H100i gtx to Noctua nh-D15

[Deleted member 24505]

Switched from a H100i Gtx aio to the noctua NH-D15, pretty impressed so far. I thought the H100 was good, but the pump was too noisy for me. Got the Am4 mount kit today, better than zip ties.
Idle temp is 28c which is easily as good as the AIO. I will update with the load temps later.

My case has 2 x 140mm on front, 1x 140mm on back and 3 x 1140mm on top. In/out/out respectively

Edit- Load temp while gaming is 56c so that is good enough for me. I don't run stress tests anymore, unless testing for OC stability.

 

[John Naylor]

The NH-D15 is a great cooler... but the Fuma is slightly better thermally (1C) and quieter (1 dbA) for half the price. Both outperform the H100i by a more significant margin

https://www.techpowerup.com/reviews/Scythe/Fuma/6.html

 

[Deleted member 24505]

Is the fan setup ok for a air cooler?

I still think AIO's are not that bad really, but i guess there is a chance of failure, and my PC is on all day whether i am in or out. It was the pump noise that decided it for me, my fans where quiet but the pump noise was not enough.

 

[Vario]

Fuma is slightly better

The guy already bought the NH-D15.

Is the fan setup ok for a air cooler?

I still think AIO's are not that bad really, but i guess there is a chance of failure, and my PC is on all day whether i am in or out. It was the pump noise that decided it for me, my fans where quiet but the pump noise was not enough.

You might get another degree or two of temperature reduction with a second fan on the front of the heat sink.

 

[HUSKIE]

I always use custom water cooling much than air and AIO cooling.

 

[Deleted member 24505]

I always use custom water cooling much than air and AIO cooling.

I would use a custom loop again, but cannot afford to buy the whole lot in one go, so this will certainly do till then.

 

[NdMk2o1o]

I would use a custom loop again, but cannot afford to buy the whole lot in one go, so this will certainly do till then.

@HUSKIE has a full setup going for cheap in the BST

I would add another fan for a few more c and just "cause"... it looks better

 

[xkm1948]

Using a D15 but going AIO soon. D15 was good but wanna try some 280 AIO for a change.

 

[Deleted member 24505]

Using a D15 but going AIO soon. D15 was good but wanna try some 280 AIO for a change.

Imo stick to the D15, you will barely see a difference with the AIO except pump noise

 

[John Naylor]

The guy already bought the NH-D15.

Yes, I know ... extra knowledge is a good thing .. whether for next purchase or to assist another user

 

[Deleted member 24505]

Yes, I know ... extra knowledge is a good thing .. whether for next purchase or to aassist another user

Actually I swapped my h100i for the D15, the pump noise was bugging me.

 

[John Naylor]

Using a D15 but going AIO soon. D15 was good but wanna try some 280 AIO for a change.

If ya go that route, I would normally say avoid the CLC route and get an OLC (open, expandable loop) that would allow adding a water block for MoBo and / or GFX card ... plus they are all copper so no galvanic issues. Downside is Swiftech and EK seem to be giving up on 140mm AIOs reasons I can't understand

 

[JL87]

Yes, I know ... extra knowledge is a good thing .. whether for next purchase or to assist another user

No disrespect meant, but you're cherry picking.

69 for the fuma vs 70 for the d-15 (non-s) on the aida64 oc but on the prime oc test the d-15 is 69 and the fuma is 70, so in that aspect I can't really say it beats the d-15, trades blows/ties at best.

I'll agree it's better price-wise, but performance-wise I'd say they are both equal depending on which chart you want to look at, keeping in mind a 1 degree difference either way would be within margin of error for testing purposes. (It's even stated on the 1st line of the review.)

 

[John Naylor]

Actually I swapped my h100i for the D15, the pump noise was bugging me.

Yes, I know .... See Post No. 2 . I don't see the point in 2 x 120mm AIOs w/ aluminum rads since they don't bring anything to the table, the better air coolers top them thermally and embarass them acoustically. I'm like you ... if i can tell that the PC is on with my ears, Papa ain't happy.

