Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

theorist9

macrumors 68040
May 28, 2015
3,880
3,060
Apple specs the M1 iMac as producing 7 dB at both idle and under light operation (wireless web surfing) from a "typical operator position" (see screenshot at bottom). 7 dB is really quiet (acceptable noise floor for a recording studio is 20-30 dB). The figure is not meaningful without knowing the "typical operator postion", which they don't specify, but since it's got a display we can guess it's no further than ~24". Given that, the thing should be really quiet under light load.

Alas, and in practice, the only way for you to know if there's a problem with your unit is to find a way to compare it to a second one.

Others can confirm, but I've read the reason the Apple Silicon MBP's can be completely quiet at idle is that their fans can turn off completely.

Also, a general comment on Apple engineers and Apple product design: They're experts, and should be respected as such. So you shouldn't criticize them without giving it serious consideration. But that doesn't mean Apple product design is beyond reproach when it comes to thermals and noise (I say product design, since engineers can be overruled). Those who make the final design decision, whether it's the engineers or the managers, are human, and can make mistakes and questionable judgements like anyone else.

For instance, when they released the last-generation 27" Intel iMac in late 2014 (the "Retina" iMac), its max thermal output with the upper-end chip was 288W, which led to high fan noise under load (https://forums.macrumors.com/thread...or.2307038/page-6?post=30436381#post-30436381). Many have complained of this with subsequent iMacs of that generation, even though their TDP's were lower; it's also been mentioned by Lisa Gade of Mobile Tech Review, who I've found be very reasonable in her assessment of Apple products. She described it as sounding like it's "taking off". The point is Apple engineers knew what TDP they'd have to support from the start, and yet Apple didn't produce a cooling system sufficient to ensure quiet operation.

And a quiet cooling system for the iMac was readily achievable by Apple. We know this, because we have the iMac Pro, which has an even higher TDP, yet has a reputation for quiet operation. The difference in cost is small—an extra fan, some more plastic ducting, and probably a larger heat sink (and maybe some more copper tubing) (even at retail, 1/4" copper tubing is $1.20/ft; maybe you'd need special alloys or purity, but still...). So I'd say the 27" Retina iMac's thermal design represented a bad judgement on Apple's part when it came to usability and quiet operation. They could have readily, and cheaply, made it much better.

So let's respect those Apple engineers and product designers, but let's not beatify them, or ascribe to them some sort of papal infallibility ;).



1652148824352.png


 
Last edited:

sunny5

macrumors 68000
Jun 11, 2021
1,838
1,706
Have you considered the possibility that you're the one being stupid here, not the multiple engineers and managers who created and signed off on those designs?

Fans don't have to be "connected" to anything if the airflow they create is guided by ductwork so that it passes through the heatsinks it needs to cool. Guess what exists in both of those computers, when fully assembled? Apple's designers take full advantage of controlling the shape of everything in the machine, so you have to look at all the components seen in a teardown and think about how they fit together when assembled, rather than just knee-jerk reacting to a single picture of a mostly disassembled machine.

But hey you saw those pics posted somewhere on a forum of PC overclocker d00dz who know it all and they made fun of them so their dimwitted criticisms MUST be true, right?
LOL, Mac Pro 2013 is a great example why Apple engineers are stupid. Have you ever built PC desktop before? What Apple did is putting a case fan WITHOUT CPU fan for a desktop. Fans require heat pipes to transfer heat efficiently but M1 iMac does not have such things. Airflow is worst than directly-connected fan. I guess you dont even know the basic of fan cooling after all.
 

mr_roboto

macrumors 6502a
Sep 30, 2020
856
1,866
LOL, Mac Pro 2013 is a great example why Apple engineers are stupid. Have you ever built PC desktop before? What Apple did is putting a case fan WITHOUT CPU fan for a desktop. Fans require heat pipes to transfer heat efficiently but M1 iMac does not have such things. Airflow is worst than directly-connected fan. I guess you dont even know the basic of fan cooling after all.
I've built several PCs, yes. I've also done solder rework on a $50,000 USD circuit board (yes, the patient survived). And I've designed boards. Sometimes assembled them myself, sometimes worked with an assembly house. And I've written logic for FPGAs and ASICs. And I've written software too.

Airflow is airflow. Doesn't matter whether a heatsink is "connected" to a fan; all that's important is the actual mass flow and velocity numbers through the fins. The heatsink doesn't know or care whether that flow is coming directly off the blades of a fan, or through some ductwork.

The only reason why air cooled build-your-own PCs almost always have a fan mounted directly on the heatsink is that there are no standards guaranteeing airflow through specific paths inside an ATX case. Since heatsink manufacturers who serve the DIY market can't count on anything, they just do it themselves, and that means the fan is direct attach.

That all changes when the box and the things inside the box are co-designed by employees of the same company. Lets them be as creative as they want to be, within the budget they're given.

Here's a learning experience for you: crack open a few 1U rackmount servers and tell me how many times you find a fan directly bolted or clipped to the CPU heatsink.

You could even crack open a Dell desktop PC. I haven't looked inside one recently, but IIRC back in the Core 2 days they all had ducts with the fan located far from the heatsink. Worked fine.
 
Last edited by a moderator:

sunny5

macrumors 68000
Jun 11, 2021
1,838
1,706
I've built several PCs, yes. I've also done solder rework on a $50,000 USD circuit board (yes, the patient survived). And I've designed boards. Sometimes assembled them myself, sometimes worked with an assembly house. And I've written logic for FPGAs and ASICs. And I've written software too.

