Intel Core 2 Duo & Extreme Power Consumption

FrgMstr

FrgMstr

Just Plain Mean
Staff member
2FA
Joined
May 18, 1997
Messages
58,681
Intel Core 2 Duo & Extreme Power Consumption - Are you wondering if Intel’s new Core 2 Duo and Extreme are raging infernos like their predecessors? We show you what kind of wattage you can expect to draw so you know what PSU is in your future. And we throw in AMD's Energy Efficient 3800+ as well.

What we are going to do today is actually very simple. We have tested the very similar systems and measured just how much power it takes to run each system under different work loads. We do this by plugging our power supply into a device that sits between the power supply and the wall socket. This device will report to use the wattage that the system is using. We then take that data under different loads and simply report it to you here.
Click to expand...
 
I still say Prescot wasn't as hot as everyone says <_<

Glad to see Intel finally lower power consumption. Should make for some good PR with the energy crisis and all.

The low power usage should make for some good overclocking. What are the chances of an overclocking article?
 
No power saving features are being used in their idle state as many gamers turn this feature off in the BIOS due to negatively impacting full power usage with some systems.
Click to expand...

What on earth is this? Turning off power saving features like AMD's cool'n'quiet has NO impact on performance - the processor speed ramps up whenever the processor is under load - this myth has been debunked over and over again on the web. Benchmarking the idle power consumption after turning off the power saving features designed by AMD and INTEL makes these benchmarks almost totally useless.

I'm very interested in the difference in idle power consumption between these two processors and wish that you had at the very least offered two sets of idle benchmarks - one with power saving features and one without. Could you please add a set of idle power consumption numbers with the power saving features enabled??

EDIT: Also why no benchmarks whatsoever of the 2MB cache Core2 variants in any of these articles? I think most of us are choosing between a 6400 and a 6600 here and feel like you've really left a lot of us out in the cold. I frequent this site a lot and am really disappointed that the two things I was most interested in - idle power consumption & the performance difference between the 2MB and 4MB Conroe models - were either improperly covered or simply ignored. On top of this there's no overclocking results? It took you four seperate articles to say... well... nothing we haven't all seen before. Much more comprehensive non-gaming benchmarks have been floating around the web for months now. In the future either make a commitment to properly covering all aspects of a product or stick to what this site has always been good at - gaming benchmarks and overclocking results. Sorry for the rant, but I really feel like all four of these articles could've been condensed into a single one that was shorter, better written, and contained more information. Your core audience of DIY computer enthusiasts. I know you're trying to expand this by dividing your site into four seperate sections, but try to remember us. I don't give a hoot about Ghost Recon on the XBOX360, but I sure would've liked to see some budget Conroe overclocking numbers from the [H] because that's what this site used to be so good at.
 
InorganicMatter said:
I still say Prescot wasn't as hot as everyone says <_<
Click to expand...

I dunno about that...mine didn't get tamed until watercooling. Of course, it may have just been me, but the evidence is out there...
 
heh one can argue a pc is idle much more then its at and kind of load. so the AMD low idle watatge is a greeat thing imo

also it gets even better with cool and quiet on so intel has stopped cooking beef with there cpu's but amd still wins this round i think.
 
InorganicMatter said:
I still say Prescot wasn't as hot as everyone says <_<
Click to expand...

Believe it, Prescott's get damn hot. I upgraded last year from a Northwood to a Prescott, and now when I do any sort of video rendering that lasts more than a few minutes, I start getting warning beeps from the heat sensor on my motherboard, and that's with 8 fans in my case. I can't wait to build a new system around one of these Core 2 Duo CPUs.
 
JethroXP said:
Believe it, Prescott's get damn hot. I upgraded last year from a Northwood to a Prescott, and now when I do any sort of video rendering that lasts more than a few minutes, I start getting warning beeps from the heat sensor on my motherboard, and that's with 8 fans in my case. I can't wait to build a new system around one of these Core 2 Duo CPUs.
Click to expand...
Well I've got a Prescot, and while I won't deny it's not hot, it's just not as bad as people made it out to be.

