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

[Brother Esau]

Thank you JWilson Yah Polygon contacted me a number of months ago to test a Mod Bios for the Board but I just did not have the time

[justapost]

Another suicide run at 3.4GHz 1.6V * 100,2%. I rearanged watercooling and have ~8°C better cpu temps now.

No one on hwbot yet?

[Final8ty]

Another suicide run at 3.4GHz 1.6V * 100,2%.

How come you got FSB & not HT
And cache Level 3 2048KB & not 2 MBytes
with cpuz 1.46

[justapost]

How come you got FSB & not HT
And cache Level 3 2048KB & not 2 MBytes
with cpuz 1.46

It's a screenshot from the validation webpage, cpu-z shows HT (see previous page). I could not capture a screenshot because the machine powered off during validation file upload.

[Final8ty]

It's a screenshot from the validation webpage, cpu-z shows HT (see previous page). I could not capture a screenshot because the machine powered off during validation file upload.

was it a PSU shut down or MOBO CPU heat.
The Mobo CPU version is softer most of the time

[justapost]

was it a PSU shut down or MOBO CPU heat.
The Mobo CPU version is softer most of the time

My Odin is on RMA so I can't say if it is the psu. I don't think heat is an issue with 1.575V i had ~65°C running prime95 at 3.2GHz (failed after a minute).
The system powered off and booted again, after overheating poweroff the system stays off i think.

[illogical]

Well my 9850BE died for no apparent reason.
I had installed a Digital TV tuner card and on the reboot after driver installation my monitor didn't trigger, had the dreaded C1 error.
Swap the RAM, no change, threw in a 4400x2 posted fine, put the 9850 in the board from which the 4400 had been removed, no post.
Overclockers UK have now had my dead Phenom 9850BE since 25th June, apparently it's with AMD, now AMD can't tell me anything without a RMA number and Overclockers won't give me the number as they say it's private .
How the hell they deduce that a number relative to them sending my property to a 3rd party can be private beats me.
Their attitude on the phone stinks, their email replies are glib and robotic
Never ever will I use them again, I'll be sticking 100% with Aria UK from now on.

On another note has anyone any idea when we might see some commercial 790 boards with SB700.?

[Oese]

In Germany the new DFI LP UT 790FX-M2RS with 750sb is already listed at alternate and I called them yesterday, hearing they should be available this month!!

[illogical]

In Germany the new DFI LP UT 790FX-M2RS with 750sb is already listed at alternate and I called them yesterday, hearing they should be available this month!!

Can you give me a link to the site, I would like to try to pre-order one, I don't mind paying extra for shipping etc.

[illogical]

Can you give me a link to the site, I would like to try to pre-order one, I don't mind paying extra for shipping etc.

The Sb600 is doing my head in, everytime I load a disc it hogs the whole system until the disc has been recognised or whatever, this is regarless of SATA or IDE they are equally as bad.
Sorry this was intended as an edit to the previous post, obviously I clicked the wrong tag, I used to being able to delete such double posts, says I looking around for the Admin controls LOL.

[iocedmyself]

Ok, to whomever asked about my 4400x2 toledo core, no i did not mean 2.6ghz. I meant 3.6ghz. I would not express my grief over it's frying if it was only a 400mhz OC lol. 1400mhz however is something to get misty eyed over. 1400mhz on air cooling is something to lock yourself in a dark windowless room listening to the Cure while you cut yourself giving control over to full blown emo mode.

I actually have a scucide screenie of it at 3,783mhz but i think it was at 1.55v or something obcenely high to be running without water. So if i was a little more obsessive i may have been able to get it 4ghz....which i was actually planning on attempting after my Phenom rig was built if the chip were ok. Ugh now i am sad again.

Anyway

I have not come across anyone else with this stepping, and i'm only about 98% sure that it's BPIW and not BPLW. Either way if anyone else has a stepping with either combo i'd love to know what the expierience has been. Knowing little about the AM2+ platform regarding OCing and seeing that people have had chips fry from increasing the RAM and NB voltages i'm a little nervous about pushing the limits.

