Jump to content
Corsair Community

H80i Very High Temps during stress testing


Recommended Posts

Hello! I've been using H80i for 5 years and it has been cooled very well until a week later when I decide to run Prime 95 v26.6 small ftts. It started around the early 70s. It quickly roses up to the high 70s after the first minute. It hangs there until it reaches 80c-81c. It still hasn't rosed up until I stopped at the 5 minute mark.

 

I'd aslo tested Aida64 using FPU+CPU+Cache+Memory and it starts in the mid 60s and by 30 secs it quickly rose to 77c then drops to high 60s..then rosed to 80c. The spikes are fast. It just spikes up and down. It reaches to 82c and temps are spiking in between low and high 70s until I stopped at the 5minute mark. During the test I decide to put my hand on the back of the rear fan and I somehow couldn't feel hot air coming out from the radiator. I'm not sure if this is normal for stress test, so I decided to run RealBench 2.43.

 

I started RealBench 2.43 stress testing using 32gb of memory and 15minute duration. It starts in the mid 60s. I watched the temps very closely and it hangs in the late 60s in the first minute. Then during the 4th minute mark, it crawls to 73c then 74c the next second. Temps are still in the low 70s until the 10th minute mark where the top core reached 75c. I went to the back of the pc and feel my hand next to the rear and I can feel a bit of warm air but not as much. At the end of the test, it still lingers on in the early 70s and the max temp is 75c.

 

I'd also tried using the 4gb version, same duration 15 minutes as it starts in the late 50s, then spikes to the mid 60s then roses up until 71c at the 2:26 mark. 4th minute in, temps rose to the early 70s until it roses to 76c at the 6 minute mark. Temps were still in the low to mid 70s til the end of the test where the max temperature is 76c on core 0 and CPU Package and IA Cores.

 

I really don't remember what makes load temps so high compared to 3 months ago where temps don't exceed 71c during Prime95 for 24 minutes as well as Xtu Benchmark where temps also don't exceed 71c. I'd figure it might be Xtu software, even though I didn't touch the setttings. Or it could've been the cold weather that could drop those load temps. I did uninstalled Xtu and went to bios and use "Load Optimized Defaults". Then, I change the case fan speeds, set ram to xmp, then turn on intel virtualization and save the settings and exit the bios. Somehow temps are still the same.

 

For games, I tested TF2, Genshin Impact and GTA V. TF2 and Genshin Impact's temps were in the high 40s to mid 50s. Gta V went mid 50s to early 60s, max temp was 63c. I never thought that temp can be that high, though last time it was like the low 50s during the first few years of using h80i and my system (which was bought from Ironside Computers around May of 2015). Dang that game was CPU intensive as all hell.

 

My settings were all set to auto, with no multi core enhancement and CPU ratio is auto as well, except ram settings to use XMP profile. Ambient Temps varies between 73f to 77f as I live in a room that is converted from a garage to a bedroom. My fan setup as 1 200mm fan as intake, 2 1x120mm as intake. 1 200mm fan top as exhaust and my h80i fan placement to push-pull out to the rear end of the pc.

 

I've been googling around about this as well as talking in tech support on PCMR discord and one told me it is time to change the paste and reseat the CPU, as unfortunately I couldn't do that because I don't have a thermal paste and my family would told me no. So I guess I'm out of luck. Also, I do have a bad fan bearing on the back of the radiator that makes clicking noise sometimes. Does this has to do with high temps?

 

So it leaves me with few possibilities..

 

1) Thermal paste wasn't good, thus need to clean old paste and reseat it which for me deemed impossible due to family choices.

 

2) Bad bearing in fan - I don't know if that could bring high temps

 

3) Maybe pump? I dunno. Pump seems to work okay as temps are 27c - 33c (though if left alone for a while it was 24c-28c)

 

I know I'd probably answered my own self but I definitely want your opinions on this. I'd really appreciated, thank you! :)

Link to comment
Share on other sites

1) If you have not changed the TIM in 5 years, you are probably due for a re-application. There is no definitive time period it should last and mileage varies, but 5 years is plenty long.

 

2) No, just noise. Small changes in fan efficiency bring small changes in liquid temp.

 

3) Not the pump. Pump problems are serious and will result in temp problems at idle, let alone max load.