Of course, that's not to say I won't be getting Core 2. Once school is paid for, I'm going to step back and budget some dough for Conroe.
 
Those numbers are seriously weird. I understand that the power consumption is that of the entire system, but how does this go with the 3.5 GHz overclocks on air? Or the TDP reported by Intel, I know we're not supposed to take that at face value, but I don't think Intel straight up lies about that stuff...
 
Donnie27 said:
You guys sure you had power C2D power savings features turned on correctly?
Click to expand...
No read the article, they actually turned them off.

As well they inserted a Low Wattage X2 in there.
 
gwai lo said:
Those numbers are seriously weird. I understand that the power consumption is that of the entire system, but how does this go with the 3.5 GHz overclocks on air? Or the TDP reported by Intel, I know we're not supposed to take that at face value, but I don't think Intel straight up lies about that stuff...
Click to expand...
These numbers are in line with "system power draws". They didn't isolate the CPU itself, it is obvious.
 
I know that and I said that, I meant the system power draw should be somewhat lower for the Conroe rigs, no?

If processor power draw is lower, the entire system power draw is lower?
 
coldpower27 said:
No read the article, they actually turned them off.

As well they inserted a Low Wattage X2 in there.
Click to expand...

:rolleyes: The question meant as a kind of sarcasm/:rolleyes:
 
Donnie27 said:
:rolleyes: The question meant as a kind of sarcasm/:rolleyes:
Click to expand...
Sorry, I can't tell just based on read of the context of the words alone, you didn't have any sort of smiley face there...

This isn't like real life where I cna hear the tone of your voice man.
wink.gif
 
coldpower27 said:
Sorry, I can't tell just based on read of the context of the words alone, you didn't have any sort of smiley face there...

This isn't like real life where I cna hear the tone of your voice man.
wink.gif
Click to expand...

No man, my bad, I screwed it up! :(
 
The HOCP numbers don't reflect what everyone else is getting.

Turning off power saving features makes little sense when testing power consumption in the real world.
 
ssnseawolf said:
The HOCP numbers don't reflect what everyone else is getting.

Turning off power saving features makes little sense when testing power consumption in the real world.
Click to expand...

Turning that off would impact teh idle numbers if you allowed enough idle time for the CPUs to scale down. As for load numbers, I have seen where leaving the power saving features on can impact this as well and was why it was turned off.

I am sorry if our content does not agree with your standards of testing. There will be plenty of others to read today. :)
 
I'm wondering if Conroe was made for energy saving features. The other sites use energy saving features and get lower wattage and higher framerates. Is their anyway you guys could try a test with energy saving features? Both of power and of games?
 
The power saving features only affect the idle consumption, and then only after the cpu has had time to clock down.

On the flip side, it cracks me up how 'into' power consumption AMD fanb0yz can get, even while their running their SLI or qSLI rig consuming 300W in vid cards alone. :rolleyes:
 
headala said:
The power saving features only affect the idle consumption, and then only after the cpu has had time to clock down.

On the flip side, it cracks me up how 'into' power consumption AMD !!!!!!s can get, even while their running their SLI or qSLI rig consuming 300W in vid cards alone. puLEEZE! :rolleyes:
Click to expand...
Yeh it's kinda amusing. Though considering a 7900 GTX SLI configuration will consume 170W on the video cards alone, with the X1900 XT crossfire consuming 220W or so.

Quad SLI on the other hand consumes wattage in between these 2 setups.

Simple put an SLI/Crossfire configuration will eat all the savings of the CPU.
 
ElMoIsEviL said:
Kyle.. no offense.. but your content is biased and doesn't agree with THE standards of testing. It's not his standards is THE. They're worldwide. You are purposely ensuring that the tests favor AMD (cast them in a better light for whatever reason). In ALL the tests released today, you're the only site that has come to these conclusions.

I am ashamed of HardOCP right now.. yes.. I feel sorry for you guys. I don't know what influence pushed you to turn into a Tom's Hardware of old.. but whatever vendetta you're on should not show up in journalistic articles.
Click to expand...

That's really uncalled for! I don't agree either but that's too much.
 
coldpower27 said:
Yeh it's kinda amusing. Though considering a 7900 GTX SLI configuration will consume 170W on the video cards alone, with the X1900 XT crossfire consuming 220W or so.

Quad SLI on the other hand consumes wattage in between these 2 setups.

Simple put an SLI/Crossfire configuration will eat all the savings of the CPU.
Click to expand...

You don't think it is a just a little strange that the only part of the tests where AMD seemly didn't do so poorly has over 400 posts? The ones where they clearly were trounced (Favorite AMD user word) has about one tenth? ;)
 
Donnie27 said:
You don't think it is a just a little strange that the only part of the tests where AMD seemly didn't do so poorly has over 400 posts? The ones where they clearly were trounced (Favorite AMD user word) has about one tenth? ;)
Click to expand...
Oh of course.
tongue.gif
There is no reason to harp on the result we know to be true.
wink.gif
It's the results that aren't in line with the methods of other websites, that people will complain about.

If they had throw in a Athlon 64x2 3800+ & Pentium D 915/945 I would have been perfectly fine with I myself. As it stands this "gaming" review hasn't told me any information which I didn't know prior till know, so it is of limited use to me.
 
The CPU idle consumption should be the same for all Core 2 chips b/c IF YOU ENABLE SPEEDSTEP (which you should) they should all be running at 1.6Ghz (6x multi) when idle.

BTW for comparisons:
Hard OCP - did not enable speedstep
Anandtehc - did enable it
Toms Hard - benchmarked with it enable & not enabled
 
Going back and running these power numbers with Speedstep and I will update the article as well when I am done.

Here are intels idle

X6800 - 172 (193 before)
E6700 - 171 (179 before)

Doing AMD next.
 
chrisf6969 said:
The CPU idle consumption should be the same for all Core 2 chips b/c IF YOU ENABLE SPEEDSTEP (which you should) they should all be running at 1.6Ghz (6x multi) when idle.
Click to expand...

Your point rings true by my numbers. :)
 
I work at at a datacenter where I have tenants (MMORPG Companies---that you have heard of and likely play their games) who are running SMP dual core optys at the moment. I asked them if they would be switching to intel in light of the new platform and lower power reqs (power is VERY expensive in a datacenter).

They told me 'no' because they said that intel simply moved the power savings from the CPU and moved it to the chipset. Please keep that in mind when comparing conroe and X2/FX. The total consumption for conroe will equake the sum of the cpu and chipset. AMD only has the CPU to worry about. Also, please know that watts are watts and heat is heat. I would rather have to cool one CPU than worry about keeping NB and CPU cool.

Thanks,

M
 
Mysogonist said:
I work at at a datacenter where I have tenants (MMORPG Companies---that you have heard of and likely play their games) who are running SMP dual core optys at the moment. I asked them if they would be switching to intel in light of the new platform and lower power reqs (power is VERY expensive in a datacenter).

They told me 'no' because they said that intel simply moved the power savings from the CPU and moved it to the chipset. Please keep that in mind when comparing conroe and X2/FX. The total consumption for conroe will equake the sum of the cpu and chipset. AMD only has the CPU to worry about. Also, please know that watts are watts and heat is heat. I would rather have to cool one CPU than worry about keeping NB and CPU cool.

Thanks,

M
Click to expand...

Not completely true.

Yes, an Intel chipset MIGHT take a little more power, but while looking at total power consumption for a system the Core 2 systems are still lower.

Unless, you're talking about Woodcrest, adding those FB Dimms (several watts each dimm) + chipset then that might be an issue.
 
