Dark Rock 4 owners - i need re-assurance.

Danny75 said:
I've managed to overtighten a Noctua NH-D15S in the past so the system wouldn't start until I loosened the screws a quarter-turn or so. Don't ask me how I managed to fumble that... I guess I was expecting the spring screws to just stop turning beyond a certain point.
Click to expand...
It's screws should have had stoppers from the start.
Could be simply that massive weight being the issue.
 
Cheers @orbitalwalsh - but, that's a Mortar and no m.2 drive installed. The Tomahawk worked with a standard SSD - but i have a nagging suspicion over pressure on the m.2 slot.
 
EsaT said:
It's screws should have had stoppers from the start.
Could be simply that massive weight being the issue.
Click to expand...
Thoughts on my bar over screw fitting theroy, @EsaT?

Couldn't build in the end- kids then more work - so I have time to take a consensus of opinion...
 
Plec said:
Thinking of buying the Noctua NH-U12S - i can only get the Cromax for next day delivery so £10 more for a cooler you're never going to see :/
Click to expand...
NH-U12S has become more than little overpriced since its release compared to competition, like cheaper Mugen 5.
That "baby" is indeed fitting description for its size comapred to what coolers you can get for same money/cheaper.

Also its NF-F12 fan is one of the crappiest I've touched in 25 years of PC hobby.
At least precedessor's NF-P12 fan had OK nice sound signature and only vibrating roller motor, which I considered as bad individual/sample at the time.
Until NF-F12 had equally bad vibration level and also added restless sound profile on top of that.
 
doyll said:
Spring pressure on IHS from base of cooler is determined by design, same as mounts without springs. All of the top tier cooler companies design their mounts to applie the right pressure. It doesn't matter if they have springs or not, mounting pressure is most important part of mount design. The shorter the spring is compressed, the higher it's rate of pressure right up to when screw bottoms out and stops turning, same as tightening mount with no spring. Both are designed to apply same amount of pressure when screws are tight. Hope that makes sense.
Click to expand...
They definitely aren't comparable.
Springs adds lot more tolerance into keeping pressure optimal.
Without springs there's only minimal tolerance between loose mounting and very tight.

For example Scythe changed to spring loaded mounting after those cases of Skylakes bending under some coolers.
https://www.pcgamer.com/intel-skylake-cpus-are-bending-under-the-pressure-of-some-coolers/
https://www.tomshardware.com/news/skylake-cpus-damaged-by-coolers,30690.html
 
EsaT said:
NH-U12S has become more than little overpriced since its release compared to competition, like cheaper Mugen 5.
That "baby" is indeed fitting description for its size comapred to what coolers you can get for same money/cheaper.

Also its NF-F12 fan is one of the crappiest I've touched in 25 years of PC hobby.
At least precedessor's NF-P12 fan had OK nice sound signature and only vibrating roller motor, which I considered as bad individual/sample at the time.
Until NF-F12 had equally bad vibration level and also added restless sound profile on top of that.
Click to expand...
Agree (i'm not happy) - but looking for quick/temporary solution to test socket with a fitting i can trust for another (stock levels disappearing - can't source a Mugen from reputable place - will look again later tonight.

Thoughts on my bar vs screw theory?
 
EsaT said:
OcUK still has two Mugen 5s in stock, in RGB version which is still lot cheaper than NH-U12S.
https://www.overclockers.co.uk/scyt...-cpu-cooler-with-pwm-fan-120mm-hs-04c-sy.html
Without RGB wires attached it wouldn't cause any "light pollution" and under that plastic cap it's standard Mugen 5.
Click to expand...
Price not really an issue (within reason) - good shout re unicorn vomit version. (i'd never hear the end of it)

EsaT said:
There are also others like Alpenfohn Matterhorn.
Click to expand...
Hadn't thought of that one - i would get a NH-U14S/NH-D15 if i trusted the socket - but need to establish a base. So, i'm going to install the stock cooler tomorrow, fit it into the case and if it works, i may chuck on the DR 4 - or pussy out and order a cooler with a more controlled method of installing.

Thanks, @EsaT.
 
EsaT said:
They definitely aren't comparable.
Springs adds lot more tolerance into keeping pressure optimal.
Without springs there's only minimal tolerance between loose mounting and very tight.

For example Scythe changed to spring loaded mounting after those cases of Skylakes bending under some coolers.
https://www.pcgamer.com/intel-skylake-cpus-are-bending-under-the-pressure-of-some-coolers/
https://www.tomshardware.com/news/skylake-cpus-damaged-by-coolers,30690.html
Click to expand...
They both do the same job. Like I said, it's mount design. Some are designed to with springs, some are not. Both accomplish the same goal of mounting cooler with correct seating pressure.

