Should my temps be lower?

[Lays]

My load temps were like 72~ with the crappy TIM I used previously, I replaced it recently with ic diamond 7, and remounted my block etc.  It's mounted properly, I tightened everything the way it should be.

 

 

 

all 8 of my sycthe gt's are at 100% (1450rpm) on 2 480mm 60mm thick radiators.

 

 

 

NO load on my GPU's, (both are at 25c )

 

 

 

Under gaming load the cpu is around ~48-55 whilst both gpu's are 38-42.

 

 

 

[kinkywink]

dat voltage... maybe its just because its a diff cpu but i thought above 1.35 wasnt a good thing to do

[Lays]

dat voltage... maybe its just because its a diff cpu but i thought above 1.35 wasnt a good thing to do

 

 

1.35v-1.4v is intel's recommended max for sb-e yeah.

[kinkywink]

1.35v-1.4v is intel's recommended max for sb-e yeah.

okay  well at least your temps arent 80+ right?

[DoubleY]

okay  well at least your temps arent 80+ right?

Imagine his temps on like a 212 EVO

[.:MARK:.]

Those are good temps, I see no reason to change anything at this point.

[jscho]

They aren't that bad, what are your ambients?

[Lays]

@JK Vengeance @.:MARK:.

 

~20c and what I meant is, with the  960mm of rad space I have should it be more in the 40-50 load temp range? 

[warmmilk]

@JK Vengeance @.:MARK:.

 

~20c and what I meant is, with the  960mm of rad space I have should it be more in the 40-50 load temp range? 

 

once you get past a certain point with rads space, and 960mm is well past that point, your temps don't go down any further.  the difference is you can spin you fans slower and have a quieter system while still maintaining great temps.  and the temps you posted up are great for what they are.  

 

You gotta keep in mind, the best you can do is bring down your water to room temp, it doesn't matter if you have 17 480 rads with 3000 rpm GT's on them, the water won't drop below the temp of the air thats going through the rad.  and the temp of the water is what pulls the heat away from your cpu and gpu's

[Lays]

once you get past a certain point with rads space, and 960mm is well past that point, your temps don't go down any further.  the difference is you can spin you fans slower and have a quieter system while still maintaining great temps.  and the temps you posted up are great for what they are.  

 

You gotta keep in mind, the best you can do is bring down your water to room temp, it doesn't matter if you have 17 480 rads with 3000 rpm GT's on them, the water won't drop below the temp of the air thats going through the rad.  and the temp of the water is what pulls the heat away from your cpu and gpu's

 

 

I was just curious, because some guy the other day posted a thread on here with a 3930k, he didn'/t mention overclocking though.  He said his load temps were 42-48 with a 480 monsta, and a 360 monsta + a 360.

 

Although he doesn't mention he tested with prime... my temps are the same as his in games, I wonder if his "load" was gaming

 

http://linustechtips.com/main/topic/108774-build-log-project-rampage-ii-caselabs-sth10-tri-sli-780ti-full-wc/

 

http://linustechtips.com/main/topic/108774-build-log-project-rampage-ii-caselabs-sth10-tri-sli-780ti-full-wc/page-3

[warmmilk]

I was just curious, because some guy the other day posted a thread on here with a 3930k, he didn'/t mention overclocking though.  He said his load temps were 42-48 with a 480 monsta, and a 360 monsta + a 360.

 

Although he doesn't mention he tested with prime... my temps are the same as his in games, I wonder if his "load" was gaming

 

http://linustechtips.com/main/topic/108774-build-log-project-rampage-ii-caselabs-sth10-tri-sli-780ti-full-wc/

 

http://linustechtips.com/main/topic/108774-build-log-project-rampage-ii-caselabs-sth10-tri-sli-780ti-full-wc/page-3

 

sorry, those threads are to long to read.  but you said he didn't mention overclocking, and based on those temps its safe to assume it's not overclocked.  

 

save your OC profile in your bios and reset back to defaults and check your temps.  they'll be somewhere around there as well.

[Lays]

sorry, those threads are to long to read.  but you said he didn't mention overclocking, and based on those temps its safe to assume it's not overclocked.  

 

save your OC profile in your bios and reset back to defaults and check your temps.  they'll be somewhere around there as well.

 

 

Alright, I guess now that I think about it, heavily overclocked 39xx chips easily use over 200 watts, so I suppose I shouldn't be complaining about high 50 low 60 temps considering it's using like 220 watts lol

 

 

Tyvm btw

[airdeano]

My load temps were like 72~ with the crappy TIM I used previously, I replaced it recently with ic diamond 7, and remounted my block etc.  It's mounted properly, I tightened everything the way it should be.

 

yeah, that is about right on 1.350v and on 1.45v about 70°ish. different TIM pastes

need some curing. MX-4 and NT-H1 usually 6hrs for full flow. also depends on

ambient and block restrictions and a plethora of things. but for basics, your dead

money. if you think having many fans will lower even more, there comes a point of

diminishing returns. more capacity gives you longer time towards temperature

stability.

only way to find out how much more efficient the system might be would be to have

faster fans, but that usually gets into more noise. might try the Rosewill/Akasa 120mm

fans with deeper static and less noise at 12v (1300rpm) or if you have some SP120

with some fan speed to check and see if there is any more in the system.

i don't see an issue with what is given.

[Lays]

yeah, that is about right on 1.350v and on 1.45v about 70°ish. different TIM pastes

need some curing. MX-4 and NT-H1 usually 6hrs for full flow. also depends on

ambient and block restrictions and a plethora of things. but for basics, your dead

money. if you think having many fans will lower even more, there comes a point of

diminishing returns. more capacity gives you longer time towards temperature

stability.

only way to find out how much more efficient the system might be would be to have

faster fans, but that usually gets into more noise. might try the Rosewill/Akasa 120mm

fans with deeper static and less noise at 12v (1300rpm) or if you have some SP120

with some fan speed to check and see if there is any more in the system.

i don't see an issue with what is given.

