Anandtech Power Consumption test - 780G v GF8200 v G35

smilingcrow · Post by **smilingcrow** » Fri Apr 18, 2008 3:58 pm

Thereâ€™s a nice review of power consumption for the 3 chipsets at idle and whilst playing HD video formats at Anandtech.

50W at idle may not initially sound that low but when you consider that GF 8200 system is idling 25W lower than the G35 system it looks stunning. Iâ€™ve managed 50W at idle with a good spec G35 system which suggests the GF 8200 would achieve very low idle power figures with a PicoPSU.

yuu · Post by **yuu** » Fri Apr 18, 2008 4:40 pm

8200 hotter as a chip, but less power for being single chip solution.
we'll see g45 65nm performance soon. p35 is 90nm chipset, it's not to be compared with 8200, even more with asus board, these are 20+ W hotter lately, or i'm wrong.

ntavlas · Post by **ntavlas** » Fri Apr 18, 2008 7:51 pm

I would like to see how the intel version of the 8200 compares when it comes out. I would imagine it would need a little more power since than it`s amd counterpart due to the separate memory controller.

smilingcrow · Post by **smilingcrow** » Fri Apr 18, 2008 11:47 pm

yuu wrote:We'll see g45 65nm performance soon. p35 is 90nm chipset, it's not to be compared with 8200, even more with asus board, these are 20+ W hotter lately, or i'm wrong.

According to the SPCR review you are wrong; itâ€™s showing idle as low as 49W without BIOS under-volting which is very good for a G35 board.

smilingcrow · Post by **smilingcrow** » Sat Apr 19, 2008 5:48 am

ntavlas wrote:I would like to see how the intel version of the 8200 compares when it comes out. I would imagine it would need a little more power since than it`s amd counterpart due to the separate memory controller.

The Intel platform really needs a low power chipset that supports HybridPower otherwise itâ€™ll be at a distinct disadvantage at idle versus AM2+.

Have Nvidia released drivers for HybridPower support yet? Hardly any cards support it currently, not even the 9600 GT, but I suppose their next range will.

Edirol · Post by **Edirol** » Sat Apr 19, 2008 12:18 pm

At the end of the article, Anandtech seems to imply that the 780G has a better performance/watt ratio. Still waiting for the last part of their review between the 3 chipsets before I make my purchase.

walkingjohn · Post by **walkingjohn** » Sat Apr 19, 2008 2:36 pm

This looks good for the 8200 vs. 780G, but it's not quite as apples-to-apples as I'd like: one board is biostar, one gigabyte; one atx, one matx; different audio, but same LAN chips; different numbers of usb headers apparently. Wish I knew how much difference that could make. 7-10W seems like a lot though.

PASware · Post by **PASware** » Sun Apr 20, 2008 4:52 am

Why did Anandtech used the Gigabyte 780G motherboard? Gigabyte motherboards aren't so power efficiÃ«nt like Asrock and Biostar

If Anandtech used the Asrock 780G then the powerconsumption would have been lower then the 8200.

smilingcrow · Post by **smilingcrow** » Sun Apr 20, 2008 6:08 am

PASware wrote:Why did Anandtech used the Gigabyte 780G motherboard? Gigabyte motherboards aren't so power efficiÃ«nt like Asrock and Biostar. If Anandtech used the Asrock 780G then the powerconsumption would have been lower then the 8200.

Would you please give a link to data backing up your analysis for people that havenâ€™t seen this comparison.

Anandtech werenâ€™t trying to compare every board on the market just giving a quick snapshot of the GF8200.

rei · Post by **rei** » Sun Apr 20, 2008 5:15 pm

have nvidia chipsets overcome older data corruption/lack of driver support/stability/compatibility issues? i'm sure these would be more serious issues than 25 watts. sorry to go off on a tangent.

smilingcrow · Post by **smilingcrow** » Mon Apr 21, 2008 6:09 am

I emailed the author about the G35 power data as it seemed disproportionately high and was skewing the results; he didnâ€™t reply but the G35/E2200 at idle data has now been reduced from 84 to 74W. Below is an extract from my email outlining why I thought the G35 data seemed wrong:

â€œThere does seem to me to be a glaring anomaly in the power data for the G35 which doesnâ€™t match data recorded by SPCR, myself and a previous review by yourself.

