**Official DFI LanParty UT 790FX-M2R Review/Overclock/Guide Thread**

[serial099]

So after some swapping/troubleshooting I figured out that somehow one of my ramsticks is bad. Or at least it would appear that way. Have a pair of 1gb corsiar xms2 sticks and when I put one of them in (we'll call it stick a) it boots fine but when I put stick b in (either by itself or with stick a) it stops booting and gives me a c1 error. Strange it went bad after I tried to run speed fan.

I've never experienced that where a piece of software seems to have such an effect on a piece of hardware. Going to try and tweak out my bios settings today on that board, but being I'm down to 1gb of ram atm might not get too impressive of numbers off of it.

*edit* BTW, double-checked and this one did have the 11/22 bios on it just to clear up that mystery

[Brother Esau]

I don't know how it is these days but Back in the DFI NF4 days We all avoided Corsair Ram on DFI motherboards like the plague!

[serial099]

well, this is the first time I've gotten either corsair ram or a DFI board. I figured that Corsair was a good name brand and DFI being a good name brand, so it'd mix well together.....wrong on my part. I notice you have OCZ, do those play nice with DFI? Other brands that are on the up and up?

Got a $50 gift certificate at Newegg to spend, so I guess I'll just have to put that towards some new RAM...*sigh*

[Brother Esau]

People deny it but its my belief that DFI boards are tailored towards OCZ Ram Primarily, also OCZ works very closely with DFI and I have yet to get anything but the best results with DFI/OCZ combo

Yes these FLEX 9200 on this board will do 1190mhz @ 2.12vdimm in Bios

I would recommend the FLEX 9200 or the Reaper 1050

I do not see much in the way of selection for what you might want at Egg Head as I would recommend OCZ, Mushkin, G-Skill , Crucial and from what I see with current availability you would probably fair the best with the Crucial right now for what they have at new egg

[Rammsteiner]

Those Flex's look nice yeah, problem however is the huge heatspreaders which won't fint below the TRUE.

Did some more testing with the RAM btw. I tried the 10x351 thingy with dividers and somehow my system holded very much longer, looks like AMD still loves running dividers.

Though didnt manage to hit anything near 293Mhz, more like 280 which results in 3.4Ghz...

Also a few quick tests at 280Mhz in Everest showed a huge drop in performance vs 251Mhz 4-4-4. So Im happy as it is now. Maybe an updated version of the 03/01 BIOS without the issues in it will solve it a little. Maybe more Vdimm will solve it but I dont feel like running much higher than 2.25V on 2.1V rated sticks with standard heatspreaders (not big sized ones). I do have a fan over them but still.

Time to move onto the GPU's again and do a little pencilmod I'd say. And of course wait for a better BIOS and maybe it will improve things.

[serial099]

Thanks for the recommendation. I'll poke around the egg and see what (if anything) I can come up with for a replacement. I think I want to go with a couple of 2gb sticks at PC6400 or better (probably better). The OCZ selection on there with the above restrictions is slim pickings ATM that's for sure.

[Brother Esau]

Yes this Ram with this Board has its limitations if Air Cooling that for sure! I just Barely make the grade with my FLEX / Tuniq Tower 120

[=[PULSAR]=]

I hit a very nice 305mhz on my Team Xtreem and finally hit the limitations of cpu running 1:1 (305x12) Guess it could have went higher was only @ 2.2V

[KTE]

so i posted for anyone (particularly KTE) to laugh hysterically at one of my settings and say "THAT'S UR PROBLEM !!!" because i feel a lock-up is innevitable as it has always been when running reference clocks above 234

Lockup is not inevitable unless the system is unstable and no one laughed at you neither shouted. Your CPU was unstable as is a known characteristic for Phenom to show in this way but you kept puzzling over why and I informed you, either lower it down or add volts if you really want a stable oc, because it's unstable. Known this behavior since November with Phenoms, nothing new. I get it with the new 9600 BE too, the second its over 2.712G. I need to shoot up from 1.32V to 1.392V to get it stable then, i.e. no lockups whilst idling. The more load you put on Phenom, the better it seems to do but the minute you bench memory/idle 20 mins, you'll see it freeze if you have too low volts. If I boot 2.812G at 1.3V it does the exact same.

Brother: Change of plan. Can't get hold of DFI board, tried another MSI RD790 today and the CPU didn't do any better at all, so I'm giving it a miss until B3 now.

[madfaze]

Yes this Ram with this Board has its limitations if Air Cooling that for sure! I just Barely make the grade with my FLEX / Tuniq Tower 120

what is the size of those fan stick on your mosfet?
is it runnin on a high speed (rpm)?

