**Unified** DFI LanParty UT 790FX-M2R-M2RS / Phenom /Data Base Thread**

[stocius]

Title Changed any input?

I like the "unified" bit, I've finally found a home on the net after being lost in the fanboy wilderness on other forums

[keithlm]

I've finally found a home on the net after being lost in the fanboy wilderness on other forums

Now that is funny. Fanboy wilderness.

Do they eat their own? (I'm guessing "YES".)

[stocius]

Now that is funny. Fanboy wilderness.

Do they eat their own? (I'm guessing "YES".)

YES and fanboys are generally 13 yr old fathers of 3 living on gov't dole, they DO . Even after stating I prefer AMD for what I believe are solid reasons and NOT bashing Intel (though I have solid reasons for doing so IMHO) I would get flamed to no bloody end and I wans't trying to convert anybody. It's like an Intel verbal jihad.

Intel is responsible for global Jihad. and I'm an infidel. I am sure of it.

The above statement is facetious. And I apologize for going off topic, I just had to vent, won't be doing it again

[justapost]

U guys must love these nazi jokes. I can laugh about a good hitler joke also, but this kraut jokes and clapping adolfs always make my day.
Please stop it and I'll remove my Howie avatar.

[stocius]

U guys must love these nazi jokes. I can laugh about a good hitler joke also, but this kraut jokes and clapping adolfs always make my day.
Please stop it and I'll remove my Howie avatar.

Wasn't making fun of Nazis (I don't like Nazis tho who does?) just rabid Intel fanboys.

[MotF Bane]

Is Everest fully safe to use now? I seem to recall something about BIOS corruption when using certain programs such as Everest; also my M2R killed two Phenoms just when starting up Everest. I'm a bit wary of using it now that this thing just started working with an M2RS and new Phenom.

[justapost]

Is Everest fully safe to use now? I seem to recall something about BIOS corruption when using certain programs such as Everest; also my M2R killed two Phenoms just when starting up Everest. I'm a bit wary of using it now that this thing just started working with an M2RS and new Phenom.

My 9500 stopped working on day one. I moved an disc from an 770 build to the M2R and the system allways powered down when i started everest. It simply did not post (stuck at C1) after a freeze (during ATI driver installation with monitoring software running).
With an fresh install updated bios and an 9600BE everest worked fine till now. But I'm using monitoring software more carefull now. I only run one at a time (exception everest and cpu-z sometimes together). That way it worked fine for weeks now.
Sandra caused problems with 9850BE's a about week ago. Maybe aGeom and Keithlm can comment on that.

[justapost]

New Beta Bios available for the M2R 4/25

http://www.dfi.com.tw/Support/Downlo...FLAG=B&SITE=US

Major Reasons of Change:
1.Fixed BIOS DRAM Voltage detecting error.

[MotF Bane]

What version of AMD OverDrive works with the M2RS? 2.07 tells me to kill myself, and 2.0.17 informs me that the processor is not AMD.

[Brother Esau]

For the AMD Overdrive ver 2.0.17 you need Sammi's .engine dll to recognize the 9750 BE you can download that from techpowerup.com just copy the engine dll into the AMD Overrive folder and overwrite the existing on ..Clapping Adolph by special request

[AlabamaCajun]

I dropped it in tonight and so far I've crashed trying to get back to 2.7G but it's voltage related are the temps are way down. I'm sure it's running 1.25V so I just need to bring it up to 1.268 again and 2.7G should be doable. I want to see 2.8G at least but hope to be able to go for 3.0G.

[justapost]

I just updated to the 4/25 beta bios and measured vdimm. I get the exact same results as with 416.

After first start at stock settings i reverted to my last testing setup booted into windows, started everest and bam power off.
Powered off the psu restarted and everest starts and works again flawless.

I currently test 2.9GHz/1.33125V CPU 2.2GHz/1.25V NB under load the DMM reades 1.376V so the ~0.04V voltage increase is still there.
With ambients around 20°C CoreDamage heats up the cpu to ~58°C at those settings.

EDIT: Made a mistake, had set 2.04V for the mem in the bios, will edit my post from yesterday. The DMM reads 2.037V in the Bios Healt Page it's read as 2.01-2.03V and in Everest 2.02-2.03V.

I just tried to boot at 3.0GHz with 1.4-1.425V CPU-VID set, both test froze during windows startup and I get idle freezes at my previous settings now 2.9GHz/1.33125V. Upping the voltages to 1.35V CPU-VID and 1.275V NB-VID does not fix those freezing issues so i dropped the cpu multi one notch for the moment.

