• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Anything wrong here?

Wingless92

Limp Gawd
Joined
Jan 22, 2011
Messages
333
Wanting to get some input on my setup.

i7 920 @4.2Ghz,

2012-12-01103441.jpg


2012-12-01103531.jpg


EK Supreme HF Copper Acetal Block, 1/2ID 3/4OD tubing

D5 pump

This radiator
http://www.frozencpu.com/products/1...120mm_Radiator_LC-RADI360S.html?tl=g30c95s161

With 3 Gentle Typhoons in a push config. I can't do push/pull just not enough space.

IMG_7184.jpg


Temps are 30c-35c idle and 60c in game. Anything else I can do to get better temps? Better radiator? Not sure if I can get better than that.
 
Those aren't bad temperatures, especially for an x58 chip. You could try lowering the voltages a little (though you've probably done this). Or stick a bunch more fans externally (which you said you can't). The only things left are remount the heatsink (though that seems good), better TIM (which wont make that much difference maybe 1-2c) or add some more radiators or if those are 1450rpms, go for 1850s. Or switch them all to pull and another fan to vent might help. :D

Or swap if for a 950 which have lower temperatures. But theres not really a problem here to be solved. :D
 
Last edited:
dont leave everything on auto and control more things yourself. I am not familiar with OC intels systems lately, and I know they are veyr much more sensitive to things being changed on them, but, alot of things on auto will generally push them far harder then needs to be, and usually results in higher temps.

Like for example, the cpu-PLL voltage is supposed to be max of what 1.75v and yours is nearly 1.9v that is a huge jump(probably more in windows, even my 2 asus boards bios screen and windows are vastly different)

Also the Qpi frequency should be locked at 4.8Gt(which if I recall yours is as well)

Personally I would just go back through, tweak what you can in regards to voltages, speed etc, as 60c loaded is stull very good(they can handle 67.9c Tcase which is temps just above the cores, not the core themselves)

And just looking at the picture you have there, I notice off the top of that bottle looking thing, the one pipe going off the top into a slot to motherboard area, is there not a way to make this go down into the hole off the left, instead of the top, this might give you that little bit of extra space for push/pull for the rad?
 
Forgot to say that this stable as hell. I can throw anything at it, no crashes, random reboots, nothing.

As for the fans they are the 1850's. Thanks to Tsumi for that, lol.

Not sure if a better radiator would help or not.

Just canged the cpu PLL voltage to 1.75. I will leave it there and see what happens.

Thanks for the input.
 
Last edited:
On my 920 system, if I didn't increase the PLL to 1.88v, I would get instability. Practically every guide I read says to increase the PLL to 1.88v and leave it there.

60 C at 4.2 ghz and 1.3v vcore is very good. My 920 got to 62 C on a 3x140 radiator at 4.0 ghz 1.275 vcore, and 70 C at 4.2 ghz 1.315 vcore (probably due to me running 6x4gb RAM).

The only thing I can see you playing around with is the QPI voltage. You might be able to get that down to 1.35 or 1.3v. I got mine to run at 1.3v, but I was running my RAM at 1200 mhz.
 
Set the PLL to 1.88. Gonna run that for awhile and see how it does.

Tsumi to the rescue again, lol.
 
Thanks Tsumi. I primed the computer for the last 8 hours and it hit 65c. Gonna leave it alone, more than I could of asked for.

Thanks again for the tips and guide.
 
I am SO jealous of your voltages. If my QPI voltage is anything under 1.4v, i get 124 crashes. My 5649 is running at 1.3v on the cpu, so i know its not that. I'm going to turn my PLL up to 1.88 tonight to see if that helps.


Like for example, the cpu-PLL voltage is supposed to be max of what 1.75v and yours is nearly 1.9v that is a huge jump(probably more in windows, even my 2 asus boards bios screen and windows are vastly different)

Not for x58. Stock voltage on the CPU pll is 1.81. 1.9 is a safe max for most. 2.0 is doable, but not for longterm.
 
Last edited:
http://www.evga.com/forumsarchive/tm.asp?m=100494809 here is a nice list of voltages what they do etc.
http://www.tomshardware.com/forum/267465-29-question a thread with helpfull suggestions
http://www.overclock.net/t/1302808/x58-qpi-volts-whats-safe according to title
http://www.techpowerup.com/forums/showthread.php?t=108526
http://i4memory.com/f55/intel-core-...oltages-uncore-memory-frequency-ratios-23430/ max recomended for i7 900 series chips

I guess the only thing to point out, is alot of folks and Intel themselves said it is folly to push it to "close" to max voltages, as it may last a little while, it could just up and die. But hey, if it works for you, awesome :)
 
On my 920 system, if I didn't increase the PLL to 1.88v, I would get instability. Practically every guide I read says to increase the PLL to 1.88v and leave it there.