As for Scythe changing an the Skylake bending problem, that whole thing about Skylake was nothing but a media grab to get more viewers with no real supporting data to support the claim Skylake had weaker PCB. There is a video test showing a case built system boxed, dropped, even thrown into ceiling and falling onto concrete floor while running with only problem was loss of image on monitor because cable came loose. But it did seriously damage case and PSU housing among other things. ;) I can't remember who did it or would post link.

The pressure difference in the springs like Noctua uses is not much. The springs look neat but the bigger shaft of screw through spring stops screw so spring has same tension as mount without springs have when their screws are tight .. end result is same amount of mounting pressure.

I've done loads of cooler testing on on i7 6700k with many brands of coolers and have had no problems at all with cooler mounts causing any flex issues. This includes some of the older no spring Scythe cooler mounts.
 
Last edited:
Useful, info - on my phone now heading back home. Knackered.

Decided to mount stock cooler and test tomorrow. If all is tickety-boo then I may put the previously installed DR4 back on, as mentioned.

@orbitalwalsh, searched out some similar builds, using a NVMe coupled with a DR4 on a Tomahawk so there's proof out there that combo works (thanks again, bud). I may give it a day of testing with stock and then apply a DR4 that's been a staple for a lot of builds for a couple of years. Had the motherboard fault not been so ambiguous I wouldn't be trying to second guess the build in-between down time.

Just need to find time tomorrow to build and put theories to test...
 
orbitalwalsh said:
Get hold of bequiet and see about quality control of the unit you have . Will see if I can dig out that departments email if you can't find it
Click to expand...
Possibly, but suspect i will get a generic reply that they're happy with the design of their units. Plus, the 2 units i have look fine and we all seem to be in agreement that the gap, before tightening, between bar and retention bracket seems standard - take a look ----> https://imgur.com/75LJZvl

I've stripped it all down - and packed up suspect board.

Now back to square one again - fresh board unpacked, CPU in - think i'll install stock cooler - get it up and running, hopefully, and decide from a position of relative strength...
 
Grabbing an hour - quickly assembled and inside the case (no point breadboard as cleared all other components) and testing with stock cooler.

Switched on, setup BIOS - and now beginning install process and then driver install...

*Christ, the stock cooler is noisy!
 
This is like the silicon version of groundhog day hell...

Starting on driver install - this is where it craps out, if it's got a problem...
 
Installed:
  • Chipset
  • GPU drivers
  • Lan
  • Audio
  • Core Temp
  • Cinebench
  • Aida64
And about 10 restarts - never got this far before installed in the case so the other motherboard definitely had issues regardless if this develops some weird alternative neurosis.

Looking good, so far, no flashes, strange delayed starts - although early days the original board would have corrupted the drive by now.

EDIT: Slay the Spire works - hardly the most taxing game, but another successful restart. Cinebench score was 200 lower than with DR4. All restarts OK, so far. Going to install FarCry 5 and run benchmark for comparison plus I may get #1 son to play it for an hour as this is better than synthetic stress testing. Then perhaps Arma 3 - but will take hours to DL with mods - but worthy test.
 
Last edited:
doyll said:
Good news! Thanks for letting us know it's working .. at least so far .. finger still crossed for you.
Click to expand...
Thanks, @doyll.

orbitalwalsh said:
Still working ?
Click to expand...
It is thanks, bud - but with stock cooler :/ (You can imagine my torment.)

Still, seems OK and very obvious that it doesn't have the previous neurotic issues. Not played anything in anger yet - but may get the chance tomorrow - only crunched data.

Pretty much decided to avoid the DR4 fitting (may change my mind) and will attach a Mugan 5, (when back in stock) as it's one of the easiest/gentler fittings for a large cooler - more so than Noctua. So, with current events I'm going to play it safe and wait for Mugan 5 to have stock again and then swap out this insanely noisy stock cooler. (Why we advocate the stock cooler instead of cheap 120mm cooler - if funds are there - as a viable option for the sake of ~£20 seems nuts now that I've lived with this for a day. I know i'm fussy with regard to noise - but, honestly, it sucks.)

*If times were different i would strap on the DR4 - but the Mugan is a safer option and i'll hopefully have silent system 'relatively' soon. Maybe not as silent as i would like - but needs must as the devil drives, and i need a stable machine sharpish without the nagging uncertainty of instability occurring in the future (rarely gets used for games).
 
You must log in or register to reply here.
Back
Top Bottom