Which motherboard company is more efficient ASUS or Gigabyte

[Apoptosis]

Going Green? Which motherboard company is more efficient, ASUS or Gigabyte?

Asus and Gigabyte recently had a spat about power consumption a couple months back that got some people wondering which motherboard was really the most energy efficient. We took a look at the real numbers to see who comes out on top and the results might just shock you. Read on to see which motherboard maker comes out on top when we try out ASUS's EPU and Gigabyte's DES equivalent.

So, in the end, who is the winner? I would have to say first off that the winner is the consumer. Less power being used means less money flying out the door to the utility company, and that is a good thing for the pocketbook, especially with gas prices headed the direction they are. I would also say that the environment is a winner. I am not going to get all politically correct. In fact, no matter what your view is, you would have to admit that using less power is a good thing. Just the fact that we are even talking about this subject is a good thing. We have a responsibility to take care of the earth, and this certainly can help.

Article Title: Going Green? Which motherboard company is more efficient, ASUS or Gigabyte?
Article URL: http://legitreviews.com/article/736/1/

[gwolfman]

Let me play the advocate here (though I think saving energy is important too).

So, taking into account that the biggest difference in power at idle is less than 20 watts (from both boards, but I'll be generous) from turbo (o/c mode) to max power savings, let's do some calculations.

Assuming you want to keep your PC on 24/7 (of which most of the time it's idling):

20 Watt Hours * 24hrs/day = 480 W*h a day

480 W*h/day * 30 days/month = 14,400 W*h = 14.4kWh

14.4kWh/month * 11.4 cents/kWh (in San Joaquin valley in California) = 164.16 cents/month = $1.64 month in total money saved

I don't know about you, but I wouldn't make any purchasing decisions based on "energy savings". That's only a saving comparable to (6) 100 watt light bulbs on for one day during the whole month. Not a whole lot, but I guess it's a step in the right direction.

But if it's on every hour of the day for a whole year, I guess it adds up to ~$20 USD. Enough for you and a friend/spouse to grab some fast food and have enough for an ice cream. :p But in reality, it was only 16 watt difference, or 80%, which would be only ~$16.

[DMB2000uk]

So it's not much when it's just one computer, but imagine when this tech takes off, and it's implemented across a whole school or company.

Dan

[gwolfman]

Oh yeah, definitely. But still, this are usually found in higher-end motherboards than what you find in a normal office/school so it'll be interesting to see how long it take before it's implemented in OEM parts (i.e. dell/HP PCs).

[gwolfman]

Either way, interesting article. Especially seeing why Asus was so angry at Gigabyte for the accusations and how unfounded they were. Again, assuming the tests done here are average results.

[Jobastion]

I notice that in the Gigabyte table, there are absolutely no changes in processor frequency between the default settings (which I assume is with DES off), and with any of the levels enabled. (That is to say, it idles at 2Ghz and ramps up to 3Ghz under load with or without DES) That leads me to believe that all of the procesor throttling seen in the Gigabyte testing is the natural Intel SpeedStep.

Combined with that, and the screenshot of "Gigabyte Dynamic Energy saving on... " having the little dot besides CPU throttling colored red... which I believe = off, that makes me wonder, was the test run without that feature enabled, or does the feature just not work?

[gummi.cezar]

I believe Gigabyte. - Asus savings are not because of EPU but FSB changing.
Read this:
http://www.tomshardware.com/news/asus-g ... ,5348.html

[DMB2000uk]

And what if they are?

They are still using less electric, and isn't that the point of all this energy saving race?

Dan

[RedDwarf]

I notice that in the Gigabyte table, there are absolutely no changes in processor frequency between the default settings (which I assume is with DES off), and with any of the levels enabled. (That is to say, it idles at 2Ghz and ramps up to 3Ghz under load with or without DES) That leads me to believe that all of the procesor throttling seen in the Gigabyte testing is the natural Intel SpeedStep.

Combined with that, and the screenshot of "Gigabyte Dynamic Energy saving on... " having the little dot besides CPU throttling colored red... which I believe = off, that makes me wonder, was the test run without that feature enabled, or does the feature just not work?

