• 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.

5ghz stable 4930k Ht-on.

I agree and the new refresh is still to be TIM.

If this POS could do 4.6GHZ with the low voltages some get here I would Delid and go for 5GHZ.

I am sorry I never kept the Z68+2700k but I got the update itch be it late on this (Dec/Jan).

The new Z79 Mobos do not bring that much to the table so I can drop a new CPU in here as I already have the Bios for them.
 
jh30uk said:
I agree and the new refresh is still to be TIM.

If this POS could do 4.6GHZ with the low voltages some get here I would Delid and go for 5GHZ.

I am sorry I never kept the Z68+2700k but I got the update itch be it late on this (Dec/Jan).

The new Z79 Mobos do not bring that much to the table so I can drop a new CPU in here as I already have the Bios for them.
Click to expand...
Tbh as warm as my cpu can get, at 4.2ghz it was faster than my old 3770k at 4.5ghz. Im not interested in z79 either tbh. Ive just recently switched cases and i like the formula board, system is running great. Maybe another gpu or a new monitor and im set.
 
TI have a Dedi soundcard and Asus for some reason when back to ATX over EATX with Z68 so will be tight if I went SLI

Also cannot use top CPU PCI-E slot as of 140mm cooler again due to being ATX not EATX (measured both my Z68 and Z87)

This Mobo in many ways is a step backwards with no ESATA etc, some reviews said same about the Z77 to Z87 (Z77 being nicer).

Hope they get it right with Asus 97 ROG but I doubt it.
 
But this is the axiom; can you run P95/IBT for hours on end and your system still not be stable? YES

Therefore what does it prove? Only that it can run P95 etc for the specified length of time. Which does give you a certain level of stability, which is fine for many that, as it happens, correlate this to mean 'true' stability.

I myself have had a system run 12 hours of Prime95 no problem only to fall over a couple of hours into a gaming session, with encode in the background. Which is why I say Prime95 is better than nothing but better still...

Actually, I'd be more inclined to suggest looping some Video benchmarks if you want to test for stability in this way.
 
Proves more than not doing so and it BSOD days/weeks later what is so hard to understand?

A MOT on a car does not mean it will not break in two a mile down the road leaving MOT station and its only legal as you leave premises not done the road.

Nobody has said a Prime stable PC is 100% solid and will never fail in other use but tis a start.

How many times above have I stated I know Prime is not going to test the same 3D side of CPU that a game bench like 3Dmark/Heaven will?

I run them all and mem test too then use my PC as normal till it acts up or I want to do more OC'ing then retest.

In the 13+years I have used Prime, the only ones who do not like it are the ones who know their PC will not pass it.
 
Last edited:
Few loops of asus real bench, intel xtu, super pi for ram does me. This is with ram, at 2600mhz 1t, not the 2t from the factory.
 
I can run 500 passes on IBT several times in a row with no errors, start prime95 and can crash within couple of mins, I can also mess about in the bios and have my pc pass a few hours run on prime95 but then fail IBT while crashing shortly into a gaming session.

I can also have my pc setup where it will fail both prime and IBT yet will let me game gor 12+ with no errors or crashes at all. Which asks the question what IS a stable oc pc???

The thing I find crazy in this thread is the questioning of a guy who holds world records for oc so obv 8pak must know a bit about what he's doing.

Until there is an industry standard program for stress testing people will have there own opinions on what's best way to stress a machine. As I said going by posts in here prime and IBT didn't give me a stable pc end of
 
Well running 500 passes several times of IBT shows you are doing it wrong to start with, used to be default to 3 now 10 but 5 is more than enough esp. with a lot of Ram where it takes time to do each pass if its set up correctly (bottom option in stress level to use all Ram).

Did you even read posts above, when is Prime/IBT going to stress same 3D aspect of a CPU as a game or 3DMark/Heaven does?

That is why you run those benches also and nothing will make sure it does not fail doing other tasks but its better bet its stable.

A few mins running a OC on LN2 is not real life use and will not pass Prime/IBT.

There is no industry standard and OC'ing was not supported and was geeky a while ago (I had to solder VR's to Mobo's/GPU's etc), but prime has become the tool most use/trust use and IBT is based on an actual Intel made tool.

I had an AMD rig long ago and I passed every test in the book both benches and real use like encoding movies etc that I did do back then and it would still crash Windows Explorer when I opened my Web Browser 9 times out of 10.

It was not stable and I had to up voltage to the CPU 1 more notch and back then that was 0.25v jumps so I was not ideally happy but it did not effect heat by more than 2C so lived with it stable.

end of.
 
Last edited:
jh30uk said:
Prime and IBT interest many and has become the standard for stability.

IMO it is not stable if it cannot pass them, who cares about a few secs/mins benching?

Well done OP, but a few screenies would help but TBH there is no point in telling lies and only lying to yourself, other will be jealous hint.