 

 

I won't be turning away from my sycthe GT's anytime soon, especially not after coming from nf-f12's and seeing how much quieter they are comparatively :3

 

Someone mentioned push/pull to me yesterday, but I really don't think it'd make a difference.  At 75%-100% fan speed I can put my hand above the top rad and feel lots of air coming through, so I don't think static pressure is a problem.   I think it'd make a bigger difference at 40-60% speed though?

 

And yeah, I like my temps a lot, I was just wondering if they were supposed to be lower with the amount of rad space I had.

 

Tyvm for the reply

[airdeano]

And yeah, I like my temps a lot, I was just wondering if they were supposed to be lower with the amount of rad space I had.

 

Tyvm for the reply

 

misconception.. more fans does not always equal lower delta temps. on a CFX/SLI/CPU

loop, once you get to 480+240 and higher, the point of lowering deltas gets to be harder.

as the system is pretty efficient from the coolant transfer to air (hence the fan speed test),

next is to look into block efficiency. but spending hundreds more to loose 5° gets to be

'spensive and almost boarders insane. but that is what enthusiasts and their water cooing

rigs do, test the extent of how low is low for a water cooling loop, before sub-zero chillers

and such.

one thing i can say, yes, the Rosewill/Akasa are more quieter than F12 and GT-15. i'm

trying the performance numbers now.

[Lays]

misconception.. more fans does not always equal lower delta temps. on a CFX/SLI/CPU

loop, once you get to 480+240 and higher, the point of lowering deltas gets to be harder.

as the system is pretty efficient from the coolant transfer to air (hence the fan speed test),

next is to look into block efficiency. but spending hundreds more to loose 5° gets to be

'spensive and almost boarders insane. but that is what enthusiasts and their water cooing

rigs do, test the extent of how low is low for a water cooling loop, before sub-zero chillers

and such.

one thing i can say, yes, the Rosewill/Akasa are more quieter than F12 and GT-15. i'm

trying the performance numbers now.

 

 

One thing I was sorta pissed about the nf-f12's was, they're really only quiet at like 40-50%, but you can still hear them perfectly fine if you're near your pc.   With my ad900x's on my head I could still hear the slight hum they make.  It makes no sense to me why people recommend them so much, yes they're great performing, but only being somewhat quiet at 40-50% makes no sense to me, what if you're in a usage-scenario where you need temps to be a bit lower and need to run them at 70-100%, quietness goes out the window.

And at 100%, they're NOWHERE near quiet, so freaking loud.   

 

Even with my motherboard running them at like 550rpm, it's still audible

 

It makes much more sense to me to get fans that are not only quiet at 40%, but also at 100%.  I'm willing to lose 3-4 degrees for that, hence why I love my GT's so much

[Monty007]

What size tubing are you running? I switched to 1/2" ID 3/4 OD tubing and increased my pump speed ever so slightly and I noticed a significant temp difference. Larger volumes moving over the heat exchangers makes a difference. Well it did for me at least. Idle temps: CPU 18-22 on all 4 cores. GPU idle at 18-19. Load temps with prime 95 CPU is maxed at 68. Under 3d mark test and heaven I haven't seen more than 50 ever. I game in triple monitor surround and haven't seen GPU temps over 50 ever.

My $0.02

[warmmilk]

What size tubing are you running? I switched to 1/2" ID 3/4 OD tubing and increased my pump speed ever so slightly and I noticed a significant temp difference. Larger volumes moving over the heat exchangers makes a difference. Well it did for me at least. Idle temps: CPU 18-22 on all 4 cores. GPU idle at 18-19. Load temps with prime 95 CPU is maxed at 68. Under 3d mark test and heaven I haven't seen more than 50 ever. I game in triple monitor surround and haven't seen GPU temps over 50 ever.

My $0.02

What was the temp difference when you stepped upto 1/2" tubing? And what size tubing did you have before?

[Lays]

What size tubing are you running? I switched to 1/2" ID 3/4 OD tubing and increased my pump speed ever so slightly and I noticed a significant temp difference. Larger volumes moving over the heat exchangers makes a difference. Well it did for me at least. Idle temps: CPU 18-22 on all 4 cores. GPU idle at 18-19. Load temps with prime 95 CPU is maxed at 68. Under 3d mark test and heaven I haven't seen more than 50 ever. I game in triple monitor surround and haven't seen GPU temps over 50 ever.

My $0.02

 

 

I would but I don't think it'd make that much of a difference considering temps are already so low.  Plus I'd rather not spend another like 200$ on fittings

 

I'm using 3/8 5/8

[Monty007]

What was the temp difference when you stepped upto 1/2" tubing? And what size tubing did you have before?

Between 3-7 degrees. Both in idle temps and load temps.

[warmmilk]

Between 3-7 degrees. Both in idle temps and load temps.

and what size tubing did you have before the 1/2"?

[Monty007]

and what size tubing did you have before the 1/2"?

3/8

[Leonard]

I am sure you can get lower temperatures but using offset voltage which would lower the 1.3v you have to about 1.2 something.

[BrightCandle]

Just thought I would mention my 3930k@4.4Ghz 1.35V is running at load about 53C after its warmed up. Its a loop with 2x 360mm + 1x480mm MCR rads with Gentle typhoon's at ~800rpm and 2x 680's. Low 50's is about normal for this CPU with that overclock. The GPU temperatures at idle are basically the same as are my load temperatures. It looks relatively normal and healthy to me. Try turning your fans down and see what it gets to then, you may find you don't gain much in temperature.