M1X Mac mini's Leaked Schematics. Fake or?

[deconstruct60]

I don't think that ethernet port should be present if the mac mini would have that new magnetic charging port.

That design gimmick probably won't work well for 10GbE. ( the smaller screen iMac M1 don't have to cover that feature)

Conceptually the Mini could have two possible Ethernet ports. Apple could go to the 10GbE being standard. If you want a second then use the one on the power supply. ( the Ehthernet controller is on the motherboard regardless. Moving the port farther away just creates RF drama have to deal with. )

10's of thousands of Minis get used as servers. Putting the primary Ethernet socket on the power brick is a mess.
(some folks might what to deal with that as easy way to attach to a admin/management network or seperate SAN/NAS from Internet versus some dock or USB ethernet dongle. )

Pushing the power supply out into a power brick is just a 'ballon squeeze' if placing these in a rack. If have to now place the brick and the mini side by side then pretty good chance that isn't any thinner than if left the power supply in the system and have bigger issue eyeballing which ethernet line goes to which system.

 

[sirio76]

100% fake, anyone with some design experience will tell you this.

 

[deconstruct60]

Nobody has said they are compromising on thermals. We don’t know what the thermal envelope is of the new design. All I was saying is the current mini is designed to handle a 65W processor and the M1 is only 15W. Even if they cut the thermal envelope in half, it would still be more than enough to sustain max load on the M1.

Apple has gotten more thermal designs wrong far more often than they have gotten them right. With their track record having some cushion is a good thing.

If this "we don't need any air vents at all" design even was suggested by Apple Industrial labs then even more true. If that is what they are gearing up to ship.... same old broken record.

 

[deconstruct60]

It really looks like J. Prosser will be right this time.

Remember this rumor?

High-End Mac Mini Said to Feature Thinner Design With 'Plexiglass' Top, Magnetic Power Port

Last week, Bloomberg reported that Apple is working on a high-end version of the Mac mini with additional ports and featuring the same Apple silicon...

www.macrumors.com

Or this is just regurgitated rumors with a new "Mock up" schematic attacked to it. Prosser isn't really validated when his stuff just shows up in the rumor echo chamber again.

If flipped those two Type A ports to Thunderbolt 3 and those vertical slots are just air vents and slap a M2 (follow on to M1 , same size , similar thermals ) then that would be more believable. Basically the laptop , two port logicboard tweaked to fit in a box with no keyboard or screen attached. ( plastic / plexiglass top for RF placements that probably make more sense in most deployment contexts. ) [ Perhaps the audio jack is just missing on the right side of back end ( on other side of HDMI). Or Apple is 'brave' and think folks just pass audio through HDMI and/or Thunderbolt docks. ]

if basically a slightly tweaked laptop logic board than attaching it to a laptop power brick would line up.

However, I highly doubt that is where the "large die" M-series ( M1X , M2X ) powered Mini is going.

 

[leman]

Apple has gotten more thermal designs wrong far more often than they have gotten them right. With their track record having some cushion is a good thing.

Do you have any examples in mind? I think Apple thermal designs are generally top-notch, but they are focused on overall utility and not delivering lowest possible temperature at the expense of other properties.

 

[Jorbanead]

Apple has gotten more thermal designs wrong far more often than they have gotten them right. With their track record having some cushion is a good thing.

If this "we don't need any air vents at all" design even was suggested by Apple Industrial labs then even more true. If that is what they are gearing up to ship.... same old broken record.

A bit of a slippery slope. We don’t know what they’re doing. Apples issues with thermals also stems from Intel not being able to keep up on their progress/node process as Apple had anticipated. Now that Apple has their own designs they can tailor their thermals to their processors better than before.

My comment was simply saying the thermal capacity of the current mini is more than triple, maybe quadruple what is required of the M1. Like I said, even if they cut the thermals in half, and the M1X is double the TDP, it still would be enough. We can complain about lack of thermal capacity if the time comes.

 

[Carlson-online]

Looks fake.

1. There is no ventilation there at all. All the 'holes' are for sockets or bottoms. The presentation of the bottom looks to be missing any inlets. The back has now outlets That makes little sense if the M1X is 2-3x higher thermal output. ( If more than double the number of GPU cores then not going to get a doubling of wattage consumed. Likewise if double up on RAM modules then path lengths go up incrementally over a doubling). Doubling the TBv3 sockets is a doubling of power there. (MBA don't have four sockets) . The 10GbE still generates substantively more power burn off than 1GbE do.

Cooling by air capacity is about how many cubic feet of air you move per second. If there is no input or output then amount moved is probably going to be pretty close to zero. Lowering the thermal capacity is one thing. Dropping it zero is silly. ( Apple has done silly thermal things before so not impossible. )

Completely getting rid of the vents is a bonehead move. The size of the hole doesn't have to change to lower the cooling. Just blow air through the holes at a slower rate will reduce the cooling capacity. That buys less noise which is a good enough "win".


"Apple will just passively radiate everything through the chassis". The problem with that is multiple over several contexts.

a. stacking Mini. ( a super cheap cluster is to just stack 2-3 minis on top of each other. ) . If the top and bottom are radiators then are radiating into one another. In a 3 stack the one in the middle is being radiated to on both sides. You have put it in a sandwich if heat sources. That isn't good.

