Question Air Cooler for a 8700k.

[Samuel SC]

Good day, I don't want to go for liquid cooling, so I'll need an air cooler.
The price doesn't matter, the only thing I ask is that it does not interfere with my ddr4 (4 x 8 Corsair vengeance RGB pro).

These data can help:
Case: Corsair obsidian 500d rgb.
CPU: Intel core i7 8700k.
Motherboard: Asus rog maximus XI hero (wifi).
DDR4: Corsair vengeance rgb pro 4 x 8.

My main use is videogames and video editing. I don't plan on overclocking for a couple of years, until my processor becomes a bit slow for the new requirements of the games.

 

[rigg42]

I actually ran a DR4 in pull with a 2700x for a while for the exact same reason as the OP. It was about a 2c warmer than push with prime95 26.6 small fft. I suspect the Fuma 2 will do a bit better than that without the compromise. It's also cheaper and on par quality wise. I can let you know how it performs when I get my hands on an x570 and a ryzen 3000. I have a Fuma 2 sitting here that is waiting to be used in my new system.

 

[fry178]

If you want to stay with air, look at the arctic freezer series.
Pretty silent, good cooling and around 30$ a lot less than what others charge for similar or even less performance.

https://pcpartpicker.com/product/99DwrH/arctic-cooling-cpu-cooler-acfz13

Aio have a couple of advantages you cant get from an air cooler (unless a sealed "box" covering the cooler connecting it to front/back exhaust so no heat escapes in the case)
Not only do they dump the heat outside the case, lowering in-case temps noticeable,
and that helps lowering temps of other components like chipset and gpu.
They can also run fans at lower rpm without impacting perf as much,
as i prefer absolut silence at idle (inaudible when no other sounds in the room),
and my 110i with 2 artic fans (800rpm/0.08 sone) keep the cpu at max 67C while running 3h of prime and 100% cpu load. Not happing on air.

Corsair covers their units with a 5y warranty that covers ALL parts that get damaged if it fails.
Name 3 brands that offer that for air.

I forgot to plug my pump back in after cleaning my loop (custom, not aio) and my 3770 at idle load running for 30min didnt even go past 70C, with air it would have not been more than 5min before it would have shutdown because of getting to hot.

And with less airflow needed to cool the rig, less dust build up.
Then again, to some degree possible with air, but usually at similar cost,
as good fans/coolers arent cheap.

 

[Karadjgne]

The i7-8700k is actually a very cool running cpu, for an i7, all of the CoffeeLake cpus stock power draws at full load come in at under rated TDP. The 8700k is rated at 95w but 100% loads only pull 87w. That's at stock, it can also hit 150w+ at a full core 5GHz OC, which is very common on that cpu. The Darkrock 4 is a 200w rated cooler. It'll have no issues with keeping temps in check. Might not be the absolute lowest, but that does not matter at all, the cpu could care less if it's 40°C or 60°C, it runs the same. Any consistent temp under @ 70°C is well within specs, has no bearing on performance, only on a benchmark number.

With most pastes, it's a personal choice. I've been using Noctua NT-H1 for years, along with Grizzly Kryonaut. They are top performance pastes, but are more like a grease than a paste. They clean easily, spread far easier than most thicker pastes, do not dry out like AS5, unless you change coolers they are a 1 shot deal, you'll retire the pc before needing repaste. The DH-1 that comes with the DR4 is also a good paste, I've never used it. Many of the premium brand coolers like beQuiet and cryorig and Phanteks and Noctua all have very good pastes, so I'd not even bother looking for better as better will be an expensive (it's cheap really) sideways move for no gains.

 

[dphotog]

Yes, the Noctua NH-U12A would be great, and does clear the ram, however, the last time I saw one in a retail store it was $139 and that was over a year ago. Brand new on EBAY, $149.

is newegg an option for you? only $99 for the NH-U12A I got mine in its box awaiting for a new juicy build this week.

 

[Karadjgne]

Oh yes, I like newegg. It must be a case like the Cryorig H7 when it first came to the States, you could only get it in one store in Canada, and you had to catch it at the right time. It literally sold out in minutes. When it was $30 and the hyper212 was still $25, the H7 was in such high demand, pre-orders had back-orders as high as a month wait. Prolly why I've not seen the NH-U12sA in stores for a year, even if $99 is steep. My Cryorig R1 Ultimate was $60.

 

[rhysiam]

OP -> feel free to ignore this post if you're not intersted. I might be starting us down another tangent here.

I just wanted to address that claim above about water loops handling passive operation better than air...

Spoiler: Passive air vs Passive water

I forgot to plug my pump back in after cleaning my loop (custom, not aio) and my 3770 at idle load running for 30min didnt even go past 70C, with air it would have not been more than 5min before it would have shutdown because of getting to hot.

I know this thread keeps getting off track and I'm potentially doing that again, but I can't let this go without a response. Your post makes some great points and I wasn't aware of Corsair's 5yr all-encompassing warranty, so that's good to know. But with respect, that paragraph quoted above is demonstrably false.

Air cooling will cope far better passively than any standard water loop will. Heat pipes don't require power, so the mechanism transferring heat from the CPU to the fin array continues to work unimpeded. With a water loop, the pump ensures the water is physically moving and thus transferring heat from CPU to the fin array. No pump means no water flow. Movement of heat is drastically reduced. Let me put it this way, if there was a simple way for a tube with static water in it to move heat more effectively than a heatpipe, every air cooler would immediately ditch heatpipes and use tubes filled with water instead.

People have built entirely passive PCs with off-the-shelf air tower coolers.
Add just a little airflow through the case and you can actually run sustained loads on mainstream CPUs without issue.
I'm sure a better search could yield more reputable results and benchmarks, but here's some evidence:

-> With a single case fan and no fans on the CPU cooler here's a build with a i7 5960X :

View: https://www.youtube.com/watch?v=RXZrWqCT7R0


-> Again, a single case fan no CPU fans, here's an old i7 2600K running sustained Prime95(!!) @ 60 degrees(!!):

View: https://youtu.be/8_ZawCcWXXk?t=259

 

[Karadjgne]

Hah, Linus cheated. Turned the rear exhaust fan into a cpu fan, it's not passive, just semi fanless, and has 2x intakes in front, intake at rear and only exhaust is 140mm up top in front of the cpu cooler.

 

[Bearmann]

...Do you think that the temperature reached by this cooler and the noctua nh-u12a is very different?
If I make the decision to buy dark rock 4, Is there any termal paste that recommends a better quality that reduces a few degrees or that the one that brings the package is good?
Greetings.

The U12a is a better cooler than the Dark Rock 4, but that's not to say that the DR4 won't be sufficient for your needs.

Noctua NH-U12A Review | bit-tech.net

Embarassing AIO coolers with sheer quality.

bit-tech.net

I would use whatever paste comes with your cooler. It usually makes minimal difference. The Noctua NH1 paste is very well reviewed, by the way, and now has been replaced with the NH2 (which I assume is better.)

 

[Samuel SC]

OP -> feel free to ignore this post if you're not intersted. I might be starting us down another tangent here.

I just wanted to address that claim above about water loops handling passive operation better than air...

Spoiler: Passive air vs Passive water

I know this thread keeps getting off track and I'm potentially doing that again, but I can't let this go without a response. Your post makes some great points and I wasn't aware of Corsair's 5yr all-encompassing warranty, so that's good to know. But with respect, that paragraph quoted above is demonstrably false.

Air cooling will cope far better passively than any standard water loop will. Heat pipes don't require power, so the mechanism transferring heat from the CPU to the fin array continues to work unimpeded. With a water loop, the pump ensures the water is physically moving and thus transferring heat from CPU to the fin array. No pump means no water flow. Movement of heat is drastically reduced. Let me put it this way, if there was a simple way for a tube with static water in it to move heat more effectively than a heatpipe, every air cooler would immediately ditch heatpipes and use tubes filled with water instead.

People have built entirely passive PCs with off-the-shelf air tower coolers.
Add just a little airflow through the case and you can actually run sustained loads on mainstream CPUs without issue.
I'm sure a better search could yield more reputable results and benchmarks, but here's some evidence:

-> With a single case fan and no fans on the CPU cooler here's a build with a i7 5960X :

View: https://www.youtube.com/watch?v=RXZrWqCT7R0


-> Again, a single case fan no CPU fans, here's an old i7 2600K running sustained Prime95(!!) @ 60 degrees(!!):

View: https://youtu.be/8_ZawCcWXXk?t=259

Don't worry, the more I can learn better, everything is useful.

I think my options are reduced to Noctua nh-u12a and Be Quiet! dark rock 4.
The noctua uses 2 fans and has 7 heatpipes, also in my country it costs $24 dollars more tan the DR4 (but remember that I said that the price doesn't matter).
The DR4, on the other hand, has only one fan (which I should change location and you mentioned that it would reduce its performance slightly) and 6 heatpipes. Also, as you mentioned, it's a bit bigger and should work better.

I found this video that compares the temperatures obtained by these 2 coolers, where the noctua gets at least 4 degrees difference.
I don't know how reliable this channel is but at least it has the 2 coolers.
What do you think?

 

[Samuel SC]

I'm sorry, forget to attach the video, here it is:

View: https://youtu.be/g4qm85Hlq2I

 

[Samuel SC]

The U12a is a better cooler than the Dark Rock 4, but that's not to say that the DR4 won't be sufficient for your needs.

Noctua NH-U12A Review | bit-tech.net

Embarassing AIO coolers with sheer quality.

bit-tech.net

I would use whatever paste comes with your cooler. It usually makes minimal difference. The Noctua NH1 paste is very well reviewed, by the way, and now has been replaced with the NH2 (which I assume is better.)

Thanks, I think I'll go with the Noctua nh-u12a, I just want to wait to know what @rhysiam thinks.

 

[Karadjgne]

4°C is nothing to a cpu. While benchmark numbers put the U12A ever so slightly ahead, for all intents and purposes they are equitable, especially according to the cpu. Which leaves 5 other criteria.

1. Dependability. Eh, equitable, both highly rated, top quality coolers, you'll retire the pc before they quit working.
2. Looks. Ok, the beQuiet got the Noctua beat by miles, that silver/tan literally blends into no color scheme well.
3. Value. For the intended purpose, beQuiet wins again. With pretty much equitable results, beQuiet wins again. $75 vrs $100+. That's in the States.
4. Brand loyalty. Some ppl simply follow along like lemmings, Noctua has the best reputation for cpu performance bar none. For quality, performance, silence, dependability, nobody tops Noctua in every catagory, only tie in some. So definitely a Noctua win there.
5. Adaptability. The U12A uses standard 120mm fans. In the event of failure, personal tastes, whatever, you can swap that fan for any other 120mm fan. No such luck with the 135mm beQuiet, you are pretty much stuck with factory.

So there's where you are at. 2 very good coolers, either will do the job, but it'll be on the gimmicks as to which fits you better.

 

[Bearmann]

4°C is nothing to a cpu. While benchmark numbers put the U12A ever so slightly ahead, for all intents and purposes they are equitable, especially according to the cpu. Which leaves 5 other criteria.

1. Dependability. Eh, equitable, both highly rated, top quality coolers, you'll retire the pc before they quit working.
2. Looks. Ok, the beQuiet got the Noctua beat by miles, that silver/tan literally blends into no color scheme well.
3. Value. For the intended purpose, beQuiet wins again. With pretty much equitable results, beQuiet wins again. $75 vrs $100+. That's in the States.
4. Brand loyalty. Some ppl simply follow along like lemmings, Noctua has the best reputation for cpu performance bar none. For quality, performance, silence, dependability, nobody tops Noctua in every catagory, only tie in some. So definitely a Noctua win there.
5. Adaptability. The U12A uses standard 120mm fans. In the event of failure, personal tastes, whatever, you can swap that fan for any other 120mm fan. No such luck with the 135mm beQuiet, you are pretty much stuck with factory.

So there's where you are at. 2 very good coolers, either will do the job, but it'll be on the gimmicks as to which fits you better.

1. be quiet coolers have a reputation for scratching easily
2. Noctua has a reputation for excellent customer service. Be quiet has a reputation for... service needs improvement (Well recent reviews seem good. Maybe that has changed)
3. Noctua 6 year warranty. Be Quiet 3 year warranty.
4. Many report that the DR4 is difficult to install. Noctua is known for it easy to install SecuFirm2 mount.
5. The be quiet DR4 has too much play on the heat spreader according to some. Noctua is known for it easy to install SecuFirm2 mount.

(those were supposed to be 6,7, 8, 9, & 10 but it keeps changing my numbering)

You would think that I am against the be quiet cooler. I am not. I just report what I read. OK! OK! After reading all those user reviews, I'll admit it. I'm against the Be quiet cooler, especially for someone who states they are not concerned with the cost.

Look carefully as Amazon co-mingles the reviews (i.e. combines different heatsinks.)

https://www.newegg.com/p/13C-001F-0...yMark=False&IsFeedbackTab=true#scrollFullInfo

https://www.amazon.com/quiet-Dark-R..._cr_arp_d_product_top?ie=UTF8#customerReviews

https://www.newegg.com/p/13C-0005-00187

https://www.amazon.com/Noctua-NH-U1...e=all_reviews&pageNumber=1#reviews-filter-bar

Pay once, cry once.

 

[Bearmann]

Another quality cooler that I think would work is the Scythe Mugen 5 Rev. B. Probably a few degrees warmer than the U12a, but would still work well for the i7-8700K.

https://www.scytheus.com/aircooler-mugen5

https://pcpartpicker.com/builds/by_part/8GBrxr

https://www.amazon.com/Mugen-Rev-CP...&s=gateway&sr=8-1-spons&psc=1#customerReviews

 

[rhysiam]

Hah, Linus cheated. Turned the rear exhaust fan into a cpu fan, it's not passive, just semi fanless, and has 2x intakes in front, intake at rear and only exhaust is 140mm up top in front of the cpu cooler.

Yeah, I'm not for a second claiming that either of those example are truly "passive" operation. Not even the biggest air coolers like the D15 or DRPro4 can handle high end CPUs under sustained load with no active airflow. I was just objecting to the claim that a CLC with no pump running will handle that scenario better. They won't!

Not that it's really relevant for this discussion anyway (which is why I put the whole thing under a spoiler).

 

[Superlp12]

Let me share my experience as someone who has gone through a number of corsair aio and ended up with the Noctua D15. I had the 115i in push pull configuration on 8700k(with 4 noctua fans). At 4.7ghz it hit 87c with fans at around 80%.. With the Noctua D15 with the two default fans, I hit 84c with fans at 70%. The cooling capacity of the Noctua D15 is actually limited by the intel ihs. When I delidd with liquid metal applied, i was able to hit 77c max at 5ghz. All test were done running prime 95 small fft. The Noctua D15 cools better, runs quieter, cheaper and more reliable.

 

[fry178]

always depends on the config/setup.
just the options i have to setup the rad (front/top/back/bottom) and which way the fans work (intake/exhaust) will have a big impact on temps.
there will be a big difference if i mount it in the front as intake or on the top (eating the gpu heat), not even talking about having the same (amount) of airflow will impact temps.
and when i see how many ti owners have throttling issues (mainly 20xx series), there is a reason why they are better off with an aio...