Hows this OC look?

[jdoggg12]

http://i12.photobucket.com/alb...%20or%20chops/cpu1.jpg

Hows this look? I'm going to be ordering a tuniq tower soon to bring the temps down, but i figure as long as they stay below 60c its ok for now. I hope the voltage isnt too much, i hear these chips like to sip power.

 

[Pantlegz]

looks good, what mobo are you using? I can't get my 780i stable at 450x9 at less than 1.40v

 

[jdoggg12]

IP35-E i cant get my ram past 1000mhz so its making things tricky to maximize my OC.

This is the ram im using


Orthos has been running Small FFTs since i posted... 3.5hrs no errors yet

Once i get the aftermarket cooler i'll see if i can OC to 4Ghz at less than 1.3V vcore. I can do it at 1.3 now, but the temps will go to 63c while running orthos. Stock cooler is pretty weak i guess.

 

[Pantlegz]

I'm not sure about that board, and the BIOS but if it has an option like evga's 780i you should be able to run FSB/RAM in sync. With the ram at 1000mhz the QDR would be 2000 and you should be at 500x9(4500Mhz) which I dont' see happening on air but you can try .

 

[jdoggg12]

I dont want to hit over 60c under load on stock cooling otherwise i'd try that now

 

[jdoggg12]

Quick question... would the Tuniq tower be a better/quieter choice than the Thermaltake 9700?

 

[manimal]

I use a lapped Tuniq and have a lapped 9700 that I bought before I knew any better. Now the Zalman does a good enough job on one of my quads with moderate voltage. With 45NM chip things would be even better with the 9700 but I just cannot justify the price tag even today after the price has dropped a bit. The Tuniq can be had for less than 40 bucks now shipped so that is much better deal IMO.

I get 52-54 load temps on my E8400 at 1.42 volts with a lapped Tuniq.

 

[Dadofamunky]

Looks good overall. SpeedFan seems to have a compatibility problem seeing both cores on your Xeon. Core voltage is well within tolerance too AFAIA. I think it looks good!

Get CPUID's Hardware Monitor as a second temp monitoring option. Temp measurements on PCs are notoriously inaccurate, and SpeedFan is way too forgiving.

But overall, I think you're in good shape. When you get a better cooler you might be able to shoot for the Big 4.

 

[jdoggg12]

CPUID takes the TjMax of 105, yes? I've been using Real Temp b/c it is shows TjMax 95 which is supposed to be more accurate for the 45nm C2Ds. The difference will be 10c between real temp and cpuid/core temp. Am i wrong in believing the lower temp is more accurate?

 

[jdoggg12]

So i got my Tuniq Tower in.. here are some pics for those interested.

Boxed
Side 1
Side 2
Stock E3310 heatsink vs Tuniq Tower
VS from different view
Seated on the mobo
In the case
In the case 2
In the case 3

 

[Dadofamunky]

Originally posted by: jdoggg12
CPUID takes the TjMax of 105, yes? I've been using Real Temp b/c it is shows TjMax 95 which is supposed to be more accurate for the 45nm C2Ds. The difference will be 10c between real temp and cpuid/core temp. Am i wrong in believing the lower temp is more accurate?

Just from a personal standpoint, I always accept the highest reading I get from a range of measurements. But SpeedFan is basically junk at this point. Your CoreTemp is much mroe accurate. Now, I want to get further edumacated on the specific temp readings and their meanings, but by the time I get to these postings I'm already worn out from work and all I want to do is play... figuring stuff like that out sounds too much like work to me at the moment.

I think you make a good point. I believe the most pessimistic readings because it gives me more of chance to OC and still safeguard my system.

With that Tuniq, you should be able to take a shot at 4.0 GHz!

 

[myocardia]

Originally posted by: jdoggg12
CPUID takes the TjMax of 105, yes? I've been using Real Temp b/c it is shows TjMax 95 which is supposed to be more accurate for the 45nm C2Ds. The difference will be 10c between real temp and cpuid/core temp. Am i wrong in believing the lower temp is more accurate?

Read this thread, and you'll understand how it works on these chips.

 

[jdoggg12]

Originally posted by: myocardia

Originally posted by: jdoggg12
CPUID takes the TjMax of 105, yes? I've been using Real Temp b/c it is shows TjMax 95 which is supposed to be more accurate for the 45nm C2Ds. The difference will be 10c between real temp and cpuid/core temp. Am i wrong in believing the lower temp is more accurate?

Read this thread, and you'll understand how it works on these chips.

Ok... i dunno if its just really confusing or if i'm not getting it.

But from what i read....

Scale 1: Duo
E4x00: Tcase Max 73c, Stepping M0
E2xx0: Tcase Max 73c, Stepping M0
E8x00: Tcase Max 72c, Stepping C0
E8x90: Tcase Max 72c, Stepping C0
E6x50: Tcase Max 72c, Stepping G0
E6540: Tcase Max 72c, Stepping G0

-Tcase/Tjunction-
--70--/--75--75-- Hot
--65--/--70--70-- Warm
--60--/--65--65-- Safe
--25--/--30--30-- Cool

And looking at real temp 2.24 it says its using a TjMax=95c would that mean that it's reading my temperatures ~20c too high? since the stated Tjunction max temp is ~75c?

 

[jdoggg12]

Well I OCed to 4Ghz and max temp showed 63c in Core temp which is cooler than it was at 3.7Ghz with the stock cooler.

I'm still trying to decipher that article on temperatures.

 

[ther00kie16]

Did you lap the heatsink? Lapping usually lowers the temp by a few degrees.