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

[Brother Esau]

hey Brother Esau is it possible to get the OCZ DDR2 PC2-9200 FlexXLC to work at 1066mhz without that spd modification.

WIth the newest bios of course with the microcode update ?

No But I can send you the SPD File that I used but you use it at you're own risk of course

But I have to admit allot have joined the ranks with owning this Board and giving AMD a chance with the new line up of their technology and the thread is really starting to take shape more and more as new users join in and participate. Very exciting Indeed !

You fellas will see as anticipated D.F.I will come out swinging hard on the up and coming Bios Release just you wait and see

[justapost]

@BortherEsau: Have you tried an 11x nb multi with an 14x cpu multi? If so what voltages does the cpu need for that?
Here it's 1.3125/1.275V.
Did you do any special memory tweaking to get this 1.275V at 2.8GHz stable?

[Brother Esau]

@BortherEsau: Have you tried an 11x nb multi with an 14x cpu multi? If so what voltages does the cpu need for that?
Here it's 1.3125/1.275V.
Did you do any special memory tweaking to get this 1.275V at 2.8GHz stable?

None at all Achmin just patience and logic through simplicity is all that was needed One thing I can say as a hunch but I think this is very true when I say that HT Bus Overclocking Is Not needed and kid=nd of a moot point when everything can be clocked independently I think only a assumption at this point is that people go wrong with the Phenom and this board by being locked into the K8 Clocking frame of mind and how we applied overclocking the old way. As RGone said you must let go of the old thoughts and start with the new when approaching overclocking Phenom and I have to agree with him on that most definitely

P.S.....Does anyone have a Updated Bios Template from Bios 1/30/2008 or onward that I can have a copy of so I don't have to redue and Edit the 11/22/07 Bios Template I made for all of you nice folks in this thread way back when I was the Only Man on Planet Earth that owned this Board

JustApost.....Give me you're Bios Template or No Soup For you! JK....Send me that Template if you would Achim and I'll post my settings

[Slagathor]

...


Sorry for snivveling so much.....

I posted my issue at DFI globul support forum.

I was having a cigarette craving, one of the worst since I quit 2 weeks ago,
I will shut up now and enjoy what I got till I get this resolved.....

[Undersea]

I'm so pissed something just fell in my CPU socket:







So I used the stock cooler for a little break in because the "mini" ninja is a motherless ***** to install.

I'm going to play some 2142 tonight as the first test.

[justapost]

None at all Achmin just patience and logic through simplicity is all that was needed One thing I can say as a hunch but I think this is very true when I say that HT Bus Overclocking Is Not needed and kid=nd of a moot point when everything can be clocked independently I think only a assumption at this point is that people go wrong with the Phenom and this board by being locked into the K8 Clocking frame of mind and how we applied overclocking the old way. As RGone said you must let go of the old thoughts and start with the new when approaching overclocking Phenom and I have to agree with him on that most definitely

Well I got my first phenom end of last december and had started ocing a month before with x2's. So no old habits in the way here.
You're right ht bus overclocking is useless, but lowering em gave me no benefits here also.
The NB is not really independant from the cpu frequency. At 2.8GHz she needs ~1.3V to run at 2.2GHz. At 2.9GHz ~1.4V are required here.
I tested the 9850BE in the M3A two days ago and found that the third core is the weakes. The board has a downcore feature in the bios, so I tried it with just two cores. Result 3.0GHZ/2.2GHz at 1.35V/1.3V prime stable for six hours and no idle freeze (30mins app usage).
Currently testing an 4850e here and I miss the independant memory clock.

JustApost.....Give me you're Bios Template or No Soup For you! [/B] JK....Send me that Template if you would Achim and I'll post my settings

The only template like thing i have is for 3-07 and already in this thread. I simply CnP that stuff into code-tags.

[keithlm]

Keithlm had a corrupted bios after running sandra.

PLEASE NOTE: That turned out to be "Sandra was running when my Phenom blew up" and it was not the BIOS.

I thought it was the BIOS at the time. I didn't have another AM2 chip to put into the socket to test it out... so I RMA'ed it to Newegg. After waiting 4 days for shipping, Newegg received it but they were out of the M2R model; so they just refunded what I paid.

