NotebookForums.com › Forums › General Notebook Discussions › Notebook Forums - General › Turion 64 vs. Pentium-M on the desktop - total system power consumption compared
New Posts  All Forums:Forum Nav:

Turion 64 vs. Pentium-M on the desktop - total system power consumption compared

post #1 of 37
6/25/05 at 12:02pm
Thread Starter 
The following is a fair comparison of the total system power consumption for Pentium-M and Turion 64 on the desktop. The power consumption of Turion 64 has been tested using the DFI LANParty UT NF3250GB motherboard (NF3250GB), while the power consumption of Pentium-M has been tested using the Asus P4P800-SE motherboard (865PE) with the Asus CT-479 adapter - both comparable desktop systems. Here are the results:

Total System Power Consumption - Idle:

Pentium-M 2.7GHz O/C (865PE): 118W
Pentium-M 2.13GHz (865PE): 111W
Pentium-M 2.0GHz (865PE): 110W
Turion 64 2.0GHz O/C (NF3250GB): 97W
Turion 64 1.8GHz (NF3250GB): 97W
Turion 64 1.6GHz (NF3250GB): 96W

Total System Power Consumption - Full Load:

Pentium-M 2.7GHz O/C (865PE): 143W
Pentium-M 2.13GHz (865PE): 127W
Pentium-M 2.0GHz (865PE): 125W
Turion 64 2.0GHz O/C (NF3250GB): 130W
Turion 64 1.8GHz (NF3250GB): 128W
Turion 64 1.6GHz (NF3250GB): 123W

Based on these numbers the “idle” power consumption of Turion 64 is lower than that of Pentium-M, whereas the “full” power consumption of Pentium-M is slightly lower than that of Turion 64. The O/C'd Pentium-M also draws much more power (21/13W idle/full load) than the O/C'd Turion 64. I find this very interesting, don't you think?

References:
http://www.gamepc.com/labs/view_cont...70ct479&page=4
http://www.gamepc.com/labs/view_cont...urion64&page=4
Reply
Reply
post #2 of 37
6/25/05 at 1:51pm
Do you work for AMD? I swear every single one of your posts is about how AMD is better than Intel. We get it, you like AMD and hate Intel. Please stop being so damn annoying.
Reply
Reply
post #3 of 37
6/25/05 at 2:16pm
He's providing information. I don't find it particularly annoying.
Reply
Reply
post #4 of 37
6/25/05 at 2:38pm
Quote:
Originally Posted by snorre
The following is a fair comparison of the total system power consumption for Pentium-M and Turion 64 on the desktop. The power consumption of Turion 64 has been tested using the DFI LANParty UT NF3250GB motherboard (NF3250GB), while the power consumption of Pentium-M has been tested using the Asus P4P800-SE motherboard (865PE) with the Asus CT-479 adapter - both desktop systems. Here are the results:

Total System Power Consumption - Idle:

Pentium-M 2.7GHz O/C (865PE): 118W
Pentium-M 2.13GHz (865PE): 111W
Pentium-M 2.0GHz (865PE): 110W
Turion 64 2.0GHz O/C (NF3250GB): 97W
Turion 64 1.8GHz (NF3250GB): 97W
Turion 64 1.6GHz (NF3250GB): 96W

Total System Power Consumption - Full Load:

Pentium-M 2.7GHz O/C (865PE): 143W
Pentium-M 2.13GHz (865PE): 127W
Pentium-M 2.0GHz (865PE): 125W
Turion 64 2.0GHz O/C (NF3250GB): 130W
Turion 64 1.8GHz (NF3250GB): 128W
Turion 64 1.6GHz (NF3250GB): 123W

Based on these numbers the “idle” power consumption of Turion 64 is lower than that of Pentium-M, whereas the “full” power consumption of Pentium-M is slightly lower than that of Turion 64. The O/C'd Pentium-M also draws much more power (21/13W idle/full load) than the OC'd Turion 64. I find this very interesting, don't you?

References:
http://www.gamepc.com/labs/view_cont...70ct479&page=4
http://www.gamepc.com/labs/view_cont...urion64&page=4
Wouldn't it be a better comparison to use the Turion benchmarks against a P-M setup that does not use an adapter? Since the purpose is to get a good idea of Turion power use against P-M power use for notebooks it would be better to compare the P-M benches when it is in the 855GME chipset, not the 865PE chipset, correct?

I can understand using the only 2.0Ghz P-M chip against a 2.0Ghz Turion, however the large differences in total system power draw between the 855 and 865 chipsets needs to be addressed.