 

 

Frankly I don't see a clear problem from the data. If you start a Prime 95 run and the instant 1 second in temp is 70C, then it slowly rises to 80, the +10C increase is the coolant temp rise. That +10C is pretty much in line with expectations. Obviously the initial jump to 70C is the larger part of the overall temp, but that is not something you can do much about except by lowering voltage or somehow improving conductivity. The last part is hard, but if your TIM is a cracked mess, it will have an effect on the initial second 1 value.

 

You can monitor cooler efficiency through the Link or CUE apps. This is the H80i Temp value in either program and its also referred to as coolant or liquid temp. This is the value you are attacking with fan and pump speeds. Neither of those can reduce the immediate temperature impact of applying 1.xx volts to the underside of the CPU. The heat must be conducted through the CPU before the "cooler" takes it away. In that sense, it is more like a waste heat disposal device. You are preventing the CPU temp from being worse, rather than actively "cooling it".

Link to comment
Share on other sites

1) If you have not changed the TIM in 5 years, you are probably due for a re-application. There is no definitive time period it should last and mileage varies, but 5 years is plenty long.

 

2) No, just noise. Small changes in fan efficiency bring small changes in liquid temp.

 

3) Not the pump. Pump problems are serious and will result in temp problems at idle, let alone max load.

 

 

Frankly I don't see a clear problem from the data. If you start a Prime 95 run and the instant 1 second in temp is 70C, then it slowly rises to 80, the +10C increase is the coolant temp rise. That +10C is pretty much in line with expectations. Obviously the initial jump to 70C is the larger part of the overall temp, but that is not something you can do much about except by lowering voltage or somehow improving conductivity. The last part is hard, but if your TIM is a cracked mess, it will have an effect on the initial second 1 value.

 

You can monitor cooler efficiency through the Link or CUE apps. This is the H80i Temp value in either program and its also referred to as coolant or liquid temp. This is the value you are attacking with fan and pump speeds. Neither of those can reduce the immediate temperature impact of applying 1.xx volts to the underside of the CPU. The heat must be conducted through the CPU before the "cooler" takes it away. In that sense, it is more like a waste heat disposal device. You are preventing the CPU temp from being worse, rather than actively "cooling it".

 

1) I'll probably see about it, although I remember my bro-in-law had alot of trouble installing the h80i on my old build. Last time when he tried to reseat a liquid cooler that came from my old build from IbuyPower, it made it run to 8gb out of 12gb of ram and the post says "detecting dram settings", something like that then it reboots like 3 or 4 times and finally it posts (x58 build). I honestly don't want to fall the same faith again on my current build. If my h80i decides to die, I wonder what's a great cooler that is also easy to mount to the cpu? I'd figure it could be air cooling but if the mounting on the block on newer liquid coolers are improved, I would love to go liquid cooling again as my case was haf 922. I can't decide which path to go next. Its pretty medium and my ram is G.skill ripjaws 1600mhz ram. If I go air cooling, I'd fear that monster coolers, like BeQuiet! Dark Rock Pro 4 would have trouble with my ram and case.

 

2) Oh, I see. The RPMs on the stock fans from h80i were the same as I have the z97 system around 5 years ago, though increased more slightly as far as temps go. I didn't know fan speeds affect liquid temps. If that case fan decides to get worse, I've been thinking of getting Corsair ML120s as it has the similiar fan speeds as stock, though its a shame that Corsair SP 120s are out of stock. Dang!

 

3) Oh okay! Gotcha! The pump hasn't made any bad noises yet besides the bad fan-bearing. I do recall the liquid cooler (I think it might be an asetek cooler) had temp problems and it got me concerned until 2 years later when I bought the h80i cooler on my x58 build. Temps were in the 50s idle and it can go as high as the late 70s. Yeah, that was bad until I realized that it has those power saving settings, speedstep, and turbo boosts were disabled as it was mildly overclocked by IbuyPower. Man, it had bad oc on start on posts and it reverts to stock settings. For me, I don't know anything about OC to begin with. :X

 

So I'm guessing these temps are in line with my h80i despite my ambient temps from 77f-73f from my window ac unit, which in case they're fine? I do read the Intel CPU Temperature guilde on Tom's Hardware by CompuTronix and according to him, he reccomends the ambient temps around 22c and here's the base line of temps and the load: Here's the normal operating range for Core temperature:

 

"80C Hot (100% Load)

75C Warm

70C Warm (Heavy Load)

60C Norm

50C Norm (Medium Load)

40C Norm

30C Cool (Idle)"

 

So what he's saying is below 80c are safe for everyday use. Right now as I'm watching a twitch stream on my second monitor while typing on my main monitor, temps are around 30c-34c at about 14% load which is normal. For medium load is I regularly play TF2 and Genshin which yields around mid 40s through mid 50s, 51c-60c if I watch a twitch stream or youtube video. Though both of those combined will yield around 50% usage. GTA V was the most CPU intensive game, which yields similarly as I would play both TF2 and Genshin while watching a twitch stream. I've tested GTA V around 30 mins and the max temp was 64c. So technically, these temps are good on stock settings on H80i, right? Same with synthetic benchmarks in the high 70s to 82c?

