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

**Oese** · 01-20-2009, 11:36 PM

hehe i'll gonna test another cpu as well...

campbell, i dont think i can add much to the NIC latency discussion because i'm not sensitive enough. i play online mmorpg and that runs fine, ping also showing ok numbers, dont know about latency like you probably feel in fast fps...

whatever i'll care about voltage now and post here/report to rgone...

OK vcore:

1,35v set in bios gives 1,358v DMM in idle (load not measured), 1,344v cpu-z and 1,34v bios readings.

1,375v set in bios gives 1,38v DMM idle/1,394v DMM load, 1,376v cpu-z (idle/load) and 1,37v bios (idle)

1,4v set in bios gives 1,406v DMM idle/1,42v DMM load, 1,392v cpuz (idle/load) and i think 1,39 or 1,4v bios then...

so board still overvolts slightly even if not upping vcore using the %-settings, but i think well within acceptable limits. i measured a slight positive vdroop using the dmm, which is not reflected by cpu-z readings..

vdimm:

set in bios: 2,01v; DMM: first: 2,014v/later 2,03v Bios: first 2,01v/later 2,04v Everest: 2,03v (after boot up of course)

set 1,9v; DMM first 1,933v/later 1,937v; BIOS first 1,92v/later 1,93v; Everest (after boot up) 1,94v

set 2,19v: DMM first/later: 2,196v; Bios first/later: 2,19v; Everest first/later: 2,19v

So Vdimm is ok as well. Seems like after bootup things stabilize a bit, more on the higher vdimm level, but thats apparently not the case using higher vdimms of for example 2,19. Here, vdimm seems stable at max 0,006v overvoltage (thats only 6mV or +0,2% above the set level..

very small error, 2 pro mille)

When using lower vdimm, the board overvolts a little after it has warmed, but as this is the case only at lower vdimms it should be no problem and 0,03v above desired level is tolerable (about 2% error).

**bldegle2** · 01-21-2009, 06:45 AM

second one almost goes the 4gig, I refuse to pump massive volts, so I settled on 3.82gig and 1.44vcore......idles nicely @25*c, full load on this chip gets around 47*c, it scales very well to 3.9 and runs the 3.8+ with no bickering at lower vcore than the first cpu and is steady @3.8+, the first one actully boots into windows @4gig, but struggles to be comfortable (read completely stable) over 3.72gig, and at 1.525vcore.....number 2 almost gets into windows @4gig.....each one seems to have the same max clock, one handles the scaling mucho batter....

it is what it is, i like numba 2 just simply because the temps are so much more manageable, and the Delta spread is only 22*c.....

all this is with Vista Ultimate 64bit with a totally trashed install, bandaided together so many times from tweaking mishaps (system restore sometimes is not your friend, heavy tweaking can render previous backups useless).........after I check out this CPU for a few more days and completely finalize the settings, a fresh dual boot install (Vista and Win7) is in order for sure....

laterzzzz.............

**Brother Esau** · 01-21-2009, 07:08 AM

Olli , bldegle2 ...................Can the both of you have success with beyond 1.8ghz HT Set? Sometimes it works for me other times it does not.

Problems with HT above Rated 1,8ghz
With first Initial set above 1.8ghz to 2.0ghz Save / Exit computer will exit Bios but not initiate Re- Boot Sequence. Computer Needs to be Manually Powered OFF allowed 10 seconds wait interval before Manual Power Up and at that Point The Boot Sequence Initiates Properly. Sometimes Not! EZ- Clear is needed at that Point and Start over again (Sigh). The odd thing is that I have been in Windows Normal Boot from Save / Exit in Bios @ 2ghz and have had it to 2.4ghz in Windows running EVEREST Testing..............So I am lead to believe this is a Bios Issue not a CPU Specific issue on my end.

Just so you know I only Overclock via Bios Never with AMD Overdrive!