Using the total system power comparison with the 855 chipset, the P-M uses ~20W less than the Turion at idle and at load, while the P-M is even at higher CPU clocks:

http://www.gamepc.com/labs/view_cont...urion64&page=4
Reply
Reply
post #5 of 37
6/25/05 at 5:18pm
owned.
Reply
Reply
post #6 of 37
6/25/05 at 5:24pm
No, just a question.

snorre is too good to be pwned!
Reply
Reply
post #7 of 37
6/25/05 at 5:25pm
i stopped trusting gamepcs reviews when they released that article showing that a p-m at the same freq as a turion was like 50% faster. BS. Turned out they had a buncha problems with their testing methodology.


edit: i must be blind or something, but where did snorre get his chart from the link he provided?
Reply
Reply
post #8 of 37
6/25/05 at 6:01pm
Thread Starter 
Quote:
Originally Posted by Karma
Wouldn't it be a better comparison to use the Turion benchmarks against a P-M setup that does not use an adapter? Since the purpose is to get a good idea of Turion power use against P-M power use for notebooks it would be better to compare the P-M benches when it is in the 855GME chipset, not the 865PE chipset, correct?
Yes, but if you take a closer look you'll notice that the Turion 64 uses a NVIDIA nForce3 250Gb chipset that's like the Intel 865PE also is a desktop chipset. The adapter dosen't make any big difference here, as a matter of fact it's the only way to test a Pentium-M on a motherboard with a desktop chipset. So the test setup for both CPUs are very comparable, they both use desktop chipsets as well as the same type of memory, graphics card, hard drive, etc.

Quote:
Originally Posted by Karma
I can understand using the only 2.0Ghz P-M chip against a 2.0Ghz Turion, however the large differences in total system power draw between the 855 and 865 chipsets needs to be addressed.
Correct, but the same goes for Turion 64 too. We really don't know the differences in total system power draw between the nForce3 250Gb and a mobile chipset (e.g. ATI Radeon Xpress 200M) either. That's why a comparison between Pentium-M on the 865PE to the Turion 64 on the nForce3 250Gb really make sense here.

Quote:
Originally Posted by nark
edit: i must be blind or something, but where did snorre get his chart from the link he provided?
I got the numbers from both of the links I provided, I just rearranged them in this post they so they make more sense for a direct comparison.
Reply
Reply
post #9 of 37
6/25/05 at 8:29pm
I feel that now that the F4000's are out in force, we can get much better numbers from individuals that benchmark their notebooks with DDR400. Specifically, 3DMarks, SiSoft Sandra, SuperPi at 1 and 2 Million, etc. Compare those numbers with the Asus Z71V/Z70V, which is almost exactly the same but with a P-M instead of Turion, and DDR2-533 instead of DDR400.

It's difficult to get those numbers from the guys who have the notebooks. Apparently they're too buzy playing games on them!
Reply
Reply
post #10 of 37
6/25/05 at 8:49pm
You're posting test results for mobile processors on desktop motherboards. That doesn't tell me a thing about how they will perform in a notebook. GamePC messed up their testing once again. Fine - I now know how these processors fare against each other in a desktop. But what about heat yield, battery life, clockspeed & power management? Still don't have that info if you're relying on desktop mobos to test the units.
Reply
Reply
post #11 of 37
6/25/05 at 8:56pm
Quote:
Originally Posted by snorre
I got the numbers from both of the links I provided, I just rearranged them in this post they so they make more sense for a direct comparison.
yeah, but if u look at your second link, the 2.13 ghz p-m is putting out 101 watts at load. I doubt that the 2.0 ghz p-m is putting out more watts than the 2.13 ghz p-m. I know the chipset is different(855 vs 865), but then, that tells you how much of an impact the chipset has on the power usage. So in fact, the difference that you see between the p-m in the 865 chipset vs the nforce3 turion could be entirely due to the chipset (and probably is, the nforce3 could in fact be using half the power of the 865 chipset). I dont think you should be comparing the system's entire power usage, but the chip's individual power usage. Perhaps gamepc should learn from anand's example.