 

Though I'm beginning to think around the time I bought the PC around 2015, my ambient temps are probably around mid 65f from my a/c as I tend to run my room chilly at the time. Temps were steady in the 40s and early 50s in WoW, Overwatch, TF2, and GTA V. As time went on, I decided to put my ac on 76f and I didn't even think these load temps reach in the low 60s when I play GTA V around a day ago. Then again, I regularly play TF2 at the time so mostly it was around mid 40s through early 50s. I'm 100% sure that ambient temps plays the major part as well as maybe TIM's age (which I assumed the paste was pre-applied). For now, I feel like I'll just accept those temps and move on.

 

Unfortunately the PC I had in 2015, the builder didn't put corsair link cable on the h80i block, though the link cable was put away in a ziplocked bag inside the box from my dresser. I honestly didn't know it was even necessary so I just go along with it. Though I'll try to plug that one in if I have the time to check out the liquid temp.

 

Thank you, C-Attack! :)

Link to comment
Share on other sites

  • Corsair Employees
Hey there, thanks for reaching out. Your temperatures aren't really that out of line. I would for sure repaste your CPU with new thermal paste. Be sure to clean off the old paste well, and check that the fan and radiator is clear of dust/debris.
Link to comment
Share on other sites

Hey there, thanks for reaching out. Your temperatures aren't really that out of line. I would for sure repaste your CPU with new thermal paste. Be sure to clean off the old paste well, and check that the fan and radiator is clear of dust/debris.

 

You're welcome! :)

 

That's good! I'm glad that these temps I've experienced are in good range, though I'd forgot to mention that my cpu is at stock settings which means no overclocking. With stock settings I'm using since I bought the PC, are the temps okay for h80i? I'd apologizing for repeating the same question.

 

I do clean my PC every three months by using air can spray to remove dust out of the radiator as well as the inside of my case. I'll try to clean out the case fans and cpu fans as well next time when its due to cleaning. I'll try to clean the paste off the cpu and applying a new one if my temps starting to get worse. Once again, thank you very much, Corsair Travis! :)

Link to comment
Share on other sites

  • Corsair Employees
Your temps look good to me, if it's been over a year since you redid the thermal paste I would go ahead and redo it though. Keep in mind that these synthetic tests put an unrealistic load and are intended to show a worst-case scenario.
Link to comment
Share on other sites

Your temps look good to me, if it's been over a year since you redid the thermal paste I would go ahead and redo it though. Keep in mind that these synthetic tests put an unrealistic load and are intended to show a worst-case scenario.

 

Actually, I haven't done cleaning and reapply thermal paste since I bought the pc in 2015, though I'll do that if my cpu temps start to get worse. I guess I'll live with gaming with late 50s to early 60s temps. :)

 

As your input on synthetic tests, I guess its true that these tests put extra heat and voltages no matter what good cpu cooler you use. Then again, I guess I can call these low 80s temps in prime 95 v26.6 acceptable too along with 77c in xtu benchmark, and the rest of the benchmarks I've used with similiar results.

 

Once again, thank you, Corsair Travis, for your input! :)

Link to comment
Share on other sites

6 years on the same paste, you will see improvements, particularely if it's the default one.. it will probably be caked on that IHS. But those temps are far from terrible.

 

Yeah, before that it was toppin at 71c on Prime95 v26.6 on december but now it reaches in the lower 80s. I'm 100% positive that ironside computers used the preapplied thermal paste that comes in the h80i. I dunno how long does that thermal paste last though.

 

I did a test on Cinebench R20, as I started it it starts in the late 60s, then it slowly climbs to the late 70s when the test is finished results in cores: 1) 79 2) 78 3) 75 and 4) 67 and CPU Package and CPU IA Cores are both top 79c.

 

I'd also ran Cinebench R23, as I started the test, the temps jumps to the late 60s and it slowly climbed up to the early 80s during the test. At the end, the highest temp was 84c. It was pretty bad. :(

 

I still haven't gotten around to reapply new paste yet as I still don't have it in the house.

