i5-4670K help to get stabile

[Shou Miko]

i received my i5-4670k today, even running at stock settings one or two cores keeps failing in Prime95

anyone know what settings i should use with my Asus Maximus VI Gene board?

i flashed bios with the newest bios 0711 to see if that helped and so far it didn't.

Description MAXIMUS VI GENE BIOS 0711
1.Revise help string in BIOS.
2.Fix XMP button issue
3.Improve system compatibility.

i use my 4x4gb kit (16gb in total running 1.5v that's stock from Corsair), could it be that my memory needs more voltage or is it just that i need to change some voltage values for the cpu?

 

[Sasqui]

i received my i5-4670k today, even running at stock settings one or two cores keeps failing in Prime95

i use my 4x4gb kit (16gb in total running 1.5v that's stock from Corsair), could it be that my memory needs more voltage or is it just that i need to change some voltage values for the cpu?

That sucks!

That may be the culprit. Did you load XMP profile?

 

[McSteel]

What are the temperatures like? Maybe your cooler isn't making proper contact...
I don't think you can blame this on your RAM, so I'd explore other possibilities. You could verify this by running the small FFT test, which uses no RAM, only the CPU L2 + L3 cache.
No matter how good or bad a CPU is for overclocking, it shouldn't fail with a computation error at stock in prime95, unless it's either defective or overheating...

So try to eliminate heat as a potential issue.

 

[Shou Miko]

That sucks!

That may be the culprit. Did you load XMP profile?

yap XMP Profile always, nuth else, too lazy, only thing i changed from the Command Rate from 2T to 1T, been using that on my i5-2500k and i5-3570k and no problems.

hmm about hit around 93c at 4ghz with auto voltage at 1.264V that's CPU-Z 1.65.1 64bit reading 0.0

but they jump from 62-93c

 

[cheesy999]

yap XMP Profile always, nuth else, too lazy, only thing i changed from the Command Rate from 2T to 1T, been using that on my i5-2500k and i5-3570k and no problems.

change that back, it'll only cause issues if you're trying to find where the problem is

 

[Shou Miko]

oh and i forgot to say i am running IntelBurnTest atm.

 

[Sasqui]

but they jump from 62-93c

Stock cooler? I assume you're experienced mounting a heatsink...

 

[Shou Miko]

Stock cooler? I assume you're experienced mounting a heatsink...

nope H100 with 2x Noctua NF-F12 PWN fans with Low-Noise Adaptors on both.

 

[Shou Miko]

my living room where my computer is atm is about 28c, bcs i can't make an airflow bcs i am having windows on one side of the apartment building i am living in, but i didn't had problems with my Ivy Bridge processor

Haswell with stock settings running prime95:

 

[cadaveca]

Try AIDA64 or IBT. I don't think full Haswell compatibility with P95 will come until v28.1 or so.

 

[cheesy999]

event viewer say anything?

 

[Sasqui]

I don't think full Haswell compatibility with P95 will come until v28.1 or so.

Now that's interesting!

 

[cadaveca]

Now that's interesting!

check the P95 forum, author is working on getting more out of Haswell last I saw, last week, I think.

At the same time, I suspect that perhaps some ram issues might be afoot...not enough info. 0x124 BSOD, even, isn't just vCPU any more. I love Haswell.

 

[Shou Miko]

event viewer say anything?

The computer has rebooted from a bugcheck. The bugcheck was: 0x0000001a (0x0000000000041790, 0xfffffa8008c7e3a0, 0x000000000000ffff, 0x0000000000000000). A dump was saved in: C:\Windows\Minidump\072313-37112-01.dmp. Report Id: 072313-37112-01.

072313-37112-01.dmp	23-07-2013 20:01:08	MEMORY_MANAGEMENT	0x0000001a	00000000`00041790	fffffa80`08c7e3a0	00000000`0000ffff	00000000`00000000	wmiacpi.sys	wmiacpi.sys+fa02da40					x64		C:\Windows\minidump\072313-37112-01.dmp	4	15	7601

i re-applied my thermal paste and i won 3c running in bios actually. temps dropped from 39c to 36c,

 

[cheesy999]

The computer has rebooted from a bugcheck. The bugcheck was: 0x0000001a (0x0000000000041790, 0xfffffa8008c7e3a0, 0x000000000000ffff, 0x0000000000000000). A dump was saved in: C:\Windows\Minidump\072313-37112-01.dmp. Report Id: 072313-37112-01.

072313-37112-01.dmp	23-07-2013 20:01:08	MEMORY_MANAGEMENT	0x0000001a	00000000`00041790	fffffa80`08c7e3a0	00000000`0000ffff	00000000`00000000	wmiacpi.sys	wmiacpi.sys+fa02da40					x64		C:\Windows\minidump\072313-37112-01.dmp	4	15	7601

i re-applied my thermal paste and i won 3c running in bios actually. temps dropped from 39c to 36c,

A page table page has been corrupted

Is the error

Sounds like it is likely to be ram related, though it doesn't rule out the cpu or hard drive

Have you tried turning the memory clock down?

 

[Shou Miko]

A page table page has been corrupted

Is the error

Sounds like it is likely to be ram related, though it doesn't rule out the cpu or hard drive

Have you tried turning the memory clock down?

still using XMP Profile no adjustments now and running Aida64 Stress test instead of Prime95.

 

[cheesy999]

still using XMP Profile no adjustments now and running Aida64 Stress test instead of Prime95.

You may want to backup from the XMP is what i'm saying, your memory controller may not like your ram as much as the last one did

stick it to 1333 or something just to be absolutely sure its not the ram

 

[cadaveca]

still using XMP Profile no adjustments now and running Aida64 Stress test instead of Prime95.

You need to also run it without any test selected other than FPU. That'll show you the heat and voltage boost that you've probably seen reported. And I don't think running one vs the other is a good idea...you need BOTH.

 

[Shou Miko]

You need to also run it without any test selected other than FPU. That'll show you the heat and voltage boost that you've probably seen reported. And I don't think running one vs the other is a good idea...you need BOTH.

i am using the System Stability Test ain't that good?

and i got my temps down to max 70c atm.

 

[cadaveca]

i am using the System Stability Test ain't that good?

and i got my temps down to max 70c atm.

No, it's not that it's not good... you need that to have load on the rest of the chip and ram. The generic-setting AIDA64 stability test is more of a "typical usage" stress test rather than a Hardcore IBT-like test. To get the IBT-like test, you need to deselect the other three options, and leave just FPU selected. P95 uses a different workload.

 

[Shou Miko]

No, it's not that it's not good... you need that to have load on the rest of the chip and ram. The generic-setting AIDA64 stability test is more of a "typical usage" stress test rather than a Hardcore IBT-like test. To get the IBT-like test, you need to deselect the other three options, and leave just FPU selected. P95 uses a different workload.

now i am running only FPU in Aida64 System Stability Test and nuth has happened yet

 

[Shou Miko]

i guess leaving my pc over night running FPU wouldn't cause any damage of some sort right?

 

[cheesy999]

i guess leaving my pc over night running FPU wouldn't cause any damage of some sort right?

Only if your motherboard has really low quality VRM's or your cooler isn't fitted properly

 

[Shou Miko]

Only if your motherboard has really low quality VRM's or your cooler isn't fitted properly

hmm, i refitted my cooler took it off, put new thermal paste on it, and so far i have been running System Stability Test - AIDA64 FPU for 37mins.

and with my case a sampled again i hit these temps:

 

[radrok]

Toasty but still within safe limits... Should be much lower though.

You might've got a "bad paste" CPU... Happened with Ivy, can probably happen with Haswell.

Also dat boobs