I have a wall socket energy monitoring device which my PC PSU is plugged into. My pc is a Q9450 CPU, GA-X48-DS5 Gigabyte motherboard, 2x2GB OCZ Reaper HPC 8500 DDR2, Gigabyte 512MB 3850 graphics card, 4 hard drives and 3 optical drives and 7 fans (CPU and six case fans).
I use the DES software and I am running it right now. The little red dot that you refer to beside the CPU throttling means that the feature was disabled and therefore the energy saving was not at it's maximum. But even though it refers to it as CPU throttling, it does not make any difference to the CPU speed. Disabling this does not turn off the Intel Speedstep or adjust the CPU frequency in any way. Intel Speedstep (EIST) and Enhanced Halt (C1E) have to be enabled in the Bios for DES to run or it pops up a window telling the user to enable these options in the Bios. So in a way it does not work, but enabling it does mean saving a very small amount of energy, 1 watt on my system. The Q9xxx series do not benefit so much from power saving such as DES like earlier CPU's did.

Another point to seriously consider is the fact that the review compared a DDR3 Asus motherboard with a DDR2 Gigabyte motherboard so the power usage would be expected to be higher with a motherboard using DDR2 due to the higher voltages used for the memory. So it was not really a fair playing field.

One more point to note about the review is the Dynamic Energy Saver Advanced that was shown earlier in the review does not work with the GA-X48-DS5 motherboard. I have tried it and it reports that the Bios does not support the software. The Dynamic Energy Saver Advanced is meant to be able to work with overclocking enabled so would be a much more desirable feature. As I understand it, this is only available on the newer P45 motherboards and only reduces the power phases and does not allow CPU throttling or undervolting when overclocked.

So in my opinion the review was very unfair and not a true comparison of Gigabyte and Asus motherboard energy saving.

[DMB2000uk]

First story on the Inq's hardware roundup, congrats Jason.

EVEN THOUGH Asus and Gigabyte came to hugging and kissing, Legit Reviews is picking the scab a bit and is trying to figure out which of the power saving feature sets is more efficient. You get to look at some of the gory slides freom the Gigabyte presentation – sex, lies and powerpoint. Jason thinks Asus’ offering is better, but they are both improvements over previous generation mobos, meaning the consumer wins either way. A power-full article. Catch it here.

Source: The Inq

Dan

[Wwhat]

That's only a saving comparable to (6) 100 watt light bulbs on for one day during the whole month. Not a whole lot, but I guess it's a step in the right direction.

I think that if you are into energy saving you'd use energy efficient lightbulbs and one 100 Watt light bulb would be more like 9 energy saving lightbulbs, in other words one '100 Watt lightbulb 'can light your entire house.
So in that sense the savings are quite significant, free light in your house

[bombastic]

I am surprised that this review did not compare performance of the two boards. So what if the Asus saves more power when I don't know how much performance it's sacrificing in order to do so? I think a much fairer comparison between the boards would be in terms of performance per watt.

I agree with RedDwarf regarding comparing a DDR2 board with a DDR3 board, that's not not really fair.

[pastorjay]

I notice that in the Gigabyte table, there are absolutely no changes in processor frequency between the default settings (which I assume is with DES off), and with any of the levels enabled. (That is to say, it idles at 2Ghz and ramps up to 3Ghz under load with or without DES) That leads me to believe that all of the procesor throttling seen in the Gigabyte testing is the natural Intel SpeedStep.

Combined with that, and the screenshot of "Gigabyte Dynamic Energy saving on... " having the little dot besides CPU throttling colored red... which I believe = off, that makes me wonder, was the test run without that feature enabled, or does the feature just not work?

I have a wall socket energy monitoring device which my PC PSU is plugged into. My pc is a Q9450 CPU, GA-X48-DS5 Gigabyte motherboard, 2x2GB OCZ Reaper HPC 8500 DDR2, Gigabyte 512MB 3850 graphics card, 4 hard drives and 3 optical drives and 7 fans (CPU and six case fans).
I use the DES software and I am running it right now. The little red dot that you refer to beside the CPU throttling means that the feature was disabled and therefore the energy saving was not at it's maximum. But even though it refers to it as CPU throttling, it does not make any difference to the CPU speed. Disabling this does not turn off the Intel Speedstep or adjust the CPU frequency in any way. Intel Speedstep (EIST) and Enhanced Halt (C1E) have to be enabled in the Bios for DES to run or it pops up a window telling the user to enable these options in the Bios. So in a way it does not work, but enabling it does mean saving a very small amount of energy, 1 watt on my system. The Q9xxx series do not benefit so much from power saving such as DES like earlier CPU's did.

Another point to seriously consider is the fact that the review compared a DDR3 Asus motherboard with a DDR2 Gigabyte motherboard so the power usage would be expected to be higher with a motherboard using DDR2 due to the higher voltages used for the memory. So it was not really a fair playing field.