What I was pointing out to Vario is, now that you have the NH-D15 (and again great cooler) ..... next time around, or if you are helping another user down the road, you / he / she might want to consider the Fuma as it is slightly better thermally (1C) and quieter (1 dbA) ... and it's HALF the price. Unfortunately worldwide availability is limited.

No disrespect meant, but you're cherry picking.

69 for the fuma vs 70 for the d-15 (non-s) on the aida64 oc but on the prime oc test the d-15 is 69 and the fuma is 70, so in that aspect I can't really say it beats the d-15, trades blows/ties at best.

I'll agree it's better price-wise, but performance-wise I'd say they are both equal depending on which chart you want to look at, keeping in mind a 1 degree difference either way would be within margin of error for testing purposes. (It's even stated on the 1st line of the review.)

There's no bad choice here, so I don't understand the nitpicking . If it wasn't obvious, the "half the price" thing was THE point . Be that as it may, as to the thermal tests let me ask a question. If you are evaluating SUVs to tow ya new 450 pound Ski-Doo and trailer to the beach 10 miles away , would you put much stock in a test where they towed a 12,000 pound trailer up and over the Rocky Mountains ? It's like evaluating an particular SSD because it makes 200 zip files faster or opens 150 tabs in Chrome faster than another under consideration. I just don't see any value in tests that test "using loads that the CPU will never, ever be exposed to." To my view of the tests available, I have always felt that AIDA provides the closest fit to real world usage. I find the Prime and FPU tests of no interest or applicability, but hey that's me. I think a drop test for a phone to my view shouldn't include dropping it out a 3rd story window .

But since there were two tests, let's call it a tie .... Even had the Noc won both tests by 1C, since the *stated* concern was noise, ... that is the tiebreaker. Aesthetics, availability and fit also come into play so I'd be hard pressed to argue against anyone choosing the Noc or the Cryorig as the differences are small ... but the price isn't and with an 800 pound gorilla laying on the couch, I think we can ignore the mouse scrambling across the floor.

What I think is the main point here is that there is a common mindset that AIOs provide better thermals and lower noise levels and the test dats is in direct conflict with that view.

 

[HUSKIE]

@HUSKIE has a full setup going for cheap in the BST

I would add another fan for a few more c and just "cause"... it looks better

If someone offer me a reasonable price I can accept

 

[NdMk2o1o]

If someone offer me a reasonable price I can accept

I'd be all over it if I wasn't so poor myself just thought I'd throw your name out there as you haven't had any takers as of yet

 

[HUSKIE]

I'd be all over it if I wasn't so poor myself just thought I'd throw your name out there as you haven't had any takers as of yet

If they're still sitting on my shelf till weekend I'm taking out of sale and buried them on eBay...

 

[JL87]

There's no bad choice here, so I don't understand the nitpicking . If it wasn't obvious, the "half the price" thing was THE point . Be that as it may, as to the thermal tests let me ask a question. If you are evaluating SUVs to tow ya new 450 pound Ski-Doo and trailer to the beach 10 miles away , would you put much stock in a test where they towed a 12,000 pound trailer up and over the Rocky Mountains ? It's like evaluating an particular SSD because it makes 200 zip files faster or opens 150 tabs in Chrome faster than another under consideration. I just don't see any value in tests that test "using loads that the CPU will never, ever be exposed to." To my view of the tests available, I have always felt that AIDA provides the closest fit to real world usage. I find the Prime and FPU tests of no interest or applicability, but hey that's me. I think a drop test for a phone to my view shouldn't include dropping it out a 3rd story window .

But since there were two tests, let's call it a tie .... Even had the Noc won both tests by 1C, since the *stated* concern was noise, ... that is the tiebreaker. Aesthetics, availability and fit also come into play so I'd be hard pressed to argue against anyone choosing the Noc or the Cryorig as the differences are small ... but the price isn't and with an 800 pound gorilla laying on the couch, I think we can ignore the mouse scrambling across the floor.

What I think is the main point here is that there is a common mindset that AIOs provide better thermals and lower noise levels and the test dats is in direct conflict with that view.

