it works now with 1,36V soon a picture
i7-960 | 4515MHz | 1,39v | 21x | 215MHz | DDR-III-720MHz | off | Gigabyte UD5 | Wakü |3901B769 | Tray | siyah
http://www.abload.de/img/45ghzlinxstabil6lld.jpg
Printable View
it works now with 1,36V soon a picture
i7-960 | 4515MHz | 1,39v | 21x | 215MHz | DDR-III-720MHz | off | Gigabyte UD5 | Wakü |3901B769 | Tray | siyah
http://www.abload.de/img/45ghzlinxstabil6lld.jpg
now we are talking :D
vodka3dg | Xeon W3520 [4C 4T] @ 4510.3MHz | 1.472v | ASUS P6TD Deluxe | Water
http://img401.imageshack.us/img401/8198/98699678.th.png
http://img30.imageshack.us/img30/1802/18972836.th.png
HT was off, but i am pleased to see that this chip had more head room for overclocking; at least it's more than i was hoping to get :up:
SimpleTECH | Xeon W3520 [4C 4T] @ 4620.0MHz | 1.360 | Gigabyte X58A-UD5 | Water | 3845A935
Ha cha!
http://img3.imageshack.us/img3/319/46201360vf7d4c.jpg
wow... now thats completely awesome :D
http://img.club.pchome.net/product/m...2b02d98f36.jpg
4.8g linx 980x
Nice results everyone :up:
I have enjoyed this thread a lot, just haven't had a good new chip in a while to post any results. If there was one thing I would change in this thread, it would be a requirement to show a screen of the test while its running. With the software motherboards come with now its easy to make an average chip look like a Super clocker. Maybe I am just anal but when I see a screen of a LinX run at 4.5ghz or whatever, and with no screen taken under load it bugs the piss out of me.
Here is an example of why, just a quick run.
OMG WTF, Uber Chip 1.352v!!!!:up::up::up::up::up:
http://img59.imageshack.us/img59/2686/idlefz.jpg
Uploaded with ImageShack.us
cooling.
http://img198.imageshack.us/img198/9...0cpucooler.jpg
Uploaded with ImageShack.us
Actual voltage under load, can do better but just set this up quick.
http://img683.imageshack.us/img683/636/load0.jpg
Uploaded with ImageShack.us
No offense to anyone, a lot of hard work went into getting these results to post in this thread by everyone.
_AntLionBR_ | 980X [6C 12T] @ 4500.5MHz | 1.392 | Gigabyte X58A-UD7 | TRUE 120 Copper
http://img815.imageshack.us/img815/6691/980x45ghz4.png
Totally agree with you. Would be best if we have screens with LinX running and complete like yourself, myself and many others have done. Just make it as much no BS as possible. No offence meant to anyone obviously, but when it comes down to it, most of us don't know each other and have no way of knowing whats real, only what you show in your screen shots.
You are absolutely right, :up: but this is just a club, we are not competing here, if I got it right? :confused:
None of the results can be confirmed if not CPU-Z validated and taken s/s during LinX run as deachus said.
I'm more concerned about some very low GFLOPS, which looks to me like not totally stable, or not done with latest linpack libraries, from my experience.
Probably best solution is to leave it on everybody's conscience and honor. ;)
The updated linpack libraries actually made it harder for me to pass 4.5Ghz LinX 20 runs stable. I would say most recent linpack libraries should be a requirement at the very least.
I think that comes down to LinX needing to be updated.
My thoughts were, you see a chip that has finnished a LinX run at 4500 or higher on extremely low vcore, the natural response is "What batch# is that? ", so 20 people run out looking for a batch thinking they may gey lucky and get a super chip.
But I guess this would complicate things more.
Im working on a not so great Xeon w3550, hope to have results tonight.
@donmarkoni: you are right, its a club not meant for competing. I posted a 5ghz LinX run in a thread on another forum, a 5ghz club thread. I was told "Nice run" but because they required a 5ghz cpuz validation to be in the club, the submission was not valid..
Not the best chip in the world but finally passed 4500mhz no HT. Using as much of a 3x1gb kit as possible to test in 64bit W7, just to take a screen at 19 passes killed my gflops.
That is the only reason why I do not like opening 3 instances of CPUZ and taking a screenshot while I am running LinX because it lowers my gflops. I have not seen my gflops drop as low as yours did though. It usually drops from 66.XXXX to 65.XXXX.
If I select less that 2500mb of the 3072 it only drops down to 64gflops when taking a screen, but wanted to use as much as possible. I have become lazy and started using this 3x1gb kit to save time, will require more voltage, or at least more vtt to be stable with a 6gb kit.
I was asking myself many times too how GFLOPS are so low on some entries. :shrug:
There are only eight (excluding mine, which is not ranked yet) 80-some to 100-some GFLOPS entries from 33 Core i7 (Gulftown & Westmere-EP, 32nm) entries. :confused:
OK, I know that GFLOPS would drop all the way down to 60-some if UnCore and RAM are dropped to lowest multis possible, but what's the point? I wouldn't use it like that. :shakes:
I don't have a hexacore chip but some people may have to do that to pass 20 runs LinX to get in the 4.5Ghz club. They may not be using those settings for 24/7 use but I could be wrong:confused:
Just a quick note...
I think it will be too easy to get to 4.5GHz using Sandy Bridge (we will see, I guess). Gotta preserve some challenge! :D I am considering expanding my LinX club by adding a new "5GHz Stable" category. And this will probably be the only category SB owners can get in. ;)
Opinions?
@ Zalbard, I hope SB is so good that we will need a 5ghz LinX club, early results look good.
Xeon w3550, not so good batch#3949B156, 4508mhz, MSI XPower, 6gb Dominators, Water.
I'm up for trying :D This is a great club, something to aim at & a load of fun.
I've got an opinion on everything & could suggest all sorts of things that will turn this thread into a right shambles of opinions, Thing is I'll play by the rules whatever they are as the challenge is the fun for me.
Keep up the great work :up:
.