Congrats guys some awesome runs here that im kinda ashamed of posting mine but well here's some linx with my hypers tights at PL7
4500 Mhz / 1600 6-6-6-18 PL7 2T
http://img203.imageshack.us/img203/3...cr2pl7linx.jpg
Cheers :toast:
Sergio
Printable View
Congrats guys some awesome runs here that im kinda ashamed of posting mine but well here's some linx with my hypers tights at PL7
4500 Mhz / 1600 6-6-6-18 PL7 2T
http://img203.imageshack.us/img203/3...cr2pl7linx.jpg
Cheers :toast:
Sergio
Alright, I'm back.
The OP is fixed (without errors, I hope!). Thank you for your patience. :)
Excellent results, everyone. Especially PiLsY - awesome clocks using water cooling! :up:
I'd like to remind you that it's a LinX club (read the rules, please), and you have to list your motherboard. Takes 5 seconds, anyway. Batch listing is not compulsory, but it helps.
Also, thanks to those of you who is trying to format parts of their posts similar to the OP - it makes editing easier. :yepp:
Welcome back zalbard.
Have a screen using 4gb of memory but cannot find it, will put the chip back in next week and try again using 4gb, and try 4600-4700. I dont really like this chip as much as my others, it doesnt scale well after 4500. It is the coolest running chip I have ever had, it refuses to get warm even in LinX.
i5-655k, Gigabyte P55A-UD6, 2x1gb HyperX, Water, W7 64
Edit, this was on the P55a-UD7-now dead, maybe it will do better on the UD6.
http://img97.imageshack.us/img97/7346/capture2ng.png
Uploaded with ImageShack.us
Damn PiLsY, 3rd spot on water, that is amazing. Awesome run. :up:
Good afternoon Mr Zalbard,
Please update my Gulftown entry - I wanted to be the first linx stable 32nm over 5ghz, this is with two stage Cascade cooling. :D
Same 980x as previous - 3003B287
The lowest evaporator temp was -98, the lowest IHS temp was -57.
I don't think this is the lowest cpuv for this chip @ 5ghz - I did not spend a lot of time adjusting.
I have uncore and memory set very slow.
I have some happy snaps located at - http://www.xtremesystems.org/forums/...d.php?t=255686
Also, I didn't want to put the added stress of HT on the chip though I do think the chip would do 5ghz HT - perhaps even at these same settings.
Anyway - so screen shot -
http://img831.imageshack.us/img831/7065/67928148.png
Hi Guys! May I know how's the ranking is made on the first page.
Glad to join the club!
Well done SteveRo, it was only a matter of time me thinks :up:
Now get it done again with a bit of HT goodness thrown in :D
Nice SteveRo :up:
And yes when you have a chance give it some nice uncore/mem and HT :cool:
OK this time it's the E760 without throttling bug. Still with Venomous-X lol
I hope it qualifies this time.
sniper_sung | Core i7 980X [6C 6T] @ 4509.0MHz | 1.388 | EVGA E760 | Air | 3013A524
http://img838.imageshack.us/img838/3492/980x45linx2.png
Wow, 92 GFlops at only 4,5GHz. I am impressed here. Nice job.
I want one :(
980x (6C/12T) @ 4530 MHz - Vcore @ 1.44375v - Gigabyte GA-EX58-UD4P (rev. 1.0), bios version F13 - water - 3003B320
http://i286.photobucket.com/albums/l...53GHzVcore.jpg
From my observations disabling HT on gulftown can increase gflops. I only used stupid loose settings for qpi and mem trying to reduce pressure, cuz you know i don't have a decent cooling solution...
Because the OP doesn't accept my previous post :shakes: I had to torture my 980X again with very poor cooling
Thanks! :)
Nice job! And yeah, I'm sure you won't stop at this point, the start looks too damn good! :D
Thanks, and you're welcome! :)
I'm trying my bets to keep the thing up to date. I am sorry it's so sluggish recently, though... I was away for 2 weeks, came back and 3 days later my Corsair HX1000 dies. :shakes: It's sorted now.
Holy crumbs! Updated, and waiting for more...
