• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

**LETS SEE YOUR PILEDRIVER OVERCLOCKS - LET ME START WITH 5GHz+!!**

Gigabyte still having issues with their VRM's on high power draw processors:
http://www.brightsideofnews.com/new...e-year-with-a-bang-kills-amd-fx-8350-cpu.aspx

First X79, now AM3+.

Actually X79 wasn't first.
The gigabyte VRM woes started with Socket 1366.
I had 5 GigaByte EX58 UD3R's.
Only one is alive now.

I used these 24/7 for various DC projects.
One by one they all went down with blown VRM's
I got some of them RMA'd and the replacements also failed.

The one that is still alive is on light duty only, it is my web server.

This put me off GB big time and i use ASUS almost exclusively now.
 
However i am choosing to ignore the fact that P95 AVX tests will send the temps into the stratosphere.:)
LA LA LA never heard of it all good here, P95 what.

When I'm stability testing I turn all the fans up to 11 and I can run IBT on max no problem. If that's OK I can put it all back onto the silent or normal fan profile because I'll NEVER do anything that stressful with the PC. Even hours of encoding is much less stressful.

I've stopped using OCCT because when it fails it tends to lock up the PC and then it won't boot. For some reason it corrupts the boot sector, bloody nightmare. :rolleyes: Luckily I take regular drive images. IBT just throws up errors and I know a voltage needs to be increased somewhere. It's quick too, if it's not stable (CPU, RAM or board) it will error in the second or third run. If it does 5 runs then it'll go on and will of peaked with temps too.

There's no way you'll get me being anal about hours of stability with these apps. I've got a life and quite like actually using my PCs! :p
 
Makes quite a difference to what? Stress levels?

I'm using;

v2.54
- Linpack binaries updated (7-20-2012) + patched for AMD CPUs

Just been playing BF3 aftermath. Max CPU socket and cores 52C, max GPU 46C. You can barely hear the HTPC, so chuffed. :D
 
Last edited:
You will notice the difference.
Just overwrite the linpack libaries and put the dll in the folder as well.
 
i understand the point of overclocking for more performance out of the same value chip as it's good cost wise and makes sense, but what i don't understand it high numbers just for the sake of it unless it's stable and can be used, using nitro to keep a chip cool for 30 mins while a test runs, then the chip is bunrt out makes little sense to me other than to have bragging rights with your mates, and as most of mine would think im mad throwing a good chip away for 30 mins of testing im sorry i just don't see the point.

unless it's stable and doesn't fry egg's or smell of burning all day, if it can;t be used then it's just a number for rich kids to looks at and has no place in the real world same as nitro in car engines, 1 use and engine needs re building etc.

then again with the amount of posts in here, i'll be the odd one out im guessing again.

ah well, said my bit, time to tell me im wrong again
 
i understand the point of overclocking for more performance out of the same value chip as it's good cost wise and makes sense, but what i don't understand it high numbers just for the sake of it unless it's stable and can be used, using nitro to keep a chip cool for 30 mins while a test runs, then the chip is bunrt out makes little sense to me other than to have bragging rights with your mates, and as most of mine would think im mad throwing a good chip away for 30 mins of testing im sorry i just don't see the point.

unless it's stable and doesn't fry egg's or smell of burning all day, if it can;t be used then it's just a number for rich kids to looks at and has no place in the real world same as nitro in car engines, 1 use and engine needs re building etc.

then again with the amount of posts in here, i'll be the odd one out im guessing again.

ah well, said my bit, time to tell me im wrong again

It's an enthusiasts hobby to see what they can achieve, and I guess a challenge at the same time.

Personally, I am happy @ 1.4625v @ 4.8, not pushing anything excessively, I have good temps and all power saving features enabled still so I get the best of both worlds IMO.

I have booted into Windows at 5.2, never tested any kind of stability at this speed but was checking it could for some fun, my system is stable at 5.0 using just an Antec 620, but needs 1.55 volts to achieve this, all the power saving disabled and upping various LLC's, phase control etc, for me the 200Mhz gain, isn't worth the disadvantages, maybe with a better cooler I could achieve this with the above features enabled still.
 
Last edited:
if stable i can see the point and advantages etc

if not then for a challenge only, but if it goes wrong a expensive one?

That's the risk that comes with overclocking...it's also the overclockers choice as to whether they pushes their chip to the max or not.

I'm not sure where you got your "30 mins" from though, unless you're talking about suicide runs, something that's done for world records and they do that knowingly :rolleyes:
All my rigs are and have been overclocked passed the chips limits and all currently still work, it's a lot harder to bin a chip than people seem to think :D Trust me, i've tried :D :D
 
thats what i ment, for records and bragging rights etc, and tbh would you by a second hand cpu off someone knowing that it had been overclocked?
 
if stable i can see the point and advantages etc

if not then for a challenge only, but if it goes wrong a expensive one?

I have been running a 1Ghz overclock on my chip for more than 2 years.

Its on for 18 hours a day and gets stressed regularly, its still as healthy today as it was when it was new.

Overclocking does reduce the lifespan of a chip, yes. but how long do you need it to last? 10 / 15 years?

My mothers Athlon is 7 years overclocked and i'm itching for it to die so i have an excuse to build her a new rig, its just flat out refuses to die!
 
Last edited:
thats what i ment, for records and bragging rights etc, and tbh would you by a second hand cpu off someone knowing that it had been overclocked?

Yes, yes i would.
Would i buy a chip knowing it's a terrible clocker? Nope.

The degradation that occurs from slapping extra voltage through the chips is minimal, if the chips have 10 years lifespan from new...and you knock 5 years off that from overclocking....that's still getting well passed even the longest upgrade cycles :D
 
I have also been following that thread.
Makes a lot of sense and certainly fits with my own observations.
 
I wouldn't trust that 50p DVM, looks like he bought it from Maplin. LOL. I suspect there's a reason why his meter doesn't pick up the fluctuations in voltage that CPUZ shows. Go get a calibrated Fluke bud!! :D

Still, this is really what I wish the whole so called 'enthusiast' community would realise;

I like my FX-8350. It is nowhere near the slow-performing fail-train that it gets made out to be in enthusiast discussions where second-fastest in benchmarks means being dead-last in perception.

Folks tend to get a little too binary in their categorization and characterization IMO. And I include myself in that generalization.

Until I had one to play with first-hand, I too had this impression that not only would a 4GHz bulldozer/piledriver consume 600W but it would also perform like an Atom processor.

But the performance gap just isn't that wide, and the power usage situation (once you factor in the power used by the entire platform, LCD included) just isn't that stark in contrast between the two.
 
Last edited:
You will notice the difference.
Just overwrite the linpack libaries and put the dll in the folder as well.

Yep. Spot on bud, big difference.

ibtwithoutandwithamdbin.png
 
Back
Top Bottom