60 C at 4.2 ghz and 1.3v vcore is very good. My 920 got to 62 C on a 3x140 radiator at 4.0 ghz 1.275 vcore, and 70 C at 4.2 ghz 1.315 vcore (probably due to me running 6x4gb RAM).

The only thing I can see you playing around with is the QPI voltage. You might be able to get that down to 1.35 or 1.3v. I got mine to run at 1.3v, but I was running my RAM at 1200 mhz.

Why would you tell him to RAISE the PLL voltage for no reason? He already stated it was rock solid stability wise, he simply wants to see if he can work his temperatures down a bit. Raising the PLL voltage is going to the opposite of that:rolleyes:...

OP, those are GREAT temperatures for that setup. You are running a 45nm chip at the same speed I was running a 32nm Quad w/HT Xeon and have temperatures right around what I did, although I was running a 240mm thicker radiator.

One thing to keep in mind is that these slim, high FPI radiators perform the best with higher airflow. I have a very similar one, a Koolance 20FPI model, and Bundy's testing showed the best results with airflow over 1200-1500 rpm. If you don't mind the small amount of noise they make, keep your GT-15s running @ full speed. I have the exact same fans, and love them.

EDIT:

If I were you, I would try setting the PWM Volt Control and CPU PWM Freq to manual or disabled if you have the option. IIRC, I had those off when I had my Rampage III. There is a great chance you can squeeze another ~200Mhz out of that cpu by raising your BLK to 210. The RP3 had no problems going as high as 221BLK for me. You should be able to do 4.4Ghz with a Vcore of ~1.45V, maybe a bit less.
 
Last edited:
Yea, I had some stability issues with the PLL at 1.88v. It would BSOD and then lock up. So, I set it back to auto and it runs great. Gonna keep it were its at. Runs like a champ. I really see no reason to upgrade anytime soon. It handles everything that I throw at it with ease.
 
Yea, I had some stability issues with the PLL at 1.88v. It would BSOD and then lock up. So, I set it back to auto and it runs great. Gonna keep it were its at. Runs like a champ. I really see no reason to upgrade anytime soon. It handles everything that I throw at it with ease.
 
Yea, I had some stability issues with the PLL at 1.88v. It would BSOD and then lock up. So, I set it back to auto and it runs great. Gonna keep it were its at. Runs like a champ. I really see no reason to upgrade anytime soon. It handles everything that I throw at it with ease.
 
Yea, I had some stability issues with the PLL at 1.88v. It would BSOD and then lock up. So, I set it back to auto and it runs great. Gonna keep it were its at. Runs like a champ. I really see no reason to upgrade anytime soon. It handles everything that I throw at it with ease.
 
Yea, I had some stability issues with the PLL at 1.88v. It would BSOD and then lock up. So, I set it back to auto and it runs great. Gonna keep it were its at. Runs like a champ. I really see no reason to upgrade anytime soon. It handles everything that I throw at it with ease.
 
lol..I think there should be a button to delete posts, as the edit button I noticed doesn`t alow one to delete thier own posts :(
 
Why would you tell him to RAISE the PLL voltage for no reason? He already stated it was rock solid stability wise, he simply wants to see if he can work his temperatures down a bit. Raising the PLL voltage is going to the opposite of that:rolleyes:...

OP, those are GREAT temperatures for that setup. You are running a 45nm chip at the same speed I was running a 32nm Quad w/HT Xeon and have temperatures right around what I did, although I was running a 240mm thicker radiator.

One thing to keep in mind is that these slim, high FPI radiators perform the best with higher airflow. I have a very similar one, a Koolance 20FPI model, and Bundy's testing showed the best results with airflow over 1200-1500 rpm. If you don't mind the small amount of noise they make, keep your GT-15s running @ full speed. I have the exact same fans, and love them.

EDIT:

If I were you, I would try setting the PWM Volt Control and CPU PWM Freq to manual or disabled if you have the option. IIRC, I had those off when I had my Rampage III. There is a great chance you can squeeze another ~200Mhz out of that cpu by raising your BLK to 210. The RP3 had no problems going as high as 221BLK for me. You should be able to do 4.4Ghz with a Vcore of ~1.45V, maybe a bit less.

Well for me, raising my PLL allowed me to decrease my vcore (and increased stability in my case), which netted me lower temps. Of course every chip is different, which is why I also suggested trying to lower the QPI voltage.
 
Back
Top