Kind of went all off point here...lol

Again I don't mean any disrespect, but on more than one post you have pointed out that the Fuma beats the d-15 by ONE degree, but only by referencing one of the two tests. So saying "just" the price was the point, when you allude to thermals, acoustics and price is not all of the story.

Price-wise it does beat the d-15, so yes, let's leave that out of the equation. There is no argument here.

But as you have stated in many posts, you point out it beating it by 1 degree and in acoustics which at 25% and 50% it's 1-2 dBA quieter (at 100% both are the same), price which I previously conceded the Fuma wins. I'd be very hard pressed for anyone to be able to distinguish the 32 dBA and 34 dBA in a real world environment.

So ok, I'll concede the price and 2 dBA difference in acoustics, but for the actual performance both coolers are equal if you include the fact that 1 degree is within the tolerance of error for thermal testing.

 

[Tomgang]

Switched from a H100i Gtx aio to the noctua NH-D15, pretty impressed so far. I thought the H100 was good, but the pump was too noisy for me. Got the Am4 mount kit today, better than zip ties.
Idle temp is 28c which is easily as good as the AIO. I will update with the load temps later.

My case has 2 x 140mm on front, 1x 140mm on back and 3 x 1140mm on top. In/out/out respectively

Edit- Load temp while gaming is 56c so that is good enough for me. I don't run stress tests anymore, unless testing for OC stability.

I will be hornestly and say from the start that i am not a fan of water cooling, as there are more that can go wrong with water cooling. like pump failure, leaks and cooling head can clug up and water needs more maintenance than air cooling does. But yeah a Noctua NH-D15 is a very nice cooler. Its big but I works pretty good. You can get even lower temps if you mount a fan on the front of the cooler as well and as a option one on the rear as well, but a rear fan will only gain you like 2-4 degress depending on what fan you use.

I have the older Noctua NH-D14 and I am now fairly plessed with it after I made some changes. It can now keep my overclock I7 980X that is overclock to 4.4 GHz and pump with 1.42 volts (stock is 3.33 GHz all throw when turbo kicks in, it can go to 3.46 GHz on all cores) with in safe temp now. What I dit is to replace my old fans that whas 3 x Silverstone FM121 120 MM (capable of up to 2400 RPM) fans with 3 x Noctua IPPC fans 3000 RPM high static pressure models and replace the stock cooling paste with my new favoridt Thermal Grizzly Kryonaut. That lowered temp at the same overclock with around 8 degress celsius, but with a little noise penalty as 3000 RPM fans is not silent but I can live with that (the old fans at 2400 RPM whas not silent either and had a more annoying sound to them). Temp would off cause be way lower if the cpu ran stock, in fact I cut al most passiv cool the CPU at stock.

The old fans setup and the paste that came with the cooler.

The new fans setup, here i have also made a bit of ghetto mod so i cut mount some fans on the chipsæt.

The old cooling setup temp a max load.

Temp with the new setup. Ilde and max load. ilde CPU and case fans runs 800 RPM while at max load the cpu fans runs around 2900 RPM and case fans 2300 RPM.

 

[freeagent]

Is the fan setup ok for a air cooler?

I still think AIO's are not that bad really, but i guess there is a chance of failure, and my PC is on all day whether i am in or out. It was the pump noise that decided it for me, my fans where quiet but the pump noise was not enough.

Looks good to me, I bet if you didn't oc the cpu you could run that cooler passive, she's got some girth I like the fan that way, nice easy access to your ram if you like to play with different sets. I like the clean looks of an AIO. I never did have a leak, or pump problems with my H100 that I beat on for five years lol. I also didn't use the stock fans. Its what I replaced my D14 with. Back then, my pc was always on, no sleep, powered off for maintenance.

I'd be all over it if I wasn't so poor myself just thought I'd throw your name out there as you haven't had any takers as of yet

If you have space for a tall cooler, a True Spirit 140 Power will cool a little better than Le Grand macho RT which is as good, or better than a D15, for fifty bucks. Then add a TY-143 later to secure the win

I've heard a lot of good things about Scythe, but have never used them, mainly due to availability.

The NH-U12A is interesting, but at 130 CAD from the egg.. its not that interesting.