Asus M3A79-T Deluxe

[leoy]

@ nanohead in the overdrive folder change the maximum vcore setting to 1 in the preferences xml.
This should give u 1.55 max.
It worked for me.

Be careful but because aod seems to add voltage when u up the voltage if u are not in 1.3v from bios (original setting in aod).

[nanohead]

@ nanohead in the overdrive folder change the maximum vcore setting to 1 in the preferences xml.
This should give u 1.55 max.
It worked for me.

Be careful but because aod seems to add voltage when u up the voltage if u are not in 1.3v from bios (original setting in aod).

Thanks leoy.... I tried, but cannot open the file (or there isn't anything in it ) I can open pagesettings.xml, but no preference.xml ?


EDIT: I figured it out. Turns out that because AOD kept crashing, it never successfully wrote the files contents. I started it, hit apply on the preference page, the shut it down (all at stock speeds to prevent crashing) and it created the contents, and I then made the change and it now shows 1.5V... thanks!!!

And here you go

[nanohead]

Well, I experimented tonight with RAID 5 on the M3A79-T, and results were somewhat mixed.

Disk performance was actually pretty decent. I did a 2+1 array, with 3 old seagate 80GB SATA 1 drives. I hit 90MB/sec transfer on HD Tune, which is actually pretty good for those old drives. I need to get 3 or 4 SATA II drives to run this. Also, I want to test the behavior of the SB750 with a RAID 5 HD disk array and a standalone SATA CD drive. This could be the wrench in the works for using RAID on this board without it being a pain in the arse

On the downside, Overclocking was MUCH flakier, which is what I was gonna look for. 3.6 was attainable, but MUCH more crash prone, and getting there was not a sure thing. So there is some weirdness with the RAID filter driver and how sensitive Vista is to hairy clock rates.

Also, AOD is a horrible as ever. On the single Raptor installation I did initially, version 2.13 installed with the black skin and worked pretty well, and it showed ACC as being set to auto. When I installed the exact same version on the RAID 5 install, it installed the red skin and worked poorly, as well as not showing ACC even existing.

Its truly a strange piece of software. I took leoy's advice about hacking the preference.xml file, and on the single Raptor install, it worked great. On the RAID 5 install, it keeps resetting itself randomly, as well as occasionally deleting itself.

What a train wreck...

[kazgirl]

Yes I think in the AOD thread, there's mention that RAID stops AOD from detecting the SB750 for some reason with AOD v2.13 and that the next release of AOD should fix this, I'm hoping that it'll come out before I geta M3A79-T mobo..... hopefully in the next couple of weeks though seems availability is still an issue here in the UK.... latest availability date seems to have been pushed back yet again to 6th Oct now, its forever changing, first it was 4th Sep, then 23rd Sep then 39th Sep, then 30th Sep, latest is 6th Oct.... its ridiculous...!!!

I'm looking into ordering from Germany now....!!!

[oldpueblo]

I can confirm this, I have 2.14 and AHCI and ACC works and the skin is black.

[merkk1]

Grats Nanohead gald to see you found your self a SB750 chip mother board you like . On my Foxconn board the ACC didnt work right with AOD 2.13 but dose work with 2.14 just like Oldpueblo said .On my old DFI board using raid and not using, i saw that my rig was able to over clock better using a single drive as to the raid set-up . I was thinking that maybe the power draw on the mother board from running the raid set-up makes it harder to keep the cpu stable , Do you think that could be happen to you ? I was thinking that it may be a problem for all the AMD boards with the high power draw from a over clock Phenom cpu . Your ASUS board dose look to have the best power set-up out of all the new board out there right now .

[Jaivan]

Just received this board today so far I'm happy with the OC results got 2.9GHz so far @1.45v NB and CPU.
That 200mhz more than my DFI so far and I think with a little tweaking i can get it to 3Ghz.
Too bad mines the poopy browm color though altogether I'm quiet satisfied with this board.

I'm also wonder if your boards overvolts your voltage settings that you set in the BIOS?
For example I put 1.45v in the BIOS and overdrive says that its at 1.45v while HWmonitor and CPU-Z says its at 1.488v.
Anybody know where to check vcore with a DMM on this board?

