I have seen 2300 mark, 2400 u really need a chip which can push around 107blck+ .. Most chips can run dimms @ 2k-2200..Quote:
Originally Posted by makaka
Printable View
I have seen 2300 mark, 2400 u really need a chip which can push around 107blck+ .. Most chips can run dimms @ 2k-2200..Quote:
Originally Posted by makaka
5Ghz 1.36 in bios! Fully Stable in Prime 26.4 12hours 30mins on Small FFT'S!
http://img692.imageshack.us/img692/8...hz7117281t.png
whau, very nice! Where is the absolute maximum your chip? Did u try with example 1.5V max stability?
@bro20000 thats a awesome clock for 24/7 usage.. Really golden chip bro!
Max have taken my chip to is 5.2 1.42v runs vantage perfect just not tryed prime yet! Waiting till i get a chiller for going above 1.45 and on! Thanks am really happy having this chip Batch number L040B307!
http://img6.imageshack.us/img6/9238/...mhz686261t.png
nice oc! I know that raise bclk could damaged the cpu, but lower it to say 98 and raise multi to 51x, could it damage the cpu by reduce the bclk?
Thx
I was trying out cas6 memory Freq! Highest i can get Bclk is 107! I keep mine at 98 for 24/7!
Well I've determined my L045A998 is fully game stable at 100x50, 4988 mhz or something, at all auto voltages (which bios sets to 1.390, though I think that this is actually 1.4v, because Prime bluescreened in 30 seconds when I manually set it to 1.385v) except LLC must be set to level 1 on my UD5 (otherwise windows won't load thanks to the vdroop) and PLL overvolt enabled. CPU-Z shows 1.416v idle, 1.440v load under Rybka 4, and 1.452v-1.460v under Prime. I only run Prime for a few minutes AT MOST because the temps get too close to 90C, and I'm not going to break or degrade my chip trying to prime at high temps. Deep Rybka 4 Chess engine, with 4 cores/8T reaches about 75C max and runs forever, which is a good sign of real world stability. Black Ops doesn't even break a sweat.
No idea why LLC level 1 RAISES The vcore that much. Thought only LLC level 2 was supposed to do that...
I should note that with HT disabled, Rybka 4 reaches 66C instead of 75C.
POSTS at 5.2 ghz with all auto voltages (bios reads 1.395), but I don't even WANT to know what vcore that needs to load windows. Probably like 1.5v to run games, which I'm not going to try.
Chip runs at all auto voltages (even with LLC disabled) at 4.5 ghz...everything auto (had Overvolt PLL enabled btw).
Fantastic overclock and a wonderful chip you have there. Top job!Quote:
Originally Posted by bro20000
Well only a few people have killed there chips tbh, High BLCK + High voltage usually results in degrading then kills the chips.. Using lower blck = lower memory frequency's so its best to find a good clock where your happy with the core + memory freq..Quote:
Originally Posted by wekktor
So I have a scenario with my 2500k where at 4.9ghz and 1.385 vcore the system is linx stable (20 runs, all memory used, SP1 not loaded), OCCT blend stable (I let it run 4 hrs then stopped it), and memtest86+ stable (I let it run for 5 complete passes then stopped it). It fails prime (version 26.4) blend, usually within 1 hour with a stop 124 blue screen. It also crashes with handbrake video encoding with a stop 124 blue screen.Quote:
Originally Posted by Sam_oslo
I increased vcore to 1.395 and now it encodes video without errors so far, but still bluescreens with prime blend. It also bluescreens and freezes in games at this increased vcore.
At 1.415 vcore I'm still crashing with prime blend.
So what is the logical next step if I want to keep running at 4.9ghz? Keep increasing vcore until I'm prime blend stable? Or would prime small fft's be more diagnostically useful? My memory (2 x 4gb gskill sticks) in all cases is at 1600mhz with stock timings and voltage.
Attempting to achieve overclocked real world stability with these Sandies is taxing my feeble brain. With previous generation intel chips, it seemed to me that linx and OCCT stability were usually within a couple of vcore notches of prime blend stability, and I don't recall ever having stability issues if I could prime blend for 12+ hours without errors. I wonder if I should still be aiming for prime blend stability as my gold standard of real world stability. I did try loading up SP1 and using linx with avx, but my temps on air cooling (mid 80's) made me uncomfortable and I stopped before achieving the requisite vcore to be linx + avx stable.