Airflow is airflow. Doesn't matter whether a heatsink is "connected" to a fan; all that's important is the actual mass flow and velocity numbers through the fins. The heatsink doesn't know or care whether that flow is coming directly off the blades of a fan, or through some ductwork.

The only reason why air cooled build-your-own PCs almost always have a fan mounted directly on the heatsink is that there are no standards guaranteeing airflow through specific paths inside an ATX case. Since heatsink manufacturers who serve the DIY market can't count on anything, they just do it themselves, and that means the fan is direct attach.

That all changes when the box and the things inside the box are co-designed by employees of the same company. Lets them be as creative as they want to be, within the budget they're given.

Here's a learning experience for you: crack open a few 1U rackmount servers and tell me how many times you find a fan directly bolted or clipped to the CPU heatsink.

You could even crack open a Dell desktop PC. I haven't looked inside one recently, but IIRC back in the Core 2 days they all had ducts with the fan located far from the heatsink. Worked fine.
Then you prove yourself that you know nothing about it. The temperature difference between Air and Pro already proves that you are wrong. And yet, you are clearing ignoring how Apple made Mac Pro 2013. How come you aren't talking about the maximum temperature for M1 iMac compared to MBP with a fan connected to M1? Why not make wireless fans for all Macs? You see, you have no idea what you are saying and far from being an expert.

Comparing server rack to a AIO desktop is already idiotic. They already meant be like that but M1 iMac? That's stupid. Beside, the airflow for M1 iMac is not even efficient at all.

You just wanna admit that Apple is the best even if they make stupid design decisions after all.
 
Last edited by a moderator:
  • Haha
Reactions: jdb8167

leman

macrumors Core
Oct 14, 2008
19,521
19,677
LOL, Mac Pro 2013 is a great example why Apple engineers are stupid. Have you ever built PC desktop before? What Apple did is putting a case fan WITHOUT CPU fan for a desktop. Fans require heat pipes to transfer heat efficiently but M1 iMac does not have such things. Airflow is worst than directly-connected fan.

The current Mac Pro also doesn’t have any CPU fans and yet it can sustain 250W on the CPU alone indefinitely. The CPU fan is only there so that the heat sink can be made smaller. Check out fully passive CPU coolers such as NH-P1, they exist and work very well. Similarly, the 2013 model did not need a CPU cooler because all the components shared a single radiator. The cylinder Mac Pro was not unlike a water cooled PC, just that you had heat pipes instead of water.

And regarding the M1 iMac… it’s a 25W design. It literally generates less heat than some android phones.

I guess you dont even know the basic of fan cooling after all.

Look who is talking ?
 

Apple Knowledge Navigator

macrumors 68040
Mar 28, 2010
3,692
12,912
I’ve owned a 2013 MP since 2014. It’s never overheated. I think you are equating the GPU failures that some 2013 Mac Pros experienced with general overheating. I don’t think that is accurate.
The GPU failures were a result of the overheating. It's obviously all dependent on how you use the device, since FCP users were the most vocal about it.
 
  • Like
Reactions: AlphaCentauri

jdb8167

macrumors 601
Nov 17, 2008
4,859
4,599
The GPU failures were a result of the overheating. It's obviously all dependent on how you use the device, since FCP users were the most vocal about it.
I can’t find any articles on overheating being a general problem. Apple communities, Reddit, and MacRumors there are complaints but it didn’t seem to enough of a problem to warrant a mainstream article.
 
  • Haha
Reactions: sunny5

pshufd

macrumors G4
Oct 24, 2013
10,149
14,573
New Hampshire
I can’t find any articles on overheating being a general problem. Apple communities, Reddit, and MacRumors there are complaints but it didn’t seem to enough of a problem to warrant a mainstream article.

I've heard complaints about heat from Mac minis but the newer ones use integrated graphics and I've not heard of Intel Integrated failing due to thermals. We've had a bunch of logic boards replaced in MacBook pros over the years but those had discrete graphics. I think that some iMacs failed due to discrete graphics too.

Stuff may run hot but it's within design tolerances. One other problem that I've run into is fan sensors failing which can result in overheating and abrupt system shutdown.
 

Apple Knowledge Navigator

macrumors 68040
Mar 28, 2010
3,692
12,912
I can’t find any articles on overheating being a general problem. Apple communities, Reddit, and MacRumors there are complaints but it didn’t seem to enough of a problem to warrant a mainstream article.



https://www.reddit.com/r/macpro/comments/fpphtj

And that was just a quick Google search. Apple was quick to delete discussions on their own communities website (which they do for other problems with their hardware).

The fact is it overheated during maximum load and that was a result of poor engineering.
 

jdb8167

macrumors 601
Nov 17, 2008
4,859
4,599



https://www.reddit.com/r/macpro/comments/fpphtj

And that was just a quick Google search. Apple was quick to delete discussions on their own communities website (which they do for other problems with their hardware).

The fact is it overheated during maximum load and that was a result of poor engineering.
Yes, I found those user written posts. What I didn't find and apparently you didn't either, was anything that was written up in the tech press. Generally if a problem is widespread it eventually reaches the press. Since the 2013 Mac Pro was available until 2019, I would expect a general problem to be covered. An obvious recent example is the butterfly keyboard.
 
  • Haha
Reactions: sunny5
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.