SPCR were measuring 56W for the Asus P5E-VM HDMI at idle albeit it with a lower spec setup:

I have measured 53W at idle for a Gigabyte GA-G33M-S2H with an E2140 (M0), 2GB (2x1) DDR2-800, Samsung 500GB SATA, Samsung SATA DVD+-RW.
This is the older G33 chipset but I canâ€™t see that the G35 at idle is going to consume much more power than the G33 and SPCR seem to confirm that.

In a previous article of yours you looked at the same Asus G35 & Gigabyte 780G boards but compared them against an Asus GF8200 board and made the following comment:

â€œAs far as power consumption goes during H.264 playback, the AMD platform averaged 106W, NVIDIA platform at 102W, and the Intel platform averaged 104W - too close to really declare a true winner.â€

oberbimbo · Post by **oberbimbo** » Wed Apr 23, 2008 10:07 am

These idle values seem ridiculous.

I have a Asus M3M78-EMH (GF8200) with a BE-2400, 1GB Kingston, Etasis EFN-300 and a 500GB Samsung SATA disk. The system idles at about 35W. And the EFN300 is a rather old design.

Rather sounds like they forgot to enable Cool'n'quiet to me.

smilingcrow · Post by **smilingcrow** » Wed Apr 23, 2008 10:14 am

oberbimbo wrote:These idle values seem ridiculous.

I have a Asus M3M78-EMH (GF8200) with a BE-2400, 1GB Kingston, Etasis EFN-300 and a 500GB Samsung SATA disk. The system idles at about 35W. And the EFN300 is a rather old design.

I agree. I wonder if the inclusion of two HD optical drives pushed the power draw up a lot?
4 sticks of RAM with the voltage bumped also adds a bit but it still seems very high.
The power supply is not quite so efficient by today's standards but it's in no way a dog.

oberbimbo · Post by **oberbimbo** » Wed Apr 23, 2008 10:21 am

I forgot to say, I also have a Benq DVD Writer in there. So that can't be it alone.

Just for completeness, with two instances of burnk7, the machine draws 93W AC BTW. (All values are on 230V)

smilingcrow · Post by **smilingcrow** » Wed Apr 23, 2008 10:27 am

oberbimbo wrote:I forgot to say, I also have a Benq DVD Writer in there. So that can't be it.

I was referring to the two optical drives that were installed that support Blu-ray burning and HD-DVD reading. The lasers are supposed to be fairly high power but at idle I imagine they should be okay. Just a thought.

oberbimbo · Post by **oberbimbo** » Thu Apr 24, 2008 5:16 am

One way or another, without Bluray it's possible to go far below their quoted numbers.

I don't currently consider BD to be of much use considering what the discs retail for, anyway.

Bembotto · Post by **Bembotto** » Fri Apr 25, 2008 3:13 am

Is the Asrock Nvidia GF8200 chipset (K10N78FullHD-hSLI R3.0) fully compatible with RmClock? What is the minimum voltage for the CPU?
Thz

yuu · Post by **yuu** » Sat Apr 26, 2008 10:07 am

http://www.tweaktown.com/articles/1400/ ... index.html

how did they do that, looks horrific

frank2003 · Post by **frank2003** » Sat Apr 26, 2008 10:15 am

yuu wrote:http://www.tweaktown.com/articles/1400/ ... index.html

how did they do that, looks horrific

Maybe I'm reading the graph out of context, the caption says:

Power Usage - Watts
(Higher is Better)

smilingcrow · Post by **smilingcrow** » Sat Apr 26, 2008 10:19 am

yuu wrote:http://www.tweaktown.com/articles/1400/ ... index.html
how did they do that, looks horrific

The site might be called tweaktown for a reason; theyâ€™re all tweaked out of theirs heads judging by that review.

â€œThe other thing to remember is that our test system is bare minimum â€“ only a 7,200RPM SATA-II single hard drive is used without CD-ROM or many cooling fans.â€

smilingcrow · Post by **smilingcrow** » Sat Apr 26, 2008 10:22 am

frank2003 wrote:Maybe I'm reading the graph out of context, the caption says:
Power Usage - Watts
(Higher is Better)

Thatâ€™s probably their general philosophy on life.

Itâ€™s probably all a joke.