Please see if situation can be duplicated on your ends..>>>

Thank You..........Campbell

**Oese** · 01-21-2009, 08:51 AM

hey campbell,

unfortunately i cannot test this because i have non-be 920, and with that chip 2ghz ht would be higher then 1.8ghz nb so thats not possible. perhaps i'll get a be 940 in the next days (depends what stepping they have here) and i can test then...

did you try upping the nb to 2.2ghz to get it out of reach of the ht freq? perhaps that solves the problem?

**Brother Esau** · 01-21-2009, 08:56 AM

Originally Posted by Oese

hey campbell,

unfortunately i cannot test this because i have non-be 920, and with that chip 2ghz ht would be higher then 1.8ghz nb so thats not possible. perhaps i'll get a be 940 in the next days (depends what stepping they have here) and i can test then...

did you try upping the nb to 2.2ghz to get it out of reach of the ht freq? perhaps that solves the problem?

Hey, Olli yes naturally I always keep NB above HT Frequency

**Oese** · 01-21-2009, 09:23 AM

ok then..

dunno if its got to do with this: seems my chip doesnt like high ht's at all. the 9750 had no problem running 2,4g ht, the pII somehow doesnt do this...

**Astennu** · 01-21-2009, 11:04 AM

I dont know if i can put my questions here but i dont know a other good place.

I got the DFi JR 790GX M2RS mATX (cant put the a ATX board in this case) (running a Phenom II 940 BE)

First i have problems with most biosses and booting the board. Cold boot no prob but reset sometimes gives me problems with a 52 error on the display indicating a problem with the memory check. I tried different settings and other mem modules. But i keep having the same problem if i reset or reboot the system. Sometimes there is a 85 error. Something indicating a problem with the cmos check.

Anyway i'm running the latest 19-01-2009 bios (beta). But i think the OC is a bit poor. On a SB600 GA DS4 board the CPU would run 3650 MHz on 1.4v. Now i can only get 3400 @ 1.35v. The DS4 was on XP 32 Bit SP. The DFi is at vista ultimate x64 because somehow XP is giving problems with the raid drivers. I guess its possible that Vista and especially x64 is less table then the Win 32bit XP ? or is there something wrong with my board ? Acc aint working its only making the stability worse.

And my NB cant run at 2400 MHz 2200 is the max i can get. I tried 2300 but that was not 100% stable. i expected a bit more from this DFi board.Is there anyone that has some tips for me ?

**ChristmasGT** · 01-21-2009, 12:15 PM

Hey there, I didn't know if I could post in this thread or not but here goes.

I have had an 2 M2RSH's and have had issues with not being able to power the computer off, then second I lift off the power button the computer starts right back up, I RMA'd the first board, then the second board did it, so I tried different PSU's and even a Different case and nothing changed. is that a problem for anyone else? I had to RMA my second board because the PWM's started to make really high pitched noises under even a small amount of stress.

Both had the latest BIOS and I quadruple checked all my grounding and switches, Ideas?

**Koekerwauz_NL** · 01-21-2009, 01:52 PM

do you have a logitech g15 V2 keyboard.
If so, it's a know problem. Go to the dfi support forum and search your answer.

**ChristmasGT** · 01-21-2009, 04:00 PM

Wow oddly enough yes I do! tyvm for the heads up i'll check out!

**panzerchaos47** · 01-21-2009, 06:40 PM

Does anyone know how to get two Gskill DDR2-1000 running at 1066 for 9850? I copied most of the EPP #1 config and tried to increase the TRC to 42 with CPU tweaker (BIOs is limited to 41) but still get errors.
I'm testing the same ram on P35-T2RL it seems to better than 790FX-M2RS + 9850
Settings is 533 Mhz 5-5-5-15-64 (tRFC) @ 2.0v while 790FX-T2RS isn't stable on 533 Mhz 5-5-5-16-42 (tRC) @ 2.3v

**Brother Esau** · 01-21-2009, 06:59 PM

Already told you my thoughts on that. But see any way

**Oese** · 01-21-2009, 11:25 PM

try higher trfc's >= 195ns AND LOWER VOLTAGE 2.1v is the sweetspot for powerchip ic's more is contraproductive...

look at dfi club for a clue on why the onboard graphics dont work...

