Jump to content

unofficial DUAL CORE opteron thread


racinjimy

Recommended Posts

Well thats exactly what i have ccbwe 0550 upmw

and its crapping out every time it hits 40c :mad:

On the good side tho its stable at 275*9@ 1.2*104%vcore (only vcore i can manage without hitting 40c :mad::mad::mad::mad::mad::mad: )

any more vcore lowers my max stable oc cuz of that weird 40c problem

and any more htt requires more vcore.

I WANT WATERRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRR

 

You got a strange CPU. You've either got two really weak cores or your CPU is not making good contact with the HS. The CPU should be able to get up ~50C+ before it fails. You're not even close.

 

When I removed the IHS from my 3500+, my waterblock wasn't making good contact with the core and it would easily crash even with slight OC. Even though you removed the 5V floppy power from the mobo, you're still is experiencing the same heat 40C problem. I think you have a contact problem because 40C should not crash the CPU.

 

There is a very low probability of both cores being severly anemic. Secondly, even the weakest cores should be able to handle at least 50C.

 

How are you monitoring your CPU temp? I'm betting that your core temp is much higher than 40C. Somehow, your IHS is not doing its job. I have a feeling that the WC system will help but it won't be the miracle cure unless you improve your contact.

 

Good luck

 

cjoe

Share this post


Link to post
Share on other sites

  • Replies 940
  • Created
  • Last Reply

Top Posters In This Topic

the ihs is already off didnt change anything (other then lowering temps a few degrees)

im using speedfan for temp monitoring

altho the floppy conector didnt change this behavior it did drop my temps ALOT as that mosfet was hitting i assume close to 50c+

Share this post


Link to post
Share on other sites

the ihs is already off didnt change anything (other then lowering temps a few degrees)

im using speedfan for temp monitoring

altho the floppy conector didnt change this behavior it did drop my temps ALOT as that mosfet was hitting i assume close to 50c+

 

Even though the mosfet is very hot, its not detrimentally affecting your OC. The 5V floppy power is supposed to increase your board stability by providing more voltage to the board. You may actually reduce your OC by removing that power.

 

My Expert is under reporting CPU temps by ~10C for Opterons; so, I have MBM add 10C to the CPU temp.

 

If you do have 2 anemic cores, your WC solution will only help you marginally. After removing your IHS, you gained a slight FSB increase. WC'ing will help you stablize your temps but good air cooling is comparable.

 

What are your PWMIC temps? Have you considered just blowing some fans across your mobo to see if that helps.

 

Have you tested your VCore with a multimeter? (Shot in the dark)

 

You may have hit your OC max. Clint could only achieve 2.5G on his 0551UPMW. On the positive side, you're hitting 2.5G @ 1.25V. :nod:

 

Keep us updated on your progress.

 

Good luck.

 

cjoe

Share this post


Link to post
Share on other sites

Removing the ihs didnt do much of anything ocwise.

removing that connector tho droped the temps significantly and stabilised my oc with an extra 5 fsb

If the dfi is reporting bad temps (10 degree offset) that could explain some of my problems.

50-51c is darn near close to my tcase max.

Pwmic is in the mid 40s i have pretty good flow right now i have 3 120mm case fans at high speed on my p180 and my cpu fan is at 120cfm.

darn thing sounds like a jumbo jet!

 

I would find it VERY odd that i can only get a marginal speed increase from water tho.

I mean heck if i cant get a good fsb boost from going from 1.2 to 1.5 vcore i got one heck of a problem! lol

 

I'm gonna try the cpu in a asus board this weekend to see if i dont have any mobo issues.

Then i could always try ocing with the window open on a cold day (-10°c) but id be afraid of condensation at those temp differences.

Share this post


Link to post
Share on other sites

Removing the ihs didnt do much of anything ocwise.

removing that connector tho droped the temps significantly and stabilised my oc with an extra 5 fsb

If the dfi is reporting bad temps (10 degree offset) that could explain some of my problems.

50-51c is darn near close to my tcase max.

Pwmic is in the mid 40s i have pretty good flow right now i have 3 120mm case fans at high speed on my p180 and my cpu fan is at 120cfm.

darn thing sounds like a jumbo jet!

 

I would find it VERY odd that i can only get a marginal speed increase from water tho.

I mean heck if i cant get a good fsb boost from going from 1.2 to 1.5 vcore i got one heck of a problem! lol

 

I'm gonna try the cpu in a asus board this weekend to see if i dont have any mobo issues.

Then i could always try ocing with the window open on a cold day (-10°c) but id be afraid of condensation at those temp differences.