[Daveburt714]

Just received this board today so far I'm happy with the OC results got 2.9GHz so far @1.45v NB and CPU.
That 200mhz more than my DFI so far and I think with a little tweaking i can get it to 3Ghz.

Hey Jaivan, long time no see..
Gratz on the new board man!

I don't have this board, but I would suggest that you might get better results by lowering your V's a little bit until you find what's best, especially the IMC volts. Just a suggestion, work your way up...

[pinoyxpryde]

Just received this board today so far I'm happy with the OC results got 2.9GHz so far @1.45v NB and CPU.
That 200mhz more than my DFI so far and I think with a little tweaking i can get it to 3Ghz.
Too bad mines the poopy browm color though altogether I'm quiet satisfied with this board.

I'm also wonder if your boards overvolts your voltage settings that you set in the BIOS?
For example I put 1.45v in the BIOS and overdrive says that its at 1.45v while HWmonitor and CPU-Z says its at 1.488v.
Anybody know where to check vcore with a DMM on this board?

you should have no problems with 3ghz. I run 3.2 at 1.37v cpu and nb. I noticed that cpu-z will move the core voltage on me in windows, when I was testing at 1.45 it would jump between 1.456 to 1.479 and go back and forth

[nanohead]

I can confirm this, I have 2.14 and AHCI and ACC works and the skin is black.

Hey, where did you get 2.14 beta? I'd like to try it with the Frankenstein machine I'm hacking together

[merkk1]

it link in the AOD poll thread 3 page i think

[MrMojoZ]

http://download.amd.com/Desktop/AOD_214_Setup.exe

[nanohead]

Thanks.... Got it

[Jaivan]

Hey Jaivan, long time no see..
Gratz on the new board man!

I don't have this board, but I would suggest that you might get better results by lowering your V's a little bit until you find what's best, especially the IMC volts. Just a suggestion, work your way up...

Yeah been busy since school started...

Anyways thanks for the tips but i've already tried that 3ghz its a no go on anything lower than 1.4v even at that is crashes shortly after boot up.
I'll try lowering the IMC if my current overclock is unstable and see where that takes me.

Right now I'm testing 1.5v on the CPU @ 3Ghz and 1.45v on the IMC @ 2.2ghz.
*fingers crossed* hopefully its stable... so far 15mins and counting.

Update: Seems like it was the IMC that was holding my OC back running it at stock speeds now and CPU at 3Ghz with 1.45v

[nanohead]

Yeah been busy since school started...

Anyways thanks for the tips but i've already tried that 3ghz its a no go on anything lower than 1.4v even at that is crashes shortly after boot up.
I'll try lowering the IMC if my current overclock is unstable and see where that takes me.

Right now I'm testing 1.5v on the CPU @ 3Ghz and 1.45v on the IMC @ 2.2ghz.
*fingers crossed* hopefully its stable... so far 15mins and counting.

Hey, did you set ACC to auto per chance? I can let you know what my BIOS settings are that got me that great 3.6 clock. I'm gonna start transplanting tomorrow so the 9950 will go in at some point. Still have to work out the whole disk configuration, cuz Asus farged up the RAID/IDE/SATA config in BIOS.....

[Jaivan]

Hey, did you set ACC to auto per chance? I can let you know what my BIOS settings are that got me that great 3.6 clock. I'm gonna start transplanting tomorrow so the 9950 will go in at some point. Still have to work out the whole disk configuration, cuz Asus farged up the RAID/IDE/SATA config in BIOS.....

Yeah ACC is set to auto in the BIOS I think that my chip is just a bad clocker.
I had a 9850 that could do 2.9 on a sb600 before it died.
This is actually my third 9850 lol.

[Milans]

Why you set AUTO? Try to find optimal value. It's different for every CPU.

[justapost]

Anybody know where to check vcore with a DMM on this board?

Take a photo of the violtage regulator chip. Figure out what chip it is ISL or ST google for the chip and normaly you'll get the specs. Then look out for a pin called Vsen and try to follow the circuit path to the next soldered joint. Measure there.
If you can not follow the ciruit path simply try a few points around the Vsen pin.

Sitiin on the ASRock AOD790GX since a few days. It's sensor chip reports up to 0.1V higher voltage than the DMM.