@ campbell: i test with a 940 now and i have no problems booting with 2,2 and 2ghz ht set in bios. did not fail once since i got the processor today morning..

**bldegle2** · 01-22-2009, 05:37 AM

Brother Esau,

Ht Link @2.6
NB @2600

I went to bed last night about 10PM, left unit on @100%, 5.5 hours later it finally spontaneously rebooted (I could tell by checking my "up" time in task manager this AM), my OS is hosed so bad from several months of bandaid repairing the event viewer is not working at all except to tell me it is not loaded and is unavailable, so I could not access the event to figure out the offending softwares/hardware problem....I noticed that when I set the HT to 2.4 and left the NB @2600, I too had problems with booting sometimes.....

Also, regarding the 1.8 HT default HT with the PII, AMD had to set the default that low to get the CPU's out to retail, it is to be improved with future pours and the AM3 steppings, I heard this directly from someone who knows about this stuff....so, it is the luck of the draw on the HT link working at higher settings on all of the PII's, some are weaker than others....

the 2.6, 2600 setting is working so far....

I do have a question regarding the AMD stepping in the bios, mine defaults to P0, read reports of changing to P1 to help with the PhenomII, does anyone have input and a quick explaination??? I changed it to P1 just now, seemed to be thinking a bit more on a reboot, will do further testing...rebooted several more times, back to normal boot times with the P1 setting....still need to understand what it does...

laterzzzzz......

**Brother Esau** · 01-22-2009, 06:55 AM

These are taken from PM sent to RGone for Taiwan:

Problems with HT above Rated 1,8ghz
With first Intial set above 1.8ghz to 2.0ghz Save / Exit computer will exit Bios but not initiate Re- Boot Sequence. Computer Needs to be Manually Powered OFF allowed 10 seconds wait interval before Manual Power Up and at that Point The Boot Sequence Initiates Properly. Sometimes Not! EZ- Clear is needed at that Point and Start over again (Sigh).

Here's some issues with this Bios in addition to what I have already stated.

There is a Definite Problem (On This Board) Flashing Bios with USB Flash Drive! Flashes Fine and is very Fast compared to Floppy But.... For some reason although the Flash went fine on 1/16/2008 as well as 12/26/2008 - 12/01/2008 each and every single time Flashing Different Bios with USB Flash Drive there is apparent USB Board issues there after such as USB Keyboard Lock Ups repetitively some times takes as many as 4 Manual Reboots to get it initiated and some Times EZ-Clear has to be Performed and even that does not work at times. I have taken the Time to Flash 12/01/2008 , 12/26/2008 1/16/2009 Bios With USB Flash Drive and each time Flash has went Fine But USB Issues are Present. Now I have also Flashed each and every Bios above mentioned with my usual Method through Floppy with Manual Syntax Parameters with Dr Flash instead of D.F.I AUTOEXEC.BAT File and each instance with all Bios Releases that I mentioned the USB Board Issues disappeared.

Now for just 1/16/2009 "Beta".......There is a Issued Present with Optimized Defaults Loaded where the Sil Image Post screen will hang and not go beyond and requires a manual restart. I usually disable this feature as I am sure that most do but it there is a Problem that exists.

Also EZ-Clear has issues Please look into it.

There is a Issue with The OCZ Ram in my sig with Tony's SPD File not Really a issues but this is a Huge annoyance and has been going on with these Bios since 5/5/2008..... Example,,,, If you enter Bios with Optimized Defauilts Loaded with this Ram and the Phenom SPD installed Bios Reads as such>>>

(1) Auto Set:
1066 DDR Value
400mhz
AUTO: Primary Timings: 5-12-3-4-2-14

(2) Now at that point if Raise Vdimm to 1.97 = Actual (1.95)

And set Manually 1066 Mode and Proper Primary Timings 5-5-5-18-23

Save / Exit once at Desktop CPUZ reads 5-7-7-30-23 >>>>>Re-Boot / Enter Bios / DRam Page Timings in Bios also Read 5-7-7-30-23

This Problem is a Annoyance!!! because to get the Proper Reading and Memset in Bios for RAM you must First Raise Vdimm to 1.97 Set Manual 1066 Mode / Leave Dram Page @ Auto Save /Exit Let the DMI Data Pool Build / Update C'Mos and Re- Boot Back into Bios at that Time and only at that Time Can you Manually configure Ram Timings where they will actually Be applied to Bios Properly with Corresponding Reading @ Desktop in CPUZ.