It is split into several categories depending on the micro architecture and chips used. Then it is just a matter of who has the highest possible CPU clock.
I think joining is a good indicator of the quality of your components and your overclocking skills, but if you wish to compete and beat a few people on the list by having a higher CPU clock, why not, it's good fun! :D
Hell, yeah, it sure does! :yepp:
Agreed! :up:
Some people may be offended by you saying that VX is "poor cooling". :p:
Jokes aside, there is nothing wrong with a little bit of challenge and following the rules that were initially set. Not to mention that it makes the membership more valuable and the achievement more significant. ;)
Unlucky on the psu dude - at least it should be under warranty? 7 years on those isnt it?
And the whole point of this is that its hard. If you cant stand the heat dont try and join! Ive run all my chips at 90+ to get my place - suffering no ill effects either. Over 90c in linx puts hairs on your chest!
Please welcome me to the club:
Olly_K | Core i7 980X [6C 12T] @4511.2 | 1.423 | EVGA 762 | Custom Water| 3003B396
http://www.ollyk.com/4.5.jpg
Going to get the ram tightened up and try and drop some volts
I know this is stupid but this is actually stable with linx, the ss was taken after a few times of rebooting as i was adjusting sub timings etc..
http://img227.imageshack.us/img227/4...ysoserious.jpg
@John - nice low vcore!
yes very low vcore. John can you show a proper screenshot with Linx...I have never seen a 930 come anywhere near that clock/volt combo...20+ linx stable
There are some newer low vid 930's out. I picked up one from microcenter this past weekend. Its possible Johnny87au got 4500 stable on low voltage. My watercooling met its match with this chip. This is not linx stable but the chip boots at 4500-4600 at very low voltage. I am working on stability now but the temps are a problem.
http://img227.imageshack.us/img227/7814/4500930.png
Uploaded with ImageShack.us
The packing date is june of 2010, maybe there will be some good bloomfields still to come.
I am having a heck of time with temps, I have only tried a full run of LinX at 4.0ghz so far and temps were almost 60c at 1.184v. My other chips are much higher vid and temps were so much easier to control. Im not sure I like the low vid, reminds me of my q9650.
for some reason my LinX keeps stoping after the first test to what seems like ghost errors...
i7 980x @ 4.5ghz (39x115) w/ 1.4125v, qpi - 1.475v (12gb memory), PLL - 1.61v, dram - 1.643v
data link @ 2777mhz and uncore 2425mhz (testing with my G.skills, not Dominators @ 1613mhz 7-8-7-24-72-2N)
every other stability tests i have go through fine but linx just hangs...
it's not temp either cause i'm averaging around 68*c on water..
but every test done (which i did atleast 20 single tests to get an average), it averaged beteen 58 and 62 Gflops w/ hyperthreading enabled... then again i'm not sure how much that counts cause it was 20 single tests with 1 lap each, not 1 test with 20 laps.
EDIT: one thing i think could benifit this program... and just about any other stability program period is if it freakin tells you what the error is or is caused by... LinX will not finish completely stable for me even on 100% stock clocking. i then set it to ignore errors... test completes and says "completed with errors" but doesn't tell you how many errors there are or whether or not they are heat errors or critical voltage errors or anything...
at the same time Vantage tells you your gflops anyway in it's CPU tests. test 1 is in Mflops and test 2 is in Gflops. i can further confirm this by looking at the vantage test as it's going, test 1 i average around 5600-5900 Mflops and test 2 average around 57-58 Gflops.
batch #3004B176, looks like 4400 will require around 1.328v load, still working on it.
One from me.
i7 920 | 3922A423 | R3E | 1.45v | h2o
http://i903.photobucket.com/albums/a...c/d25ef2fa.jpg
Will take some pics tomorrow guys.. As for timings i didnt even tweak them as i losened them, Im sure i could hit cas 7 quite easily as thats what their rated At... Ive got a unique batch number really havent seen many around, its #3938b285
Playground - nice result but looks like youve got some throttling going on there. Im not familiar with asus x58 boards so cant really advise beyond that but it should be fixable in bios. Maybe totally disable speedstep/eist/c-states? Just a guess. Forum search on the intel forum should sort you out for sure :). Doesn't mean its not stable btw, its likely a current throttle not temps or anything :).
