• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

ASUS GeForce RTX 3070 Noctua OC

Joined
May 2, 2017
Messages
7,762 (3.05/day)
Location
Back in Norway
System Name Hotbox
Processor AMD Ryzen 7 5800X, 110/95/110, PBO +150Mhz, CO -7,-7,-20(x6),
Motherboard ASRock Phantom Gaming B550 ITX/ax
Cooling LOBO + Laing DDC 1T Plus PWM + Corsair XR5 280mm + 2x Arctic P14
Memory 32GB G.Skill FlareX 3200c14 @3800c15
Video Card(s) PowerColor Radeon 6900XT Liquid Devil Ultimate, UC@2250MHz max @~200W
Storage 2TB Adata SX8200 Pro
Display(s) Dell U2711 main, AOC 24P2C secondary
Case SSUPD Meshlicious
Audio Device(s) Optoma Nuforce μDAC 3
Power Supply Corsair SF750 Platinum
Mouse Logitech G603
Keyboard Keychron K3/Cooler Master MasterKeys Pro M w/DSA profile caps
Software Windows 10 Pro
My take is that it's more down to the mass and thermal transfer capacity of the heatsink. IME, fan speed doesn't contribute much above a certain threshold. (Granted, these experiences aren't with high-end parts.) For instance, on a typical CPU single-tower cooler with a fan that runs 700-2500 RPM, temps under load don't drop appreciably above about 1500 RPM. On the modest dual-fan GPUs I've run, that value can be more like 1800-2000. But I've never had a scenario where running the F on an HSF at full tilt has produced anything other than extra noise. You're just chasing the diminishing returns at the far/wrong end of the efficiency curve at that point.
But that's the thing, a thinner fan necessarily needs to run faster to achieve the same static pressure as it simply doesn't have the room for the blade geometry to create that pressure otherwise. And the thicker your heatsink (assuming the same fin density), the more static pressure is needed to make air flow through it effectively, so a bigger heatsink with a too-thin fan risks being no improvement at all. The reasons why differences aren't massive is simply because most fans these days are quite okay, but assuming similarly competent designs and some flow restriction, a thicker fan will always provide more airflow, allowing for lower speed. Transfer capacity of the heatsink doesn't matter if there is no airflow to remove the heat, after all. You're right about diminising returns though - the balance between heat source temperature, heat source thermal energy output, heat sink transfer efficiency, heat sink area, and airflow through the heatsink gets pretty muddled once you get a decent amount of flow going, as at some point you start cooling the heat sink (and thus also source) to such a degree that transfer becomes somewhat less efficient - a higher thermal delta makes for easier thermal transfer, after all. So, ironically, having too good a cooler makes your cooler worse, in a way. This is also why huge honking air coolers generally don't massively outperform smaller ones at low heat loads - there simply isn't enough energy to make them usefully hot, allowing the smaller heatsinks to dissipate heat more efficiently.
2,500 regular gpu fans I mean. (80 or 92mm?) . Yes they are 'loud' but again once you adjust its all the same. Noise should not be such a big deal on gpus as most of the time they'll only be spinning once you're gaming and by then you have gun blasts in your ears anyway.
People throw around the word 'silent' too easily, or like in this review, you have to put your ear directly to it to hear it. That is bogus. I tried this years ago to build a silent pc and wondered why a 'silent' gpu was making noise I could easily hear.
Someone said its silent because '23dB is well below most people’s noise floors.' Thats like when people used to say the eye can only see 60 hz yet now suddenly everyone wants 144hz screens.
You're right about the '23dB is well below most people's noise floors' thing. Noise adds up - it's all vibration energy in the air, after all - the question if it adds enough noise to be a noticeable increase. The log scale of dB(A) explains why a low number might not be audible, as 23dB might simply not be enough of an addition if your noise level is already sufficiently high. If your noise floor is 41dB then those 23dB represent something like a 1,6% increase in sound energy, while over a 32dB noise floor it's a 12,5% increase - much more likely to be noticeable. Of course that's a gross simplification that ignores the frequency spectrum entirely, but it goes some way towards explaining that misconception still. If there's lots of noise already, a little more isn't likely to be noticeable, and if it's very quiet, even the quietest fan will be audible, as it's adding some noise.

In my case, I haven't had an air cooled GPU in my main PC since 2015, so I'm a bit out of touch (I do have a Sapphire RX 570 ITX in a secondary/travel PC though, but that PC is pretty noisy for other reasons), but to me any fan running above ~1500rpm starts getting audible-bordering-on-annoying - but that also depends a lot on the noise profile. My two Arctic P14 intake fans are very clearly audible at 100% (~1550rpm measured in Aquasuite), but not annoying at all thanks to their low-pitched and spread-out sound signature. They're on a level where they never bother me despite my use of open-backed headphones. I also have just one fan running at idle (typically around 900-1000rpm), which makes the PC inaudible over outside traffic or wind noise from the vents in my home office - it blends perfectly into the background noise of the room. I used to have Gentle Typhoons on my previous 240mm radiator, but they were annoyingly grindy at idle, which got on my nerves over time. Switched to some Silent Wings 3s, which were nicely quiet, but didn't deliver on performance (1400rpm 120mm fans just wasn't enough). These P14s are noisier at full tilt, but the pleasant frequency spread makes that a non-issue.

As for 2500rpm fans ... well, the NH-L9i in that travel PC gets plenty loud, and that's one of the quieter slim 92mm fans on the market. There might be GPU fans of the same diameter that make less noise, but they would also push less air, which would mean an NF-A9x14 in the same place could most likely run quieter while delivering the same performance. Either way, no 2500rpm fan is acceptable for long-term use for me - I could never get used to that type of noise, and I can say from experience that those kinds of droning noises makes tinnitus much, much worse. Mine is typically unnoticeable, but long-term exposure to loud fans makes it quite annoying.
 
Top