• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Trouble oc'ing Opty 165 on Asus A8V Dlx

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
mjw21a

mjw21a

Member
​
Joined
Sep 29, 2004
Guys, I just ordered in a new Opteron 165 CPU however I've found the same oc limitations as with my old 3200+ venice. Can't run stably at more than 233MHz.

I'm wondering whether something in my setup could be the issue, or more likely, BIOS rev and my lack of knowledge with overclocking socket 939 kit.

I'm running 3x 512MB PQI DDR 400 modules and 1x 512MB KingMax DDR 466 module. I don't forsee the KingMax giving any issues as it's higher rated the my PQI RAM (which is basically cheap, nasty stuff).

I'm running the Asus A8V Deluxe mb, not quite sure of the revision though as CPUz lists is as Rev 1.xx, which isn't terribly useful.

Any suggestions would be welcome. I would like to get some decent performance out of this setup.

Also, if anyone has a modified BIOS for this board it would be appreciated.

Thanks in advance guys.
 
The A8V I used to run got sketchy on FSBs higher than 280, my raid would get flakey at that speed too, plus my sound would start to crackle, the PCI lock worked fine too. I dont think you will hit the CPU's max on that motherboard.
 
You could also try taking out two sticks of ram and see if the Opty prefers one pair of dimm slots over the other.
 
Please don't double post, if you need a thread moved ask a mod.
 
Maviryk said:
You could also try taking out two sticks of ram and see if the Opty prefers one pair of dimm slots over the other.
Click to expand...

I suppose its possible that a better oc could be achieved with just two modules also.

Are there any tricks to overclocking Athlon 64's Opterons guys? I don't know anoything about dividers. All my previous overclocks were done with P3's, P4's and Athlon XPs.

Avg said:
Please don't double post, if you need a thread moved ask a mod.
Click to expand...

Sorry mate, couldn't figure out how to move it.
 
mjw21a said:
I suppose its possible that a better oc could be achieved with just two modules also.

Are there any tricks to overclocking Athlon 64's Opterons guys? I don't know anoything about dividers. All my previous overclocks were done with P3's, P4's and Athlon XPs.



Sorry mate, couldn't figure out how to move it.
Click to expand...
I have a lot of experience with this board so hopefully I can help you out. First, running 4 sticks of memory, and not even the same models at that, is definately not going to help you. If you're stuck at the same FSB on two different chips, it's highly likely your mobo or the memory hindering you. Put just two matched pair in for the time being. CPU-Z never gets the revision right with this board as my v2.0 was always seen as 1.xx as well. Next, the memory divider are screwed up in the BIOS, since revision 1011. 1009 was the last BIOS that actaully worked right with all the memory dividers. I'm guessing you have 1017(same as 1017.002 listed in the CPU support page) or 1018 beta? You have to use the 3:2 "divider" in order to have your memory start at 166mhz. 5:3, which should be the 166mhz divider actaully seems to start off at like 180mhz. If you were ever able to boot and check CPU-Z you might see this. I'm pretty sure the 4:3 starts you at 133mhz, though mathmatically it should be 150mhz. Never needed a divider that low to remember for certain. I'll also note that I've never seen a modded BIOS for this board.

Finally, even if you get things going well, running the FSB(HTT) higher than 262mhz will cause slowdowns with devices run by the chipset. It's noticeable when doing hard drive benchmarks and I"m not 100% sure what causes it. Once you get the BIOS dividers figured out, you can obviously try OC with more certainty about what's going on. Let me add that even though Asus lists the 1017 as needed for support for my Opteron 148, I used 1009 with no issues at all. Maybe it would be different with dual-core, but it wouldn't hurt to flash to that BIOS and see if it works. If it doesn't recognize the CPU, just flash back to the current one. Good luck.
 
Last edited:
mjw21a said:
I suppose its possible that a better oc could be achieved with just two modules also.

Are there any tricks to overclocking Athlon 64's Opterons guys? I don't know anoything about dividers. All my previous overclocks were done with P3's, P4's and Athlon XPs.



Sorry mate, couldn't figure out how to move it.
Click to expand...
I have a lot of experience with this board so hopefully I can help you out. First, running 4 sticks of memory, and not even the same models at that, is definately not going to help you. If you're stuck at the same FSB on two different chips, it's highly likely your mobo or the memory hindering you. Put just two matched pair in for the time being. CPU-Z never gets the revision right with this board as my v2.0 was always seen as 1.xx as well. Next, the memory divider are screwed up in the BIOS, since revision 1011. 1009 was the last BIOS that actaully worked right with all the memory dividers. I'm guessing you have 1017(same as 1017.002 listed in the CPU support page) or 1018 beta? You have to use the 3:2 "divider" in order to have your memory start at 166mhz. 5:3, which should be the 166mhz divider actaully seems to start off at like 180mhz. If you were ever able to boot and check CPU-Z you might see this. I pretty sure the 4:3 starts you at 133mhz, though mathmatically it should be 150mhz. Never needed a divider that low to remember for certain. I'll also note that I've never seen a modded BIOS for this board.

