be quiet! Dark Rock Pro 4 CPU Air Cooler Review @ [H]

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,532
be quiet! Dark Rock Pro 4 CPU Air Cooler Review

The Dark Rock Pro 4 promises two things right up front, "NO COMPROMISE SILENCE AND PERFORMANCE." be quiet! even says it in all caps, so it must mean business. Other marketing speak is "Silent Wings," "Funnel-shaped," "high-performance," and "easily installable." We will put it all to the test on our highly over-volted Ryzen 7.
 
  • Like
Reactions: Aenra
like this
What is more surprising to me is how good the Corsair h100i does compared to the h150 Pro

Great review, thanks..
 
Last edited:
I find it interesting that going dual towers doesn't give a noticeable improvement over the Reeven Ouranos or Noctua NH-U14S.
 
I find it interesting that going dual towers doesn't give a noticeable improvement over the Reeven Ouranos or Noctua NH-U14S.
Ditto - I wonder if the gap in the center allows for more warm air to escape the path of the fans' pressure and recirculate inside the case, rather than exit smoothly out the back of the case following the pressure path. It seems ludicrous that adding ~10% more mass and a second fan (compared to the Ouranos) nets such a small temperature change - I guess BeQuiet! decided to take the 2db instead, which is a good play if you ask me.

That $/degree graph is nutso for Reeven, though. Like, to an almost fishy degree. Am I really going to pay $90 for the DeRP-4 instead of $55 for the Ouranos for the added pleasure of having to source low profile RAM? I don't think the ~1-2C and ~2db is worth it (although I could be convinced on the noise).

I imagine that I could buy two new quiet PWM fans for the Ouranos (because who wants a yellow fan?), spend about the same $90 and have the best of everything - cooling performance, low noise, any DIMMs I want, and a color scheme that is no longer shit.

Question for you, Mr. Gouge - what would you think of doing a periodic (like, once per year, say?) article where you took the top 2 air coolers, replaced the included fans with [insert poll result] and re-test?
 
Last edited:
^ It's kind of why Noctua no longer offers the NH-D14, which also had barely any gains over the NH-U14S
 
I thought the D14 was simply replaced by the D15?

What is more surprising to me is how good the Corsair h100i does compared to the h150 Pro

Great review, thanks..

The 150 pro is limited by the low RPM fans, I believe they top out at 1050 rpm. Yes, it gets cooler than the larger 150, but it's also 10dB higher, which is nuts.
 
Last edited:
Question for you, Mr. Gouge - what would you think of doing a periodic (like, once per year, say?) article where you took the top 2 air coolers, replaced the included fans with [insert poll result] and re-test?

I'd be down for that, I could take say the Ouranos and test it with some other fan options and see how it changes, would be difficult on the be quiet! stuff though, as the non standard fan sizes with the odd mounting holes (they're in the center of the fan and not on the faces, the fans sold by be quiet! seem to have a little spacer there to mount in a standard way if you look) would make it difficult to install a different fan on them. Noted that in the Dark Rock 4 review as we thought it may be the fan limiting performance to an extent, and I was going to re-test with a Noctua fan, no go without some zip tie jerry rigging.

(because who wants a yellow fan?)

My personal rig lol

IMG_20180330_155130.jpg

The 150 pro is limited by the low RPM fans, I believe they top out at 1050 rpm. Yes, it gets cooler than the larger 150, but it's also 10dB higher, which is nuts.

You are 100% correct. The 150i noise levels were not unpleasant at 100% with my face 2' from the side of the case. The 100i at 100% could be heard over 20' away
 
"can be done with minimal profanity" - perfect. Nice review, liked the sound comparison stats.
 
You are 100% correct. The 150i noise levels were not unpleasant at 100% with my face 2' from the side of the case. The 100i at 100% could be heard over 20' away

It's why I picked up the 150 instead of the 115. Yes, I could tune the fans myself, but I never want them audible at all, so...
 
I thought the D14 was simply replaced by the D15?



