Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
First review about sound... fan noise
- Thread starter Cape Dave
- Start date
- Sort by reaction score
Keep in mind this is all core load 100% without a break. I wonder what scenarios people would have to blast this much load at all times?sounds a bit like the old intel macbooks
somewhere between that and a miniature quiet vacuum.
I have no experience in rendering using the GPU side of it so I am curious to learn more!
I duplicated every track 4 times to sweat the CPU even more. Also threw on a Fab filter limiter and put the most CPU intensive setting on.
The fans on the mini didnt even ramp up. Around 48-50C degrees. Safe to say this one is all good for music production IMO.
Sounds exactly like my old Intel 16” MBP only that it’s a Desktop. Probably will skip on the Mini and go 14” MBP M4 Pro (Base Model) instead . The Mini M4 Pro imo is too expensive for what it is (M4 MBP is 550 more expensive in my country) and the 14” MBP to me is the better value, especially since there seems to be no significant difference in thermals.sounds a bit like the old intel macbooks
somewhere between that and a miniature quiet vacuum.
I may also pass on the new m4 mac mini. I was contemplating the base m4 because I thought it would be perpetually inaudible like the m2 / m2 pro, but that's not true. I may consider the m4 studio when reviews come out once released / early next year. It's a shame though because the overall idle / moderate load energy use of the studio is considerably more than the m4 mini, and I don't really need/want that extra power and heat. I'd gladly buy an m4 in the old m2 mini enclosure or an m4 / m4 pro in a studio enclosure (all things equal in terms of keeping thermals the same). I guess Apple got what they wanted -- segmentation of their desktop market and a possible push of more users to the studio (which has been their most lagging computer in terms of sales).
I do think they will get more folks that are first time desktop buyers or swtich-overs from PC who don't care so much about or aren't into the weeds about fan noise like us -- but that's probably just a good marketing campaign and the price-points for the higher spec base hardware that would have driven people even if they just modestly changed the case design to retain the dB profile of the prior m2 mini
I do think they will get more folks that are first time desktop buyers or swtich-overs from PC who don't care so much about or aren't into the weeds about fan noise like us -- but that's probably just a good marketing campaign and the price-points for the higher spec base hardware that would have driven people even if they just modestly changed the case design to retain the dB profile of the prior m2 mini
Last edited:
I was getting fan noise just by doing a MacPorts install. The port for some reason needed Rust installed, which was then built from scratch. It looks like there were 8 to 10 instances of clang compiling at times, so I would assume that all 10 of the performance cores were being used. What's weird is that installed the "same" port on a couple of other Mac's and don't remember the porting process taking an hour to finish. I'm wondering if this has anything to do with MacOS 15.x.x being more persnickety about executables being downloaded, so needed compilers are built rather than downloaded.
It's really useful to use the fan noise data graph here:
www.notebookcheck.net
I realized that the multiple boxes under the device selection (i.e. comparing m4 mini to m2 studio) are single measurements -- the rightmost measurement of SPL is in fact the dB for one of the tests in the graph below -- so for example if you cross out each of the 5 for the m4 mini and then....
If you go back into the mac studio review:
www.notebookcheck.net
It's harder to decipher which of the 4 measurements correlate to each of the measurements in the table below, but it's clear that almost every one has the same dB spectrum curve with an average SPL / of 25dB meaning that under low, medium, or high/max load the unit stays almost at the same low RPM as idle. This means the unit is extremely effective at dissipating heat and never likely changes the fan rpm / noise curve.
Comparison of m4 mini and m2 studio:
I don't think it's a fair assumption to say "hold out for the m4 studio if you want a silent machine"... it's been mostly silent in all past m2 tests, but it remains to be seen how much more power draw they'll decide is tolerable to get their marketed increase in performance values -- those may very well push the studio into the audible range under load if they don't counter with a design change or if there isn't still more headroom in the current form factor. We are seeing the net wattage continuing to increase from m2 to m4 in performance cores (and GPU cores) under full load, so it's not just a matter of efficiency but also how much they're playing with the thermal envelope.
Apple Mac Mini M4 review - Smaller, faster and louder
Notebookcheck has tested the new Apple Mac Mini with the M4 SoC, 16 GB RAM and a 256-GB SSD.
www.notebookcheck.net
I realized that the multiple boxes under the device selection (i.e. comparing m4 mini to m2 studio) are single measurements -- the rightmost measurement of SPL is in fact the dB for one of the tests in the graph below -- so for example if you cross out each of the 5 for the m4 mini and then....
