Opteron 180 stuck @ 234

[mrbadnback]

The settings that you gave me before are all the same TRC @11, TRFC@14, Max Async Latency @.o8ns, DRAM Response Time @ Normal, Read Preamble Time @ .06ns and that is all running with the RAM @ 180 9/10.

Why am I changing to the different settings? Since I am at 180 9/10 RAM divider now. Is it to lower the multiplier to see if I can manage a higher OC on the lower multiplier?

I will have to try the different settings after XMAS since I will be at my Sister's till then and I'm not taking my system. I couldn't believe just by taking the RAMS VCORE that was set by default to 2.66V and then setting it to 2.74v that it actually let me get out of the lower 230FSB and allowed me to use the 180 9/10 setting on the RAM. It also gave me the option to go from the 3-3-3-8-settings to 2-3-3-6 2T. So far today my idle is @ 26C which for me is normal. Even @ the stock settings it was running at 24C-26C idle so it's not doing bad for a very cheap $39 case that I purchased 2 years ago that came with a 350W PSU.

[Sharp]

Is it to lower the multiplier to see if I can manage a higher OC on the lower multiplier?

Yes, if 250x10 works try 250x11 then try 250x12. < which = 3Ghz.

[mrbadnback]

Do I set it at 166 and x10 and start at 200FSB and work up or just set it to x10@250 and see what happens. If I have to start at 200FSB@x10 and work up do I do that with every time I raise the multiplier? OR do I just keep the FSB @ 250 and just switch the multipliers?

I just want to make sure so I don't screw it up like last time.

[Sharp]

With your current settings but start at 166, 250FSB and x10.

 

If it is stable just change the multiplier.

[mrbadnback]

Just one quick question before I start changing the multiplier. Just out of curiosity why am I doing it this way? Why don't I just lower the RAM to 166 with the multiplier at x12 and try hitting 250FSB that way? I just want to know what I'm doing for future OCing of my other system. It's a DFI expert also with a X2 4400+, G.Skill DDR500 RAM and a OCZ PowerXStream 600w PSU. It's my next project.

[Sharp]

If you set you fsb to 300 and memory to 166 that means your memory will try to run at 250Mhz (500DDR).

 

At 250Mhz you may find that the vcore voltage needs to be increased.

At 300Mhz you will need a lot more and you will start to go over the safe limit of 1.45v. (it may not even work)

 

The point is to slowly work your way up.

If you can't get 250Mhz stable then it will never work at 300Mhz.

[mrbadnback]

Sorry about that Sharp! I wrote it wrong, what I wanted to know is why start at 166 x10 @250FSB and test it then up the multiplier to x11 times then if it passes up it to x12. I know if I get it to x12 x250 its at 3.00GHz which would be GREAT. BUT what I wanted to know is why start at 250 x10 then go to x11 then to x12. Do I just set it to 166 RAM 250FSB x10 then test? Then if it passes do I just go straight to x11 250FSB or do I start with a lower FSB and try to get to 250?

[Sharp]

Why 250FSB,

Well because you are currently at 234, and 250 is only 16Mhz more.

250Mhz x 10 is less than your current OC but puts more stess on the core and memory controller.

 

If the CPU does not pass this then it could mean two things.

1, CPU cannot handle 250FSB.

2, Needs more voltage.

 

Next when you go to 250Mhz x 11 this will be closer to 2.8ghz.

So it has matched your current OC but at 250FSB.

 

Then at x12 you are seeing if it will run at 3Ghz.

Increasing the multiplier is easier than increasing the FSB.

 

 

Do I just set it to 166 RAM 250FSB x10 then test? Then if it passes do I just go straight to x11 250FSB

Yes.

[mrbadnback]

Thanks for the fast reply! I will have to wait till Sunday to start this but I will get back to you when I find out the results.

[mrbadnback]

Ok I just tried 166 RAM divider with x10 @250 and it would boot part way through the BIOS startup but would just reboot before it got to the windows startup. I then just changed the multiplier to x11 and it booted into windows with no problems so I just came here to report what I have found so far. I am going to try going to x12 now and will report back soon.

Well I just upped it to x12 and it made it through the BIOS startup but as soon as it got to the windows startup I got the BSOD. ROFL I knew that was going to happen. Now what's my options just upping the VCORE and trying or is staying at the x11 multiplier and increasing the FSB an option? Which is the best of the two choices?

[Sharp]

Let memtest run for a while with it on x11.

If you get errors up the vcore to 1.45v.

And memtest again.

 

If it turns out to be ok increase the multiplier to x12 and run memtest.

[mrbadnback]

Memtest ran fine with x11 so I tried memtest @ x12 and it ran through 3 passes and then I ran 10 passes each of #5 and #8 and it ran fine. I tried to boot into windows and I get the BSOD. So I went back into the BIOS and set the VCORE to 1.35v @107.2% = 1.447v and I still get the BSOD just as it gets to the windows startup. So I guess that is about as far as this 180 is going to go. The next setting would take me to 1.476V @109.4%.