I found a local place that had the Gigabyte MA790FX-DS5 for the same price as Newegg. I picked that up yesterday. I had NO JOY with the Phenom. It is fried... so another RMA to Newegg for replacement. In the meanwhile I picked up a 4800+ to use for the 2 weeks or so that I'm waiting for the RMA. Yet more money. <sigh> OTOH: It is nice to have a 2.5Ghz dual core CPU so I can compare benchmarks against the Phenom. X2 versus X4 at the same clock.

DFI BONUSES: I like some of the features of the DFI board a lot better. Sound on the DFI is a Realtek ALC885 instead of the ALC889A. (Thus DFI has better quality.) Plus the DFI has a better chipset cooler, maybe a better voltage regulation, 2 network cards, and a socketed bios chip. The only thing Gigabyte has I like is the "idea" of the dual bios and the "Q-Flash". ALSO: It has a parallel printer port! MAJOR BONUS! I can resurect my Epson FX-80+. Or NOT.

I will probably end up buying a new DFI board in the future; It will be whatever replaces the M2R since I'm now not in a hurry.

[justapost]

PLEASE NOTE: That turned out to be "Sandra was running when my Phenom blew up" and it was not the BIOS.

I thought it was the BIOS at the time. I didn't have another AM2 chip to put into the socket to test it out... so I RMA'ed it to Newegg. After waiting 4 days for shipping, Newegg received it but they were out of the M2R model; so they just refunded what I paid.

I found a local place that had the Gigabyte MA790FX-DS5 for the same price as Newegg. I picked that up yesterday. I had NO JOY with the Phenom. It is fried... so another RMA to Newegg for replacement.

Oha, my fault, thank you for clarifying, thought the phenom worked in the g bt mobo. But you could not sort out that the bios in the DFI got corrupted?

In the meanwhile I picked up a 4800+ to use for the 2 weeks or so that I'm waiting for the RMA. Yet more money. <sigh> OTOH: It is nice to have a 2.5Ghz dual core CPU so I can compare benchmarks against the Phenom. X2 versus X4 at the same clock.

I bought an 4850e for a new workstation at work, too bad you don't have the DFI mobo, that would have been an nice comparison of two 2.5GHz X2's.

DFI BONUSES: I like some of the features of the DFI board a lot better. Sound on the DFI is a Realtek ALC885 instead of the ALC889A. (Thus DFI has better quality.) Plus the DFI has a better chipset cooler, maybe a better voltage regulation, 2 network cards, and a socketed bios chip. The only thing Gigabyte has I like is the "idea" of the dual bios and the "Q-Flash". ALSO: It has a parallel printer port! MAJOR BONUS! I can resurect my Epson FX-80+. Or NOT.

GBT-780G chipset also has onboard connectors for a serial and a parallel port. We have alot of special Epson printers using serial and parallel, so this will be my first choice for office workstations.

[KTE]

Thanks for the details guys
About temps, the DFI based boards are known to be very buggy in this department. They keep adding different offsets for some reason.
About multis, many report buggy multi and special add voltage changes, if any even with the 4/15 latest. I've seen the problem before.
About C1 errors with many different RAM, DFI has this problem on the X38 too.
About CPU-NB linked Achim, they are independent, both voltages and planes are unlinked - its the same as your Core voltage and HT voltage is unlinked, different source. I had verified with Sami late November on here, that one should not affect the other in voltage/speed. I have never observed any link yet to date, stable at x voltage was stable at x voltage even if I increased/decreased CPU speed. Recheck your RAM timings are equal on all runs though but what you're saying confuses me

What voltages do you guys need for your stable settings? also only list settings where system has been on for at least 8-9 hours not under full continuous load. P95/EVEREST/Sandra/OCCT/Orthos/AOD doesn't show stability for any Phenom I've had so far and many users are reporting the same across many boards, it does show load stability though. Most of my completely unstable settings can be ran for a few hours without failing when using for Firefox etc after some stability testing. From 1.4v 3069MHz all the way down to 2800MHz. Stable is daily use without problems.