Why u think so? What throttling? :(
As I remember all speed step and c-states disabled.
only thing i can see is gflops not being consistent, but that also can be due to opening programs or surfing the web while running linx. sometimes linx is flaky and you just dont get a good run.
It's not "throttling " at all. What we are seeing is just about normal for Linx.
Don't forget guys, Linx is flakey at the best of times and Gflop's go up and down like a yoyo especially when you are on the limit.
Agreed with the above. About the cleanest Gflops run I've ever seen is the one linked in my sig. That is with no pagefile and no user activity on the system at the time.
Sorry yeah shouldve given more detail. Its varying too much. Even opening and closing programs wouldnt give a variance like that. Linx doesnt give up thread time easily.
I never get more than a 0.1 variance across the whole run (sometimes the first loop is a bit off, but after that rock solid). All my submissions on this thread have a variance of less than 0.1 from loops 2-20.
Really early 10 loop run - 0.04 max variance 1-10.
Current i7 - 0.02 variance 2-20.
P55+i3 - 0.08 variance 1-20
P55+8gb+550+244bclk - 0.04variance 1-20
P55+8gb+4600mhz i3 550 HT on - 0.01 variance 1-20
Its throttling for sure :). Or you have AV running - lol. 3 Gflops swing is massive. Not criticism, just trying to help ;).
Can anyone explain noob what that throttling is pls:)
what about everest running in background?
Welcome to the club Olly, nice chip.:up:
Pilsy, I looked back a few pages at other runs, and his gflops are kind of up and down. If I run linx and plan to save the screen to post here at XS I dont surf the web or do other things that would hurt gflops. Normally I will chat on msn or surf the web though while it runs so I dont stay bored for an hour or so. I have had some pretty ugly runs with gflops not consisent, when that happens I normally start it over.
Try a short run with the same settings without running Everest and CoreTemp at the same time, just run one of the temp programs, dont web surf or open anything, just lest it run and see if gflops are better, no need to do 20 passes just do a quick run of say 5-10. On my R3E board if I have Everest and RealTemp open at the same time RealTemp show my CPU multi go back and forth between 20 and 21 like im losing turbo mode.
If you dont care to run it again, it says 'Finished without error" and that is a pass so no need to stress it anymore unless you want to.
BTW Congrats, nice run.:up:
Yeah exactly same thing happening here;) allrigh I'll do another run than.
thx m8t ;) but could be better with HT On.
I think should be allrigh now ;-)
1.5v
http://i903.photobucket.com/albums/a...c/56dde604.jpg
Very nice run, temps are good for that vcore. Gflops are better to.
owikh84 | Core i7 920 [4C 4T] @ 4504.4MHz | 1.424 | Asus Rampage III Extreme | Water | 3914A113
http://img33.imageshack.us/img33/7273/97493170.jpg
Yeah nice run there playground :up:. Anyone noticed this seems to affect Classys and P6T's more than other brands? Mybe just co-incidence.
When im running linx for here I usually use diagnostic mode. It gives much steadier results and eliminates software as the cause of instability. The only app I leave running is realtemp.
Start > msconfig > diagnostic mode and reboot.
Start > msconfig > normal startup to change back.
Thx;-) testing 4.6 now.
Nice hint m8t.
No problem, hope it helps :).
92-95c peak temp is where things go unstable in linx with every i7 ive built (lost count now - lol). That equates to about 85-88c prime95 peak temp and about 75c gaming. So about what you get on the retail cooler @ stock in a poorly ventilated case. People forget what circumstances these chips are designed to operate under for 3 years+ and start to freak when they see over 80c. It'll tell you when its not happy by erroring in linx.
