-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Noob with X2 550BE and Asus A485TD-V EVO, basic questions
- Thread starter stowenrat
- Start date
- Thread Starter
- #62
I think I might have a problem with my PSU, it is an old that I used on my last system so it might be a bit tired.
I get quite a wide variance in what CPU-Z says the core voltage is, for example I just tried an overclock using 1.45v as it is quite cold in this room and it was just a quick test, CPU-Z was fluctuating from 1.408 - 1.424 - 1.440 - 1.456 and it does this at lower voltages too.
I am using LLC of 22.5% becuase when I use auto (50%) it varies just as much if not more and if I use a lower LLC (like 9%) it can spike too high.
There is no disabled selection for LLC, just 0-100% and auto, would disabled be 0% or 100%?
I get quite a wide variance in what CPU-Z says the core voltage is, for example I just tried an overclock using 1.45v as it is quite cold in this room and it was just a quick test, CPU-Z was fluctuating from 1.408 - 1.424 - 1.440 - 1.456 and it does this at lower voltages too.
I am using LLC of 22.5% becuase when I use auto (50%) it varies just as much if not more and if I use a lower LLC (like 9%) it can spike too high.
There is no disabled selection for LLC, just 0-100% and auto, would disabled be 0% or 100%?
- Thread Starter
- #64
Load line calibration.
This site reckons 19% is best for my mobo http://www.hardwarecanucks.com/foru...4a785td-m-evo-am3-motherboards-review-18.html
They say100% is basically the same as disabled which is weird but confirms my own experience.
This site reckons 19% is best for my mobo http://www.hardwarecanucks.com/foru...4a785td-m-evo-am3-motherboards-review-18.html
They say100% is basically the same as disabled which is weird but confirms my own experience.
Last edited:
QuietIce
Disabled
- Joined
- May 7, 2006
- Location
- Anywhere but there
I've always considered ±0.02v to be quite acceptable - I wouldn't worry about it ... 
- Thread Starter
- #68
It's not looking like I got a lucky CPU, I might be stable at 18x200 at 1.400VDD, 1.250VDDNB but even going to 1.450VDD I am crashing out at 18.5x200, not just failing the stress test but computer crashing.
Is there anything else that can help stability apart from VDD, VDDNB, HT (currently on 1600), my ram is OK and I am using LinX for quick stability tests which doesn't even stress the ram anyway.
I really would have thought I could get 3.7ghz with VDD 1.45 and VDDNB 1.225.
Meh.
Is there anything else that can help stability apart from VDD, VDDNB, HT (currently on 1600), my ram is OK and I am using LinX for quick stability tests which doesn't even stress the ram anyway.
I really would have thought I could get 3.7ghz with VDD 1.45 and VDDNB 1.225.
Meh.
- Thread Starter
- #69
I think I have a problem at higher voltages in that when I set VDD to 1.45 I get hardly any higher readings in HW monitor than when it is set at 1.400! They really are basically the same.
There is a setting I just found called VDDA which can, hopefully, counteract this. http://vip.asus.com/forum/view.aspx...1&model=M3A79-T+Deluxe&page=1&SLanguage=en-us
I will test that out later.
There is a setting I just found called VDDA which can, hopefully, counteract this. http://vip.asus.com/forum/view.aspx...1&model=M3A79-T+Deluxe&page=1&SLanguage=en-us
I will test that out later.
- Thread Starter
- #70
My fave thread ever! http://www.overclockers.com/forums/showthread.php?t=329627
I had that board and BH5 ram and I just copied what Jess did, lol.
I sold that ram on Ebay recently for like £10, was worth loads at one time! I forgot to put a a reserve limit on the auction.
I had that board and BH5 ram and I just copied what Jess did, lol.
I sold that ram on Ebay recently for like £10, was worth loads at one time! I forgot to put a a reserve limit on the auction.
QuietIce
Disabled
- Joined
- May 7, 2006
- Location
- Anywhere but there
Some OC'ers will set VDDA to 2.6v right at the beginning and sometimes even higher. Nothing wrong with that. Can't say as I've heard of anyone going as high as 2.8v but I guess it's worth a shot.
How are your load core temps at 18x200 @ 1.40 vCore ...?
How are your load core temps at 18x200 @ 1.40 vCore ...?
- Thread Starter
- #72
I used 2.7 and it didn't help at 1.45, still dropping down to 1.40 levels. Not good.
I am a bit wary of 2.8 and since 2.7 did nothing I can't see 2.8 helping either.
It looks like I am stuck at 1.40 and the OC I can get there is 3.6ghz.
My load temps at 1.40 vcore and 1.225 vddnb were 53c max with the stock cooler and lowish ambient.
I am not getting the 212 Plus now, I am getting the Arctic Cooling Freezer 64, should be here this week.
I am a bit wary of 2.8 and since 2.7 did nothing I can't see 2.8 helping either.
It looks like I am stuck at 1.40 and the OC I can get there is 3.6ghz.
My load temps at 1.40 vcore and 1.225 vddnb were 53c max with the stock cooler and lowish ambient.
I am not getting the 212 Plus now, I am getting the Arctic Cooling Freezer 64, should be here this week.
QuietIce
Disabled
- Joined
- May 7, 2006
- Location
- Anywhere but there
The temps are what's hanging you up, I'd almost bet on it. That 50-55°C zone is where most CPUs start to fizzle in their OC ...
- Thread Starter
- #74
- Thread Starter
- #75
Something weird happening now, I have previously been passing 100 runs of LinX, 18x200 with 1.40 vcore, 1.225 vddnb, 2.5 vdda, 19% LLC.
I decided to see if a combination of 1.45vcore, 1.2250 vddnb, 2.7 vdda and 9% LLC would stop my CPU-Z vcore readings dropping to 1.4 levels...it did hold it a bit higher....
BUT
I had chosen a lesser overclock of 17x200 to do the test because I wasn't trying to push it I just wanted to see what the vcore readings under load were going to be....
it failed LinX within 30 seconds, that is before temperature could be an issue.
So a lower overclock with more voltage fails? Why?
I decided to see if a combination of 1.45vcore, 1.2250 vddnb, 2.7 vdda and 9% LLC would stop my CPU-Z vcore readings dropping to 1.4 levels...it did hold it a bit higher....
BUT
I had chosen a lesser overclock of 17x200 to do the test because I wasn't trying to push it I just wanted to see what the vcore readings under load were going to be....
it failed LinX within 30 seconds, that is before temperature could be an issue.
So a lower overclock with more voltage fails? Why?
Last edited:
- Thread Starter
- #76
So I went back to 18x200 1.40 - 1.225 - 2.6 - 19% and it passed 20 rounds of LinX.
So, something about 1.45 vcore, 2.7 vdda and 9% LLC is bad news for some reason even though my vcore readings looked good and it didn't even have time to over heat.
So, something about 1.45 vcore, 2.7 vdda and 9% LLC is bad news for some reason even though my vcore readings looked good and it didn't even have time to over heat.
QuietIce
Disabled
- Joined
- May 7, 2006
- Location
- Anywhere but there
You're just proving what I've been saying all along with the Phenom II's: it's all about balance. That's such a hard concept to get across, though, and it's impossible to prove, so I seldom mention the word. Sounds too much like Zen OC'ing ...
Similar threads