(Sigh) Message to long cannot submit and must send Split Message Robert.

Part 2:

Also still From Clear C'mos 2nd PCI-E Bios Menu Value is Random with either X2 Value or X4 Value

More Weirdness that has been present in these Bios Release for over Half a year Now if not a almost a Year already is that just Like the Ram Situation above the same situation is also Present when trying to set
CPU Value O.C First must Diable Cool & Quiet / Save Exit / Reboot / Build D.M.I Data Pool / Update Cmos / Re-boot Back into Bios and then configure CPU / NB / HT Overclock ....If try to do this all at once you have Wrong Voltage set in Bios ......Please Fix these issue with this its getting very old and more work then anyone should have to do to overclock a Motherboard and its been going on for almost a year now~!

Problem with HT above 1.8ghz SAVE/Exit = No Re-Boot Intiate........This is a Bios Problem as some times it works sometimes NO!

Robert,

Please pass this info onto Taiwan,

As I had stated prior in PM there is a issue with C1 error to C'mos after beyond 1.8ghz HT Link Frequency is Set Save / Exit does not initiate Reboot "Sometimes"

Well after trouble shooting and investigating this issue further I decided to see if this applied when 1.8ghz set on Ht Link Frequency and Raise HT Bus Speed with lower CPU / NB Multipliers applied and RAM taken out of the Equation . There is NO Issue what so ever in HT Link Frequency Running Past Default 1.8ghz when Raised via HT Bus! Started Process@ 210 HT Bus and all the way up to 238 HT Bus and each and every time it Reboot and Wrote to C'mos just Fine and each and every time Rebooting from Windows or Bios NO Manual Power Down or C'Mos Reset was needed either.
However ....In my Testing I have also discovered a New Issue that Exists with Bios HT Bus Will Not Work Past 237 HT Bus anything past that Results in B.S.O.D even with HT Link Frequency set @ 1.6ghz , RAM @ 400 Ratio , And CPU and NB well below already Tested Stable Clocks with same Voltage set for Stable clocks at Higher CPU / NB / HT Frequencies.

**Oese** · 01-22-2009, 07:21 AM

last issue, ht...

some people from forumdeluxx complain about issues with high ht reffreq just like you are, i simply dont now whats the cause.

did you set htt boot up clock and cpu host clock the same when you try upping the ht ref frequency?

the other thing: getting wrong vcore setting (lowest p-state vcore) in pwm section and no boot after save/exit happens to me as well some times but can be cured by simply hit reset and let it boot again. dos this work for you campbell?

**Brother Esau** · 01-22-2009, 07:33 AM

Originally Posted by Oese

last issue, ht...

some people from forumdeluxx complain about issues with high ht reffreq just like you are, i simply dont now whats the cause.

did you set htt boot up clock and cpu host clock the same when you try upping the ht ref frequency?

the other thing: getting wrong vcore setting (lowest p-state vcore) in pwm section and no boot after save/exit happens to me as well some times but can be cured by simply hit reset and let it boot again. dos this work for you campbell?

