Am I getting this mobo puzzle right ?...

All about them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Am I getting this mobo puzzle right ?...

Post by jones_r » Tue Aug 14, 2007 10:47 am

Please read the following information, and see if you agree with my final conclusion:

At a thread named "Two 690G builds", Frank2003 compared the following motherboards, and got the following results:

1. Biostar TA690G-AM2 (latest 605 BIOS)

2.1ghz @ 1.056V (orthos stable 24hours): idle: 36W, load: 63W
1.0ghz @ 0.88V, 31W, idle (under CCPUID control)
1.8ghz @ 0.976V, 55W, load (under CCPUID control)


2. Gigabyte GA-MA69GM-S2H (latest F3a BIOS)

2.1ghz @ 1.056V (orthos stable 24hours): idle: 52W, load: 72W
1.0ghz @ 0.88V, 44W, idle (under CCPUID control)
1.8ghz @ 0.976V, 65W, load (under CCPUID control)

Now, according to this comparison, my pick was definitely the Biostar. The Gigabyte motherboard looks *horrible* in comparison.

But this isn't the end of the story.

At a thread named "My Biostar TF7025-M2 Review: Biostar Raises the Bar", Amourek found the following results for the TF7025 mobo:

1.28Ghz, 0.850V: 43W idle, 53W during Orthos (Small FFT)
2.00Ghz, 1.025V: 47W idle, 67W during Orthos (Small FFT)
2.71Ghz, 1.275V: 57W idle, 102W during Orthos (Small FFT)

Now, this addtion, by itself, doesn't relate in any way to Frank2003's results, BUT, things become interesting when you add the following piece of information to the puzzle:

At a thread named "My Biostar TA690G Review: mATX Overclocking Gem", Amourek found the following results regarding the Biostar TA690G (which is the same mobo that Frank2003 was using), using *the same* parts that he (Amourek) used for the Biostar TF7025 review:

5.0 x 253Mhz = 1.27Ghz, 0.900V, 62W, 71W
7.0 x 253Mhz = 1.77Ghz, 1.100V, 64W, 83W
9.5 x 253Mhz = 2.40Ghz, 1.275V, 70W, 108W

*************************************

Now, by extrapolating using the information given here, and using Frank's results as reference, here is the power consumption for all three boards, on the same scale, for ~0.9V @ idle:

Gigabyte GA-MA69GM-S2H => 44W
Biostar TA690G-AM2 => 31W
Biostar TF7025-M2 => 21.5W

And again, the system that these figures relate to, is Frank's system, which consists of:

* AM2 Brisbane 4000+, 2.1ghz
* 2 x 1GB DDR2 memory
* Seagate 750GB 7200rpm SATA drive
* 120W picoPSU with 80W brick

Now, what I find almost impossible to believe, is that once you replace the seagate desktop hard drive, with a laptop hard drive, the Biostar TF7025-M2 result should drop to ~10W !!!, this is lower than what you get with VIA EPIA systems !.

papakoo
Posts: 71
Joined: Fri Aug 10, 2007 4:18 am

Post by papakoo » Tue Aug 14, 2007 12:11 pm

The difference between the seagate desktop hard drive and a laptop hard drive is not 11.5Watts...

Looking datasheet of hdd 3,5 and 2,5'' the difference is about 6W
(also i dont understand how you are coming to the last system power consumptions
Gigabyte GA-MA69GM-S2H => 44W
Biostar TA690G-AM2 => 31W
Biostar TF7025-M2 => 21.5W ) I dont think that any of them is true.

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Tue Aug 14, 2007 12:21 pm

I couldn't find the full datasheet so I assumed the Seagate takes around 10W idle, in comparison to around 1W idle for the laptop.

Anyway, even if the difference is 6W, it still means the Biostar TF7025-M2 should take 15.5W at idle (with voltage ~0.9v), which is simply amazing (especially taking into account it is so much packed with goodies).

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Tue Aug 14, 2007 12:34 pm

deleted
Last edited by jones_r on Tue Aug 14, 2007 12:44 pm, edited 1 time in total.

