Temp check.. High ambient @ 30C; is 2700K 4.8Ghz/1.355v 74C LinX w/ Raystorm decent?

[ZeroOne]

Just set up a new loop with:
Raystorm
MCP655 variable @ 4/5
2 separate MCR220-XP's
Corsair SP120s (2350rpm / 63cfm)
EK Spin Res
in a Vengeance C70

The ambient temp in here is a very warm 30C. Just wondering if someone could verify these numbers and give an opinion on whether they look right. Seems warm to me, but this house is hot and it is throwing me off.

CPU is 2700K, currently @ 4.8Ghz, 1.355v

Temps at 30C room:
idle: 35-40C
load (LinX): 74C

Used an old Arctic Silver 5, but wasn't sure about the mount. Re-did it once. Does lapping the cpu still help these days? Last build went for almost 6 years, so I'm "out of the loop" a little.

[jomama22]

Just set up a new loop with:
Raystorm
MCP655 variable @ 4/5
2 separate MCR220-XP's
Corsair SP120s (2350rpm / 63cfm)
EK Spin Res
in a Vengeance C70

The ambient temp in here is a very warm 30C. Just wondering if someone could verify these numbers and give an opinion on whether they look right. Seems warm to me, but this house is hot and it is throwing me off.

CPU is 2700K, currently @ 4.8Ghz, 1.355v

Temps at 30C room:
idle: 35-40C
load (LinX): 74C

Used an old Arctic Silver 5, but wasn't sure about the mount. Re-did it once. Does lapping the cpu still help these days? Last build went for almost 6 years, so I'm "out of the loop" a little.

That is a bit warm even with those ambient temps. I would say you should be sitting around 60-65*c, 70 being the Max. I would buy some new tim and try again. AS5 does have a bit of set time (100-200 hours I believe) but that will not drop your temps 8*c.

[Beriphent]

LinX and Intel Burn Test are quite more intensive than Prime 95. If the chip was an IB, it would be a bit more in line. Arctic Silver is also a bit dated...go for MX-2 or MX-4 TIM.

[Jansa]

MX-4 supposed to be good with GPU-s. Made yesterday remount and temperatures went up by 5 degrees Celsius (under load with 130W CPU overclocked)
Made one more remount with MX-4 and still seems the same. Previously had some old Prolimatech (bundled with aircooler) now MX-4. As currently testing loop had made about 10 remounts with Prolimatech, constant results. Noob in watercooling but just badly surprised how much can be difference So speculating on my results, next TIM I go for and very soon should be PK-1

[jomama22]

I use timtronics grey ice 4200. Its a great paste and does a great job at water cooling. To put this in perspective, I have a 3960x at 1.38v @ 4.7 and also use a raystorm and see Max of 63* in Intel burn and prime. 1 core goes as high as 67 while all others stay below 61 and this is with greater than 30* ambient temps

[OCAUfedexpress]

seems way high for a raystorm

your delta temp is 44c which is crazy high for a 2700k with such low volts

maybe you using the wrong mounts and the cpu isn't touching the block?

take the block off, get some blu tack and stick it on the cpu and mount again, take it off and see if it is really flate lol!

just remember to clean the cpu and block after

Just set up a new loop with:
Raystorm
MCP655 variable @ 4/5
2 separate MCR220-XP's
Corsair SP120s (2350rpm / 63cfm)
EK Spin Res
in a Vengeance C70

The ambient temp in here is a very warm 30C. Just wondering if someone could verify these numbers and give an opinion on whether they look right. Seems warm to me, but this house is hot and it is throwing me off.

CPU is 2700K, currently @ 4.8Ghz, 1.355v

Temps at 30C room:
idle: 35-40C
load (LinX): 74C

Used an old Arctic Silver 5, but wasn't sure about the mount. Re-did it once. Does lapping the cpu still help these days? Last build went for almost 6 years, so I'm "out of the loop" a little.

[OCAUfedexpress]

MX-4 supposed to be good with GPU-s. Made yesterday remount and temperatures went up by 5 degrees Celsius (under load with 130W CPU overclocked)
Made one more remount with MX-4 and still seems the same. Previously had some old Prolimatech (bundled with aircooler) now MX-4. As currently testing loop had made about 10 remounts with Prolimatech, constant results. Noob in watercooling but just badly surprised how much can be difference So speculating on my results, next TIM I go for and very soon should be PK-1

not sure with MX-4 but other TIM like artic silver need like 200 hours of set time before they work properly :P

[OCAUfedexpress]

LinX and Intel Burn Test are quite more intensive than Prime 95. If the chip was an IB, it would be a bit more in line. Arctic Silver is also a bit dated...go for MX-2 or MX-4 TIM.

Wrong mate!

What has being dated got to do with performance? lol!

[hyperdoggy]

PK-3 is out, and thur far is better than PK-1 from what i can tell, not by much but i already favor PK-1 over AS-5 from personal experience. My results is very close to what Skinnee's test results. So if you were to get a new tim to see if that improves your temp, start off from checking out some reviews, Vapor and Skinnee does an awesome job of stuff like that(which both are sticky on the watercooled section of the forums).