I'd be interested how much voltage that board can handle the GBT790GX handled up to 1.725V fine at 3.3GHz, on the ASRock I primed for a few hours at 1.65V. Temps where still close to 60°C on water.

[kazgirl]

JUst a word of caution to folks in the UK who are looking to buy a 125W Phenom 9950BE....

I've been looking around for one to pair with the M3A79-T, and it seems that a lot of UK retailers are mistakenly lising, the 140W Phenom using the HD995ZXAGHBOX part #, and not the HD995ZFAGHBOX Part #.

The ZX is supposed to be the 125W 9950BE and the ZF the original 140W 9950BE, but so far a few retailers that list the ZX part # are actually selling the 140W 9950BE, I almost placed an order for one when I found a retailer with stock, but at the last second noticed that it was am original 140W 9950.

Just to let ppl know...!!!

[Jazzman]

JUst a word of caution to folks in the UK who are looking to buy a 125W Phenom 9950BE....

I've been looking around for one to pair with the M3A79-T, and it seems that a lot of UK retailers are mistakenly lising, the 140W Phenom using the HD995ZXAGHBOX part #, and not the HD995ZFAGHBOX Part #.

The ZX is supposed to be the 125W 9950BE and the ZF the original 140W 9950BE, but so far a few retailers that list the ZX part # are actually selling the 140W 9950BE, I almost placed an order for one when I found a retailer with stock, but at the last second noticed that it was am original 140W 9950.

Just to let ppl know...!!!

I have the ASUS M3A79-T mobo and have been waiting for the 125 watt 9950 BE. However, the ASUS cpu listing does not list the 125 watt version. It lists "Phenom X4 995Z (HD995ZXAJ4BGH),2.6GHz,140W,rev.B3,SocketAM2+,Quad-Core". It clearly states 140 watt.

As a result, I thought we would have to wait for a new bios version to accommodate the 125 watt 9950. BTW, from what I can find this cpu will not be available until tomorrow, Sept. 26. You can backorder from Zoomfly and others, but I don't think it will ship until tomorrow

[nanohead]

Why you set AUTO? Try to find optimal value. It's different for every CPU.

Coming from the only guy to beat me in ripping.org, I'm all ears

How did you go about doing this? I found that Auto gave me the most consistent overclocks so far. Did you do it by core? or the whole CPU itself

[pinoyxpryde]

I found my 3.4ghz OC would not even post with ACC set to auto but setting it manually made it all the way into windows.

To find out what core needs how much ACC I like to set a OC that I know will boot into windows but will fail p95 within a couple minutes. This is usually always a single core. set that core to 2% then start OCing till you can't make it into windows. At that point I add one notch to every core, this resulted in the max OC I could get at a given voltage.

2%/2%/4%/2% is what my ACC is set to

[Tom128]

I'm also wonder if your boards overvolts your voltage settings that you set in the BIOS?
For example I put 1.45v in the BIOS and overdrive says that its at 1.45v while HWmonitor and CPU-Z says its at 1.488v.

I think every board I have OCed with did that. Then again I have only used ASUS for my performance boards, maybe they do something in some way to cause it. Hard to really know what you are sending out w/o a volt meter anyways

[nanohead]

This board is just freakin awesome... Nothing else need be said.

This was a casual run, I'm starting the transplant to a proper water cooling system tomorrow. But it made it through 3DMark06 with no problem. This run was with 2 x 4870s. I'm gonna try the same with my 3 x 4850 machine and see what happens.

Just amazing to see a Phenom based system doing this..........

FYI, this looks like the fastest Phenom score in Futuremark!!!! Zowee

We're catching the Intel guys!

[Jaivan]

I found my 3.4ghz OC would not even post with ACC set to auto but setting it manually made it all the way into windows.

To find out what core needs how much ACC I like to set a OC that I know will boot into windows but will fail p95 within a couple minutes. This is usually always a single core. set that core to 2% then start OCing till you can't make it into windows. At that point I add one notch to every core, this resulted in the max OC I could get at a given voltage.

2%/2%/4%/2% is what my ACC is set to

Nice thanks for the tip gonna try that now

Update: Well I found out that my 2nd core was the weak one but it was still unstable at 2%/4%/2%/2% so I upped it to 2%/6%/2%/2% stress testing to see how it goes now.