Printable View
u can see it on the box and chip where it was manufactured...Quote:
Originally Posted by i_max2k2
for testing max multi its very easy, just disable HT and leave 2 cores 2 threads, put 1,55 on bios and start 100x50, x51, x52....till u see a flashing cursor...meaning that its not entering to load windows no matter what, put internal PLL auto or enabled to do this
have fun !
cheers
Sergio
Oh if thats the case then I think mine was made in Malaysia. I'll try this tonight.Quote:
Originally Posted by andressergio
Thanks!
So I had the craziest situation happen this past weekend. I decided to move my current (not so great) 2600K from my MIVE main rig to my media server paired with a P8P67LE and go buy a new 2600K hoping for a great clocker.
I recorded some Batch# to look for, specifically Andressergios #L045B836 and Hazzan #3103B326 before going to MicroCenter St. Davids. When I got there they had all the Batch# readable through the glass case so I started to check them out and compare to the batches I was looking for. They had about 12 in the case that where all 3115BXXX batches and in the back corner was one 3103B306. I was excited as this was very close to Hazzan’s 59X chip so I bought it and headed home to test it out.
I pulled my existing CPU and put it in the P8P67LE. I put the new chip in my main MIVE rig (Bios 1850) and it booted right up. I wanted to find max multi so I went back to bios, put in 55X, 56X, 57X, 59X with memory at 1866Mhz instead of my standard 2133 and put vcore at 1.56v. It booted windows all the way up to 56x no problem. It posted and tried loading windows but just before it gets to welcome screen it locks up on 59X multi. At this point I am pretty sure I got a great chip.
I restarted and put multi back to 56X with 1.565vcore and memory to DDR3-2133 to test SuperPI 1M and 32M real quick. It booted into windows fine and ran 1M fine, 6.733s 1M. I then tried running 32M and about half way through benchmark my monitor went black. I thought it tried to reboot meaning it needed more vcore so I turned off PSU via switch and then board would not post at all. I tried switching to second bios, tried clearing cmos and no matter what I did the Board would not post at all. I swapped memory, one stick, 2 sticks, etc but nothing. I pulled the chip and put my old 2600K back in and the Board booted into windows no problem at 4.8Ghz. I then put the new 2600K in the P8P67LE board and nothing, no post and no boot.
Within 3 hours of me buying the chip and me thinking I got a great clocker, I took it back to MicroCenter and they exchanged it for a new one. They let me choose any Batch# I wanted and even let me look at their stock in the back to see if I could find another 3103B3XX chip but no luck. That chip was the last one and all they had was a ton of the 3115BXXX chips and one 3108B401 chip which I bought. I can get the new chip rock stable, LinX 25 runs maxmem at 4.8Ghz 1.44vcore. This indeed is not the great clocker I was looking for. Additionally, I am unable to get this new chip stable at 5Ghz LinX up to 1.475vcore and it won’t bench past 51X multi. I have a water setup that runs great and keeps my chips temps under 80C LinX stable.
So, I am looking for feedback. Did I really kill the 3103B306 chip with only 1.56vcore? I see people benching all the time at that voltage on water so how did I kill my potentially great clocker while doing a quick 32M run? I will probably try to sell my new chip at a discounted price (only about 4 hours on it) and try another batch in hopes of a great clocker.
Just trying to figure out what I did wrong and how I killed that chip with 1.56vcore and no temp issues?
did u read what i post mate ? same happened to me on MIVE with my L045B---i was on old bios and tried to enter windows at x59...then i upgraded to 1850 and max multi becomes x56...i dont know if its chip or just BIOS error...Dumo told me that he had to set manually all cores to x57 for ex but i did it and no luck same blinking cursor that means no go...Quote:
Originally Posted by deachus
if anyone can clarify this would be nice for us
btw mine is L045B386 i think i made a mistake posting the #
cheers
Sergio
Yeah, I don't know, something just didnt seem right. I have been doing this for at least 11 years and by no means am I a professional but I take every possible precaution and always want to research/test to know how things work and why things happen. I cannot come up with an answer on this one. I also had all cores set manually when benching at 56X when the monitor went black and then refused to post no matter what I tried. Chip was definitely dead but how or why I cannot explain.
