MacBook 16-inch Fan Noise

"no modern cpu will shut itself down because of heat even if ran without the fan. "

What you're saying is incorrect. Attached is a support document from Intel. Their processors absolutely shutdown when necessary.

raimiss wrote:

i’m wrong, yes modern cpu indeed has failure shut down in them, buuuut they will downclock first. From all the cases that i know that had thermal paste failure on a laptop (it can dry out and crack) it will downclock to a stall, but it won’t shut down. Shutting itself down is extremely edge case that only happens on catastrophic failure of components. Saying device does not overheat unless it thermally shuts itself down is like saying you are fine with your body temp unless you die and we all know its not the case.

You believe heat throttling indicates a system is broken, where in fact it is part of modern thermal management. This is the reason why people who "overclock" PCs install ever more elaborate liquid cooling systems to dissipate excess heat that would otherwise be mitigated by slowing the CPU down from what would otherwise be occasional speed peaks.

So once again, if a system reaches thermal shutdown, that may or may not indicate a problem; older MacBooks (not Pros) did this as a matter of course if you ran something like Photoshop with an external display attached. I know from experience you would generally see a thermal shutdown message like the one I posted above in less than an hour.

However, if the system does not reach that point, that does point to good thermal design as the system is managing itself, even if fan noise and CPU speed are not what you want them to be.

jc_9 wrote:

I tested my same setup (one Youtube playlist + typing text) in bootcamp Windows 10 on the 16" and it was about the same thermals, but a little bit better than MacOS

Bootcamp Windows 10: • 60°C - 2700RPM - dGPU 15W

Mac OS 10.15.4 Catalina: • 65°C - 3000RPM - dGPU 19.5W

The issue is only present with the external displays.

You mistakenly believe the values reported for temperatures are equally accurate; I mentioned previously that the sensors are not calibrated for accuracy, the intent being that the operating system would apply correction factors appropriate for each individual processor as calibration levels are to be determined by experimentation at the factory.

Remember also that even at the power levels you cite, that's still far from the specified maximum TGP of 50W for the Radeon Pro M GPUs:

AMD Radeon™ Pro 5000M Series

If you have access to AMD or Apple design documents, feel free to post links to them; somehow I doubt you do, either.

Your comments that it is ridiculous for the computer to use "X" amount of power to do "Y" are also just pulled from thin air, unless you can show design documents showing this particular power draw is unexpected for AMD Radeon Pro 5000M GPUs under these conditions.

The TGP of 50w though a design maximum, shows 20w is well below that and the fans are handling the thermal load appropriately.

The fact that the power draw is well below their published TGP shows that though it may be unexpected to you, it's within an expected range.

Grant Bennet-Alder has been great about posting details about VRAM clocking and how more power is needed to drive displays and resolutions requiring heartbeat or where the entire screen must be refreshed each display interval.

Now, as far as the system being throttled down and "working worse" - can you show me the benchmarks showing that less actual work is getting done when an external monitor is attached (and drawing higher power) as compared to an older Mac running a slower CPU and slower GPU?

Or is "works worse" simply "the fans are louder than I would like?"

raimiss wrote:

Lol. I'm just following laws of physics where more work results in more energy (160hz is more work than 60hz). You somehow try to prove that apple intentionally designed the machine to break those laws. Moreover you expect everyone to accept that ridiculous claim with no evidence. I am more than happy to accept that if you provide some evidence. If not, i'll just stick to basic physics and call that broken.

Do you know what is actually happening there?

I could see a situation where the GPU actually has to work harder to drive a 60 Hz monitor than a 160 Hz monitor if 160 Hz is closer to the native speed at which its clock drives the VRAM and the slower 60 Hz rate requires multiple refresh cycles to be sent to the VRAM so the image can render properly.

I'm not saying that is what's happening, but it's one possibility.

Refresh rates and GPU clock speeds can create some interesting harmonics issues.

f nothing else, it consumes more electricity, which by itself is definition of "working worse". Or is that also considered being "better" in your world?

Why is drawing more power "working worse?"

If it's drawing content, I don't care if it takes 1w or 100w, if it displays what it is supposed to.

I suspect that's how most people would define "working" when it comes to a GPU.