I reverted back to 416 and still get the freezing issues. After the update i had installed Amd Power Monitor. To sort out os problems i'll reinstall winxp at 2.9GHz.
However after i reverted i loaded my last bootable settings in the bios and the mem voltage is now at 2.0V again. To compensate the overvoltage issues it seems they apply simply a one notch lower setting in 425. Therefore my yesterdays post was correct in terms of vdimm measurement.

Winxp reinstall also froze i'll go back to 2.8GHZ/1.3125V now.

[Rammsteiner]

Yes! I got 12x230 stable!

Then I tried 12x233 (very close to 2.8Ghz), but it froze everytime it entered Windows. But then I did something, as I was OC'ing the HT a little, to add a little HTT Volt and it didnt froze! I ran Prime for a good 40 minutes.

However, than all USB devices shutdown, screen went black and the system didnt respond to anything, not even power/reset buttons. So had to use PSU switch.

But why would HTT Volts help? So high wasnt the HTT running, and the HT multi was at 9. Might the CPU stabilize a little from adding HT Volts? And what's safe HTT Volt range anyway.

But well, Im gonna try and see what I can get out of my RAM at this point. Be it at DDR800 and pretty tight timings or DDR1066.

[justapost]

If I want to go save I use ~1.3V for all the chipset voltages. Ever tried WinXP-32bit Rammsteiner?

[illogical]

Off Playing a GIG in Manchester tonight, setting off now 170 mile journey so will post details later, but for now here's the screens.

[Rammsteiner]

If I want to go save I use ~1.3V for all the chipset voltages. Ever tried WinXP-32bit Rammsteiner?

Well, problem is that Ive an illegal copy of XP. And stupid enough I cant find a working SP2 patch anymore. It all the times says Ive wrong language.

As I need SP2 for a few drivers etc it's pretty pointless for me to use it anymore.

Besides that, if it's not stable under XP64 it's afterall not stable on any OS. Maybe it would take longer to crash on XP, or maybe it wouldnt at all because less registers. But I prefer it to have it 100% stable at all circumstances. And XP64 is my very first legal copy I ever bought so I'd like to make use of it as well.

Just tried a few quick DDR1066... HDD corrupted. HDD corruption sucks arse really because you need to downclock everything again and re-install. Othertimes I had BSOD's at loading Windows. I tried upto 2.31Vdimm, but a no go. Tried playing with same drive strengths settings I used on X2 which were helping a lot, but no matter, didnt help. I think it needs some more Volt on something else, but I dont know whether that should be the NB or CPU.

On the otherhand, maybe DDR1066 with 230HTT (DDR1226.67) might be a bit too much fantasy. I won't give up yet though, but I wont add more Vdimm since the board already overvolts pretty much.

Once I repaired the OS again Ill add some more Vcore and Vnb and see if that helps. I set all timings 5-5-5-18-T2, tRFC 195, unganged and everything else auto. But since these DIMM's are actually the same as their DDR1200 counterparts which are rated for 2.2Vdimm, I think I need to change something else really. After all 4 DIMM's might be stressing IMC's a lot as well.

If it still doesnt help, Ill try DDR800 3-3-3, otherwise 4-3-3.

[EDIT]
Proof my RAM rocks, although non of these achieves are mine:
http://forums.techpowerup.com/showthread.php?t=58501
But mostly all 'N' Axe RAMs get such insane results. That's why I think I most likely should change something somewhere than that my RAM just cant do it.

[EDIT #2]Proof Phenom can hit high DDR speeds as well:
http://forums.techpowerup.com/showthread.php?t=58172

Grrrr lol, wtf am I missing

[justapost]

On the otherhand, maybe DDR1066 with 230HTT (DDR1226.67) might be a bit too much fantasy. I won't give up yet though, but I wont add more Vdimm since the board already overvolts pretty much.

Once I repaired the OS again Ill add some more Vcore and Vnb and see if that helps. I set all timings 5-5-5-18-T2, tRFC 195, unganged and everything else auto. But since these DIMM's are actually the same as their DDR1200 counterparts which are rated for 2.2Vdimm, I think I need to change something else really. After all 4 DIMM's might be stressing IMC's a lot as well.

If it still doesnt help, Ill try DDR800 3-3-3, otherwise 4-3-3.

[EDIT]
Proof my RAM rocks, although non of these achieves are mine:
http://forums.techpowerup.com/showthread.php?t=58501
But mostly all 'N' Axe RAMs get such insane results. That's why I think I most likely should change something somewhere than that my RAM just cant do it.