[Jaivan]

Yeah of course it is Logans a good dude and did allot of hard work putting that Tutorial together for folks with out Floppy Drives to use Just remember its better to do the double flash method that I describe and all you need to do is Edit the AUTOEXEC.BAT File as Loggan says and add /E to the Switching Parameters at the end this way you will be brought back to A:\ Prompt so that you may repeat the Process.

I am just partial to my Old Dinosaur

Thanks i'll make sure to add the /E the next time I flash the BIOS and so far the 4/15/2008 BIOS is also stable at my end @ 2.9GHz with 1.39v

[Rammsteiner]

What voltages do you guys need for your stable settings? also only list settings where system has been on for at least 8-9 hours not under full continuous load. P95/EVEREST/Sandra/OCCT/Orthos/AOD doesn't show stability for any Phenom I've had so far and many users are reporting the same across many boards, it does show load stability though. Most of my completely unstable settings can be ran for a few hours without failing when using for Firefox etc after some stability testing. From 1.4v 3069MHz all the way down to 2800MHz. Stable is daily use without problems.

Im wondering if we should look for stability by changing Mhz? I guess it's more Voltage related. Just my thoughts though.

[Aussie FX]

Guys I have come acros another guy who has killed this motherboard and his 9850. He is using Linux and installed a program called lm-sensors. Apparently that program is used to monitor temperature in Linux. Unfortunately there seems to be a theme appearing here.
Just thought you'd like to know.

Also I notice complaints about AOD not working in this board with the 9850.
My Gigabyte DQ-6 and the 9850 play together great, so it's not an AOD fault.

[KTE]

Im wondering if we should look for stability by changing Mhz? I guess it's more Voltage related. Just my thoughts though.

Do you mean oc in windows whilst you're already on an oc?

Also I notice complaints about AOD not working in this board with the 9850.
My Gigabyte DQ-6 and the 9850 play together great, so it's not an AOD fault.

Eggsactly!

With the latest version, problems are with the BIOS mostly, the chips and PLLs are the same. I know on the MSI RD790 and the Abit 770 it is working great.

[Rammsteiner]

Do you mean oc in windows whilst you're already on an oc?

Nah, get a pretty stable oc (with Prime for example), then let it idle and see. If it's unstable, reboot and try different Voltages then the first attempt (one at a time of course) and see whether upping or lowering a certain Voltage helps idle stability.

[Rammsteiner]

http://www.theinquirer.net/gb/inquir...ge-phenom-9900

Very interesting read... Oh boy (yes I know it's old but it's of use though).

[aGeoM]

PLEASE NOTE: That turned out to be "Sandra was running when my Phenom blew up" and it was not the BIOS.

Coincidence or not, my 9850 died after completed all benchs in SiSoftware Sandra XII 2008 SP2 (14.20a), then I restarted the system to set 12*250 for 3000, in witch I was but at 15x200, to compare, never booted again, [C1] error.

1st i tough was the bios, since it was the last thing I did, EZ-clear CMOS, jumper clear, and then hot-flashed, no way C1 all the time, Voltage was: for CPU-VID=+105.14 and for NB=+103.7, I could not figured out what happened.

As soon I put the Phenom 9500 in, system booted fine. 9850 back again C1 error, the poor thing had the last moments of joy running Sandra.

ATM I'm playing with my hybrid DFI-Expert/Venus nForce 2200 Pro + Opteron 170, so I haven't test the new bios, don't know how long it will take to have an 9850 again...

[Aussie FX]

I wonder who is going to end up wearing the heat for this saga.

Sorry to hear the news aGeoM.

[justapost]

About CPU-NB linked Achim, they are independent, both voltages and planes are unlinked - its the same as your Core voltage and HT voltage is unlinked, different source. I had verified with Sami late November on here, that one should not affect the other in voltage/speed. I have never observed any link yet to date, stable at x voltage was stable at x voltage even if I increased/decreased CPU speed. Recheck your RAM timings are equal on all runs though but what you're saying confuses me

Man I know they use separate voltages since Juli/07. I'm just reporting what i found. Both are on the same die so they share heat at least. Also a higher cpu frequency results in a higher nb usage.
If the nb does really not need a millivolt more it whould mean that it's better to test the NB at higher cpu frequencies because instabilities occure there earlier.