"And you believe that system that's throttled down and works worse than same hardware in older model (2019 15.6") is expected behavior for a machine that is advertised as the one having an improved thermal design. "

Your mistaken thinking that improved thermals means quiet fans, it does not. Apple was hit hard about the throttling in the 2018 MBP and that's why they improved the thermals in the 16" to reduce or completely eliminate throttling to maintain constant performance, which they have. Nowhere in their advertisement or descriptions do they say "Whisper Quiet Fans under ALL Circumstances". I would argue you find many laptops with performance dGPUs making such a ridiculous statement in advertising.

raimiss wrote:

Let's just all agree to ignore common sense and accept that:

1) GPU is expected to draw 4x more energy when doing 2.5x less work.

2) GPU that can output much better performance at 50w compared is EXPECTED

to use reach 50% of its power rating when idling. All previous generations of same manufacturer would use 5-10% of that.

Both huge leaps.

1) The GPU might or might not be expected to draw more energy, but as there are no specs listed anywhere for this there is no reason to come to a conclusion one way or the other.

However, if it does, that doesn't mean it's broken.

2) 50% of the TGP of 50w is 25w, and no one has come up with that number while the system was completely idle or, in fact while in active use as far as I know. To say no prior GPU used say even 15w at idle would take a lot of research to determine. The same family of GPUs on a card for tower PCs draw over 50w at idle.

However, if it did, that doesn't mean it's broken.

In all seriousness, I personally returned the base 16" because of the initial concern but still need a powerful laptop running macOS that can be silent while casually browsing with a podcast opened on a second screen. It would be really nice if someone with a computer hardware engineering background would chime in answering these 2 questions:

did you experience the problem yourself of extreme fan noise or was it just from this thread that gave you that concern?

I've been in Zoom meetings for the last few weeks with and without 2 additional 4k screens attached and I never had one person complain of any fan noise. At the computer I heard them just barely and checked the speed went up to 2500rpm.

You talk about needing a Macbook that can be quiet and still host another screen. I have a 2015 Macbook Air and it seemed to be quiet all the time, unless I was exporting from iMovie or CaptureOne, but general use you never heard it at all. I bet you can pick one up cheap now since it doesn't have the retina display.

I don't think there is an option to unsubscribe out of a thread but I think you can unsubscribe from all of the Apple Community emails.

There is a bug at present where you cannot easily unfollow a thread.

I remember someone mentioned this bug on Apple website a few months ago, now I'm a little concerned, if Apple can't fix a simple bug on their website, how can we expect them to fix a more serious issue with the MacBook Pro that involves two companies (AMD & Apple).

Grant Bennet-Alder wrote:

denizcan wrote:

...

When using with internal monitor dimmed, MBP consumes 8W. However, when I attached external monitor, the baseline goes to 33W. 25W for what?

...

When running ONLY the built-in display and not editing (or running Chrome, which thinks it is a high-powered graphics app) only the Intel Integrated graphics chip and system RAM are used.

When you connect an external display, the discrete AMD graphics ship is activated, regardless of whether you are doing high-powered graphics. But that in itself does not cause the large increase you are seeing.

<<25 Watts FOR WHAT?>>

to run the [private] display RAM, 4GB of smoking hot GDDR6 RAM, at full power, to support the those two displays (the built-in is a heartbeat-refresh display) at once.

Dimming the internal display does nothing for its incessant need to fetch the scan lines needed to refresh the display, changed or not, 60 time a second, on a very strict timing schedule. Readers have discovered if you shut off the built-in display, such as by closing the lid, the display RAM can sometimes run at half power, shaving 10 watts off the power consumption is some cases.

I think you missed the point.. 8W is with no external monitor, internal one is dimmed.. I take this as a baseline.. If I increase the brightness it goes above 20W..

With external, in clamshell mode, that means the system switched to eGPU, it adds 25W!!! However, in bootcamp, with exteral monitor or not it stays around 21W's.. This gives less than 4 hours battery life which is not acceptable as well..

With what you said. Smoking hot GGDR6 RAM, full powering eGPU, why does it consume less in Windows? The circuit is the same..

My 5 year old gaming laptop with nVidia eGPU consumes around 15W when idle.. When I forcefully activate nVidia it goes to 20W for some period, then it returns to 15W as well.. Unless I do something in the application that uses nVidia, it does not add extra to the power budget.