[EDIT #2]Proof Phenom can hit high DDR speeds as well:
http://forums.techpowerup.com/showthread.php?t=58172

Grrrr lol, wtf am I missing

Few screenies with high ref HT's and the 4-07 bios.


Best I could archive was ~630MHz with 5-5-5-18 timings can't find the screenshot atm.

UPDATE: System now even froze at 2.8GHz/1.3125V during sp3 install and once again during sp3 uninstall -> new reinstall.

[Rammsteiner]

Hmmm, I know my RAM can do 500Mhz 4-4-4.

Ill test further though to get DDR1066 with 230HTT. But well, Ill say. Shouldnt try to kill the DIMM's as they're not made anymore

[Rammsteiner]

Justapost, just saw you've same stepping as me...

So Im not the only onew getting crap OC's, looks like we've got a crap stepping.

Im running 12x230 CPU, 10x230 NB, 9x 230 HT. Vcore and Vnb are 1.3V, everything else default. Maybe that helps a bit. It Primed for 6.5 hours Large FFT's. I didnt test further with lower Volts for any of those.

As I said before, 233 HTT made system freeze at Windows boot. HTT Voltage bump made it Prime for 40 minutes but then it froze very hard. But Im not giving up yet

[KTE]

Any of ya have a comparison of Sandra Hardware>Processor>scroll to bottom where it mentions Control Temp with a comparison to EVEREST Sensors?

Would be very helpful in seeing what offset DFI is applying to temp registers and to which temps exactly.

Sorry to hear of those freezes, exactly same as mine - but I ended up needing 1.36v for 2800 perfectly stable

[justapost]

@KTE I tried Sandra. In idle the control temp was 35°, everest showed 28°C cpu and 33° core temps.
I started prime95 hit reload button in sandra and the system froze. Sandra is still dangerous ground on the DFI and my 9850 already acts weird, so i'll not look any deeper into this right now.
Instead i'll verify my 2.8GHz/1.3125V 2.2GHZ/1.25V prime results first and take a break meanwhile.

BTW: DMM already reads 1.355V under load at 2.8GHz.

@Rammsteiner: You know the steppings of those two hitting 3.4GHz? Give 3-07 bios a try, was much more stable with higher ref HT's.
I successfully ran 260MHz with the 9850 and that bios. I used 2x1GB modules in the slots 2 and 4 in 800 mode.

[Rammsteiner]

No clue what steppings those high clocking Phenoms have. They were posted a few times but there are like a lot of HUGE Phenom threads now, no clue where to look.

Also, what Drive Strengths did you use to achieve such high DDR speeds? Did you have CS/ODT Pin Fine Delay on or off?

[Iketh]

well something has gone wrong with my setup (check sig, it runs wcg anytime im not rendering video)... been working flawlessly past several months, so much so that i haven't bothered with new bioses or visiting any tech sites....

board gave me a warning, it rebooted randomly on 2 occassions in the last 3 days and got stuck somewhere on bootup (i didnt open case to check the code)... hitting reset returned everything to normal

now this morning i wake up to the same situation, but it's not booting at all... open case to see 8.8. in code... i tried flipping battery upside down for 15 mins.... been messing with it for 2 hours, nothing is working

should i give up and get an RMA? EDIT: i need to do something ASAP, i have work that's not getting done

[KTE]

Thanks Achim.

As I expected, that's still the same situation as earlier DFI BIOSes, they need to sort it out because some load temps here are actually in the 70s at least - the temps are reading lower than the CPU registers are showing, by a few degrees at minimum. The diode [CPU in EVEREST] will always be lower than Core temps, if not, BIOS has a problem with giving incorrect temps. In your case, they're adding negative offsets to read lower than actual temp values, which is misleading for all and mainly for the temp controlled HSF.

I've not had a problem with Sandra in anything yet, many different builds now. Just finished the total benchmark suite with 4450e including Java. Probably a DFI thing.

Yea, you need to find fully stable settings first. Sandra will crash if it detcts instability very quickly, it has done for me for a while now.
Then you can maybe try disabling SMBus/DMI and other unnecessary options access in Hardware>Processors>Options and check if it makes a difference [should do] whenever you want, enough to retrieve Tctl values.

[Brother Esau]

New Beta Bios available for the M2R 4/25

http://www.dfi.com.tw/Support/Downlo...FLAG=B&SITE=US

Major Reasons of Change:
1.Fixed BIOS DRAM Voltage detecting error.

I will wait for the Next Beta Candidate myself as I take Tony's word on it when he says that its not as good as 4/15 4/16 Beta