Finally, even if you get things going well, running the FSB(HTT) higher than 262mhz will cause slowdowns with devices run by the chipset. It's noticeable when doing hard drive benchmarks and I"m not 100% sure what causes it. Once you get the BIOS dividers figured out, you can obviously try OC with more certainty about what's going on. Let me add that even though Asus said I also needed the 1017 for support for my Opteron 148, I used 1009 with no issues at all. Maybe it would be different with dual-core, but it wouldn't hurt to flash to that BIOS and see if it works. If it doesn't recognize the CPU, just flash back to the current one. Good luck.
 
tuskenraider said:
I have a lot of experience with this board so hopefully I can help you out. First, running 4 sticks of memory, and not even the same models at that, is definately not going to help you. If you're stuck at the same FSB on two different chips, it's highly likely your mobo or the memory hindering you. Put just two matched pair in for the time being. CPU-Z never gets the revision right with this board as my v2.0 was always seen as 1.xx as well. Next, the memory divider are screwed up in the BIOS, since revision 1011. 1009 was the last BIOS that actaully worked right with all the memory dividers. I'm guessing you have 1017(same as 1017.002 listed in the CPU support page) or 1018 beta? You have to use the 3:2 "divider" in order to have your memory start at 166mhz. 5:3, which should be the 166mhz divider actaully seems to start off at like 180mhz. If you were ever able to boot and check CPU-Z you might see this. I'm pretty sure the 4:3 starts you at 133mhz, though mathmatically it should be 150mhz. Never needed a divider that low to remember for certain. I'll also note that I've never seen a modded BIOS for this board.

Finally, even if you get things going well, running the FSB(HTT) higher than 262mhz will cause slowdowns with devices run by the chipset. It's noticeable when doing hard drive benchmarks and I"m not 100% sure what causes it. Once you get the BIOS dividers figured out, you can obviously try OC with more certainty about what's going on. Let me add that even though Asus lists the 1017 as needed for support for my Opteron 148, I used 1009 with no issues at all. Maybe it would be different with dual-core, but it wouldn't hurt to flash to that BIOS and see if it works. If it doesn't recognize the CPU, just flash back to the current one. Good luck.
Click to expand...

You wouldn't happen to know where I can get the 1009 BIOS would you. I've found plenty of discussion saying its the BIOS to use, but can't find a location to download it from.....

By the way, I'm running the 1018.001 Beta BIOS.
 
I am running the same board , make sure your not running the hdd on the VIA controller either , use the promise controller due to the VIA not stable when ocing
also remember to lock the pci lock , i am using the 1017 bios with no problems
@ 2.91 ghz
I also ran a opty 175 @ 2.6 with no problem
 
Opteron 1100 said:
I am running the same board , make sure your not running the hdd on the VIA controller either , use the promise controller due to the VIA not stable when ocing
also remember to lock the pci lock , i am using the 1017 bios with no problems
@ 2.91 ghz
I also ran a opty 175 @ 2.6 with no problem
Click to expand...
I found the Promise controller to be affected the same way the VIA controller was so you might as well stay on the VIA since it isn't bandwidth limited like the Promise.
 
Ok, I've got the 1009 BIOS downloading. Will give that a go, if that doesn't work will try removing some RAM, then will try the promise controller.

I've noticed that regardless of what memory divider I try under 1018.001, there is no effect on the RAM.
 
Serlant said:
if theres a flexibility option, turn it off
Click to expand...

Flexibillity? Do you mean Cool 'n Quiet? I havent actually turned that off, its on.
 
no, its a ram setting i think it lets it change the ram devider to what it thinks is "best" all i know is that with it turned on it sets the ram to a certain devider regardless of wether ive got the cpu devider on 100/133/166/200 like its on the 100 devider right now but because of flexibility option the rams at 199.9mhz where as if it was using the 100mhz devider that i set the ram would be at 155mhz.

(yes i talk jibberish ;) sorry cant help it)
 
It seems to be setting the RAM to 149MHz at present, that doesn't change regardless of what I set the divider to (t least, thats what CPUz reports). Need to getcreate a boot disk to flash my ROM back.


By the way, I now love Artic Silver Ceramique thermal paste. Using this has dropped about 2-3C off the temps I was getting with the same setup and AS5.

I'd heard that this stuff was better with later surfaces whereas the AS5 was better for smaller but now I believe.
 
tuskenraider said:
I found the Promise controller to be affected the same way the VIA controller was so you might as well stay on the VIA since it isn't bandwidth limited like the Promise.
Click to expand...
If you run the VIA controller and you have a raid array it will corrupt your hdds if you oc
I went to the promise side and had no problems , the performance loss using promise is nothing
 
You must log in or register to reply here.

Similar threads

D
Second GPU failing ASUS Maximus Hero XII
Replies
10
Views
923
dparisi
D
Al_bundy
Alucryl - 3DFX - CNC - Retro pc
Replies
2
Views
393
EarthDog
EarthDog
funsoul
Need build oc help (5900x, 3600 b-die, 3090ti)
Replies
16
Views
2K
funsoul
funsoul
R
How to undervolt AMD-FX8350+ASUS Crosshair V Formula-Z and Overclock?
Replies
0
Views
2K
Raskaipika
R
J
Overclocking on an Asus P5E3 Work Station Professional
Replies
14
Views
1K
PolRoger
PolRoger
Back