 

I do recall installing xtu for the first time to benchmark back in december and temps top it off at 71c. I haven't touch anything in the advance settings or anything. I decided to uninstalled it through programs and features and didn't encountered that until I started prime 95 and seeing the increase of load temps in the late 70s and early 80s. Does xtu modifies the bios? Even if its uninstalled the settings still applies? I even used load optimize settings and temps are still the same. I do have the first xtu result uploaded on hwbot and compared to the first one and the last one, and the first one has 88% compatibility even though it was my system.

 

Even if I uninstalled xtu through programs and features, is there anything I can do to get rid of some of the remnants?

 

I'd apologized if that one was confusing but I've been assuming that something in xtu could be the cause of the load temps.

Link to comment
Share on other sites

I use XTU for overclocking, and from what i saw, it doesn't keep any change active after you reboot the PC. you can apply an OC, close XTU and it will be in effect untill you shut down the PC. at reboot, it will be at whatever your bios settings are.

It doesn't write anything in bios AFAIK.

 

your temps on Cinebench R20 kina points at a little paste issue, possibly.

12°C between hottest and coldest core is a bit much. It's not unusual either on those older CPUs but still a bit high.

 

But again.. your temps aren't abnormal :)

Link to comment
Share on other sites

I use XTU for overclocking, and from what i saw, it doesn't keep any change active after you reboot the PC. you can apply an OC, close XTU and it will be in effect untill you shut down the PC. at reboot, it will be at whatever your bios settings are.

It doesn't write anything in bios AFAIK.

 

your temps on Cinebench R20 kina points at a little paste issue, possibly.

12°C between hottest and coldest core is a bit much. It's not unusual either on those older CPUs but still a bit high.

 

But again.. your temps aren't abnormal :)

 

Oooh, that makes total sense! I guess Xtu is pretty much application setting so that's good. I tried using the first profile I uploaded on hwbot though it but honestly, that didn't make a difference in temps sadly so that one concludes it.

 

A little paste issue, eh? I do have a screenshot of prime 95 26.6v smallftts compared to a xtu benchmark which I've done a few minutes ago. Both have like 9c-10c difference on first and last core. Only difference is the ambient temperature maybe since the first one was taken on december which sometimes my room gets really cold during winter. (maybe water temp comes to play too? I still haven't plug that link cable yet ><)

 

At least I'm glad that you think my temps aren't bad, even in auto stock settings. Like I said before, if temps in gaming started to get worse I'll buy the thermal paste as well as plugging the link cable in the mobo.

 

And btw, since one of the radiator fan bearings went bad, what are the best fan choices for h80i as of now? The SP120s are sold out unfortunately and the corsair ML120 fans would look very good on the h80i. What do you guys think?

Clipboard01prime95v26.6maxtempsfor25mins.thumb.png.72de18b8148ac573068b9f0bf1e334cc.png

Clipboard01xtustresstest@75froomtempnewfansetup3.thumb.png.1e3eec6c72d60cc2c22835823c44cd99.png

Link to comment
Share on other sites

Prime95 is good to tell you in two seconds if you have a serious problem.. it will crash a bad overclock or overheat a CPU with cooked paste, bad mounting, dead pump etc..

it's a cooling endurance run more than a realistic AIO performance test.

 

As for fans if you don't care about RGB, yup, a pair of ML120 would be perfect in term of cooling if you're not worried about noise. You would need to create custom fan curve with this one. when the AIO kicks it close to full speed, you'll expect to see the computer take off :p

For radiator use i personally prefer the arctic P120. Corsair has yet to release a fan with that kind of sweeping blade which i find to work really well on radiators. But there's a ton of options. Also check the newly released SP elite, but i don't think they will be any easier to find unfortunately ^^'

Link to comment
Share on other sites

  • 2 weeks later...
Prime95 is good to tell you in two seconds if you have a serious problem.. it will crash a bad overclock or overheat a CPU with cooked paste, bad mounting, dead pump etc..

it's a cooling endurance run more than a realistic AIO performance test.

