Actually regression, like 13700 at default and hitting 86C at 100% fan speed.
It was the bloody GT's. I guess their motor is just too big and with the ram height clearance issues it just ended up sucking real bad due to the actual airflow not hitting enough of the heatsink's surface area. You can see the difference here:
View attachment 331434
It's like what? 20% bigger motor on GT?
Switched to stock fans:
PPT/TDP/EDP - R23 score - Temp
100/70/100 - 14400 - 72C
122/82/124 - 14788 - 82C
142/95/140 - 14384 - 82C
Now the middle fan sits flush with the cooler cover while front matches the top of the heatpipes.
@FreedomEclipse @stinger608 - since you asked:
Current setup is listed in specs
Switching from Dark Rock 4 to Phantom Spirit 120 SE:
Temperature wise it's maybe 1 or 2 degree difference between DR4 and PS120 SE under normal operation. Dark rock 4 managed pretty much identical temps while being quieter due to lower max RPM and way quieter fans. With DR4 I kept the fan at 70% (ca. 850 RPM) at idle (below 55C) and 100% at load (70C) anything between that would have the speed adjusted by FanControl based on the Tcore. With PS120 SE I keep it at 50% until 60C and 100% at 70C - I'm still trying to figure out optimal settings to avoid audible ramps. When playing Squad I hover at around 57-62C.
Like I mentioned before when I swapped PS120 SE stock fans to XPG Vento Pro's I saw an increase in temperatures for some odd reason (+4C vs stock fans) - Cinebench R23 multi temps were around 86C with fans blasting at 100%. R23 scores with stock fans are nearly identical to those using DR4, temperature peaks at 81C. Scores are within run to run variance (around 14800 with 122/82/124 power limits).
Idle temps are hovering around 33-35C, nowhere near what other people claim to see but I'm running my case fans at around 600 rpm at the front to keep the system whisper quiet when not gaming. 750 rpm at the back when GPU temp is below 65C, then it ramps to around 930 rpm. Sure I can drop idle temps to 26C but that's with balcony door open and near 0 temps outside but who in their right would do that?
I also tested my MX-5 vs the FT7 paste that came with the cooler but temps were pretty much identical (82C with MX-5 vs 81C with FT7 when running 10 min R23 multi). Sure, it's MX-5 that's over year old but there's no drastic difference so I wouldn't say that it's a matter of paste. It ain't about application either, because for MX-5 I used X with dots between the corners and for FT7 I used the dot in the middle.
One bizarre thing was that it was really hard to take off the damn heatsink when I wanted to repaste. The two screws that are used for mounting simply wouldn't release until I basically pushed on the top end of the mount and only then I was able to unscrew the top one and then follow up with the bottom screw. While DR4 bar setup isn't optimal at least it never had me fighting it when I tried to take it off.
TLDR:
- built quality is OK, fans feel cheap and light but aren't terrible (still, no comparison to stock DR4 fan) - they're whisper quiet up to 50% (around 900 rpm) and are ok at max speed of around 1500 rpm,
- mounting system caused issues on disassembly, a big no-no for me,
- ram clearance is far from optimal, had to raise the fans by about 7 mm in order to clear the Ballistix module sitting in slot 2, DR4 had no such issues and these modules aren't even THAT tall
- if you're already have a decent air cooler don't bother swapping out, gains are minimal in my opinion, especially on 5800X3D with its 3D cache layer sitting between the core and the IHS,
- if you're build new system and are budget oriented it's an OK option if you're trying to save some money.
I might end up selling it at some point and returning to DR4. I would like to give AIO a try but sadly PM01 RGB has bullshit clearance for AIOs on top (the height of rad+fans can't exceed 33 mm in total).