- Uncross the 2nd, you get the measurement with 30.3dB/SPL which equates to the test under the Cyberpunk 2077 Ultra; so this would be the curve for gaming / GPU load.
- Uncross the 1st one, you see the spectrum curve for max-load (45.1 dB)
- Uncrossed the 4th one, you get the curve for 24.1dB (basically the routine default low RPM of the device before ramping up) -- the third and 5th measurements are also similar.
If you go back into the mac studio review:
Apple Mac Studio 2023 review - The M2 Max also impresses in a desktop device
Apple has updated its desktop computer, the Mac Studio, which is now fitted with the faster M2 Max or M2 Ultra SoCs. We have tested the base model with the M2 Max and the 30-core GPU as well as 32 GB RAM.
www.notebookcheck.net
It's harder to decipher which of the 4 measurements correlate to each of the measurements in the table below, but it's clear that almost every one has the same dB spectrum curve with an average SPL / of 25dB meaning that under low, medium, or high/max load the unit stays almost at the same low RPM as idle. This means the unit is extremely effective at dissipating heat and never likely changes the fan rpm / noise curve.
Comparison of m4 mini and m2 studio:
- Idle/low/average base RPM of m4 mini (4th checkbox) to the m2 studio -- it's about the same.
- GPU / gaming load of m4 mini (2nd checkbox) to the m2 studio, you get a 4.6dB increase in mostly lower and mid-range on the m4 mini.
- Max load of m4 mini (1st checkbox) to m2 studio, you get a 19.4dB increase across the entire spectrum, but weighted more toward mid frequency.
I don't think it's a fair assumption to say "hold out for the m4 studio if you want a silent machine"... it's been mostly silent in all past m2 tests, but it remains to be seen how much more power draw they'll decide is tolerable to get their marketed increase in performance values -- those may very well push the studio into the audible range under load if they don't counter with a design change or if there isn't still more headroom in the current form factor. We are seeing the net wattage continuing to increase from m2 to m4 in performance cores (and GPU cores) under full load, so it's not just a matter of efficiency but also how much they're playing with the thermal envelope.
Last edited:
It continues to suggest that not all M4s have the same fan/sound performance. This is getting to be fairly mysterious.
@pastrychef, what's the background noise like in your environment? Would you be able to hear soft/medium loud fan noise?
Thats exactly my impression as well.
Totally !
I´ve set yesterday my M4pro to / energy setting / High power (and forgot about it)
this evening, the fans started sometimes to spin full steam. 4350 RPM or so. Just for very short moments. just 1-2 seconds. The noisefloor reminded me on intel-MBPs, or my old Mini from 2012 IF that would spin like a jet.
The sound slightkly different. more mid rangy ( the frequenzy)
I don´rmemember the CPU temperature, but it´s usually around 72-73° under "my" full load.
my main app runs single threaded. And i´m even not excatly on very high loads within that.
This all is totally misterious from what i see.
Two camps (it seems)
Both camps not just single persons, but several.
What can cause that ?
i mean, it´s not as if the fan itself was crap, and we´d get a crappy whiny noise on some units.
it´s much related to the airstream, respectivly the RPM itself.
i´m not happy. I want a Studio Max
Normally, it's far from "silent", but when did the test, I did try to quiet down the room but shutting the window (which is where most noise comes from). I've had other computers I have heard the fans (HP EliteDesk 800 G4 DM, HP ProDesk 400 G6).
I should also add that I sit approximately 4 feet away from the mini. When idling, I don't hear it even when I put my ear within about 4 inches away.
Also, my mini was made in Vietnam, if that makes any difference...
My bet is that the thermal paste between the die and the fan heatsink has different amounts or not properly applied. From the teardown videos it looks like way to little thermal paste. It is only 3 small dots where as I would argue an even spread of paste would be more efficient transferring heat away from the die.
Am waiting for Mac Mini to arrive. In meantime had an experience with my iMac Pro which may be relevant. The machine was always totally silent until this year. Thought it was long in the tooth.
Turns out the issue was related to iCloud and Onedrive. iCloud had an upload bug, and Onedrive wasn't getting indexed properly by Alfred. Also various sync issues relating to files being stored in the cloud and not locally.
Fans were routinely 1300-2500 RPM and audible even at the low level. Even fairly light workloads would make the fans spin up. Anyway, diagnosed issued and sorted it, fans are now constantly at 1100 RPM and inaudible. Hard to make the fans go through normal use.