If your cooling can take it try messing round with matching vcore and vtt +-10mv. That gets you some extra stability at the real edge but REALLY pumps out the heat. Each chip is slightly different, but all ive used like vtt +-10mv of vcore right when you're right at the edge to gain a little linx stability. Those runs that failed at loop 4 or 5 tend to complete ;). Note that its matching vtt to vcore, not the other way round. This wont let you reduce vcore, but rather gain a couple of mhz stability at your current vcore setting. Dropping uncore and mem multi counteracts the temp increase a bit, but your gflops score will drop massively. Were just going for stable linx here, not efficiency so give that a go :). 2t command rate got me an extra 20mhz or so, cpu pll down to 1.6 got me another 20mhz. Im right at the limit for heat though so anything i do to reduce temps gives me a benefit. Hoping for more benching in the workshop in winter ;).
I've got everything cooled in one loop. Cpu, nb and gpu so I think my temps at 1.5v Are really good.
Trying to cool down room at bit;-)
PiLsY: any links to these small nuances/hints/knowhow of i7 o/c in some guides/FAQs?
That's experience m8t ;-)
Doh its 16th loop at 4600 but don't have time to wait for the end :-o :(
taxi :'(
lol I hate when that happens. Still, 16 loops says its likely a banker, so all good.
Like you im running a single loop - 2 x gtx260, mb and cpu on a PA360 and a PA240 (10x1800rpm yates). I maxed out heat wise at 4.675ghz 1.52v. 4.7 would continually fail on about loop 17 or 18 when core0 hit 96c. Dumped the gfx cards out my loop and went mad on fans for a weekend and made it to 4811.6, but with low uncore and memory speed. Your temps look bang on for a well setup single loop, I reckon youve got the heat capacity for a bit more ;). Looks like your cpu isnt quite as good as mine but I'm putting my money on about 4.65 totally maxed out. Id be happy with that for a 920!! My best "only" did 4420mhz (c0 stepping - best d0 out of 5 only did 4.38ghz :lol:). Had to get a 950 to get over 4.5ghz, but lucked out and got a golden one.
I gave this a try today, last night it failed at 1.408vcore after 15 passes. Think I can still get vcore a notch lower. 8 threads is giving me a hard time, the chip just runs hot so this is with ht disabled. I could have used more memory I guess.
i7-930 #3004B176, Gigabyte X58A UD7 bios F7j
http://img69.imageshack.us/img69/3784/4620930chip.png
Uploaded with ImageShack.us
http://img42.imageshack.us/img42/6955/4620pass.png
Uploaded with ImageShack.us
Nice low vcore there - if you can control the temps it should be good for 4.7+ @ 1.55vcore... Looks like a really hot one you got there tho. Low vid?
i got 4.7 ill get some ss's :D
http://dl.dropbox.com/u/9421425/4.5G...0%20passes.JPG
2 hours out of the box.
Thank you.
I tried last night to get 229x21 but having a weird issue. I set everything up in the bios, get to windows and everything seems fine until linx starts and my multi drops to 12, It was running fine but at 2.7xxghz. I didnt notice at first and was excited about the run until I saw the gflops lol. Only happens on this chip, my 920 doesnt do this on the same board. Temps are a bit high but nowhere near the point where it would throttle. Tried clearing cmos and starting over but did the same thing again. :shrug:
Good job, guys! :up:
@ owikh84: you have a higher frequency run with the same chip already, so I have not replaced it. I can if you want, though.
i7 920 D0 HT on, Stepping 3938Bxx (cant remember)
214.5 x 21=4504mhz 1.432v
Asus Rampage Gene II
Adata DDR2000x
Water - H50 pump/block-> Stock 120 Rad.->MCR220w/Res. with 1/4 ID tubing, 6 Rosewell 120mm 2000rpm fans push/pull 78f ambient 87c peak
All memory used, 21 run successful
Made it :D
http://i27.tinypic.com/2vubjno.jpg
i for the life of me can't seem to get LinX to ever continue past the first run without errors... yet every other program *prime95, memtest, super pi and RealTemp GT* all go through with flying colors every time... i don't get it >.<
at the moment i'm running 4.5ghz cpu and 1900mhz mem w/ 12gb and it's been running solid all day with several different games...
GA-H55M-S2HP mobo
http://img830.imageshack.us/img830/7016/dgzs.jpg
I know the LinX time ...
Onboard graphics...
Good afternoon everyone, today is a warm end-of-July day in Virginia.