Mysogonist said:
I work at at a datacenter where I have tenants (MMORPG Companies---that you have heard of and likely play their games) who are running SMP dual core optys at the moment. I asked them if they would be switching to intel in light of the new platform and lower power reqs (power is VERY expensive in a datacenter).

They told me 'no' because they said that intel simply moved the power savings from the CPU and moved it to the chipset. Please keep that in mind when comparing conroe and X2/FX. The total consumption for conroe will equake the sum of the cpu and chipset. AMD only has the CPU to worry about. Also, please know that watts are watts and heat is heat. I would rather have to cool one CPU than worry about keeping NB and CPU cool.

Thanks,

M
Click to expand...

Two things make those moot points.

1. You can't set a bare processor out on a Desktop or in a Rack and run it. AMD's main supported chipsets use the same or more power than Intel's. nVidia's are the worse BTW.

2. Whole systems results are the only real thing to look at.

Only FBDIMM makes this even close for AMD. 3rd parties will do systems with DDR2 sooner or later. AMD can't tweak the platform because it is not their's, advantage Intel.
 
coldpower27 said:
Oh of course.
tongue.gif
There is no reason to harp on the result we know to be true.
wink.gif
It's the results that aren't in line with the methods of other websites, that people will complain about.

If they had throw in a Athlon 64x2 3800+ & Pentium D 915/945 I would have been perfectly fine with I myself. As it stands this "gaming" review hasn't told me any information which I didn't know prior till know, so it is of limited use to me.
Click to expand...

I know and agree.
 
ssnseawolf said:
The HOCP numbers don't reflect what everyone else is getting.

Turning off power saving features makes little sense when testing power consumption in the real world.
Click to expand...

LOL... yea, I like this guy. So he test's for "real world" gaming performance to minimize the differece between AMD and Intel (I actually agree with this BTW), and he tests NOT for "real world" in power so he can minimize the difference between AMD and Intel. Yea, I'm think there is some pretty strong bias and/or advertising money in play here. :rolleyes:
 
Here is the answer to your conspiracty theories. The article has been updated to reflect this.

1152827378JaZngEz9pl_2_1.gif


Still this has not changed our conclusions though....
Intel has fixed their broken power consumption legacy and the Core 2 now fits inside a power envelope that will give it a very competitive performance per watt position. At the same time, however, I would highly suggest keeping your eyes on AMD’s low wattage / energy efficient processors for those projects that require a noiseless solution.
Click to expand...
 
InorganicMatter said:
Cool. Nice to see a [H]uge website like this replying to the requests of their community. Thanks.
Click to expand...

It was my mistake to leave them out. Honestly, I just always turn that stuff off by default as I have had days of testing ruined by it in the past. Just did not loose that mentality when it was obviously important that I did. Sorry, screwed up, won't do it next time.
 
coldpower27 said:
Oh of course.
tongue.gif
There is no reason to harp on the result we know to be true.
wink.gif
It's the results that aren't in line with the methods of other websites, that people will complain about.

If they had throw in a Athlon 64x2 3800+ & Pentium D 915/945 I would have been perfectly fine with I myself. As it stands this "gaming" review hasn't told me any information which I didn't know prior till know, so it is of limited use to me.
Click to expand...

Not me, I wouldn't have been fine with older Pentium D. This is a Conroe Launch and as I just posted to Kyle, I think the 3800+EE is out of place here without there being lower speed and power Conroes as well. I'm sure they'd have wacked the 3800+ silly LOL!

Then when the 4000 series launch you'll get a whole new set of figures to sort out. The FSB isn't going to hurt it that much performance wise but will help power usage system wide. Slower but still fast RAM, Slower but fast enough FS and Sytem BUS will all save power, hell it'll be almost like running a lap top.
 
I recall Kyle mentioning somewhere that the reason no overclocking results are posted is that they have not yet obtained samples of the lower end Core 2 Duo processors. I assume that this is the same reason we haven't seen these lower processors tested here.

