i5-2500K/2600K batch number and o/c results

[crossg]

It's not that it reports the wrong freq, but that's what your base clock before turbo is. It's also possible that 3dmark isn't able to stress the clock all the way up.

Nevertheless, the need for a non-loop intensive stress test is necessary to make sure you don't get chokes from the decode. I assure you this isn't an easy job; there are teams of people in semiconductor companies involved with just creating tests to break the system, a team which I am a member of

Agreed.
The reason I mentioned this was my 875K, 980x and 2600K all report the wrong clocks. The i7 950 reported the correct OC clock. I don't use turbo in the other chips.

[ugotd8]

It's not that it reports the wrong freq, but that's what your base clock before turbo is. It's also possible that 3dmark isn't able to stress the clock all the way up.

Nevertheless, the need for a non-loop intensive stress test is necessary to make sure you don't get chokes from the decode. I assure you this isn't an easy job; there are teams of people in semiconductor companies involved with just creating tests to break the system, a team which I am a member of

Well, I've only understood half of what you have said so far, but thank you very much for shedding some light on this. I've noticed the peculiarities of stress testing on this new chip and I'm frankly a little lost as to when/how to call an overclock stable.

I see people on other forums hammering their chips with voltage trying to get Prime95 to pass. I on the other hand am holding off until this is sorted, it just doesn't make sense to me that on the 1366 platform the linpack test was the killer (highest load) and now on the 1155 platform that has miraculously switched around and now Prime95 is the killer.

[KURTZ]

5K!

watercooled onto a P8P67 ... not so bad ... don't you?

ps. it's a sample ...

[Ccaution]

Not bad at all - nice bbs too

[famish99]

Well, I've only understood half of what you have said so far, but thank you very much for shedding some light on this. I've noticed the peculiarities of stress testing on this new chip and I'm frankly a little lost as to when/how to call an overclock stable.

I see people on other forums hammering their chips with voltage trying to get Prime95 to pass. I on the other hand am holding off until this is sorted, it just doesn't make sense to me that on the 1366 platform the linpack test was the killer (highest load) and now on the 1155 platform that has miraculously switched around and now Prime95 is the killer.

Feel free to ask about what you don't understand.

Different architectures have different failing mechanisms; on Nehalem the execution units (add, multiply, load/store) were the bottlenecks, but Sandy Bridge it seems the instruction decode is the bottleneck.

Linpack is designed to be more execution stress whereas Prime95 could possibly require more decode stress (especially if you choose blend instead of small fft).

I cannot verify this (still waiting for my Sandy Bridge), but if more people have problems with Prime95, that might be the goto test until more evidence comes in.

[flesheatinvirus]

I used Prime95 on my 2600K. 4+ hours of blend at 5G was no problem. batch L040B208

[ugotd8]

I used Prime95 on my 2600K. 4+ hours of blend at 5G was no problem. batch L040B208

You also have one of the nicest WC setups I've seen, and your chip never gets above 60C right ?

Very frustrating.

[ugotd8]

Feel free to ask about what you don't understand.

Different architectures have different failing mechanisms; on Nehalem the execution units (add, multiply, load/store) were the bottlenecks, but Sandy Bridge it seems the instruction decode is the bottleneck.

Linpack is designed to be more execution stress whereas Prime95 could possibly require more decode stress (especially if you choose blend instead of small fft).

I cannot verify this (still waiting for my Sandy Bridge), but if more people have problems with Prime95, that might be the goto test until more evidence comes in.

I'll be interested to see your thoughts once you get your SB. A 2600K I assume?

In the meantime, there are others that have noticed this:

http://www.xtremesystems.org/forums/...d.php?t=264940

Thanks again.

[flesheatinvirus]

Thank You for the rep. Yeah... That is true cooling is no longer an issue. I can hold this chip pretty low under load. My torture rack is made for builders.

[makaka]

Just got my 2600k , either i am retarded or there is somethings wrong i wasted 1 h in the bios without finding how to increase the multiplier it look locked at 34 no matter what i do i can't increase it, i played with major setting part the bios still show the cpu is at default speed , the motherboard is asus p67 pro
pls help

[ugotd8]

Just got my 2600k , either i am retarded or there is somethings wrong i wasted 1 h in the bios without finding how to increase the multiplier it look locked at 34 no matter what i do i can't increase it, i played with major setting part the bios still show the cpu is at default speed , the motherboard is asus p67 pro
pls help

Never used that new bios on the ASUSs, but... Make sure turbo is on, speedstep is on, and then set your turbo multipliers for each core.

[zoob]

L040B208 + P8P67 Deluxe 1053 + 2x4GB Vengeance
Prolimatech Megahalems Rev B + Scythe Gentle Typhoon 1850RPM
48 x 100 @ 1.335 V BIOS LLC Extreme
VCCSA 0.925
VCCIO 1.075
CPU PLL 1.80
PCH 1.05


Same batch as flesheatinvirus but not as nice results!

[flesheatinvirus]

Very nice. That is a potential killer chip my friend.

[ugotd8]

L040B208 + P8P67 Deluxe 1053 + 2x4GB Vengeance
Prolimatech Megahalems Rev B + Scythe Gentle Typhoon 1850RPM
48 x 100 @ 1.335 V BIOS LLC Extreme
VCCSA 0.825
VCCIO 1.075
CPU PLL 1.80
PCH 1.05


Same batch as flesheatinvirus but not as nice results!