A challenging day for oc for sure, especially with a Cascade for a space warmer! :rolleyes:
Even so, the IHS was staying in the -40s. :)
So Mr Zalbard, please update my 980x entry yet again.
I think to go much higher I will need much cooler weather.
I would love to get 5.3 but probably not likely till later in the year.
So anyway - drum roll please :D:D - 980x at 5.2 linx stable -
http://img823.imageshack.us/img823/267/5200h.png
Nice run, but whats with the low gflops dude? With HT off at that speed you should be over 100 surely?
Yep, I saw that too, gflops seem to be all over the place with 32nm
I have had this same 980 at over 100 - :shrug:
I want to Join.
Core i7 930 [4C 4T] @ 4620.3MHz pass linx 62 loops run on water.
Core i7 930 / Gigabyte X58A UD7 bios F6 / micron ddr3 1333 d9jnm 1GBx3 / Water Cooling
http://img831.imageshack.us/img831/1856/19878586.th.png
http://img409.imageshack.us/img409/9162/68782514.th.png
http://img695.imageshack.us/img695/7166/22088330.th.png
http://img39.imageshack.us/img39/4948/34739471.th.png
http://img695.imageshack.us/img695/7166/22088330.png
hi! Mr. zalbard please check, two post was same link. ~thx u
# mwlhr | Core i7 920 [4C 8T] @ 4503.6MHz | 1.470 | EVGA Classified E760 | Water | 3925A260
http://www.xtremesystems.org/forums/...&postcount=455
# kup | Core i7 920 [4C 8T] @ 4501.7MHz | 1.360 | ASUS P6X58D | Water | 3912B185
http://www.xtremesystems.org/forums/...&postcount=455
UPDATE!
i'm proud to anounce i ACTUALLY got successful passes in LinX! though i do not have a screenshot to post just yet, i just didn't have the time to get all 20 passes, but it's done 7 passes so far and i will post a screenshot when i can get 20 passes through, but here's the settings. i DO need to reseat my water block though, i have 2 cores that are getting MUCH hotter than the rest by around 10*c so i think my thermal paste is off a bit or i have too much preasure on one side of the block. i found that LinX wasn't passing because my QPI was too high and was causing errors.
4.54ghz / 1866mhz mem / 1.39v QPI / 1.48v CPU / 1.66v DRAM / (34x133 ratio)
980x cpu / ASUS R3E mobo / Corsair 12gb DDR3 1866 CL9
HTT Enabled
Gflops Score: average 81-82!!!
Thats current feedback throttling dude - happens by default when the cpu draws over 100 amps. There shoul dbe an option in bios to disable that. On my Bloodrage its called "Enable Current Feedback Over-ride", have a poke round in bios and see what you can find. Asus have been plagued by these multi throttling issues since the first P6 board came out. Took them ages to fix the P6T-Deluxe throttling issues.
Has anyone on the R3E thread mentioned or noticed the same thing?
SteveRo - Still blown away by that 5.2ghz run after a nights sleep. Just mental dude :eek2:
Thanks Mr Pilsy - I think I could do 5.3 if I could get my ambient temps down - might not happen until next Fall.
:up: Nice one Steve.
Must admit, i can't understand those Gflops either.
Sometimes one really has to question just how good or bad Linx really is as stability prog, because it is so inconsistant.
I was having a similar problem with my Q9650 the other day I was getting like 37 gflops and my temps were much lower. I have a feeling a thread crapped out w/o being detected by linx/IBT because it finished both tests (20 passes) but failed prime blend instantly:shrug:
Jesus SteveRo, awesome run with that 980X :up: You're thrashing it hard :D
phase cooling really helps with the high cpuv. :yepp:
cascade cooling has also allowed me to improve my PC Mark Vantage World Record to - 38684 - :D
see -
http://www.xtremesystems.org/forums/...&postcount=172
@ kit - yes but now I am able to speak from personal experience :)
@ Mr Zalbard - yep - see new WR score - http://www.xtremesystems.org/forums/...&postcount=172
@ Mr Kit - yes a little concerned but all is well so far.
ThanksQuote:
I fixed this. Thanks for mentioning it! :up:
Adding yours in a bit, too. :yepp: