Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
It just occurred to me that this thread is a good example of the way language evolves. Whilst I am definitely in the ‘words have meanings’ camp, I recognise that usage changes, and eventually meanings have to change to keep up. Every time I read ‘coil whine’ I wonder how things with no coils can exhibit coil whine, but I know that term is being used to describe phenomena with a similar sound.

It’s just like kids these days using the word literally to mean figurative, but with intensity.

I write this with no malice, and I have no intention of having a go at anyone, it’s simply a linguistic observation that I think is interesting.
 
It just occurred to me that this thread is a good example of the way language evolves. Whilst I am definitely in the ‘words have meanings’ camp, I recognise that usage changes, and eventually meanings have to change to keep up. Every time I read ‘coil whine’ I wonder how things with no coils can exhibit coil whine, but I know that term is being used to describe phenomena with a similar sound.

It’s just like kids these days using the word literally to mean figurative, but with intensity.

I write this with no malice, and I have no intention of having a go at anyone, it’s simply a linguistic observation that I think is interesting.
The literally thing figuratively drives me crazy.
 
It just occurred to me that this thread is a good example of the way language evolves. Whilst I am definitely in the ‘words have meanings’ camp, I recognise that usage changes, and eventually meanings have to change to keep up. Every time I read ‘coil whine’ I wonder how things with no coils can exhibit coil whine, but I know that term is being used to describe phenomena with a similar sound.

It’s just like kids these days using the word literally to mean figurative, but with intensity.

I write this with no malice, and I have no intention of having a go at anyone, it’s simply a linguistic observation that I think is interesting.

You aren't talking evolution but degradation :) I disagree about the lowest common denominator.

But yes, the word 'whines' fits this thread.
 
My sound : https://voca.ro/1do6Y9vo8r6I
Hello everyone,

I recently got a brand-new M4 Max MacBook Pro (absolutely loving it so far), but I noticed something a bit odd during my first intensive machine learning training session.

I’m training a custom YOLO model for object detection using PyTorch. The training loads thousands of images from SSD and utilizes MPS (Apple’s GPU API). Everything runs smoothly — no thermal throttling, the GPU usage is around 80-90%, and the fans stay quiet.

But here’s the catch: While training, every 1–2 seconds I hear a soft “tick-tick” sound coming from the chassis. It’s not loud, it’s not grinding, but it’s definitely audible in a quiet room. Almost like a faint electrical click or subtle coil whine — but not constant. Just periodic tiny ticks. It only happens during training (or other heavy SSD/GPU activity).It doesn’t seem related to fan speed (tried changing RPM via software). Activity monitor shows SSD usage at ~17%, but IOPS might be high due to frequent reads/writes. No sound during normal use.

I even thought it could be a stray hair or dust caught inside, but that seems unlikely. It sounds more like SSD controller noise or GPU coil whine under load.

Anyone else experience this? Normal behavior for high-speed SSD access or M-series GPU training load?
 
  • Wow
Reactions: Cape Dave
My sound : https://voca.ro/1do6Y9vo8r6I
Hello everyone,

I recently got a brand-new M4 Max MacBook Pro (absolutely loving it so far), but I noticed something a bit odd during my first intensive machine learning training session.

I’m training a custom YOLO model for object detection using PyTorch. The training loads thousands of images from SSD and utilizes MPS (Apple’s GPU API). Everything runs smoothly — no thermal throttling, the GPU usage is around 80-90%, and the fans stay quiet.

But here’s the catch: While training, every 1–2 seconds I hear a soft “tick-tick” sound coming from the chassis. It’s not loud, it’s not grinding, but it’s definitely audible in a quiet room. Almost like a faint electrical click or subtle coil whine — but not constant. Just periodic tiny ticks. It only happens during training (or other heavy SSD/GPU activity).It doesn’t seem related to fan speed (tried changing RPM via software). Activity monitor shows SSD usage at ~17%, but IOPS might be high due to frequent reads/writes. No sound during normal use.

I even thought it could be a stray hair or dust caught inside, but that seems unlikely. It sounds more like SSD controller noise or GPU coil whine under load.

Anyone else experience this? Normal behavior for high-speed SSD access or M-series GPU training load?
It is normal behavior and much better than other options. You should train on gaming laptops or Nvidia 5090/4090. It’s much worse constant whining and fan noise. I have M4 max, and still use my M1 Max at high GPU loads. Both make noises, and if my 4090 is also running, you pretty much don’t hear anything around. If sounds bother you, then use cloud for training and don’t bother local training.
 
  • Like
Reactions: georgekrav
It is normal behavior and much better than other options. You should train on gaming laptops or Nvidia 5090/4090. It’s much worse constant whining and fan noise. I have M4 max, and still use my M1 Max at high GPU loads. Both make noises, and if my 4090 is also running, you pretty much don’t hear anything around. If sounds bother you, then use cloud for training and don’t bother local training.
It doesn’t bother at all and when the fans kicked in I cant hear it anymore because the fans are working. So if this is normal isn’t harmful for the device. And one more thing , yours makes the same sound as mine’s?
Thanks you for the answer you make me feel relaxed !!
 
It doesn’t bother at all and when the fans kicked in I cant hear it anymore because the fans are working. So if this is normal isn’t harmful for the device. And one more thing , yours makes the same sound as mine’s?
Thanks you for the answer you make me feel relaxed !!
Mine makes sounds like yours and some. It’s not usually audible unless you sit in a quiet room. No need to be obsessive about this stuff and try listening to minutest of sounds. My M1 Max has been hammered for almost 4 years and still runs great, and so does my M4 Max MBP. I usually get AC+ for my MBP because I travel and can’t always control environment.
 
i've got coil whine on my m4 pro 14" macbook pro. i notice it in ableton, the frequency changes depending on which sample rate or buffer i select, and oscillates while rendering audio for export. probably gonna try returning to see if a replacement has the same issue because it is annoying to have to attempt to minimize this when working with audio. for example if i want to check a mix using my macbook pro speakers the coil whine is distracting.
 
  • Sad
Reactions: jabbr
Came here from Google. I just got a MBA 13" M4 and noticed coil whine in one specific scenario.

I have a OWC 10GBe Ethernet TB3 adapter that I plugged into my MacBook Air M4. When running a Speedtest, I hear an audible coil whine. It seems to happen on faster internet speeds (e.g., when Speedtest shows 7-8Gbps). When I cap my LAN speed to 2.5Gbps I only get the sound when "uploading" but not "downloading". Downloading a large file at high speeds using the same adapter also triggers the issue. Enabling power saving seems to make the sound go away. I tried a TB3 adapter from another brand which also gave the same issue.

Now, strange thing is that I cannot reproduce any coil whine in any other scenarios. Hooking my laptop up to a TB3 dock - no issues. Running Cinebench benchmark while benchmarking my SSD at the same time (I thought it was SSD related) - no issues. Running Geekbench GPU or CPU tests - no issues. Running yes > /dev/null across ten cores also did not result in any noticeable coil whine.
 
@slainbabyyc any luck with a different one?
just sent it in, they replaced my fans but didn't change the logic board. still making the noise.

in ableton, freezing the tracks (rendering the audio so that the processor is no longer under load) kills the noise. this is what used to be standard practice to kill fan noise in older macs. so basically apple silicon has still not solved the issue of computer noise for audio sensitive applications. i can live with it but oh well.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.