30-50% Less Energy Consumption for AMD Thuban

[randomkid]

Thuban may not beat the i7 930 in some benchmarks but the power savings is considerable.
Will this be reason enough for you to to choose AMD Thuban over the i7 930?

http://www.techspot.com/review/269-amd-phenom2-x6-1090T-and-1055T/page9.html

 

[Scotteq]

<shrug> Power consumption generally isn't one of the things on my personal list. It's just a (single) desktop machine, after all. And worrying about saving power on the processor side strikes me as rather nonsensical when I have a GPU setup capable of heating soup for lunch. < laughs @ the exaggeration >

Having said that, in a datacenter usage scenario it matters because the power savings is multiplied by the number of processors.


Didn't vote in the poll, since for my usage it's not a parameter I consider.

 

[notty22]

I'm not so sure there is any power savings or enough to even be noteworthy.
Looking at other reviews, that 5 % goes either way.
http://www.techspot.com/review/269-amd-phenom2-x6-1090T-and-1055T/page9.html

http://www.anandtech.com/show/3674/amds-sixcore-phenom-ii-x6-1090t-1055t-reviewed/10

http://www.legitreviews.com/article/1289/19/

 

[Shadow703793]

I don't care about power consumption. I buy CPUs based on their overclocking ability, performance (both OCed and stock), and price. A 30% energy saving is insignificant to me. For a server farm,etc that would matter. But not for me.

 

[FALC0N]

I don't care about power consumption. I buy CPUs based on their overclocking ability, performance (both OCed and stock), and price. A 30% energy saving is insignificant to me. For a server farm,etc that would matter. But not for me.

Uhhh....shadow, power usage is a component of overclocking.

 

[notty22]

http://www.guru3d.com/article/phenom-ii-x6-1055t-1090t-review/6

One word of advice, AMD processors start to really consume a lot of power once you overclock them. At 4100 MHz (all 6 cores stressed) we consumed 273 Watt, that's an additional 86 watt for an extra 900 MHz.

 

[Shadow703793]

Uhhh....shadow, power usage is a component of overclocking.

No. Heat is (and yes, power usage translates in to heat). As long as you can dissipate the heat all is good. Btw, I run WCing now, so this is not really an issue for me personally. YMMV if you run air Also, when talking about power consumption, I'm assuming full system power consumption.

 

[Wisecracker]

I'm tempted ... but I hate spending money

I would like to see some under-volted numbers on the 1055T before I make the leap.

I turned a Phenom 550 into a 950 quad at 3.1GHz+ that under-volted like a champ. Dropped power at load to 128w from 187w.

So if any of you early adopters want to crank it down it might be worth a look ...

 

[Bluescreendeath]

I'm not sure how techspot calculated their power consumption. The benchmarks I've seen from anadtech and tomshardware shows the i7 using slightly more, around the same, or even slightly less energy than the PII x6 at load.

Most of the other benchmarks I've seen shows the i7 920/930 and PII x6 neck and neck in terms of power consumption at load. That's the first chart I've seen that shows the i7 using 60 more watts. I think it is likely that techspot somehow screwed up in testing their power consumption benchmarks.

 

[EXT64]

Those i5 idles look a bit high, unless they aren't using a 5850/5870. Even OC'ed to 3.6, my i5/5850 idles at 80-85 watts.

 

[randomizer]

Xbitlabs got rather large differences as well. Note that PSU efficiency is not taken into account because system power consumption was not measured at the wall but "past" the PSU (as they put it).

 

[jimmysmitty]

I'm not sure how techspot calculated their power consumption. The benchmarks I've seen from anadtech and tomshardware shows the i7 using slightly more, around the same, or even slightly less energy than the PII x6 at load.

http://images.anandtech.com/graphs/amdphenomiix6_042610231918/22638.png

http://media.bestofmicro.com/E/K/245468/original/Power Consumption.png

Most of the other benchmarks I've seen shows the i7 920/930 and PII x6 neck and neck in terms of power consumption at load. That's the first chart I've seen that shows the i7 using 60 more watts. I think it is likely that techspot somehow screwed up in testing their power consumption benchmarks.

Thats what I am wondering as well. One or two show Core i7 using way more and most show it either a bit less, even or a bit more than Thuban.

I guess we will need to wait and see a few more reviews. Maybe THG or someone will do a in depth power consumption review because I want to know which one is right.

If they do use that much less, its a big jump. Deneb didn't use much less than Core i7 and in some cases more.

 

[randomizer]

It comes down to the system components used and to a greater extent how they tested it. Not all sites get a simple kill-a-watt and base their review on that.

 

[enzo matrix]

The power consumption savings isn't but the heat savings from the reviews I have read certainly is. So I can't honestly vote "yes" even though they are related.

 

[jimmysmitty]

It comes down to the system components used and to a greater extent how they tested it. Not all sites get a simple kill-a-watt and base their review on that.

True but most are using the same triple channle Asus mobo, HD5870 and 500GB Seagate/WD drive. I doubt it would make that much of a difference. More RAM will for Core i7 say 6GB vs 12GB but still. Doesn't make sense.

 

[randomizer]

Most sites probably don't account for PSU efficiency when taking readings from the wall either.