I don't want to concede defeat yet at 4.9ghz!!!:mad:
FWIW, I'm on a Biostar TP67XE and I was using offset vcore with LLC off and just kept having to add vcore to get prime blend almost stable. Seemed like it would never end.Quote:
Originally Posted by drnick
When I switched to FIXED vcore and LLC on I was able to get prime blend stable at 4.8GHz 1.416V load.
Of course, now the thing randomly blue screens at idle. :shrug:
This is a bit strange, but that stop 124 blue screen is suggesting RAM/IMC/Uncore-problem, but since you have passed memtest86+, your RAM should be OK.Quote:
Originally Posted by drnick
So, probably it's a IMC/Uncore-problem (that's the equivalent of "System Agent" or VCCSA/VCCIO-problem on SB). This gets more likely when adding vCore doesn't help, and Prime Blend would react to that too.
Make sure your LLC is not bugging tho. Intel recommends standard LLC.
EDIT:
Try to run a HyperPi32M, that's a good and fast test for pushing Uncore/IMC.
Try Custom FFt's in Place 1,5hQuote:
Originally Posted by bro20000
The most important is 960K test!
Here's mine!
http://www.abload.de/thumb/50032c77.png
http://www.abload.de/thumb/37609ii3x.png
52 Multi
Not really much point trying that test! System being updated tomorrow morning with G.Skill DDR3 PC3-17600 Cas 7 2200MHz RipjawsX!http://www.memoryc.com/products/desc...nel/index.htmlQuote:
Originally Posted by emertX
Thing is on G.skill website they shown with black heatsinks so am not sure what colour am getting!
http://www.gskill.com/products.php?index=374
That's bizarre...do you remember if the blue screens are a stop 124 error or something else? I'm on an ASUS p8p67 evo mobo using fixed vcore and LLC set one notch below extreme (ultra high or something like that). You were able to get linx + avx stable though, correct?Quote:
Originally Posted by ugotd8
I was wondering about an uncore/system agent issue, but I thought OCCT blend might have flushed that out too. In any event, I'll try hyperpi and report back. I could also play with VccSA and VccIO, although ugotd8 and some others have reported that it hasn't particularly helped them. Not sure what a good ceiling voltage for either of those settings is anyway.Quote:
Originally Posted by Sam_oslo
I could always turn off the LLC too and see what that does. I would have thought that a flaky LLC would cause it to crump in Linx though. Not necessarily?
Hi,
here is my i5-2500K. Batch# is L039B120. Tested on air (Scythe Ninja 3) with GA-P67A-UD4.
5001MHz Prime Custom FFTs in-place 1.5h Run:
http://www.abload.de/thumb/5001prime0m71.jpg
5501MHz Super PI 1m Run:
http://www.abload.de/thumb/5500-pi1mumnv.jpg
:wave:
In regards to BSODs, it's always been 124 with this chip. My first one was always 101. Doesn't matter what's wrong with the OC, always 124. Same board two different chips.Quote:
Originally Posted by drnick
Yes, I'm linx + AVX and prime95 blend (12K custom) for one hour stable.
I am also having weird stability issues. At 48x100 and 47x100 (haven't checked other multis) no matter what vcore I use it fails prime blend almost instantly (BSOD) but I am completely stable with all other stability tests (linx w/ avx, futuremark, gaming, etc). It's only a problem with prime95 so I just don't run it anymore, lol.
Great chip,56x no go :) ?Or bclock overclocking?Quote:
Originally Posted by Digg_de
I was ready to just ignore the prime blend instability too in light of all the other tests I could pass, but I will periodically blue screen during gaming, so I'm wondering if there might be something that the prime failure is picking up that has real world application consequences.Quote:
Originally Posted by babalouj
my friend`s cpu have same batch id but him cpu need 0.1 less voltage than me for same clock.
i have ud4 and he have p8p67 pro. batch id mean nothing or Gigabyte suck?
here my max oc batch L036B120T
http://i53.tinypic.com/33mprh1.jpg
Quote:
Originally Posted by emertX
Quote:
Originally Posted by Digg_de
I knew I saw those results before. :)
Welcome, Digg. :welcome: