I guess no one is using the Mugen5 air cooler for the 9900K so I will ask if anyone is using any similar type of Air Cooler for the 9900K which will stay un-overclocked. (factory setting out of the box/cube)

 

Looking for super quiet operation no water/AIO problems down the road.

 

Thanks

 

Terry

 

Problems would be pump failure, coolant leaks and increased general and preventive maintenance. More moving parts.

 

With air/passive not as many of the above apply. Generally just a good dusting/filter clean as required.

 

Terry

I'm using the Noctua D15 like Rob used to. I have slightly overclocked my i9 from 4.7 GHz to 5 GHz. It's been 1 1/2 years now I think and it's working perfectly.

I've been using the same Silver Arrow IB-E Extreme since 2013. Previously in an i7 4770k build and now in my i7 8086k build. I had to modify it with a dremel by removing a small portion of the fins that where conflicting with a) the DDR modules and b) the back of the video card.

BUT: I do very well with it cooling my 5.3ghz CPU at 1.4 volts. Under 100% load I see a package temp of approx. 72c. The CPU is delided and the TIM is replaced with Coolaboritory's Liquid Pro. The heat spreader and the heat sink are lapped to 3000grit and also mounted with Coolaboritory's Liquid Pro. So I was very particular about getting the best possible thermal conductivity into the assembly. You won't need to delid am i9 9900k because I believe it has solder between the die an heat spreader. Unless you want to do direct to die mounting of the heat sink doing away with the heat spreader all together. That's what I would do If I was going to do an i9 9900k build. You do need special mounting hardware for that thought. But you mentioned you are not overclocking so all that is probably a bit extreme for you.

Edited February 1, 20206 yr by Avidean

Thanks to all that replied. Currently my system has a Corsair H100i cooler that keeps the 9900K at around 29C (idle) and 38C at load running X-Plane 11.41 at 1080P thru a RTX 2080.

FPS are around the low 50s without  wx and 40-80 with wx depending on geographic location. Scenery at max or high and graphic settings all one notch left of max. Will be interesting to see the numbers once Vulcan is released to the general X-Plane 11 population.

 

Terry

Another vote for the Noctua NH-D15 here...I use that on my portable sim rig, because I was concerned that bouncing it around while on the road with a water loop could cause a leak.  It'll hold 5.0 GHz on all cores with the 9900K (HT Off) with temps in the 60s under a P3D workload.  I have mine oriented front-to-back (90 deg to the way Rob shows it) so as not to block the first PCIe slot.

As Rob mentions, it's big and heavy--I used a Coolermaster HAF XB Evo case that lays out the motherboard horizontally, with the cooler sitting vertically atop that.  And high profile memory is a problem--I had to compromise on my RAM choice because of that...from that perspective the D15S might have been a better solution.  I have had the D15 sitting brand new in my spares box for probably three years now, so decided to go with what I had.  I was pleasantly surprised.

Terry,

I am using a Deep Cool Assassin III (slightly cheaper here than the Noctua NH-D15). However, I am using a Ryzen system. Installation is incredibly simple and I am very seriously considering using another one for my Gigabyte GTX1080ti. Performance and quietness is brilliant (for me)

A good air cooler, will not get temps down as much as liquid cooling, according to most YouTubers' videos but the performance difference is not really worth debating unless you plan on very aggressive overclocking, and, the extra cost for liquid is probably not justified for some.   If they were not viable however, there would not be liquid coolers on the market, so don't rule them out. They can also look so much better with the RGB lighting available now. For me though, I think I can safely say, that I spend considerably more time looking at the monitor, rather than at my computer, so I have minimal RGB lighting and probably wish I had less.

Good air cooling is more than adequate for your unclocked system. My recommendations (for air cooling), would be Deep Cool or Noctua ... both pretty much equal, followed by Be Quiet Dark Rock Pro 4. Lots of videos and reviews out there for you.

Regards

Tony

 

19 hours ago, Rob_Ainscough said:

I used one of those massive Noctua's D15 on a 9900K setup for a work PC once ... it worked well for modest overclock and work loads.  Install was painful (literally) and RAM access and very close proximity to PCIe slot were not desirable.  I would still recommend water cooling (custom loop with quality fittings) ... what problems did you have "down the road"?  

Anyway, here is the D15:

 

 

If you had trouble uninstalling the D15 then it must have been down to your unique installation. Because installing the D15 is easy. Famous for it.

As for RAM access and close to the PCIe slot, that was because you went for the NH-D15 instead of the NH-D15S. 

 

Quote

"I would still recommend water cooling (custom loop with quality fittings)"

 

Not sure why you would do that for someone who isn't overclocking, given the cost and requirement for maintenance. 

18 hours ago, Rob_Ainscough said:

 Leaks can happen, extremely rare or end user induced,

 

Unfortunately you cant say they are "extremely rare". No figures are released by the manufacturer regarding that. What I would say is that leaks are less common than they used to be, but they most definitely do still happen. 

14 hours ago, himmelhorse said:

 

A good air cooler, will not get temps down as much as liquid cooling,

 

 

Not exactly. The best AIO's and a full blown water cooling loop will be better but a quality air cooler liker the Noctua will compete with all but the best. 

21 hours ago, Cat_Dad said:

I guess no one is using the Mugen5 air cooler for the 9900K so I will ask if anyone is using any similar type of Air Cooler for the 9900K which will stay un-overclocked. (factory setting out of the box/cube)

Looking for super quiet operation no water/AIO problems down the road.