[Brother Esau]

http://www.newegg.com/Product/Produc...82E16835705022

[serial099]

Hey Brother,

What were you trying to link to? That product says it deactivated. (at least for me)

[Brother Esau]

Fixed

[serial099]

I assume this as a replacement to the 70x70x15 fans for the transpiper?
Currently I'm using a 70mm Yate Loon fan and it seems to working out ok and undervolted it sounds just as quiet as any other fan in my system.

http://www.performance-pcs.com/catal...ducts_id=22820

[madfaze]

I assume this as a replacement to the 70x70x15 fans for the transpiper?
Currently I'm using a 70mm Yate Loon fan and it seems to working out ok and undervolted it sounds just as quiet as any other fan in my system.

http://www.performance-pcs.com/catal...ducts_id=22820

how was the speed then (rpm)?

[Brother Esau]

I have a 70mm x 15mm AMD Heatsink Fan on my Transpiper as well

[Brother Esau]

BY Popular Request

[Rammsteiner]

Hmmm,

lol.

I decided to give my 4x1GB OCing one more try. I upped Voltage to 2.3V and used (10x350)/6 and it friggin booted up. Though couldnt get it to even finish 32M for 3 tests. Upped to 2.35V still no go. At that point I decided to stop it since D9's, at least like a year ago, liked the die after a few months of high Voltage. Though Ive no clue whether that was due to bad PCB/memory controllers (actually only saw this by people with Intel boards...).

Not taking the risk but maybe 3.4V would have made it. Though the little I could test with Everest showed good results, with untweaked subtimings 36.0 ns latency, 11.5GB read/write.

But as I said I dont really feel like to go higher than 2.35V. I know the memory wont die instantly but if I could manage to get it stable I'd be tempted to run it 24/7 while Im not so sure how long the RAM would survive. They weren't really hot to touch though at 2.35V but still.

Most likely this amount of Volt is needed as Im running 4 DIMMs.

[Rammsteiner]

Anyone played with the Drive Strenght settings for RAM yet? It used to improve OC'ing but then again I heard high Drive Strength can kill D9's fairly quick. Though Ive no clue whether 2.0x is high or 1.0x is high

[justapost]

That new Sapphire board here seems to be a clone of the DFI mobo. The Layout looks identical.
Only difference it's alittle more expensive here in germany atm.

[KTE]

Anyone played with the Drive Strenght settings for RAM yet?

Yep, it can increase oc/stability. Just have a high MaxAsyncLatency value and you should be good. If drive strength is incorrect, the system should tell you at bootup when it performs the calculations.

Also, for anyone running P95 and having it stable and later getting freezes/lockups do this: open up AOD, monitor every core. keep your eye on it and what the cores are fluctuating to. I bet if it was 100% stable and you're sure of this but you start getting freezes, there must be PLL movement, i.e. 2.6G set must be shooting up past 2.7G for a quick second and causing lockups.

Just keep an eye on your cores for a while to see if there's any large fluctuation; it'll occur at idle, not under load.

[Iketh]

Yep, it can increase oc/stability. Just have a high MaxAsyncLatency value and you should be good. If drive strength is incorrect, the system should tell you at bootup when it performs the calculations.

Also, for anyone running P95 and having it stable and later getting freezes/lockups do this: open up AOD, monitor every core. keep your eye on it and what the cores are fluctuating to. I bet if it was 100% stable and you're sure of this but you start getting freezes, there must be PLL movement, i.e. 2.6G set must be shooting up past 2.7G for a quick second and causing lockups.

Just keep an eye on your cores for a while to see if there's any large fluctuation; it'll occur at idle, not under load.

lemme understand this better... u talking about 2.6GHz overclock right? and at idle a core might speed to 2.7GHz????

what does PLL stand for?

[Mathos]

Hmmm thats interesting infor KTE. I had my 9600 BE running at 2.7Ghz stable at 1.262v. Would boot fine, ran prime 95 stable for 24 hours. Stable in all benchies, like 3dmark06 and pcmark05. But, every once in a while after restarting for unknown reasons when I would go to start up Coretemp, or Everest to monitor temps it would lock those programs and refuse to load them. Granted that was also on 1.2 Bios on the k9a2 plat, so any gain from being at 2.7 was wiped out by the TLB fix. Rolled back to 1.13 to get away from TLB fix, and ended up having to knock the OC down to 2.5 at 1.25v for full stability.

Someone on Toms hardware has a theory that part of the Phenom's issue is a possible hotspot in the die, revolving around either the Northbridge or L3 cache. I do believe that the individual core readings that coretemp shows are correct. Under full load at 2.7 my cores were staying below 31c, but CPU temp according to everest was running and staying below 40c full load on all 4 cores.. So that temp difference between the cores and CPU has to be coming from somewhere.

[Mathos]

lemme understand this better... u talking about 2.6GHz overclock right? and at idle a core might speed to 2.7GHz????

what does PLL stand for?

Phase Lock Loop circuitry. It's what works your HT Ref clock at 200Mhz. If there is a jitter in the signal that it puts out, or a spike it will effect the cores as well.

[tictac]

can you uninstall your phenom and cnq driver? hope that fixed freezing issue. oh if you open more tham one program that do i/o operation like coretemp and everest that no good. just use one program ok?