All this to say is that when someone does testing we can't know what their local machine setup is like. An aggravating factor like a persistent indexing issue might not cause much noise on its own, but might leave the machine in a state where it is much easier to overheat.
My Mac Mini M4 Pro (12-core, 48 GB RAM, 1 TB SSD) has arrived.
I installed my audio system as usual: Cubase 14 Pro, around 700 plugins (UAD and UADx, Fabfilter, Softube, Plugin Alliance, Sonox, iZotope, BozDigital, Native Instruments, Kazrog, Overloud, Slate, SIR, and many more).
For testing, I opened a complex mix with 62 audio tracks (48 kHz, 24 bit, 64 Bit processing precision), virtual instruments (Toontrack EZ Drummer 3, Kontakt with Symphony 2, etc.), and around 125 CPU-intensive plugins (not a single stock plugin). This mix pushed my MacBook Pro M1 to its limits, and I had to freeze 5 tracks there.
On the Mac Mini M4 Pro, I don’t need to freeze anything, and the song runs smoothly with a CPU load of just 50%.
I let the mix loop while simultaneously running Mail, Safari with several tabs, banking software, and additional installations in the background (Spotlight indexing and iCloud synchronization were also running in the background).
The fan remained completely silent at a constant 1000 RPM, and the CPU cores stayed at a maximum of 80 to 85°C. Only when I ran several Ozone 11 AI tasks in parallel did one CPU core hit 86°C, causing the fan to speed up to nearly 1300 RPM. Even then, it remained silent — the fan in the Apple Studio Display, which is already very quiet, is louder in comparison.
The Mac Mini M4 Pro, by the way, feels cool during everyday tasks and warms up slightly while mixing a song.
Despite the demanding Cubase mix and multitasking tasks, I couldn’t get the Mac Mini M4 Pro fan to exceed 1300 RPM, and I never heard it once.
I will, of course, continue testing, but if this performance holds, I believe this is an excellent machine for music studios and music production. The many GPU cores of a Mac Studio are entirely unnecessary for DAW setups — even among the 16 GPU cores, I’m only using 12% according to iStatistica Pro.
As a noise-sensitive music producer, I wouldn’t worry about fan noise. Ideally, you should test your setup directly. If any issues arise, you can return the Mac Mini. Benchmark tests don’t reflect a user’s actual use case — if you’re maxing out your computer, as I said before, you’ve chosen the wrong one.
And yes, there may be users for whom it doesn’t work. It could be due to a migration (I did a clean install), background processes not completed, incompatible software versions, variability in fan quality, the wrong machine choice for GPU-intensive tasks, or something else entirely.
It’s also worth remembering that forums are usually filled with people experiencing issues — the “lucky ones” are busy working on their next song.
I installed my audio system as usual: Cubase 14 Pro, around 700 plugins (UAD and UADx, Fabfilter, Softube, Plugin Alliance, Sonox, iZotope, BozDigital, Native Instruments, Kazrog, Overloud, Slate, SIR, and many more).
For testing, I opened a complex mix with 62 audio tracks (48 kHz, 24 bit, 64 Bit processing precision), virtual instruments (Toontrack EZ Drummer 3, Kontakt with Symphony 2, etc.), and around 125 CPU-intensive plugins (not a single stock plugin). This mix pushed my MacBook Pro M1 to its limits, and I had to freeze 5 tracks there.
On the Mac Mini M4 Pro, I don’t need to freeze anything, and the song runs smoothly with a CPU load of just 50%.
I let the mix loop while simultaneously running Mail, Safari with several tabs, banking software, and additional installations in the background (Spotlight indexing and iCloud synchronization were also running in the background).
The fan remained completely silent at a constant 1000 RPM, and the CPU cores stayed at a maximum of 80 to 85°C. Only when I ran several Ozone 11 AI tasks in parallel did one CPU core hit 86°C, causing the fan to speed up to nearly 1300 RPM. Even then, it remained silent — the fan in the Apple Studio Display, which is already very quiet, is louder in comparison.
The Mac Mini M4 Pro, by the way, feels cool during everyday tasks and warms up slightly while mixing a song.
Despite the demanding Cubase mix and multitasking tasks, I couldn’t get the Mac Mini M4 Pro fan to exceed 1300 RPM, and I never heard it once.
I will, of course, continue testing, but if this performance holds, I believe this is an excellent machine for music studios and music production. The many GPU cores of a Mac Studio are entirely unnecessary for DAW setups — even among the 16 GPU cores, I’m only using 12% according to iStatistica Pro.