The 150 pro is limited by the low RPM fans, I believe they top out at 1050 rpm. Yes, it gets cooler than the larger 150, but it's also 10dB higher, which is nuts.
I thought the 150 has the Maglev fans, mine spin to 1800 I believe. But if true I agree with you.
 
Good review, also congrats to the author on his porn actor-sounding name; "Dick Gouge".
 
This result seems reasonable. The Dark Rock 4 result still seems sort of off though , probably was a bad sample (uneven base pictured in the review). I would expect it to be at least NH-U12S level for the combined GPU+CPU testing (i.e. at worst ~ 85-86°C for CPU only and 88-90°C for the combined testing).

I'm wondering what is a "very high wattage load"?

I've run my NH-U14S open bench with Prime95 AVX enabled and it peaks around 80°C with the average ~75°C (T_DIE) with a 30°C ambient and that's at ~1350RPM. The Ryzen 7 1700x was pushing 3466C14 tight timings , 3.9GHz all core with VCore SVI2 TFN ~1.35V , that's around 180-190W to the CPU+SOC per hwinfo64.

In my prior testing on the X370 Gaming 5, the V_Droop is higher than on my new X470 G7. The X370 G5 has about 0.02V V_Droop under Turbo LLC.

You have to remember that there is the ITE8686 and ITE8792 on the GBT X370 G5. The sensors sometimes go crazy because there's two chips with readouts.

edit: Is there any way you could return the Dark Rock 4 and get a new sample?
 
Last edited:
I find it interesting that going dual towers doesn't give a noticeable improvement over the Reeven Ouranos or Noctua NH-U14S.
It's quieter than those two while being a bit better. Also dual towers can handle higher TDP CPU better than single ones. I have tested a D15 and Macho Rev B. Both coolers performed similarly on the 8700k, but as soon as you jump to a 7980XE, the D15 is in its own league (about 10+ degree difference). I guess you can say the Dark Rock Pro 4 is "TDP bottlenecked" lol.
 
It makes sense that the Dark Rock 4 would fail then. It's rated for 200W.
 
It's quieter than those two while being a bit better. Also dual towers can handle higher TDP CPU better than single ones. I have tested a D15 and Macho Rev B. Both coolers performed similarly on the 8700k, but as soon as you jump to a 7980XE, the D15 is in its own league (about 10+ degree difference). I guess you can say the Dark Rock Pro 4 is "TDP bottlenecked" lol.

Yup the dual towers don't start separating until the much higher TDPs. Anandtech had a low fan speed portion for their 140mm heatsink Roundup in 2017 that really highlights this - particularly the 250W and 340W loads.
 
...would be difficult on the be quiet! stuff though, as the non standard fan sizes with the odd mounting holes...no go without some zip tie jerry rigging.
Didn't even catch that about the non-standard fans. That seems really, really dumb.

I don't see any yellow fan blades in that dumpster fire of a rig, so I guess you made my point?

/s
 
Boss if you see this, i haz question; the included fans, are they 3pin, 4pin, both or one each please?

(if you don't mind)
 
Boss if you see this, i haz question; the included fans, are they 3pin, 4pin, both or one each please?

(if you don't mind)

They're both PWM fans. 4 pin.
I believe 99% that entropism is correct, and looking at the picture of the connectors, seems to confirm that. I will have rgMekanic note that in the next review as that is very relevant.
 
  • Like
Reactions: Aenra
like this
I had a Be Quiet cooler for a while, the C-type one I can't remember the name of it. Worked well enough but the black finish was chipping off all over the place with just a couple of mounting and looked really bad. I gave it away eventually and now if I want air I use my trusty NH-U12S with two fans. No RAM clearance issues and its easy to install.
 
I wasn't wearing my glasses, lol...

Thanks to you both guys :p
 
Would be interesting to see actual cooler performance instead of how test system performs with different cooling systems. Your test procedure says you have thermocouple 1 monitoring air temp into cooler / radiator. It would be nice if you included those air temps / those delta temps so we could see how well one cooler performs against another instead of your test system's performs with different coolers.

I suspect AIO performance would not took to better better than air coolers if we had the cooler/radiator intake air temp deltas.
 
Back
Top