DFI LP UT 790FX M2R
JAAFB AA0812BPIW
2x2Gig GiEL DDR2 8600 @ 1066mhz 5-5-5-15 2T 2.07V
Saphire 4870 core 786mhz Mem 1090mhz

Basically have just fooled with multi's at this point, both because up until about a week ago i was still running a 939 platform and don't know skite about the cpu/htt/nb multi's and clock relationships...which makes me a sad panda.

However i've managed 3ghz (200x15) at stock 1.3v stable through two full runs each in 3DMark01/03/05/06, Vantage, PCMark04/05 and one hour and a half PCMark Vantage. It was prime stable at 3.1ghz, though as i posted previously NB was toasty and it seemed a little twitchy despite making it through prime. I managed to hit 3.2 (200x16) stock volt but windows crashed within 5 minutes. Other then setting the ram to 1066 and enabling PCI-E Gen 2 x16 in the bios everything is untouched there. So i've high hopes. I also need to beef up the NB cooler a bit as the 53c idle doesn't make me too comfortable.

So despite my reading dozens of posts i'm still at a loss of how everything ties in when OC. So please, anyone that has any advice on Voltages, Multi's Ram timing for the Giel 8600 2x2gig, any suggestions at all i would greatly apprieciate it.

[Brother Esau]

Ok, to whomever asked about my 4400x2 toledo core, no i did not mean 2.6ghz. I meant 3.6ghz. I would not express my grief over it's frying if it was only a 400mhz OC lol. 1400mhz however is something to get misty eyed over. 1400mhz on air cooling is something to lock yourself in a dark windowless room listening to the Cure while you cut yourself giving control over to full blown emo mode.
.........You gonna be alright there Brother???
I actually have a scucide screenie of it at 3,783mhz but i think it was at 1.55v or something obcenely high to be running without water. So if i was a little more obsessive i may have been able to get it 4ghz....which i was actually planning on attempting after my Phenom rig was built if the chip were ok. Ugh now i am sad again.

Anyway

I have not come across anyone else with this stepping, and i'm only about 98% sure that it's BPIW and not BPLW. Either way if anyone else has a stepping with either combo i'd love to know what the expierience has been. Knowing little about the AM2+ platform regarding OCing and seeing that people have had chips fry from increasing the RAM and NB voltages i'm a little nervous about pushing the limits.

DFI LP UT 790FX M2R
JAAFB AA0812BPIW
2x2Gig GiEL DDR2 8600 @ 1066mhz 5-5-5-15 2T 2.07V
Saphire 4870 core 786mhz Mem 1090mhz

Basically have just fooled with multi's at this point, both because up until about a week ago i was still running a 939 platform and don't know skite about the cpu/htt/nb multi's and clock relationships...which makes me a sad panda.

However i've managed 3ghz (200x15) at stock 1.3v stable through two full runs each in 3DMark01/03/05/06, Vantage, PCMark04/05 and one hour and a half PCMark Vantage. It was prime stable at 3.1ghz, though as i posted previously NB was toasty and it seemed a little twitchy despite making it through prime. I managed to hit 3.2 (200x16) stock volt but windows crashed within 5 minutes. Other then setting the ram to 1066 and enabling PCI-E Gen 2 x16 in the bios everything is untouched there. So i've high hopes. I also need to beef up the NB cooler a bit as the 53c idle doesn't make me too comfortable.

So despite my reading dozens of posts i'm still at a loss of how everything ties in when OC. So please, anyone that has any advice on Voltages, Multi's Ram timing for the Giel 8600 2x2gig, any suggestions at all i would greatly apprieciate it.

READ THIS........http://www.xtremesystems.org/forums/...d.php?t=167977 IN ITS ENTIRETY! Then Print this Page http://www.xtremesystems.org/forums/...01&postcount=5 and Set clocks by FID & DID DO NOT USE THE Multi set!

[justapost]

Update: 3GHz need 1.3625V to be stable under winxp 32bit, 1.35V failed prime95 after two hours.

Seems the chip also runs stable at ~3.19GHz 1.475V. I get cpu temps around 65°C max here.
Under linux 3.19GHz are not stable at 1.475V and not even at 1.525V.
With higher voltages the system powers off during the first minute running coredamage with temps slight above 70°C (power off temp 85°C). I can run wPrime fine up to 1.575V ~3350MHz with higher voltages I was not able to get into windows with higher clocks and voltages today.
If I run coredamage here (~1.55V) the whole system requires 414W.

wPrime 1024m 5:51.910
http://www.hwbot.org/result.do?resultId=752769

wPrime 24m 10.940s
http://www.hwbot.org/result.do?resultId=752516

Both better than all 9850be's.

[iocedmyself]

Well thank you brother esau, those links cleared things up for me quite nicely. Actually kind of embarrassed that it wasn't clicking sooner, and honestly it's a bit disheartening that OC phenoms is basically limited to multi's, vcore and tightening memory timings.

So after catching on as to what's what, i rigged up my TEC AC unit that i'm building. Ducted up the CPU to the cold box in which my TECs reside (which provide a spiffy 300w of cooling with a 25c temp drop below ambient) and got to cranking up the volts.

I finally ended up at 3.4ghz (17x200) @ 1.425v. I attribute this to the fact that having the TEC cooler ducted up put the idle temp down around 15c, but as the load temp still was up around 50c within about 2 minutes of benching, i have no desire to keep it there until i reassess my CPU cooling arrangements.

Which is good since the temp started climbing as soon as the TEC unit was turned off, crashing into BSOD far to quickly for my liking.

So for the time being i'll keep it down at 1.275v @ 3ghz.

It's really tempting to get a dually just to OC...that's probably 60% of the fun in having an AMD rig. But for now i'm content with the 68% performance boost this upgrade has given me.

Now, despite it being irrelivent to the thread...i came across my scuicide screenie from the 4400x2 toledo so i think it's only right to pay homage to it.



So very close to the 4ghz mark, though despite it a scuicide screenie it was one achieved with just a big a$$ heatsink and deafening fans galore.

Of all the AMD chips i've put past 3ghz on air cooling, this one is the most painful.

IT COULD HAVE BEEN SO MUCH MORE!

:::goes to happy place, where midget cowboy's and ninja's ride off into the sunset on their Great Dane steeds to fight in the drunken yo-yo yielding monkies in the wyoming midget/monkey wars:::

Ok, i think i've milked that enough.

Before you all scorn me as psychotic...just picture it.

Yeah...that's right it's funny. Bet no one ellse has thought of that. Though it's probably best not to read to deeply into why that is.

With that, i'm off to switch cases and stabilze voltages and replace heatsinks. So thanks for the help, ill check back later in hopes someone has discovered a hidden OC method in which i can become obsessed

[justapost]

However i've managed 3ghz (200x15) at stock 1.3v stable through two full runs each in 3DMark01/03/05/06, Vantage, PCMark04/05 and one hour and a half PCMark Vantage. It was prime stable at 3.1ghz, though as i posted previously NB was toasty and it seemed a little twitchy despite making it through prime. I managed to hit 3.2 (200x16) stock volt but windows crashed within 5 minutes. Other then setting the ram to 1066 and enabling PCI-E Gen 2 x16 in the bios everything is untouched there. So i've high hopes. I also need to beef up the NB cooler a bit as the 53c idle doesn't make me too comfortable.

Nice chip you have. An stock amd cooler fan on the pwm heatsink should keep your chipset and pwm temps below 50°C. Have you tried higher volts than 1.3V?

[justapost]

I was courious about the boards voltage limits, so I configured the system for 3GHz and applied voltages starting at 1.5V and increased in 0.025V steps till it powered off running coredamage. The limit was 1.525V. 1.5375V powered off after 2 mins.

Next thing I was interested was temps vs. voltages.
I measured power consumption and temps after 3min's coredamage.

CPU VID/Power Consumption Load/Core Temp idle/Core Temp load/Cpu Temp idle/Cpu Temp Load.

1.525V, 425W, 28°C, 51°C, 28°C, 65°C
1.500V, 408W, 28°C, 50°C, 27°C, 63°C
1.475V, 390W, 28°C, 49°C, 26°C, 61°C
1.450V, 375W, 27°C, 47°C, 26°C, 57°C
1.425V, 361W, 26°C, 45°C, 24°C, 45°C
1.400V, 349W, 26°C, 44°C, 24°C, 53°C

1.375V, 338W, 26°C, 42°C, 24°C, 61°C

Damn my h2o rearrangement benefits where gone. Since then I reapplied AS5 a dozend times without luck. The odd thing the core tems stay at ~42°C-44°C only the cpu temp differs up to 15°C between rearrangements.

Update: Can't get the temps down to where they where. Since the temps are higher now my voltage limit is now at ~1.4375V (70°C). 1.45V powered off the system. So 3.1GHz is my limit now, i'll run at 3GHz because 3.1GHz does require >1.4375V under linux 64bit.

[ea6gka]

i think i found the root of my problems , the dfi does not change the NB FID mutlis as i wish. i am entering hex 00 and i still get 2000freq which is 06 hex. latest bios in it , any ideas?

moreover , hex inputs from 01 to 05 hungs my mobo at post (error 75), also hex inputs 8 and above also results in error 75 and hungs at post.

[JWilson]

Is there any advantage to flashing the M2R BIOS to a M2RS board?

[Brother Esau]

Is there any advantage to flashing the M2R BIOS to a M2RS board?

Yes you will gain the advantage of Blowing up you're Motherboard DO NOT Flash M2R Bios to that Board if you know whats good for you!

[Final8ty]

i think i found the root of my problems , the dfi does not change the NB FID mutlis as i wish. i am entering hex 00 and i still get 2000freq which is 06 hex. latest bios in it , any ideas?

moreover , hex inputs from 01 to 05 hungs my mobo at post (error 75), also hex inputs 8 and above also results in error 75 and hungs at post.

I have been getting some funny NB Fid results as well.

I could not get above 200FSB at 3Ghz untill i put my NB Fid at 08
Now Ruining prime stable at CPUFID/0E NB/08 203FSB 3040Mhz

I can post at 205FSB NB/08, but will not post at 205FSB NB/07 or /06 even tho it should be less strain on the CPU.

[justapost]

I have been getting some funny NB Fid results as well.

I could not get above 200FSB at 3Ghz untill i put my NB Fid at 08
Now Ruining prime stable at CPUFID/0E NB/08 203FSB 3040Mhz

I can post at 205FSB NB/08, but will not post at 205FSB NB/07 or /06 even tho it should be less strain on the CPU.

You only set ref HT or also boot ref HT? I have np with an ref HT up to 260MHz here with the 9950BE.

[BulldogPO]

Can someone post good bios settings for 9850BE for Sapphire version of this board for maximum HT frequency?

[Final8ty]

You only set ref HT or also boot ref HT? I have np with an ref HT up to 260MHz here with the 9950BE.

If i just up the HTT i can only do 203.
If i up the HTT+BHTT then things proceed as expected but i never used BHTT before you mentioned it as there was no need with the 6400BE.

[Oese]

Can you give me a link to the site, I would like to try to pre-order one, I don't mind paying extra for shipping etc.

http://www.alternate.de/html/product...=Sockel+AM2%2B

here you are

[Oese]

If i just up the HTT i can only do 203.
If i up the HTT+BHTT then things proceed as expected but i never used BHTT before you mentioned it as there was no need with the 6400BE.

so if setting the bhtt right there is no problem running high htt? then i'd test some 9550 or so...