 

As for fans if you don't care about RGB, yup, a pair of ML120 would be perfect in term of cooling if you're not worried about noise. You would need to create custom fan curve with this one. when the AIO kicks it close to full speed, you'll expect to see the computer take off :p

For radiator use i personally prefer the arctic P120. Corsair has yet to release a fan with that kind of sweeping blade which i find to work really well on radiators. But there's a ton of options. Also check the newly released SP elite, but i don't think they will be any easier to find unfortunately ^^'

 

Yeah, I would definitely go with the ml120mm since there's a red led version and my caae was CM haf 922 and has red led fans. Can't go wrong with that. :)

 

Now, for the most of the stress tests I've done, my vcore voltage is auto. My mb's Auto vcore varies between 1.099v to 1.208v depending on the stress test. One time, it spoked to 1.275 during Xtu benchmark. I'm still confident that the temps are fine, but I'm wondering if these temps are okay with these voltages in auto during stress testing?

Edited by RyougaLolakie
Link to comment
Share on other sites

are these vcore or VID? if it's Vcore it's pretty standard and not high for stock settings.

But at the beginning of the bench run, if temps allow it will keep TDP at 120W for 28s and not 95w fixed. (PL1, PL2 and tau stock settings if i remember them..)

 

So, not too bad temps.. but could be better. it's a very small AIO so it will benefit a lot from all the help you can give it.. repaste, fans with better static pressure to save your ears etc..

Link to comment
Share on other sites

are these vcore or VID? if it's Vcore it's pretty standard and not high for stock settings.

But at the beginning of the bench run, if temps allow it will keep TDP at 120W for 28s and not 95w fixed. (PL1, PL2 and tau stock settings if i remember them..)

 

So, not too bad temps.. but could be better. it's a very small AIO so it will benefit a lot from all the help you can give it.. repaste, fans with better static pressure to save your ears etc..

 

Yes, that's the vcore.

 

I did ALOT of googling about this and somehow some of the peeps have that issue, no matter which aftermarket cooler. Most of the times it was from the mobo settings, others that have problems due to the TIM under the IHS and die. Though, I'd get it its a hot chip and I'm not sure if that TIM could be dried up after 5 and a half years of operating. I don't want to delid that cpu as I had no experience on it and I still don't plan on OC this on either.

 

Though, I still use 2x 200mm apvevia fans as a replacement since my stock cooler master 200mm fan bearing gone bad. The other two 120mm fans came from the cooler master hyper 212 cooler even though I don't have that cooler. To be honest though, I'd still like the sickleflow fans better if these two will survive for longer. The 200mm apvevia fans rated at 700rpm as I plugged in the first one in the molex and the other connected to the mobo. The 2x 120mm hyper 212 fans were hooked on the side case connected to the fan hub that is connected to cha_1 header and all of the case fans ran on full speed as configured in the bios. I did switch the 2x 120mm fans from exhaust to side intake and somehow that kinda increase temps by maybe 2-3c. Maybe next time when I clean the pc, I'll try to switch the 2x 120mm fans as exhaust and 1 200mm fan as intake. Maybe bad airflow? oof..

 

I'll get the ML120 red led fans if the h80i exhaust fan's bearings start to get worse. For now, sometimes the bearing makes noise sometimes it don't. Once again, thank you for telling me that my temps aren't bad.

Link to comment
Share on other sites

the 9900k doesn't use silicon TIM under the IHS. it's soldered, so you can forget about that one :)

repasting the waterblock tho takes only a few minutes. if the old TIM has dried you may spend more time wiping it off with rubbing alcohol than actually applying paste. still it's a 5 min job. unscrew, wipe wipe, squirt some kryonaut, screw and done.

Link to comment
Share on other sites

the 9900k doesn't use silicon TIM under the IHS. it's soldered, so you can forget about that one :)

repasting the waterblock tho takes only a few minutes. if the old TIM has dried you may spend more time wiping it off with rubbing alcohol than actually applying paste. still it's a 5 min job. unscrew, wipe wipe, squirt some kryonaut, screw and done.

 

Sounds kinda easy, but I'd fear that I might screw up the pins under the mobo nor cpu when taking out the h80i as well as tightening the screws. Didn't want to face to face missing ram problems again after my bro in law repasted tim on my x58 system. That is the scary part. x_X

 

 

That's good to know that 9900k doesn't use silicon tim under the IHS as well as Sandy Bridge cpus.

Link to comment
Share on other sites

there's no pins you touch at all :) you don't even remove the backplate, just losen the 4 thumb nuts and the pump goes away, that's all you need to do.

once it's off, it's just a matter of wiping the old TIM, applying new one, pressing the pump on the CPU and putting back the thumb nuts in a cross pattern, to make sure you keep the pump flat on the IHS.

