centimetre), but the total wattage made them real swines to cool. As I see it, watts per squre centimetre are more demanding on the CPU to heatsink thermal interface (paste or one-time pad), while the total power is more demanding on the heatsink itself (fins/fan) A small core with high specific figure, probably benefits more from a well finished copper contact area and the finest paste. Not adept with the extrem maths, but I've used power transistors on heatsinks, where you also have several thermal resistance figures: 1a. Junction to ambient (if using without heatsink) 1b. Junction to case 2. Heatsink to ambient - and in an unfanned heatsink, less than 1 degree per watt is one serious hunk of extruded aluminium - a figure achieved by a CPU heatsink you can hold in one hand. 3. Case to heatsink - and with the relatively large contact area, usually considered neglegible other than for the mica insulator that often had to be used To take the specic point on balance.... The Barton produces more heat than the Thoroughbred, therefore, the differential from the heatsink point of contact to ambient WILL be higher. It transfers TO the heatsink over a greater area, so the differential over the core to heatsink contact will be lower. The net result, is that the Heatsink itself will be hotter, but the core underneath might be cooler - but that depends hown much the interface area actually contributes - I'd guess it to be probably less difference than between different thermal pastes! Last edited by Matth; Sep 20, 2003 at 05:00 PM.

Sep 19, 2003, 05:50 PM	#3
Matth Flash Banner Hater Join Date: Jun 2002 Location: UK Posts: 3,426 Rep Power: 93 System Specs	I little off-track on the specific thermal power though, I'd say. The Barton core dissipates more power, but through an area that's larger = lower W/cm-sq. What matters to the heatsink overall, is the higher TOTAL. The lower value per area will mean lower loss across the thermal interface (be it pad or paste) and that is ALL. It's possible it balances, and the improvement in transfer efficiency to the heatsink counters the increased power - just depends how well a GOOD thermal paste is performing in the first place, as to how much there is to gain there.