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Tue Aug 14, 2007 12:42 pm

also i dont understand how you are coming to the last system power consumptions

Gigabyte GA-MA69GM-S2H => 44W
Biostar TA690G-AM2 => 31W
Biostar TF7025-M2 => 21.5W

I dont think that any of them is true.
No problem, I will explain :

First of all, the following two are facts (from Frank's measures. Both mobos operating with the same parts), not extrapolation:

Gigabyte GA-MA69GM - 1.0ghz @ 0.88V: 44W
Biostar TA690G-AM2 - 1.0ghz @ 0.88V: 31W

Now, regarding how I calculated the value for the TF7025 (if it was operating with Frank's parts) :

We know that Amourek got the following values using *the same* parts for both his testings (apart from the mobo, of course):

Biostar TF7025-M2 - 1.28Ghz, 0.850V: 43W idle
Biostar TA690G - 1.27Ghz, 0.900V: 62W idle

Now there are two ways to extrapolate:

1. Amourak's Biostar TA690G value, is 31W more than what frank got, and so Amourak's TF7025 value, should also be 31W more than it would have been if it was operating with Frank's parts. This put the TF7025 at 43W - 31W = 12W.

2. Amourak's Biostar TA690G value, is double than what frank got. So Amourek's 7025 should also be double. This puts the 7025 at 21.5W if it was operating with Frank's parts.

In reality, the truth should be somewhere in the middle between #1 and #2, which gives a value much lower than the 31W that Frank got with his Biostar TA690G.

The Biostar TF7025 seems to be simply an amazing performer, power consumption wise. Far far ahead of anything else we know.

papakoo
Posts: 71
Joined: Fri Aug 10, 2007 4:18 am

Post by papakoo » Tue Aug 14, 2007 10:33 pm

My friend, you missed that frank used a special PSU configuration. He wrote that "A generic ATX PSU adds 25W to the figures. " Amourek's psu was a generic one. (actualy he tested some of them)

So at frank's Biostar TA690G-AM2 - 1.0ghz @ 0.88V: 31W if you add 25W, you have 56W idle. Amourek's is a bit more higher clock (1,27Ghz vs 1Ghz, voltage is almost the same , just +0.02V) so the 62W idle is almost identical and not double. ;)

My opinion is that we can compare results in same system when making a change on it. Different systems can also be compared but keeping in mind that maybe there is one factor making them not comparable.

Cheers

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Wed Aug 15, 2007 1:02 am

My friend, you missed that frank used a special PSU configuration. He wrote that "A generic ATX PSU adds 25W to the figures. " Amourek's psu was a generic one. (actualy he tested some of them)
What exactly have I written that led you to the conclusion that I missed that frank used a special PSU configuration ?
So at frank's Biostar TA690G-AM2 - 1.0ghz @ 0.88V: 31W if you add 25W, you have 56W idle. Amourek's is a bit more higher clock (1,27Ghz vs 1Ghz, voltage is almost the same , just +0.02V) so the 62W idle is almost identical and not double
You've missed my point. What we seem to agree on, is that if you take Amourek's Biostar TA690G-AM2 (62W) and reduce ~30W (more or less the difference that the PSU make), you get what Frank got with the Biostar (~31W). Now, all I said is that you need to give the same treatment to Amourek's TF7025 mobo, meanining, that if Amourek's got 43W with the TF7025, then you need to reduce about 30W from this value, in order to see how the TF7025 will behave in Frank's system. If you do this, you get that the TF7025, in Frank's system, will get *less* than 20W, which makes the TF7025 an amazing motherboard, far beyond anything we have today.

papakoo
Posts: 71
Joined: Fri Aug 10, 2007 4:18 am

Post by papakoo » Wed Aug 15, 2007 3:50 am

Ok, but the PSU that Amourek used on 690G was not one of them used on biostar 7025. I asked him some days ago comparing the 690's PSU with others on 7025 but he couldn't test now because systems are not in his possesion.

Allied AL-D320EXP 320W PSU:1.28Ghz, 0.850V: 48W idle, 57W during Orthos
Fortron 400W *************1.28Ghz, 0.850V: 43W idle, 53W during Orthos
Enhance 500W************1.28Ghz, 0.850V: 34W idle, 41W during Orthos

So we dont know how "bad" was the PSU used in 690G. Plus we dont know how good is the Enhance 500W.
Sometimes the nature of load of a power supply might make it act different. Only if we plug-in frank's PSU in a 7025 biostar we will be sure about the "final" power consumption. Maybe ~20W, maybe not. (we all wish it is "yes").

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Wed Aug 15, 2007 3:59 am

Ok, but the PSU that Amourek used on 690G was not one of them used on biostar 7025
This is incorrect.

Amourek used, for both the 7025 and 690G, the exact same PSU - Forton Sparkle 400W.

papakoo
Posts: 71
Joined: Fri Aug 10, 2007 4:18 am

Post by papakoo » Wed Aug 15, 2007 4:07 am

He mention it as Sparkle SPI 400W PSU on 690G and as Fortron 400W on 7025. If it is the same you are right.