 

You never remove the CPU from the socket when doing that. If the pump is stuck on the cpu in case the TIM is dry, you gently twist it clockwise / counterclockwise until the dried paste gives up.

Dry paste disolves in alcohol so it's easy to remove without needing to scratch the surface.

Link to comment
Share on other sites

there's no pins you touch at all :) you don't even remove the backplate, just losen the 4 thumb nuts and the pump goes away, that's all you need to do.

once it's off, it's just a matter of wiping the old TIM, applying new one, pressing the pump on the CPU and putting back the thumb nuts in a cross pattern, to make sure you keep the pump flat on the IHS.

 

You never remove the CPU from the socket when doing that. If the pump is stuck on the cpu in case the TIM is dry, you gently twist it clockwise / counterclockwise until the dried paste gives up.

Dry paste disolves in alcohol so it's easy to remove without needing to scratch the surface.

 

Gotcha! What would happen if the paste is so dry and you can't wiggle it out?

 

I would love to do this but my family told me to leave it alone, even though I'm an adult. So I will leave it alone until the cpu fan bearing starts to get worse. My family didn't want me to take things apart from my pc, besides cleaning inside the pc every three months. Sorry if I'm seem to derail my thread for holding on for a fix. My family's choice, not mine. :/

Link to comment
Share on other sites

no worries hehe

even totally dried up paste will let the cooler pop free. it's not glue. it will have the consistency of dried plaster, and crumble just like it, in clumps. you can remove most of it with your nails and finish with isopropanol on some cotton rag, or any alcohol, it will disolve.

Just don't use paper towel, that thing will fall into bits all over the motherboard.

 

If indeed it's the paste going bad you'll be forced to do it sooner or later, but as of today it seems okay to leave it be. When you'll start experiencing thermal throttling, it will actually be safer to repaste than to let the CPU cook itself.

 

I imagine family just doesn't know what's involved in repasting, so better not venture into the unknown and mess things up :)

It's routine maintenance really. Looking up tutorial videos on YT helps a lot, you see how many people do it, and chose how you want to go about it.

Link to comment
Share on other sites

no worries hehe

even totally dried up paste will let the cooler pop free. it's not glue. it will have the consistency of dried plaster, and crumble just like it, in clumps. you can remove most of it with your nails and finish with isopropanol on some cotton rag, or any alcohol, it will disolve.

Just don't use paper towel, that thing will fall into bits all over the motherboard.

 

If indeed it's the paste going bad you'll be forced to do it sooner or later, but as of today it seems okay to leave it be. When you'll start experiencing thermal throttling, it will actually be safer to repaste than to let the CPU cook itself.

 

I imagine family just doesn't know what's involved in repasting, so better not venture into the unknown and mess things up :)

It's routine maintenance really. Looking up tutorial videos on YT helps a lot, you see how many people do it, and chose how you want to go about it.

 

I'd imagine it'll look like dried up piece clay when you talk about it. :p: I don't think I have some rubbing alcohol, which I will consider getting when I order the thermal paste (though thr noctua nt-h2 paste has thermal paste wipes included which I can use that instead depends on the paste corsair used for preapplied paste). Would cotton balls work okay to clean with rubbing alcohol?

 

Honestly though, I do look a bit on googling some more about 4790k and high temps using h80i/h100i and some people have similiar temps as I have. Their 4790ks are at stock settings.They have high temps (around 80c) when using either prime95 later versions or version 26.6v (as well as occt and aida) but they get temps from 50s and 60s during gaming as their room temps are around 75f which is what my room temps are right now. Even if I make an attempt to repaste, the temps wouldn't make a difference when running prime95 v26.6/aida64/cinebench r20 & r23.

 

I'd agree that its a routine maintenance for reapplying thermal paste. I did watch a few videos about it. And yeah, I wouldn't go there without messing it up.

 

Right now I'm thinking of a good way to optimize air as having 4 intakes and 3 outtakes on my case. Maybe that could be the case to bring down temps due to imbalance?

Link to comment
Share on other sites

When I did a build a few months ago, I used MX-4 with my H115i cooler. My CPU is about 2C above ambient even with a benchmark running. I use a very small amount of MX-4 as it spreads itself thin with thermal effects.
Link to comment
Share on other sites

When I did a build a few months ago, I used MX-4 with my H115i cooler. My CPU is about 2C above ambient even with a benchmark running. I use a very small amount of MX-4 as it spreads itself thin with thermal effects.

 

What were your load and idle temps on your new build, even in benchmarking?

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...