Thanks

Terry

 

 

My recommendation would be the NH-D15S. The "S" variant is the high compatibility versions. Offset so its further away from the PCIe slot and with better RAM clearance. It has one fan rather than two, but the additional fan on the D15 or any air cooler only contributes about two degrees due to the slight increase in static pressure. So don't worry about only one fan and you can add a second if desired.

 

However... don't just look at Noctua, because there is indeed a new kid on the block that  betters the D15S by a full seven degrees. It's the new Zalman CNPS20X.

 

The nice thing about the Zalamn is that it's no nosier than the Noctua. Noise isn't mentioned in the video, but he does talk about it in the comments section. Noise is important, many cooler manufactures including both AIO and air coolers, cheat by installing noisy fans. 

 

 

Edited February 2, 20206 yr by martin-w

I don't know. That thing has only 6 heat pipes. So does the Noctua.

My ThermalRight has 8! 😁

EDIT: and a copper heat sink

Edited February 2, 20206 yr by Avidean

4 hours ago, martin-w said:

Unfortunately you cant say they are "extremely rare". No figures are released by the manufacturer regarding that.

So how can you know for sure? Leaks seemed to be more of a problem in the early day of AIOs but I'd tend to agree with Rob that, whilst obviously possible, they are now very rare unless the cooler is mishandled. Reports of spontaneous leaks after the system has been in use for some time are few and far between when you consider the number of AIOs out there. I think that, like most things today, anyone who has (or has had) a leak tends to be very vociferous about it whereas those who have no problems tend not to go into print to say so. My experience with AIOs has been good but that wouldn't necessarily stop me considering an air cooler for a future build but it would have to perform well with a decent overclock.

16 hours ago, Rob_Ainscough said:

And how is your data point different from my data point?  It's rare, less common, doesn't happen often, 1 in 10 Million units sold, etc. ...

 

That's the point, I have no data point, neither do you. There are no data points available to us because the manufacturers don't release that information. What I have said is that I would be prepared to accept the possibility that leaks, not to mention failures in general, in regard to AIO's are less common that they used to be. I would be prepared to accept that as a possibility because it makes sense that manufacturers would address any issues with their products over time. What we can do in this respect is to look at failures reported at places like support forums and the internet in general and they appear to be less prevalent. But I would never make any definitive claims because I/we do not have the manufacturers data to make those definitive claims. I think that's quite a reasonable stance to have given the lack of accurate information. 

 

Quote

if you want real data points, contact Corsair or Koolance, I did.

 

Really, so you contacted Koolance and Corsair and they shared with you the failure rates of their products? That surprises me greatly. Care to share those failure rates? I would be surprised if a manufacturer gave you that information, information that is of a sensitive nature. If they did we have no way of knowing if it's truthful information. 

 

Quote

If AIO's are failing beyond "rare", they wouldn't still be selling them because word would get out.

 

You didn't say rare though, you said "extremely rare or end user induced".

Failure rates have never been so high as to "significantly" deter sales, they have always been relatively rare. What we are debating here is the tendency to make claims without evidence, like "extremely" rare. I wouldn't be prepared to do that without evidence to back it up, hence why my stance remains at "less common than they used to be". As for "end user induced". That's pure speculation too in my opinion. We have no manufacturer data to back that up either. All we can say is that it makes sense that some owners of the product might mistreat the product. 

 

Quote

Anyway, moving forward ... Corsair will replace any faulty AIO and any components it might have damaged provided they confirm it was their failure and not something else (aka user induced failure).  So if going AIO, check the warranty details.

 

Agreed, they will do that. Although I'm told that jumping through hoops is required, multiple photos submitted etc. So I'm not sure the process is that easy. I would be interested in the opinion of anyone that's gone through that process. As I've mentioned before on the forum, I have a friend that had a H100i fail. It failed after 6 months of use. Luckily for him no other components were damaged.

Edited February 3, 20206 yr by martin-w

12 hours ago, Rob_Ainscough said:

 

Either way you go, big air coolers or good quality AIO will produce very similar results ... the advantages I see in water cooling is:

• With a sufficient large water/coolant radiators (360 or 480) the fans rarely ramp up so noise low
• Easier to mount and don't get in the way
• Less stress on the motherboard/socket area due to less weight
• Can be extended as need (i.e. if you want to cool the GPU and/or VRM)
• Aesthetics

Custom loops with quick disconnect fittings are my preferred because it makes them very easy to extend and interchange components as needed.

If you do go air, I would recommend Noctua but there other options.  If you go water, I'd recommend a custom water loop and quality fittings OR go with an AIO that has a good warranty that covers damages.

Cheers, Rob.

 

Yes Rob, I agree, the superiority of a custom loop is known. But again, I'm not sure how that's relevant to the OP and why you would say you would "still recommend water cooling (custom loop with quality fittings)".... when the OP has made it clear he has no intention of overclocking and thus, does not require such extreme cooling. The OP also said he is looking for an air cooler and not water cooling. Then we have the expense of course, which is considerably higher than an air cooler or an AIO. 

As for less stress on the motherboard compared to a big air cooler. Then yes, that is true but as we have discussed before, it's not an issue. The only evidence you will find of a Noctua damaging a motherboard is in regard to the PC, with cooler attached, being shipped across the US. Clearly a silly thing to do. I know of no one who has ever had an issue with the weight of the D15. The D14/D15/D15S over quite  a few years now, has demonstrated that the Secufirm 2 mounting system works and that damage to motherboards is not an issue.