That said, it does seem like a serious admission, since even to the average reader of [H] (who is generally a fairly knowledgable enthusiast), these results seem a bit odd. Clearly the 3800+EE is a wonderful processor for lower-end or silent applications, but due to its absence from any other tests here, how do we know how it actually performs? Would the 6300 be a better cheap/power-efficient option? If it does require more power, how much more so? In your average silence-oriented PC (say a Antec P180 stock, with a quiet power supply), will the temperature difference between the two processors actually make a difference?

HardOCP has done a fantastic job in regards to real world benchmarking of games and programs, but I would really like to see this same approach reach out into all levels of their reviews. A power review like this seems average for a tech-site, and therefore sub-par when compared to the excellence that I have come to respect and expect at HardOCP.
 
Donnie27 said:
Not me, I wouldn't have been fine with older Pentium D. This is a Conroe Launch and as I just posted to Kyle, I think the 3800+EE is out of place here without there being lower speed and power Conroes as well. I'm sure they'd have wacked the 3800+ silly LOL!

Then when the 4000 series launch you'll get a whole new set of figures to sort out. The FSB isn't going to hurt it that much performance wise but will help power usage system wide. Slower but still fast RAM, Slower but fast enough FS and Sytem BUS will all save power, hell it'll be almost like running a lap top.
Click to expand...
Umm Donnie, that post wasn't in reference to the Power Consumption article but in reference to the gaming performance article.
 
Legomancer said:
I recall Kyle mentioning somewhere that the reason no overclocking results are posted is that they have not yet obtained samples of the lower end Core 2 Duo processors. I assume that this is the same reason we haven't seen these lower processors tested here.

That said, it does seem like a serious admission, since even to the average reader of [H] (who is generally a fairly knowledgable enthusiast), these results seem a bit odd. Clearly the 3800+EE is a wonderful processor for lower-end or silent applications, but due to its absence from any other tests here, how do we know how it actually performs? Would the 6300 be a better cheap/power-efficient option? If it does require more power, how much more so? In your average silence-oriented PC (say a Antec P180 stock, with a quiet power supply), will the temperature difference between the two processors actually make a difference?

HardOCP has done a fantastic job in regards to real world benchmarking of games and programs, but I would really like to see this same approach reach out into all levels of their reviews. A power review like this seems average for a tech-site, and therefore sub-par when compared to the excellence that I have come to respect and expect at HardOCP.
Click to expand...

Because read reviews where lower end 6300 and 6400 easily showed they were faster than 3800+.

http://www.xbitlabs.com/articles/cpu/display/core2duo-e6300.html
http://www.behardware.com/articles/623-1/intel-core-2-duo-test.html

Cut to the chase,
http://www.behardware.com/articles/623-9/intel-core-2-duo-test.html

Normal 3800+

Results were very good since the Core 2 Duo E6600 and E6400 are less hungry than the Athlon 64 X2 3800+. Strangely enough, power consumption of our E6400 was equivalent to the E6600 despite an identical voltage of 1.3V. The Core 2 Duo X6800 also doesn´t have a high power consumption, because it´s barely above the Pentium 4 631 at 3 GHz, whose performances of course are at a much different level.
Click to expand...

So sticking the 3800+ in where it is absolutely clobbered, then trying to push its low power staus while it gets clobbered is *off? If this were a Low Power 4800 or FX-60EE I'd not say anything.
 
coldpower27 said:
Umm Donnie, that post wasn't in reference to the Power Consumption article but in reference to the gaming performance article.
Click to expand...

I gave several links and didn't mean just one. Each conclusion was favorible to AMD be it 3800+ making a good Quiet PC, Hold out for AMD price cuts, and even say Conroe makes P4 look like a dino when Conroe made them all (current processors) like bad LOL! Or at least that's what I meant to say.
 
You must log in or register to reply here.
Back
Top