Most likely thats a 56X chip.Quote:
Originally Posted by deachus
One of the sign of a dead chip is when you turn on the mobo it will light up for a second and off.
Imo, don't do back to back benches @ high vcore on air/regular water cooling 4 Cores/8 threads HT on....instead reboot after each bench completed, clear cmos and start oc. setting from bios again. The temperature swing from load to idle and back to maximum load in a few second probly cause it.
Another causes are condensation in the socket (subzero cooling) and freak vcore spike on certain mobo.
Thanks for the input Dumo. I will reboot after each bench as suggested in the future. The other thing I did not do is turn HT off and set to 2cores. I was benching 4C/8T HT On when the chip died. Yes, the Board would turn on and light up then shut off and then would come back on and stay on stuck on 79 post code (CXM Initialization) but would not post or show anything on monitor. Did that on both rigs so it was definitely dead.
Now back to trying to find another good clocker:(
Actually, on mive post code will stuck at "00" for fraction of a second and mobo will turn on if you unplug the 8 pins cpu power.Quote:
Originally Posted by deachus
Did you take the battery out for a few hours and reboot again?
The last thing I tried before returning the CPU for a new one was taking the battery's out of both my MIVE and P8P67LE for 1 hour then put the battery's back in and reinstalled CPU and nothing on both Boards. No post, nothing. The only way I could get the Boards to POST was to put my other 2600K in and then everything ran great. I believe it was definitely dead.
@ deachus
I am so sorry to hear you got such a nice chip and then it died:(
Only 1.56v under water? I have taken my x55 to 1.58 4c/8t every time with no problems:shrug:
i dont think it died it happened the same to me l0ud i was able to take it to x57, x58 now i got blinking cursor if i go over x56...but its a very nice chip either way as i can bench at low volts and only takes 1,408 LOAD to be 5GHz stableQuote:
Originally Posted by l0ud_sil3nc3
i also booted to win int pll disabled on x51 now only on x50 but happy too
#L045B386 is my batch
@ andressergio
I would agree with you if I could get my Board to post again after the 56x 32M run but no matter what I did on either rig the Board would not post. I couldnt get into bios, blank screen. You can still get into windows and bios and your chip runs fine just at a lower multi so I dont think the issues are the same. Some sort of high volt rapid degradation? I dont know
so weird man, also weird what happened to me, those 2 issues like the chip settled or wathever on those multi...when i first build i was using old bios 1204 on the MIVE and installed WIN7x64 from scratch and passed post and chip tried to enter win till x59...after i flashed to 1850 its till x56, i would have to go back to 1204 and retest, but for me its a bios problem and those boots wheren't accurateQuote:
Originally Posted by deachus
try my batch you wont regret i see min x54 on very nice low volts
cheers !
Sergio
Dumo i keep not understanding the throotle issue i used Tstop and same thing, also i cant find on MIVE BIOS where on earth is Bidirectional Prochot...Quote:
Originally Posted by Dumo
can u elaborate your idea ?
cheers
Sergio
First of all, sorry to read about your 59x chip death @Deachus. I have lost a chip a while back (54x) on a MIVE that died. I then got a MIVE-B3 that was a very crudely modded B2 which also failed so I can appreciate your disappointment although a 59x chip is obviously a shocker.
@Andressergio.........Are you not talking about a function on the Gig-UD7 board in the CPU section. I can't remember it on my MIVE's either..........of course it's not unusual for me to be wrong:up:
I'm now using a GENE-Z and it is an easy mbd to use and clock. A bonus is all my ram sticks (4 sets) are in the QVL list. Also, good for me is the lack of bells and whistles that I don't need as fitted on the MIVE-Z. I might miss two PCIe-16 slots but not yet.