[KTE]

Looks exactly like the same BIOS/MB error I get aGeoM. Try another Phenom supporting board to see if it boots in that before you think its CPU, that's what I did before but it was wrong. I doubt very much its dead, ICs don't fail like this suddenly
I don't think it's a coincidence that only DFI board is getting this sudden error in high numbers especially because of its RAM problems since the start. C1 is Memory related. My board is the only one getting it for other boards AFAIK and I have a replacement I'm going to check with the same 9850 this week. If it boots fine whilst older MSI board gives C1 Memory initialization error, then its obviously a board problem and not CPU.

And mine is from the earliest released batch of MSI RD790 v1.0. The DFI is much newer though.

Yea I know what you were saying Achim, I'm trying to look for trends. I can't test higher NB because it has separate limits. For instance, 1.3v 2380 max, whether I ran it at 1800CPU or 2900CPU. The word "stable" I guess have different meanings with users. I've tested and can still test 1.2v 2200NB stable with some stability test but that would be lying to myself since I know in general usage that fails stability and there would be no point me misleading others about it.

[Oese]

stable...

[Rammsteiner]

KTE, is the Large FFT test in Prime a good way to test CPU/IMC? After all the description says it stresses everything. On the otherhand P95 is older than Phenom and thus might actually NOT stress the CPU/IMC/core enough.

I guess Memtest is too much RAM based rather than a good way to test IMC as well.

Im wondering if we get it idle stable for ~10 hours (and load too of course), we should consider it stable? I mean, a reboot for example in 10hours isn't that of a disaster. On the otherhand, it might be foolish to think instability only occurs after a certain amount of time. It might happen as well after 3 minutes after all.

I want to know though, if Small FFT's run on CPU 100% stable, would that mean settings for the CPU alone are fine? So we got to fine tune different stuff like IMC? If not, wouldn't that basicly mean there must be some ratio/link between IMC and CPU for both Voltages and speed?

The longer I think about it the more stupid it becomes. By simple CPU multiplier clocking you can get Small FFT's stable, idle it crashes. Though as mentioned earlier, every 'program' uses it's own algorithms. If you'd manage idle and load stable, even a simple mouse click could cause an instability if it activates a different part of the CPU...

Or maybe it's a build in 'CnQ' thingy that causes crashes... Im out of clues really. Worst part is, we all always considered Prime95 to be one of the most ultimate stability tests which is being rendered completely useless now, and I really hate that because I dont know anymore how to test stability. Of course your guidelines for using Everest benchies for and such it helps a little. But I dont see how that would say whether the CPU, IMC or whatever is instable. A crash could still be caused by anything.

[illogical]

The word "stable" I guess have different meanings with users.

Well for my 2 cents, stability means, it runs my normal day to day applications in the way I'm used to for example.
Doing AVI to MPEG2 conversion using MainConcept;
Downloading using IDM manager;
Burning disc to disc;
Running a print queue;
Browsing as I'm doing now.
All of these at the same time hour after hour day after day.
Right now I have CPU 2.9 MHZ-(using x14.5), RAM @ 1066 MHZ 5.5.5.18. 2T
Everything else at stock settings, so far after 9 hours of these settings not a glitch.

[Aussie FX]

Im wondering if we get it idle stable for ~10 hours (and load too of course), we should consider it stable? I mean, a reboot for example in 10hours isn't that of a disaster.

:

I know I sure wouldn't be happy with my pc crashing every 10 hours.

Stable is when it never crashes.

[Oese]

of course especially with phenom stable becomes more complex then simply primestable...

with x2 so far an really prime stable setting always seemed to fulfil everyday tasks and gaming without any error on my side...

[Brother Esau]

Lucky Me I don't have issue one with my Phenom

Good Morning one and all and Happy Festivous

Now ....Who's gonna be a Gentleman and give me the Edited Notepad File with the Updated Bios Template?