One more point to note about the review is the Dynamic Energy Saver Advanced that was shown earlier in the review does not work with the GA-X48-DS5 motherboard. I have tried it and it reports that the Bios does not support the software. The Dynamic Energy Saver Advanced is meant to be able to work with overclocking enabled so would be a much more desirable feature. As I understand it, this is only available on the newer P45 motherboards and only reduces the power phases and does not allow CPU throttling or undervolting when overclocked.

So in my opinion the review was very unfair and not a true comparison of Gigabyte and Asus motherboard energy saving.

I appreciate all the comments. Let me try and answer a few things as related to the review.

First off, the software seemed to be working just fine as I ran through the testing of the board. The screenshots given were just provided to get a look at the software. We received no error or warnings about using it on the board, and it came packaged with the X48. Is the software running at its full potential? That is a question for Gigabyte. I don't write the software, I just test it to see if the claims they make are accurate. Making it work to its best potential is their job.

Second, as I explained on the test setup page, the vdimm for each board was set to 2.0v, so the voltage was the same for each board.

I reviewed what was available to us. Like I said in the review, I liked each board, each board offers power saving features that are good for the end user. If you like Gigabyte better, that is fine, buy the Gigabyte. It is a good board. If you like Asus better, buy the Asus, it is a good board. I have had good and bad from both companies. I have no favorites other than what works best at the moment.

[pastorjay]

I am surprised that this review did not compare performance of the two boards. So what if the Asus saves more power when I don't know how much performance it's sacrificing in order to do so? I think a much fairer comparison between the boards would be in terms of performance per watt.

I agree with RedDwarf regarding comparing a DDR2 board with a DDR3 board, that's not not really fair.

We have already reviewed the boards. The Asus board was posted a few weeks ago, and includes the performance numbers from the Gigabyte. You can see it here. http://www.legitreviews.com/article/721/1/

[gwolfman]

I believe Gigabyte. - Asus savings are not because of EPU but FSB changing.
Read this:
http://www.tomshardware.com/news/asus-g ... ,5348.html

Yeah, not a whole lot. Plus it doesn't make much of a difference, so what's the point. They're just manipulating the numbers.

[Edgy]

My Asus EPU app looks a little different than the one in your review. I worked with it a bit and set the "rocket" mode up to a 17% OC by simply moving a slider bar. On my rig, anything higher was unstable OCing that way. Also, in that mode, at idle the multipler drops to 6X. Under load it jumps to 8x - 3.1GHz (Q9450). That's pretty brain dean OCing!

The other thing, just look at those two MBs. Doesn't the Asus, with its black PCB, heat pipes and layout, just look better than the GB?

[Apoptosis]

just to give everyone a heads up... I talked to Gigabyte and they wanted to stress a few things they didn't think was fair...

1) we compared two different clock speeds and since ASUS had a lower clock speed they consumed less power, but have worse performance according to Gigabyte. Our comment back to that is that Jason tested the settings in their software and we used the lowest setting possible on each program. In power saving mode who cares what frequency it is set at as you are trying to save power and obviously don't want extreme performance...

2) They said one board having DDR2 and the other having DDR3 is also not fair even though they both were set to the same exact voltage in the BIOS. Gigabyte is going to send out a DDR3 based X48 board for us to re-test.

Looks like I'll personally be re-testing this as Gigabyte wants an update and are sending out another board just for power consumption testing.

[taltamir]

On the gigabyte DEP program there is a bar that says "CPU throttling", to the right of it there is an "on/off" button, in your pictures it is red, indicating that it was off the entire time (that, and the text and charts supports that notion). clicking that would enable CPU throttling similar to the one observed with asus. You should probably try that and adjust the review.

I no longer use the DEP on my board, it crashes. (not the system, dep itself), and it is really annoying (popping up when I turn on the computer, etc), I want it as a set it and forget it setting, not something that bothers me.

[Apoptosis]

On the gigabyte DEP program there is a bar that says "CPU throttling", to the right of it there is an "on/off" button, in your pictures it is red, indicating that it was off the entire time (that, and the text and charts supports that notion). clicking that would enable CPU throttling similar to the one observed with asus. You should probably try that and adjust the review.

I no longer use the DEP on my board, it crashes. (not the system, dep itself), and it is really annoying (popping up when I turn on the computer, etc), I want it as a set it and forget it setting, not something that bothers me.

Welcome to the forums, but I think you missed a page in the review as we actually go over what that button does on page four - http://www.legitreviews.com/article/736/4/



We did use it for testing, so the chart is correct.