edit: word usage.
Reply
Reply
post #12 of 37
6/25/05 at 9:12pm
the chipset and setup has a lot to do with the power consumption doesn't it?
it is hard to judge which is better from those stats... they seem pretty close to me..
Reply
Reply
post #13 of 37
6/25/05 at 9:21pm
Thread Starter 
Quote:
Originally Posted by nark
yeah, but if u look at your second link, the 2.13 ghz p-m is putting out 101 watts at load. I doubt that the 2.0 ghz p-m is putting out more watts than the 2.13 ghz p-m. I know the chipset is different(855 vs 865), but then, that tells you how much of an impact the chipset has on the power usage. So in fact, the difference that you see between the p-m in the 865 chipset vs the nforce3 turion could be entirely due to the chipset (and probably is, the nforce3 could in fact be using half the power of the 865 chipset). I dont think you should be comparing the system's entire power usage, but the chip's individual power usage. Perhaps gamepc should learn from anand's example.
What chipset you use for testing obviously makes a major impact on the system's entire power usage, and if this is true for Pentium-M it's ofcourse also true for Turion 64. That's why comparing Turion 64 on a desktop chipset with Pentium-M on a mobile chipset dosen't make any sense. Since there currently dosen't exist any results of the total system power consumption of Turion 64 on a mobile chipset, the only fair comparison of power consumption that can be done today is exactly what I've presented here - Turion 64 vs. Pentium-M on desktop chipsets. Hopefully some proper reviews will be publicated soon that will shed some more light on this topic.
Reply
Reply
post #14 of 37
6/26/05 at 2:49am
Quote:
What chipset you use for testing obviously makes a major impact on the system's entire power usage, and if this is true for Pentium-M it's ofcourse also true for Turion 64. That's why comparing Turion 64 on a desktop chipset with Pentium-M on a mobile chipset dosen't make any sense. Since there currently dosen't exist any results of the total system power consumption of Turion 64 on a mobile chipset, the only fair comparison of power consumption that can be done today is exactly what I've presented here - Turion 64 vs. Pentium-M on desktop chipsets. Hopefully some proper reviews will be publicated soon that will shed some more light on this topic.
Since both CPUs were designed for MOBILE use, they should have been tested with MOBILE chipsets. It may be a "fair" comparison, but it's not a relevant comparison as it's not testing for what end-users will be looking for. Those tests do not properly measure power-management as exists with a notebook, nor do they test battery life, CPU throttling, etc. As such, the GamePC tests have no bearing on how the CPUs will stack up in a mobile environment - their target audience.
Reply
Reply
post #15 of 37
6/26/05 at 3:24am
Looks like some people are still not ready to understand that amd today makes better Cpu's than Intel.

The fact that most of the computer manufacturers are going with amd (especially turion ) is fact enough that turion grounds centrino.

1 for power 35W turion
2. for a mix centrino
3. For mobility and battery life Turion 25W

Hope it is clear enough to be understood

Ps: 2.2ghz turion is out.
Reply
Reply
post #16 of 37
6/26/05 at 10:13am
Thread Starter 
Quote:
Originally Posted by tntoak
Since both CPUs were designed for MOBILE use, they should have been tested with MOBILE chipsets. It may be a "fair" comparison, but it's not a relevant comparison as it's not testing for what end-users will be looking for. Those tests do not properly measure power-management as exists with a notebook, nor do they test battery life, CPU throttling, etc. As such, the GamePC tests have no bearing on how the CPUs will stack up in a mobile environment - their target audience.
Exactly!
Reply
Reply
post #17 of 37
6/26/05 at 1:47pm
Quote:
Originally Posted by devarshi84
Looks like some people are still not ready to understand that amd today makes better Cpu's than Intel.

The fact that most of the computer manufacturers are going with amd (especially turion ) is fact enough that turion grounds centrino.

1 for power 35W turion
2. for a mix centrino
3. For mobility and battery life Turion 25W

Hope it is clear enough to be understood

Ps: 2.2ghz turion is out.
i do believe noone is arguing that....for the desktop. But this is the mobile forum, where many users dont care that it takes 1 second longer to rar a file for the p-m over an amd 64, but instead care that they get 5 hours of batt life instead of 4 hours, and that their fans dont turn on every 30 seconds.

snorre: the problem IC with this comparison is twofold. The first problem, which has been mentioned multiple times is in reference to relevance, noone cares how the two perform on the desktop system. The second problem is that of how much power usage the two cpus actually have.....not the cpus AND the motherboard. The chipset, as you already know, plays quite a significant role in power usage. As I mentioned before, the nforce3 chipset might as well use a lot less power than the 865 chipset. A better comparison would have been to give the power usage of BOTH chipsets, and then give a comparison of OVERALL system usage. This way, we know the power usage of the cpus (which is what we care about) and then know the power usage of the rest of the system.
Reply
Reply
post #18 of 37
6/26/05 at 6:31pm
Are those benchmarks in the second link fine? They look like the same one I found a few weeks ago
Reply
Reply
post #19 of 37
6/26/05 at 6:34pm
I just wanted to post to have my quote next to michael's... but in all fairness unless there is some crossover motherboard that is exactly the same for amd and intel (wont happen) we are just going to have to wait until we get the bench's from the new higher end turion notebooks coming out now so stop arguing about the validity of the stats that snorre put up.
Reply
Reply
post #20 of 37
6/26/05 at 6:36pm
What Snorre put up is fine, but those game benchmarks look like the same ones I posted a little while back that caused controversy because the Intel 770 beat an AMD 3400

Edit: http://www.gamepc.com/labs/view_cont...urion64&page=1

yeah its the same, I was just curious

http://www.notebookforums.com/showthread.php?t=78669
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Notebook Forums - General
NotebookForums.com › Forums › General Notebook Discussions › Notebook Forums - General › Turion 64 vs. Pentium-M on the desktop - total system power consumption compared