Also on 690G monitor was connected through DVI and this might got some wattage. Some people says that their descrete VGA gets more hot when connected through DVI. So maybe more power consumption as well.

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Wed Aug 15, 2007 5:25 am

He mention it as Sparkle SPI 400W PSU on 690G and as Fortron 400W on 7025. If it is the same you are right.
Yes, it is the same - "Forton Sparkle 400W".
Also on 690G monitor was connected through DVI and this might got some wattage. Some people says that their descrete VGA gets more hot when connected through DVI. So maybe more power consumption as well.
First of all, wasn't the 7025 connected too via DVI ?. If not, we need someone with the Gigabyte 690G and a watt meter, to tell us whether there is a wattage change between DVI and VGA.

papakoo
Posts: 71
Joined: Fri Aug 10, 2007 4:18 am

Post by papakoo » Wed Aug 15, 2007 6:44 am

Gigabyte or Biostar TA690G?
:?

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Wed Aug 15, 2007 7:19 am

Gigabyte or Biostar TA690G?
Biostar, sorry.

So, if there is anyone with a Biostar TA690G and watt meter, we will be happy to know if there is a difference in power consumption between connecting the monitor to VGA or DVI.

If there isn't a difference, or the difference is minimal, then the Biostar TF7025 is definitely the mobo to get.

vg30et
Posts: 105
Joined: Tue Aug 22, 2006 5:14 am

Post by vg30et » Wed Aug 15, 2007 8:05 am

I admit that I haven't read thru and understood your entire thread but would like to offer some input on what I've seen while playing around with lowering idle power consumption. These may help account for some missing or extra watts you're seeing from the comparisons:

- Switching from a 3.5" hdd to a 2.5" 160GB Scorpio saved 5W. Going with a 40GB 2.5" saved another 1W
- Taking out a stick of DDR2 RAM saved 1W
- Fans at slow saved 1W over fans at medium
- Disconnecting DVD-RW saved 1W
- LAN cable disconnected saved 3W
- Windows "turn off monitor after x mins" mode saved 1W
- RMClock undervolting doesn't save all that much over stock AMD C&Q at idle. It does at 100%

Btw, The TF7050 system in my signature consumes 30W at idle.
[size=75]P180B | Asus M4A785-M | BE-2350 @ 0.912V to 1.056V | Ninja | 4GB DDR2-800 | Morex 150W Brick | OCZ Vertex 2 60GB SSD| 1x Scythe SFF21D + 1x Nexus 120mm | 29W Idle, 60W Max
[/size]

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Wed Aug 15, 2007 8:33 am

Btw, The TF7050 system in my signature consumes 30W at idle.
How is your PSU in comparison to the picoPSU, in terms of power efficiency ?

vg30et
Posts: 105
Joined: Tue Aug 22, 2006 5:14 am

Post by vg30et » Wed Aug 15, 2007 9:21 am

jones_r wrote:
Btw, The TF7050 system in my signature consumes 30W at idle.
How is your PSU in comparison to the picoPSU, in terms of power efficiency ?
No idea but it is more efficient than a Seasonic S12-380W power supply. If I recall correctly, power consumption went down ~10% when I switched over to the brick.
[size=75]P180B | Asus M4A785-M | BE-2350 @ 0.912V to 1.056V | Ninja | 4GB DDR2-800 | Morex 150W Brick | OCZ Vertex 2 60GB SSD| 1x Scythe SFF21D + 1x Nexus 120mm | 29W Idle, 60W Max
[/size]

jones_r
Posts: 140
Joined: Fri Sep 17, 2004 3:27 am

Post by jones_r » Wed Aug 15, 2007 9:29 am

No idea but it is more efficient than a Seasonic S12-380W power supply. If I recall correctly, power consumption went down ~10% when I switched over to the brick.
Well, Kentc uses a pretty much identical setup to yours, apart from the PSU, and gets 22W at idle, so I think your PSU is not THAT efficient.

vg30et
Posts: 105
Joined: Tue Aug 22, 2006 5:14 am

Post by vg30et » Wed Aug 15, 2007 10:28 am

Yep, it looks like that brick isn't all that efficient... I probably need to try a pico and 230V :)
[size=75]P180B | Asus M4A785-M | BE-2350 @ 0.912V to 1.056V | Ninja | 4GB DDR2-800 | Morex 150W Brick | OCZ Vertex 2 60GB SSD| 1x Scythe SFF21D + 1x Nexus 120mm | 29W Idle, 60W Max
[/size]

Post Reply