After quite few months of using this "flagship" Macbook Pro 16'' i9 8 Core for $4000, reading all the other people's experiences, trying everything available to reduce the fan noise nuisance (resets, turbo boost off, fan controls etc.), I must say that this laptop is almost unusable. How can Apple think that a "professional" creative work can be made sitting at some "hair dryer" machine making a constant terrible noise at your head whenever a professional application is used (FCP, audio software etc.).

I have a friend who also bought this for audio work and he is quite desperate - he can't record anything or work with it, since it constantly produces this terrible noise ...

I have had 2 MBPs in the last few years neither of which would heat up dangerously and become noisy just by plugging them into an external monitor.

"Once again you are calling a machine defective for not meeting your particular expectations"

These are not any individual's expectations. As others have mentioned countless times, these expectations are just that - expectations. You expect your computer to run quietly and normally when attached to an external monitor when when the computer is otherwise doing nothing. This one heats up and becomes noisy when you attach it to an external monitor. Please, stop being a part of this discussion William if you continue to believe that is normal - you obviously have no idea if you think it is normal. Yes, we have all posted to the Apple link that another denier notes we should do, but we also live in hope that someone will come up with a positive way to fix this obvious defect. We don't want to hear from you unless you have a way of helping the situation.

[Edited by Moderator]

William Kucharski wrote:

Since you believe the machine can't drive the displays the specs say it can, why don't you try it and if it fails, you can prove Apple is lying in their ads.

I never said it can't be done. I even previously linked this video showing it, also mentionning the author confirming the added noise (even with only a single monitor).

Personally I don't care what temperature it reaches, as long as it doesn't reach thermal shutdown. That means the machine's thermal management is working properly.

I don't care about CPU temps either, for the same reasons.

So you are fine with thermal throttling reducing the performance of your top of the line laptop..!

For example, if the fans were at full blast whether it's driving one or four monitors, it's no big deal. You don't know whether there is headroom or not, so why speculate?

No need to speculate, the dGPU consumes 18W and therefore increases the temps by 35°C (in my case) . That's the headroom the CPU could have use for higher clock rate while turbo-boosting.

As far as Rossman goes, being anti-Apple and criticizing their engineering is his thing, it's how he draws viewers and makes money.

Yeah he's like your opposite! (I only mentionned him because he seen a lot of fried logic boards on Macbook Pros.)

If it was so easy for Apple to do what he suggests and it would be so wonderful, he wouldn't have a business.

A business that only exists because of the absurd costs of official repairs. Don't you agree that it would be better for everyone to reduce the fail-rate of their high value machines?

I realise Apple doesn't care about this since most people buying their products are ok to spend the extra money for repairs / AppleCare+ anyway. As long as it's got the smallest form factor possible... for a magical experience! I'm fine with that, but not for a pro machine.

jc_9 wrote:

No need to speculate, the dGPU consumes 18W and therefore increases the temps by 35°C (in my case) .

Where did you get a 35C increase? I get 10-12 at the most for dual 4k displays.

PhotogWithMac:

talking about expectations id say mine are that a new 2020 16 inch mbp should run my favourite apps considerably more smoothly than my trusted late 2013 retina mbp.

It is tremendously better.

I've had more Macbook Airs, and MacBook Pros then I can count on my fingers (I think), and it is amazing. I do not have any issues with the 16" MBP. I've ran the tests and posted the results and it is by far better than the 2016, 2018 and the 2019 15" models in every aspect. It can export hour long FCP in 9 minutes. My 2015 MBA couldn't even come close in performance and it would get hot when exporting pictures!

Does the 16" get hot? When I have 16 (8x8) cores running - yes. The graphics card is amazing and works in perfect tandem. Even at full blast it hits about 90C. When it's done exporting and back to video editing it's about 60C

I process videos and I use Capture One for all my RAW picture processing. It is great for that. No lags and I can process the photographs so fast that I'm not waiting on it, it's waiting on me.

My temps when running my average 5-10 apps are 55-65C with dual 4k screens connected to a CalDigit TS3+ dock. This includes full metal apps like Affinity Suite, Adobe Suite, Parallels, Zoom and more. My fans are generally at 1800-2000. If turboboost is on it's still around there most of the time but will spike up to 2500 to 3000 RPM once in a while but not really noticeable in the office.

I would certainly buy it again. I'm just sad that not everyone has had the same great experience I have everyday on this machine.