I have the same batch as well, and need 1.4 for 4.8 so don't feel bad. :-)

My 2x4Gb Vengeance sticks should get here tomorrow.

You lowered your VCCSA ?

[-=DVS=-]

Batch# L040B705
MicroATX P8P67-M PRO
VenomX + 1500 RPM SFlex Fan, inside Antec P180

[zoob]

I have the same batch as well, and need 1.4 for 4.8 so don't feel bad. :-)

My 2x4Gb Vengeance sticks should get here tomorrow.

You lowered your VCCSA ?

Sorry typo on that! It's been corrected.

[PolRoger]

L040B208 + P8P67 Deluxe 1053 + 2x4GB Vengeance
Prolimatech Megahalems Rev B + Scythe Gentle Typhoon 1850RPM
48 x 100 @ 1.335 V BIOS LLC Extreme
VCCSA 0.925
VCCIO 1.075
CPU PLL 1.80
PCH 1.05


Same batch as flesheatinvirus but not as nice results!

That is a nice batch and your still ~200Mhz higher than many SB 2600K at similar volts!

[drnick]

It appears that this thread is pointless as every single 2500k/2600k can do 4.8g stable with good air cooling. There is no golden batch like i7-920 3849a

I'm not sure that is entirely correct- I can't get past 4.5ghz (45*100) on my 2600k on the asus p8p67 evo with either the 1003 or 1053 beta bios.

Batch# L041B741

I can get to 4.5ghz prime stable on air cooling (venomous-x) with 1.39xx vcore under load. I've gone as high as 1.475 vcore and still can't get into windows at 4.6ghz or above ( I can post at 4.6 ghz and above but windows freezes up during the opening splash screen). I've played with most of the settings such as phase control, LLC, internal PLL overvoltage, VRM frequency, etc, and 45x seems to be the wall for this chip- or at least a wall that 1.475 vcore with extreme LLC won't break (and I'm not willing to push more that that right now). Anyone have a similar experience?

Figures I would get a lemon...who would have thought that 4.5 ghz would be considered a less than stellar OC performance?!?

[ugotd8]

I'm not sure that is entirely correct- I can't get past 4.5ghz (45*100) on my 2600k on the asus p8p67 evo with either the 1003 or 1053 beta bios.

Batch# L041B741

I can get to 4.5ghz prime stable on air cooling (venomous-x) with 1.39xx vcore under load. I've gone as high as 1.475 vcore and still can't get into windows at 4.6ghz or above ( I can post at 4.6 ghz and above but windows freezes up during the opening splash screen). I've played with most of the settings such as phase control, LLC, internal PLL overvoltage, VRM frequency, etc, and 45x seems to be the wall for this chip- or at least a wall that 1.475 vcore with extreme LLC won't break (and I'm not willing to push more that that right now). Anyone have a similar experience?

Figures I would get a lemon...who would have thought that 4.5 ghz would be considered a less than stellar OC performance?!?

Too much vcore or maybe (most likely) memory voltage or timings.

[Beatle]

Batch# L040B307

5ghz 1,344v 8hours prime blend stable

[poke349]

P8P67 Pro
i7 2600K (Batch: L040B698)
4 x 4GB GSkill Ripjaws @ 1333 MHz

Testing stock vcore for now. I'll increase it later.

Stock vcore:
1.165v (BIOS + AI Suite)
1.264v (CPUz + AI Suite Monitor)

- Everything is left at stock except for multiplier.
- All temps were too low be noteworthy.

4.5 GHz:
IBT/LinX/Linpack: > 5 min. stable
prime95 (x64): instant fail
y-cruncher (SSE4.1): instant fail

4.4 GHz:
IBT/LinX/Linpack: > 5 min. stable
prime95 (x64): 2 min. fail
y-cruncher (SSE4.1): 1b stable (377 secs)

4.3 GHz:
IBT/LinX/Linpack: > 5 min. stable
prime95 (x64): > 5 min. stable
y-cruncher (SSE4.1): > 5 min. stable



*Currently, y-cruncher only supports up to SSE4.1. AVX support will come in a few weeks. I'll re-test then.

[FlanK3r]

awesome voltage! And your cooling?

[Beatle]

awesome voltage! And your cooling?

Noctua DH14

[C-N]

I don’t find this thread pointless, even if everyone submitted there 4.8GHz results it would still tell us which batch appears to run hot/cold low/high volt etc.

Would be nice to make the first post a list to make it easy for people to compare chips.

i5 2500K L041B517

5.0 @ 1.445
5.1 @ 1.46
5.2 @ 1.50
5.3 @ 1.57

LinX 20+ stable on air.

P.S I don’t go much on batch anyway!

[drnick]

Too much vcore or maybe (most likely) memory voltage or timings.

I thought about that- but I'm running memory (2x2gb ddr3-2000mhz sticks that worked flawlessly in another computer) at 1333mhz with loose timings. I'm pretty sure it's not a memory issue.

I also thought about the possibility of giving it too much vcore- but it won't load windows at 4.6ghz or above at vcores ranging from 1.2 to 1.475.

Any other suggestions? I really think that 4.8ghz is not a given with all 2600k chips based on my experience with this one.