For comparison, i build a system for a friend with an RS360 rad, that pump that comes with the xspc kit, old Rasa block on a Asus gen 1 P67 mobo, 2600K @ 5.1ghz, voltage on offset mode +0.045 which hits 1.33ish volt on load on his system. I recently added a 680 into his loop, his cpu idle temp is around 30ish and just cpu load is about 55-60c with prime95(should be 55 but the idiot wanted to watch his 1080p movies while priming, which loaded up the vid card since it was configured for hardware AVC decode). It was 27c ambient in his room when i tested his system after adding in the 680.

With your superior loop, i would expect your delta on load would be no more than 15-25C.

[ZeroOne]

Thanks for all of the info! I'm not sure what went wrong, but all I did was follow the AS5 instructions as usual, and didn't overdo it with the paste. Remounted it once, and the paste spread seemed decent from the first mount, when removing the block.

How far down do you tighten these springs? I'm about half way, but they are rock solid and don't seem to give much at all.

Remounted it the first time because I was afraid I used too much paste when i saw the temps. Could feel it skate around very slightly when trying get the first 2 (opposite from one another) screws in, and figured that must have been the problem.

I wish this mount were opposite, and studs went into the backplate FIRST as guides.

[jomama22]

Thanks for all of the info! I'm not sure what went wrong, but all I did was follow the AS5 instructions as usual, and didn't overdo it with the paste. Remounted it once, and the paste spread seemed decent from the first mount, when removing the block.

How far down do you tighten these springs? I'm about half way, but they are rock solid and don't seem to give much at all.

Remounted it the first time because I was afraid I used too much paste when i saw the temps. Could feel it skate around very slightly when trying get the first 2 (opposite from one another) screws in, and figured that must have been the problem.

I wish this mount were opposite, and studs went into the backplate FIRST as guides.

Do me a favor. Use real temp and tell us the temp of each core during IB or prime.

[Beriphent]

Wrong mate!

What has being dated got to do with performance? lol!

Um, not just performance, AS5 is worse in contact dependency and curing time. If you read his post, he was even contemplating lapping the CPU, hence the new TIM suggestion, even if it is going to be a 1-2C difference. AS5 gets beaten in just about every category (including performance) on skinnee's tests (which I trust more than HardwareSecrets):

http://skinneelabs.com/2011-thermal-...-comparison/2/

It may make a 1C or 2C difference, but if you're going to start talking about lapping CPUs, then it's worth it to at least change out your old TIM before you go on voiding your warranty.

[OCAUfedexpress]

Um, not just performance, AS5 is worse in contact dependency and curing time. If you read his post, he was even contemplating lapping the CPU, hence the new TIM suggestion, even if it is going to be a 1-2C difference. AS5 gets beaten in just about every category (including performance) on skinnee's tests (which I trust more than HardwareSecrets):

http://skinneelabs.com/2011-thermal-...-comparison/2/

It may make a 1C or 2C difference, but if you're going to start talking about lapping CPUs, then it's worth it to at least change out your old TIM before you go on voiding your warranty.

Why do you trust that reveiw more?

It doesn't even tell us how he tested?

[jomama22]

Why do you trust that reveiw more?

It doesn't even tell us how he tested?

"To read the how/what/why of our testing, please read any of the installments. To get our impressions of any specific TIM, please read the installment where we review that specific TIM." - from the first page of the article he posted. Not his fault it was tl;dr for you. Every link on the first page of the article explains in pain staking detail how each was tested.

skineelabs is a very revered watercooling review site for just that, his insane level of detail and analysis.

[Notechis]

Why do you trust that reveiw more?

It doesn't even tell us how he tested?

for example: http://skinneelabs.com/frostbite-heg...ntum-review/2/

[OCAUfedexpress]

well the AS5 has a 200h curing time? If I'm reading right, he only spends 10 hour testing? (which is a decent amount of time)

But that is nowhere near the required curing time for AS5?

[jomama22]

well the AS5 has a 200h curing time? If I'm reading right, he only spends 10 hour testing? (which is a decent amount of time)

But that is nowhere near the required curing time for AS5?

well i doubt he wanted to wait over a week just for one paste. he is after all testing dozens of pastes. After 10 hrs, you will get a good idea how well it will do, even arctics website says it will only drop 1-5* if done properly. But it is way more informative than hardware secrets. i myself have use as5 a few times and the most i saw temps drop over 200 hrs is 1 degree. sometimes, it was actually better in the first 20hrs than after 200.

[grasskicker509]

Thanks for all of the info! I'm not sure what went wrong, but all I did was follow the AS5 instructions as usual, and didn't overdo it with the paste. Remounted it once, and the paste spread seemed decent from the first mount, when removing the block.

How far down do you tighten these springs? I'm about half way, but they are rock solid and don't seem to give much at all.

Remounted it the first time because I was afraid I used too much paste when i saw the temps. Could feel it skate around very slightly when trying get the first 2 (opposite from one another) screws in, and figured that must have been the problem.