I think I've got a good one here, 5670 max and what looks like a good 24/7 chip...........MIV-GENE-Z (Z68)..............batch 3103B374.....on water cooling..........
Attachment 119031
It seems many of the chips that died were on the MIVE. Dumo, Hazzan, newhit, deachus to name a few, were using this board at the time the CPU died.
Dumo has tested close to 100 2600k's, the majority were tested on the UD7. The only one he has lost died while using the MIVE. I wonder how many others have lost a chip on this board?
very good mate mine its almost there but with HT stable 5GHz on 1,416 on the MIVEQuote:
Originally Posted by newhit
yes i was talking where on earth is BIDIRECTIONAL PROCHOT on the MIVE lol
cheers !
Sergio
I read some lost some chips on MIVE Dumo was one of them...its a very impressive board and volts are very accurate, u put 1,41 on bios and its 1,416 IDLE / 1,408 LOAD and thats using 75% calibration, on 100% its the same, as for my GIGA you put 1,41 on bios and depending on the bios if you use the 10 LLC levels that are really bad u can get even 1,5. So i guess if you come from GIGA and do that on MIVE u can pump lots of volts easy on the chip at first ending on a death.
of course just my opinion.
so far what i found on the MIVE its just amazing, still got my love UD7 but it takes lots of time to do the same that MIVE does easy, also my UD7 boots 107 bclk BUT RANDOM, the MIVE does 108 everytime, never goes down, calibration is amazing
cheers to all
Sergio
Looking at your results and Batch# makes me want to go nuts. Are you able to pass 5Ghz LinX maxmem 20+ runs with 4C/8T at those volts? Or just 4C/4T? Nice CPU regardlessQuote:
Originally Posted by newhit
Check this For Sale thread out http://www.xtremesystems.org/forums/...chips-for-sale.Quote:
Originally Posted by OC Nub
He is selling (2) 2600K's that are same exact batch as my dead chip (3103B306) and he is pumping between 1.68v & 1.76v through these chips at 57x100 on water and SS on UD7 no issues. And he was asking $750 each for the chips?WTF
What the hell hapened to my chip? Are you on to something? Is MIVE doing something we don't know about? I'm so confused and frustrated:confused:
It's me who sold them. I'm not sure what you want to hear? I certainly do not want to offend in any way. I used the UD7 and GENE-Z mbds to run them. I kept well away from my old MIVE because it seemed a bit risky. I know that you are all going to remind me that the GENE-Z is a scaled down MIVE but my thinking is that this mbd is a more refined baby MIVE...........anyway it's very quick and easy to use with no apparent suspect areas. (Don't mention the Skews).Quote:
Originally Posted by deachus
I'm not sure of course but is the LinX run not harder with 4C/4T? I will obviously redo the run with 4C/8T anywayQuote:
Originally Posted by deachus
I am not looking to hear anything from you, i am using your posting as examples of a theory starting above regarding a trend of MIVE boards being used when 2600K's died, that is all. I did not post/use your thread to bang on you, knock you, or tell you your Gene-Z is anything but fine. Above I posted how my 2600K (same batch as the ones you sold) died with only 1.56vcore at 56X. I think you are reading into my posting a little to much, take it for what its worth. :up:Quote:
Originally Posted by newhit
If I recall correctly when reading the LinX thread, back in first quarter 2011 when SandyBridge was new the linpack libraries used by LinX stressed the CPU harder when CPU was set to 4C/4T in bios. Since the new linpack libraries came out I believe its stated in the LinX thread that 4C/8T now stresses just as hard or harder.Quote:
Originally Posted by newhit
Make sure you download the latest LinX and Linpack libraries and then re-run using 4C/8T. Let me know if you cant find them and I will PM you the link
no deachus 2 exact batches will do dif clocks, MIVE aint doing nothing more than i said...Quote:
Originally Posted by deachus
well anyone has the right to sell his things at the cost he wants but...750 its just crazy inmo