 

Try the 623-1 bios, it gave me a 50Mhz gain from 623-2 in OC and much better stability on those tricky UCCC sticks

The latest bios from DFI that are supporting dual made things really unstable for me...a drop of 200Mhz OC and hangs and whatnot... :(

 

I have tried the lot of them and the 623-1 is giving me the best stability/OC...

Share this post


Link to post
Share on other sites

Removing the ihs didnt do much of anything ocwise.

removing that connector tho droped the temps significantly and stabilised my oc with an extra 5 fsb

If the dfi is reporting bad temps (10 degree offset) that could explain some of my problems.

50-51c is darn near close to my tcase max.

Pwmic is in the mid 40s i have pretty good flow right now i have 3 120mm case fans at high speed on my p180 and my cpu fan is at 120cfm.

darn thing sounds like a jumbo jet!

 

I would find it VERY odd that i can only get a marginal speed increase from water tho.

I mean heck if i cant get a good fsb boost from going from 1.2 to 1.5 vcore i got one heck of a problem! lol

 

I'm gonna try the cpu in a asus board this weekend to see if i dont have any mobo issues.

Then i could always try ocing with the window open on a cold day (-10°c) but id be afraid of condensation at those temp differences.

 

Just because you have good air flow in your case doesn't mean the mobo is cool enough. Get a 120mm fan and a Zalman fan bracket and point it directly on the hot areas of the board.

 

Why do you think that WC'ing will benefit your system? Most of the current high-end air systems actually out perform water systems. Water provides two benefits: low noise and ability to absorb a lot of heat (which can be dissipated later).

 

Water will help somewhat but don't expect miracles because you probably have another fundamental issue with your system. Many users are achieving 3Ghz with air.

 

In general regular WC'ing, will not increase your basic OC. It just helps stablize it until it cannot absorb anymore heat. Cool ambient temp = good water cooling. Hot ambient temp = bad water cooling. If you're using chilled water, then you may reap major gains.

 

cjoe

Share this post


Link to post
Share on other sites

Of course WC will not increase any normal oc.

ive been running air on anything else in the past with great results.

So far my problem is with temps

Ive tried the 623-1 623-3 and 704-2bta bioses with no change at all.

Tried pretty much every vcore in varios htts

And at any speed if i hit 40c im screwed.

Good thing about water...easily upgraded to chilled water ;)

specially if you have left over parts from an old AC unit or car ac unit.

And even without chilled properly setup youll stick pretty darn close to ambient.

Sure air will do that too...Up until you reach the maximum Kj/s your small fined xp-120/scythe ninja etc etc can dissipate.

Water you can always go for a complete car rad wich in theory would have enough heat dissipation to keep some pretty nasty voltages near ambient.

Yes theres more to it then the rad i'm quite aware of that. But assuming everything else is setup properly you can stay within a few degrees of ambient where some air setups would be begging for mercy.

I'm also tired of living next to a jet engine :P

Its also alot easier to mount a wb on a naked chip then a xp-120 and what not.

 

My next 2 steps will be measuring vcore myself instead of relying on the bios.

and testing on another board.

If the vcore is good and the board doesnt change anything....

I really dont see what else it could be.

Max oc? yes for that voltage maybe.

But .3v *should* give much higher.

Is it worth the effort? probably not

will i see a difference in speed? probably not

Am i satisfied with barely hitting 2.5ghz? HELL NO

Will i quit begore hitting 2.6? NO

Share this post


Link to post
Share on other sites

Of course WC will not increase any normal oc.

ive been running air on anything else in the past with great results.

So far my problem is with temps

Ive tried the 623-1 623-3 and 704-2bta bioses with no change at all.

Tried pretty much every vcore in varios htts

And at any speed if i hit 40c im screwed.

Good thing about water...easily upgraded to chilled water ;)

specially if you have left over parts from an old AC unit or car ac unit.

And even without chilled properly setup youll stick pretty darn close to ambient.

Sure air will do that too...Up until you reach the maximum Kj/s your small fined xp-120/scythe ninja etc etc can dissipate.

Water you can always go for a complete car rad wich in theory would have enough heat dissipation to keep some pretty nasty voltages near ambient.

Yes theres more to it then the rad i'm quite aware of that. But assuming everything else is setup properly you can stay within a few degrees of ambient where some air setups would be begging for mercy.

I'm also tired of living next to a jet engine :P

Its also alot easier to mount a wb on a naked chip then a xp-120 and what not.

 

My next 2 steps will be measuring vcore myself instead of relying on the bios.

and testing on another board.

If the vcore is good and the board doesnt change anything....

I really dont see what else it could be.

Max oc? yes for that voltage maybe.

But .3v *should* give much higher.

