Why does Pentium-M Desktop&AMD 64 socket 939 use so much watts throttled,gamepc test?

[TreeBark]

http://www.gamepc.com/labs/view_content.asp?id=lpcpuso&page=3
http://www.gamepc.com/labs/view_content.asp?id=lpcpuso&page=4

I am trying to understand why those systems still ~100 watts in Pentium M
and 140watts for socket 939 throttled idle tests at gamepc?

What is the the ~22 watt rating for the Amd 64 running at 1000 Mhz thottled if the system shows 140 watts. Does the motherboard and harddrive and memory still gobble energy. Does the 22 watt rating the maximum thermally wasted energy and really the processor gobbles a lot more energy than that? Where does AMD list the actual energy gobbled, rather than the thermal energy wasted?

Tomshardware wrote that the AMD socket 939 3500+ processor did not use more than 32 watts at full speed in their tests. So I am so confused as to where the other 130 watts is going to (the gamepc test shows about 160 watts with just the processor running without the video card so 160 - 32 = 128 watts going somewhere???????) Where?



Finally, why does the Mobile Pentium in the gamepc test always use 40 - 50 watts less than the AMD socket 939 processors in the gamepc tests? If tomshardware 32 watt max energy usage for the AMD64 3500+ 939 is correct, then where is this extra energy going to in th system when it has all the same components except motherboard. Is it to the nforce4 chip that uses that much energy? Or something else on the motherboard. Or does the Pentium M motherboard manage the components like harddrive and memory diffently and thus it uses somehow less energy? There is somethin fundamental that I do not understand here.

Please explain if you have some insight into where the difference in the system is?

[phextwin]

Those readings are for the entire system as i understand it.

[Merlin]

"Where does AMD list the actual energy gobbled, rather than the thermal energy wasted?"

'actual energy gobbled' is (almost) exactly = 'thermal energy wasted' ;)
each and every watt of power a cpu / any chip for that matter uses is wasted as heat (though some milliwatts might end up as rf energy)
sorry for :banana: ^^

as to why total system power consumption is still that high when at idle.. chipset, ram and drives probably use some juice even when at idle, but surely not >100W. i'm pretty sure it's because most, if not all, psu's encounter a huge drop in efficiency when they aren't loaded anywhere near their (maximum) rated power output. they use a 600W (enermax) in that gamepc test...

edit: there are some differences between the pentium-m and the a64 system, most notably of course the chipset, also the pentium-m system uses agp and only ddr333 instead of ddr400 / pcie (though i don't think pcie/agp makes that much of a difference in power usage..)

[Mikael]

I believe that the graphicscard should be responsible for 15-20W even at idle. Harddrives are antother 10-15W. Also, as said, the PSU efficiency might be as low 60% at those drains, which means that the power that the system gets is far less than those graphs show. All things accounted for, I think it looks correct.

EDIT: Although I do believe that there's something going on in that GamePC article...

"Still, the “Winchester” based Athlon64 processor consumes roughly 50-60W more than the Pentium-M at load at the same clock speed."

This is incorrect. Accounting for PSU inefficiency, the A64 seems to draw about 40W more at 2GHz. This still seems rather high, though... Looking at the "Minimum Throttled Clock Speed Idle Power Consumption" figures, we see that the slowest Pentium-M system consumes 75W. This makes for 49W if PSU efficiency is accounted for. If we withdraw 4W from this figure, it's likely that we get the power required to keep the system (excluding the CPU) running, in other words approximately 45W.

We can now compare this to the A64 system. It uses 117W in the same test, which is about 76W after the PSU. If both the Intel and AMD systems need 45W (excluding the CPU), it would mean that the throttled down Winchester at 1GHz and 1.1V needs 76-45 = 31W. I don't know about you guys, but I find that pretty hard to believe.

Anandtech's figure of 114W for a full loading 3500+ seems to agree a little more with reality. This system would draw somewhere around 74W after the PSU and would indicate that the CPU uses 30-40W (depending on the system config).

[TreeBark]

[QUOTE=Mikael]

We can now compare this to the A64 system. It uses 117W in the same test, which is about 76W after the PSU. If both the Intel and AMD systems need 45W (excluding the CPU), it would mean that the throttled down Winchester at 1GHz and 1.1V needs 76-45 = 31W. I don't know about you guys, but I find that pretty hard to believe.

QUOTE]


AMD tech spec shows AMD 3500+ at 1000MHz (throttled) has a TDP of ~20w (if I remember correctly). And people say that AMD TDP specs are some projected most ever possible design number . Why the TDP does ot match reality I do not have a clue.