I wish this mount were opposite, and studs went into the backplate FIRST as guides.

i went as far as the screws would allow with my raystorm...no problem and good temps

[ZeroOne]

Do me a favor. Use real temp and tell us the temp of each core during IB or prime.

Prime v27.7 build 2, with small FFT using AVX, temps in RealTemp 3.70 are:

Max:
66
73
74
73

That first core is always lower, after each mount.

Double checked the mount screws, and they are all the way down. Room temp is officially 29C, exactly.

Looking to replace the thermal paste. Was eyeing the Prolimatech PK-3, or Indigo Xtreme. Would rather hold off on the IX until this is sorted and normal with real "paste" first, in case something else is wrong. Hate to waste a $9 mount.

I can't imagine much else could be wrong... there is really nothing special going on here.

2700K batch # 3205B766 / D2 Revision
4.8Ghz, 1.355v

[hyperdoggy]

ZeroOne - There isn't much else that can be wrong, is a little redundant but may i ask if your 4 fans is config for push or pull. try pull(sucking air into the case) if you haven't done so. Is there any kink or tight bends in your loop? maybe in some way is restricting your flow. Your voltage, is that what you're reading off of cpuz, or what the mobo is preset at? i notice most new mobo have some type of load line calibration features, in addition to other options in the bios which may add more voltage to your system.

Edit:Another good way to test, is to undo your overclock setting and see if you get really good temps/delta in that. If you're doing +/- 5c delta idle and +/- 15c on load than your loop for the most case should be running fine. If is higher than that, there maybe something physically wrong with your loop.

OCAUfedexpress - Is it that shocking that the AS-5 can be improved upon and other tim in modern days can outperform it? 200hr burn in...i expect no review site really spent that long for a tim review, but then again hardware secrets did test butter and tooth paste, maybe they are pretty serious...lol. The reason i tend to trust personal but professional review is that they are not sponsor or money driven, they have no deadline to met or do they have anyone to answer to, thus normally would give you a unbiased review. I'm not saying you are wrong about AS-5, or your favorite review sites, but at least in ZeroOne's case, he can try a newer tim and see if it does better since he's already on AS-5.

[ZeroOne]

ZeroOne - There isn't much else that can be wrong, is a little redundant but may i ask if your 4 fans is config for push or pull. try pull(sucking air into the case) if you haven't done so. Is there any kink or tight bends in your loop? maybe in some way is restricting your flow. Your voltage, is that what you're reading off of cpuz, or what the mobo is preset at? i notice most new mobo have some type of load line calibration features, in addition to other options in the bios which may add more voltage to your system.

Edit:Another good way to test, is to undo your overclock setting and see if you get really good temps/delta in that. If you're doing +/- 5c delta idle and +/- 15c on load than your loop for the most case should be running fine. If is higher than that, there maybe something physically wrong with your loop.

Went back to stock.

Prime blend is: 50, 56, 51, 56
Prime small fft w/ AVX: 55, 58, 57, 59

This is getting somewhat annoying. I have a stock 2600K running in another system next to this one with 67C Prime with an Enermax ETS-T40. Not THAT much higher at about the same speed. Real temp: 63, 66, 66, 66.


The flow seems good.... D5 is on 4/5 and the EK spin res is really moving.

Loop order is: reservoir, pump, radiator-1 (pull, buttom of case, intake), cpu, radiator-2 (top, push, out), back to reservoir.

Going to order a coolant temp sensor just-because. Maybe that will help in figuring out what the problem is.

[hyperdoggy]

that temp looks good for stock, so your loop physically is fine. I guess check your setting on your oc. See what you can push with default voltage till it gets unstable, If you can tell me your temp at the most stable O/C @ default voltage and you can start over from there.

[OCAUfedexpress]

OCAUfedexpress - Is it that shocking that the AS-5 can be improved upon and other tim in modern days can outperform it? 200hr burn in...i expect no review site really spent that long for a tim review, but then again hardware secrets did test butter and tooth paste, maybe they are pretty serious...lol. The reason i tend to trust personal but professional review is that they are not sponsor or money driven, they have no deadline to met or do they have anyone to answer to, thus normally would give you a unbiased review. I'm not saying you are wrong about AS-5, or your favorite review sites, but at least in ZeroOne's case, he can try a newer tim and see if it does better since he's already on AS-5.

hehe, I'm not a as5 fanboi just annoyed that people keep saying it is conductive and that it's bad for CPU?

it says on the official site that it does not conduct electricity but conducts HEAT. I'm pretty sure they are 2 different things?

[OCAUfedexpress]

that temp looks good for stock, so your loop physically is fine. I guess check your setting on your oc. See what you can push with default voltage till it gets unstable, If you can tell me your temp at the most stable O/C @ default voltage and you can start over from there.

I wouldn't say they are good, my old corsair h70 was giving me around the same temps at 4.5ghz lol!

[ZeroOne]

I'm thinking of removing the Raystorm and disassembling to see if something is caught inside from manufacture. It's annoying to see these temperatures, that are on par with air, with a non-complicated, but good-hardware water setup.