I had a 5GHZ stable 2700k but this CPU cannot OC as POS batch, will not even post most time if OC'd past 4.1GHZ.
Click to expand...

jh30uk said:
Well running 500 passes several times of IBT shows you are doing it wrong to start with, used to be default to 4 now 10 but 5 is more than enough esp. with a lot of Ram where it takes time to do each pass if its set up correctly (bottom option in stress level to use all Ram).

Did you even read posts above, when is Prime/IBT going to stress same 3D aspect of a CPU as a game or 3DMark/Heaven does?

That is why you run those benches also and nothing will make sure it does not fail doing other tasks but its better bet its stable.

A few mins running a OC on LN2 is not real life use and will not pass Prime/IBT.

There is no industry standard and OC'ing was not supported and was geeky a while ago (I had to solder VR's to Mobo's/GPU's etc), but prime has become the tool most use/trust use and IBT is based on an actual Intel made tool.

I had an AMD rig long ago and I passed every test in the book both benches and real use like encoding movies etc that I did do back then and it would still crash Windows Explorer when I opened my Web Browser 9 times out of 10.

It was not stable and I had to up voltage to the CPU 1 more notch and back then that was 0.25v jumps so I was not ideally happy but it did not effect heat by more than 2C so lived with it stable.

end of.
Click to expand...

yes i did read above posts and the part highlighted in bold states by yourself if i cant pass THEM then the system is not stable. i was pointing out that i have a system that can pass 1 or the other but can never pass both at the same settings which in your eyes makes my system unstable.

bearing in mind i could run ibt ( iirc ) and it will fail within 2 mins or so yet i have managed to put in about 15 hours last 3 days on ACBF and my girlfriend has been hooked on trine 2 for a week now ( 19 hours all in ) and i have not had a single crash yet you are telling me my system is unstable
 
When l upgraded to x79 from x58 in January, all l did was 20 minutes of Prime, 10 passes of IBT, Cinebench and Asus realbench them gamed for a few hours for the final test for stability for my 4820K:4.6GHz.

I'v never been a fan of running Prime for 24 hours, etc to prove stability.

As for benching you can see my results in 3D, Heaven, Catzilla, Asus Realbench at 4.6GHz and its been running at that ever since.

The only time l got BSOD when l was tuning my ram timings.

As far as l'm aware there is no set standard for a stable PC overclocked or otherwise.

At the end of the day its up to the individual how he tests his new build for stability, them see how it runs after that.
 
Anyone who says they can run 8 hours of prime then crash within minuites of playing a game is talking utter BS.

put simply, a computer is just a giant flashy calculator, they work in the same way, they both calculating maths equations, plain and simple, in most cases 1's and 0's.

Now on to prime95, this software, throws massive sums to the CPU and memory and then waits for the correct answer, if it's unstable and gives the wrong answer, it fails, ie it pushes the cpu and memory to there max ability, what they were built to do.

Anyone overclocking without using prime and just using 3d applications, I would like to see your results with prime, bet you can't run it past the 2nd test.

I can run 3d applications on my 4770k all day long at 5ghz, but can't pass prime anything over 4.6ghz, at 5ghz, as soon as I press the start button, it's all BSODs.
 
Last edited:
Jamin280672 said:
Anyone who says they can run 8 hours of prime then crash within minuites of playing a game is talking utter BS.

put simply, a computer is just a giant flashy calculator, they work in the same way, they both calculating maths equations, plain and simple, in most cases 1's and 0's.

Now on to prime95, this software, throws massive sums to the CPU and memory and then waits for the correct answer, if it's unstable and gives the wrong answer, it fails, ie it pushes the cpu and memory to there max ability, what they were built to do.

Anyone overclocking without using prime and just using 3d applications, I would like to see your results with prime, bet you can't run it past the 2nd test.

I can run 3d applications on my 4770k all day long at 5ghz, but can't pass prime anything over 4.6ghz, at 5ghz, as soon as I press the start button, it's all BSODs.
Click to expand...

I am in agreement with what you are saying.

However there are scenarios where prime stable systems can bomb out quickly running 3D.
On modern Intels the PCI controller is on board the CPU and it can become unstable.
High clocks and multi GPU's will really stress the PCI controller.

Won't affect the majority but benchers will certainly be aware of it.
 
prime95 wont stress the pcie lane/controller like a gpu will hence crashing in games and passing prime

only on ivy/haswell
 
Ill get some screenies done after i finish work and put my daughter to bed.
I use linx,occt(linpack),prime95 and ibt. Ive always done it like this for true stability, also games don't stress your computer as much as the stability tests. I never crash in games apart from bf4 lol.
 
^^ Exactly!! Dear me!!

CPU's have not always had onboard IMC/PCI lanes etc etc The old ways of testing do not apply to these CPU anywhere near as much.

Using your machine is the best way. Intel engineers don't even use IBT or similar.
 
You must log in or register to reply here.
Back
Top Bottom