(1)....Yes Ht Ref Clock and HT Ref Boot Clock are always Same Value

(2).... No simple Restart usually takes Manual Power Down and Flipping the Switch on the Back of P.S.U to OFF Position for 10 Seconds. What ever is going on with that its not Writing to c'mos correctly and Producing C1 ERROR wich is a C'Mos Error in Post Code

**Oese** · 01-22-2009, 08:08 AM

campbell i now encountered a similar problem:

in case of too low vcore for that oc, the following does happen: reboot after bluescreen, the keyboard does not respond, no way to get into bios or restart again with ctrl-alt-del.

if i press reset then, the computer shuts down completely. i have to switch off the power supply, wait, and switch it on again to be able to boot. a certain time is needed, sometimes up to a few minutes.

I DID NOT ENCOUNTER THIS WITH MY 920!!

But similar things took place with my 9750 since i use the tagan psu. keyboard did always respond, but i had to switch off the psu and then on again after a while. I thought uit is psu related so i never had this happen with my former be quiet and i thought some kind of protection kicked in...

Also i didnt encounter the whole thing using my 920 PII.

What makes me wonder as well is that my pwms get a lot hotter with the 940 and it uses more power in idle (just a few watts more but...) and under load even though its clocked lower in idle and has same vcore assigned.

under load of course it consumes more because of the higher oc, but before my pwm's were at 44°C with the 920 @ 3640 @ 1.5v and now its at 64°C with the 940 @ "only" 3750 @ 1,47v.

I think thats a huge jump for those 100mhz...

**panzerchaos47** · 01-22-2009, 11:26 AM

Originally Posted by Oese

try higher trfc's >= 195ns AND LOWER VOLTAGE 2.1v is the sweetspot for powerchip ic's more is contraproductive...

Thanks! so far it lasted longer for 5-5-5-16-41 @ 2.1v. I did Memtest (USB) overnight for 7 hours and got total of 7 errors at test 7 and 8. This better than 100+ errors before I did that.

**Oese** · 01-22-2009, 12:04 PM

ok at least some progress... try trc 25 or 33.. 41 is too much..

**bldegle2** · 01-22-2009, 04:34 PM

Brother Esau,

I too noticed the ram timings on optimized defaults, they were really screwy, but after manually setting the timings, everything was okay....the bios did have me bufooned at first because of the memory timing weirdness, especially since it would not reboot after the bios flash, talk about heart pounding (I have a backup bios chip from my other 790 board) cleared CMOS via jumper as the EZ-Clear is non functional as you stated, saw the weirdness, changed it, then all was good.

I do all this without taking anything out of the system, I always leave my 8 gigs in and never have problems, I flash the bios even with my clocked settings, and, yes I violate cardinal rules, I focus on clocking with what I want to run and refuse to compromise to accomodate the clock, take it or leave it....

Had a few problems when first playing with the HT and NB multi, but once again, everything is smooth now, even at the 2.6HT, which surprised me because of issues with the the PII and the higher ht link......

I don't trust USB for flashing, have used the floppy drive almost religiously since day one, however, my current floppy drive is fubar, so I have to admit, I tried the 64bit version of Awdflash and twice have updated this boards' bios FROM winders, oh dread, but it worked and it worked well.....

That is about all I can report so far, the PCI standard host CPU bridge problem in Device Manager has now been corrected, and instantly my other two Sata ports (silicon image ports) 'came' back to life, thought they were goners....

laterzzzz................

**Brother Esau** · 01-22-2009, 05:30 PM

Just Flashed Back to 12/26/2008 Beta >>>>> 1/16/2009 Corrrupted my Bios Chip and I had to Re-Program it

**panzerchaos47** · 01-22-2009, 08:04 PM

Originally Posted by Oese

ok at least some progress... try trc 25 or 33.. 41 is too much..

I tested it cause errors under 30 mins on memtest.

So tried testing with max out timings
tRRD - 7
tRTP - 3*
tWR - 8*
tWTR - 7*
tRFC all to 195.0 ns and rest to auto.*

* - I tried to match EPP for Gskill and since I'm overclocking it to 1066 I max it for stability for previous 7 hour test

This time it last timed under 1 hr

I increase the CPU clock to 3 GHz it failed less than 1 hr with same timings.

I'm going test this with unganged now to see if it helps. I found out it can increase performance see if this helps. Another overnight with same timings.