As a noise-sensitive music producer, I wouldn’t worry about fan noise. Ideally, you should test your setup directly. If any issues arise, you can return the Mac Mini. Benchmark tests don’t reflect a user’s actual use case — if you’re maxing out your computer, as I said before, you’ve chosen the wrong one.
And yes, there may be users for whom it doesn’t work. It could be due to a migration (I did a clean install), background processes not completed, incompatible software versions, variability in fan quality, the wrong machine choice for GPU-intensive tasks, or something else entirely.
It’s also worth remembering that forums are usually filled with people experiencing issues — the “lucky ones” are busy working on their next song.
Thats nice.
To me the issue happens when GPU is under load. CPU is usually fine.
I've returned my Pro and got the base M4. Much quieter and I can make it work. Will see what Apple gives us in 2 years or so. For now, I've decided to save money and get the quieter machine (for me).
To me the issue happens when GPU is under load. CPU is usually fine.
I've returned my Pro and got the base M4. Much quieter and I can make it work. Will see what Apple gives us in 2 years or so. For now, I've decided to save money and get the quieter machine (for me).
I took a look at this review, but unfortunately it doesn't really help me. I couldn't find any mention of how far away they were measuring or what the noise floor was, so I have no way to compare my own measurements (from my 2018). They also don't have a review for the 2018, so I can't use that to compare.It's really useful to use the fan noise data graph here:
Apple Mac Mini M4 review - Smaller, faster and louder
Notebookcheck has tested the new Apple Mac Mini with the M4 SoC, 16 GB RAM and a 256-GB SSD.
Another topic: for those who have a 2024 mini, is it possible to control the fan speed using Macs fan control? If that's possible then the fan noise is no big deal to me, I always used to just limit it on my older minis when I needed to do an encode or whatever.
My Mini is an M4 Pro, with 14 CPU cores, 20GPU, 64 GB and 10G Ethernet, i.e.fully loaded except for the 1TB storage.
I did get the fan going with a MacPort install when it was building the Rust compiler needed to install the requested port, but the Mini has been quiet otherwise. I did feel the top of the Mini when the fan was going, noticeably warm but not uncomfortably hot.
As for fan noise in general, I haven't heard anything as loud as a fan running on 400Hz/3 phase.
Last edited:
Sure it helps. If you're serious about this, it means that you can set up a noise source that will sound much like the Mini, so you can see if it's tolerable or not. But that's too much work, when you can just buy the Mini and see for yourself. Returns until January 8th...
This has already been discussed multiple times in this thread. Yes, you can. There is a caveat mentioned (the current version can't force the fan on if it's off) which won't affect your use case at all.
I've been pushing my M4 Mac Mini Pro pretty hard with 4k video editing/exporting and videogames. I have yet to hear any fan noise whatsoever.
If anyone's worried, a good option would be to buy direct from apple and if you experience unacceptable fan noise with your particular use case, just return it.
If anyone's worried, a good option would be to buy direct from apple and if you experience unacceptable fan noise with your particular use case, just return it.
I was leaning towards the Pro for a while but pulled the trigger over the weekend on the base M4. No matter what it will be faster than my M1 16GB/1TB MBP. With the 32GB/512GB variant I ended up saving almost $1000 compared to the M4 Pro 48GB/512GB.
Maybe Apple will come out with a new 5K/120hz Studio Display alongside the M4 Mac Studio and I'll regret no Thunderbolt 5. Or maybe they won't and Thunderbolt 4 will still be as fast as I need it to be.
If they come up with 5K/120hz Studio Display I would assume that they will not limit it to only thunderbolt 5 and will find a way for others to use. It will have updated AS inside so that might do some magic also. We shall see 
Thanks, sorry I did not see it.
Are you using Final Cut? It makes use of the hardware encoders and GPU, so it tends to be quieter than, say, Handbrake. But if you're able to play video games without noise, that's a good sign.
If they come up with 5K/120hz Studio Display I would assume that they will not limit it to only thunderbolt 5 and will find a way for others to use. It will have updated AS inside so that might do some magic also. We shall see
My guess is that a hypothetical 5K/120hz Studio Display would have both Thunderbolt 5 and HDMI 2.1. IIRC HDMI 2.1 can handle 5K/120hz with DSC. I am pretty sure the M2 Pro Macs and onward all have HDMI 2.1.
