OCUK Sandy Bridge, SB-E and Ivy Bridge 5GHZ Club

5ghz.jpg


Hope this works, seems stable at 1.404v cpuz
spec's are:
GA-P67A-UD7 F6e, 2600k@5ghz,2x2gig Gskill Ripjaws 2133mhz, 6970 crossfire,Crucial C300 ssd, 1tb sata 3 main drive, 900w Ultimate psu, custom water cooled with Swift Apogee Xt cpu block, HAF 932 case.
 
please feel free to satisfy my curiosity guys :

So how many of these 5ghz+ overclocks are actually prime95 blend stable for more than 6 hours?

Its just that I'm seeing loads of sites reporting that LinX runs forever and a day, but Prime95 fails at less than 6 hours (The un-officially recommended stability run time is actually about 8 hours)

thanks in advance.
 
my 4.8ghz was prime blend stable for over 8 hours, i've just started a 5ghz so i'll let you know. I know its the recommended, but how often do you stress your system for that long in real world terms? Gaming is probably the most stress that i put on mine and i rarely do that for more than 2 hours at a time. Each to there own i suppose.
 
Hey

How are you people with watercooling keeping your chip so cool compared to my setup?

I have a Asetek Antarctica Waterblock (cant find a link that doesnt advertise, so clicky here for some pics: http://www.google.co.uk/images?hl=en&q=asetek%20antarctica%20cpu%20block&um=1&ie=UTF-8&source=og&sa=N&tab=wi&biw=1259&bih=564) which I've used on various set ups, but maybe its time for a change. How far has block design advanced lately?

Just realised that this block first appeared in 2004!

Could I pressume a nice result from something like this compared to my ancient block of copper:

http://www.overclockers.co.uk/showproduct.php?prodid=WC-012-EK

If so, how would I fit it to my P8P67 board. The 1155 boards arent in its spec (I dont think)

When I brought the block a while back, it seemed to be a good performer, so I've not kept up with any advancements that may have happened

Just for you info, I'm currently stuck at 4.4ghz @ 1.31v in bios, and I'm topping out at 72 degrees when stress testing with prime. It doesnt seem normal from what I can tell :(

I want in this club!!!! :D :D
 
i'm using this waterblock:
http://www.candccentral.co.uk/Swiftech-Apogee-XT-Extreme-Performance-1156-1366.html
Although it's a 1156 it comes with 2 backplates, the smaller one fits 1155. Just for comparison, my rad is 240mm with 2 fans pushing, xspc drive bay pump resevoir with 1/2 od pipe. Only the cpu is in the loop. My idle temp's are 30c in a warm room and my max temps are 68c in prime with just over 1.4v.
 
Thanks for that ferr250

Any chance you could do a 4.4ghz run at 1.31 volts and let me know the temps?

Your loop seems similar to mine, especially the warm room part

Pretty sure at 1.4 volts though, mine would melt :)
 
zoomee said:
please feel free to satisfy my curiosity guys :

So how many of these 5ghz+ overclocks are actually prime95 blend stable for more than 6 hours?

Its just that I'm seeing loads of sites reporting that LinX runs forever and a day, but Prime95 fails at less than 6 hours (The un-officially recommended stability run time is actually about 8 hours)

thanks in advance.
Click to expand...

I guess the short answer is not one, but who cares:)

Not being funny or anything but if you read the first post of this thread it clearly states that this thread is not a stable 5GHz thread. Rather it is for and about those that have overclocked their new SB CPU's over 5GHz and run Super Pi 1mil. It is not a thread about people who have a 5GHz overclock for their 24/7 rigs. It is a brag thread, a bit of hey I have clocked over 5GHz. A bit of fun in other words.

You want 24/7 stable systems - start the thread if it does not exist yet. That is a whole different ball game mate. Certainly this is not the thread for that.

But valid point :D
 
1 Day, quick question, and i know that this is the 'holy grail' but what would you consider to be a safe vcore for these chips for everyday use? 1.368 gets me stable 4.8ghz which is great but i'm needing closer to 1.45 for 5ghz. 1.41 gets me to 5ghz and will run 3d mark vantage etc. I'm taking into account Gibbo's thread regarding safe voltages, although he was testing a rig with high vcore but i hav'nt seen an update on that for a while. Perhaps the only way to find out is to try! I'd just love to have my rig @5ghz all the time.
 
ferr250 said:
Just ran prime (blend) for 20 mins. Temp was 62c on hotest core, 5-6c difference between this and coolest core. I'm sure push/pull fan setup would improve that.
Click to expand...

Thanks

So, its looking like around a 10 degree difference. You think its my block?

I've got a black ice pro 120 rad, with a push pull config, with vipers at full whack, with the same pump as you with the res attached. Oddly though, whether my fans are on full or at 30%, the temps dont really change?

hmmmm
 
Difficult one, but with fans making no difference i would guess that the rad is too small, just not enough surface area. It may be a combination as i'm sure your aware, but then rads are approx £30 and a good block closer to £70 so you could change the rad first. My concern with changing the block first is that it will probably be more efficient at taking the heat from the cpu and transfering it to the water, leaving your rad struggling more.
 
I would keep my CPU vCore below 1.35 volts on air for 24/7 use. If I was using a good water cooling kit. Say H70 and better then 1.40 tops. BUT and this a big but it really depends on your CPU. Some SB CPU's are running much hotter than others. And the funny thing is the hot ones seem to clock better. :D

And your voltages for 5GHz at 1.45 are spot on. 1.440 is normally needed to get into OS at 5GHz.
 
True true

Been looking into rads and it seems like mine, due to design, is not great for airflow due to the number of fins, so maybe one with more spaced fins would be better and this may explain the lack of difference that fan speed makes.

Also, I notice a massive temp drop whenever I open the door near my case. Went to the extent of directing the cool air towards my rad fans and the temps dropped by 13 degrees. This kind of tells me that the rad is doing its job as well as it can, maybe?

I could, of course, be talking kak :)
 
You must log in or register to reply here.
Back
Top Bottom