Is it worth the effort? probably not

will i see a difference in speed? probably not

Am i satisfied with barely hitting 2.5ghz? HELL NO

Will i quit begore hitting 2.6? NO

 

Yeah, I hear you and feel your pain. I'm kinda in the same boat. My core1 is pretty weak and I'm trying to stablize my system at 2.75Ghz. I'm on water and Its killing me also. I want to hit 2.8Ghz but it doesn't look promising. After 2days of burning in, I'm kicking myself for not acting earlier on a VPMW. All those CPUs are hitting 2.8-3.0 pretty consistently with low vcore.

 

BTW: Who told you that water was quiet? When I'm burning in my CPU with high voltage and high Mhz, The CPU temp will reach close to 50C. I have to turn up the fan speed to increase the cooling capacity. Its probably not as loud as regular air cooling but it can get annoying.

 

Remember when running at 100% for several hours, the radiator fan will have a hard time dissipating that much heat. If I'm running at less than 100%, I can leave the fans at 1200rpm and it is pure blissful heaven.

 

cjoe

Share this post


Link to post
Share on other sites

How big is your rad?

Bigger rad means you can fit bigger fans and do a push pull config so low rpm can dissipate more heat.

The flow of water is another quirky lil thing to tweak.

Take example on a car.

Too much water flow can cause overheat! Simply because if the water is moving too fast it cant absorb heat properly neither can it dissipate it properly in the rad.

Even the tubing can be a problem in some cases.

Atleast i finally reached a similar spi speed as my old p4. Its less insulting spending 900$ to get pretty much the same speed as my old p4 on those benches then when yer far slower then your old pos system ;)

Games have improved dramaticly tho.

I dunno i just cant setle for anything under 2.6 that was my bare minimum OC and i dont want to quit now.

Share this post


Link to post
Share on other sites

How big is your rad?

Bigger rad means you can fit bigger fans and do a push pull config so low rpm can dissipate more heat.

The flow of water is another quirky lil thing to tweak.

Take example on a car.

Too much water flow can cause overheat! Simply because if the water is moving too fast it cant absorb heat properly neither can it dissipate it properly in the rad.

Even the tubing can be a problem in some cases.

Atleast i finally reached a similar spi speed as my old p4. Its less insulting spending 900$ to get pretty much the same speed as my old p4 on those benches then when yer far slower then your old pos system ;)

Games have improved dramaticly tho.

I dunno i just cant setle for anything under 2.6 that was my bare minimum OC and i dont want to quit now.

 

I've got 2 single 120mm rads with 2 Silverstone 120mm 110CFM (variable) fans in pulling setup. Its in Pump -> Res -> Rad - > CPU -> Rad -> Vid -> NF4 -> Pump configuration. I'm using 3/8" ID tubing with an Asetek extreme pump (variable flow). I've tried several configurations and several different systems. The ambiant temp in my computer room is about 23-24C. Under non priming situations my WC is great. The computer is very quiet with a slight hum. When priming over 4hrs+, the heat load is tremendous and the CPU will reach 49-50C (system is stable). I have to increase the fans RPM from 1300rpm to 2000rpm inorder to get the CPU back down to 45C.

 

Ambient Temp: 23C+ (higher when priming)

Pump Flow: 1020L/Hr (Max)

 

LOAD LOAD

RPM IDLE Non Priming Priming (4hrs+)

------------------------------------------------------------

1300 33C 45C 50C

2000 28C 38C 45C

 

My less than stellar CPU (2.75Ghz @ 1.55V) with Video and Chipset is dumping more heat into the system than the rads can handle at low rpm. Secondly, my ambient temp is not giving me enough of a delta to dissipate more heat fast enough. So my only solution is to increase the fan RPMs which increases the DBs.

 

I can probably play with the water flow a little more but I don't think it will make that much of a difference. The pump will generate more heat if it runs faster.

 

Unless someone has a supersized radiator with a huge ambient temp delta, I think it would be very difficult to keep very low rpms (<1300RPMs inaudible). My situation is probably common when priming after several hours.

 

Anyone else want to chime in on their WC setup, temps and noise?

 

cjoe

Share this post


Link to post
Share on other sites

For those interested...

 

Dual Priming right now at 290x10 @1.3+116% (1.508 for those without calculators)

 

Weird though, I'm completely capapble opening stuff in windows, its not like its lagging at all. Task manager shows both core at full speed, so beats me. I guess 2900 is just fast enough....:)

 

Tried 3.0GHz, but it need close to 1.7 to get stable so I'm just calling it good enough here I think. Probably shoulda primed this last night instead of now... :rolleyes:

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×
×
  • Create New...