[ looks like there are some "bumper" feet on the bottom . The stand off distance there doesn't look large enough to be effective for this problem. ]

b. racking them in a data center. How do you get cool air into the MIni's. They have no fan. Would have to add extern fans to blow it in. That is a 'fail'. There will be fans to help "scoop up" hot exhaust , but this runs counter to about everything else in the rack. ( Minis are already a bit of 'odd ducks' in that the suck in air from the side ( racking usually means vertical placements. That isn't "great" but it isn't exactly hostile to the overall cooling flow. ).

[ Never mind there are 10's of thousands Mini racks slots already deployed out there in the current form factor that would need tweaking. ]

c. Very long ( dozens of hours) workloads ( e.g., 24/7 continuous build for multiple projects over different time zones ). This is plugged into a wall so the power draw can be constant and assigned to multiple users.



2. The "design" here seems to be just horizontally slicing off the air vent from the current model .

That appears "clever" right up until the point you notice that it also slices off the headphone jack. That is ridiculous. Even when Apple thinned out the iMac M1 variant they still left the headphone jack. Dumping the headphone jack from a desktop product wouldn't be "brave" . It would be "boneheaded".

3. Somewhat coupled to that is just how is the speakers sound suppose to get out of this enclosure.... or being "brave" and deleting the speaker too? Similarly to cooling you have to actually move air to make noise.


4. As another post mentioned those TBv3 ports are packed too tight. There is Radio frequency bleed and power coming off 40 Gb/s ports.

5. Radio out for multiple antennas where? The top is now "glass" ? ( that won't help much with thermals dissipation). Somewhat goofy placement but Apple places major antenna placment in the bottom.

Mac mini Late 2018 Teardown

The future is now! Apple’s once-neglected Mac mini is coming in hot with a brand new, cutting edge, long awaited … processor upgrade? And a couple...

www.ifixit.com

[ step 3. ]

If they "nuke' the inlet mechanism where do the antennas go?


The Mac Mini is already pretty thin. With x86_64 chips, it has run quite noisey because it is pretty thin now. "Even thinner" isn't going to buy much here.

 

[PsykX]

M1X Mac mini's Leaked Schematics Show a Generous Number of Ports, a Magnetic Charger, and More

A leakster claims to show the M1X Mac mini schematics, showing how the compact machine will different from its predecessor

wccftech.com

Personally, I don't trust LeakApplePro...

Thoughts?

He hates Prosser and released basically the same thing 2 ½ months after him.
I think it's the real thing, but to see those ports so close to each other is rather surprising. It did not really make sense on Prosser's rendering either.

 

[jav6454]

M1X Mac mini's Leaked Schematics Show a Generous Number of Ports, a Magnetic Charger, and More

A leakster claims to show the M1X Mac mini schematics, showing how the compact machine will different from its predecessor

wccftech.com

Personally, I don't trust LeakApplePro...

Thoughts?

Those three pictures show basic renders, not engineering level schematics. I doubt they are true. I can see the ports somehow making it given that the MacMini is a desktop. But other than that, nope, it's fake.

 

[CWallace]

Even if this is real, it clearly is not a production schematic. So the closeness of ports and lack of cooling doesn't really concern me as these drawings are not an accurate representation of the product in terms of measurements.

 

[sublunar]

I think this does look like a regurgitation of the Jon Prosser leak with a reverse engineered isometric sketch of the exterior case for the clicks.

One of the unexpected bonuses of the current M1 mini is the (now) over specified cooling system which most people have said in effect makes it the second quietest Mac after the (fanless) MacBook Air.

The fact that the purported redesign appears to have reduced the size of the Mini to fit more with Apple's expected cooling regime will make music producers somewhat concerned if the cooling solution becomes compromised.

The discussion of the size and relative proximity of the supposed 4 Thunderbolt ports also puts a question mark over the layout of the ports (the much commented point that they are too close together and laid out vertically which TB ports on the Mini have not been up to now). Obviously lack of visible vents is also notable alongside the removal of the lower base.

All I will say is the co-location guys must have had some input into this because I would have been more than happy for the existing form factor to be retained but with the polycarbonate top to alleviate the wifi/bluetooth issues that have been reported for years.

To address the other design decisions though, the Ethernet port that's built in would be upgradable to 10 Gig, the Magsafe power option could optionally offer an additional gigabit Ethernet if specified that way.

The weird thing is that Apple could have stuck with the existing design (with material alterations) if a theoretical M1X CPU with 8 performance cores, 2 efficiency cores, and an unspecified number of graphics cores (12-16?) was coming. You could easily double or triple the expected TDP and reuse the existing form factor which surely wouldn't be a cause for concern for the co-location guys and high performance Mac users who need peace and quiet might appreciate a form factor that stayed quiet too.

If, however, the Mini is getting an iMac style MagSafe external power supply I would offer the suggestion that for co-location this is easier to replace than taking an M1 Mini into a workshop to replace the built in power supply - a lot less downtime. We've already seen that the power supply is rated for more than the iMac 24" needs so it's almost like it's ready to be repurposed on other Mac models which would be a nice economy of scale. It seems even more over-engineered to be grafted onto a Mini which arguably uses no more power than the iMac.

It would be a pity if an M1X mini was getting a potentially needless redesign to reduce the cooling capacity just when a more powerful CPU is being released and it ends up producing additional fan noise to annoy users who may professionally need silence. If such a unit continued to have HDMI sleep and bluetooth/wifi USB3 interference issues there would be many groans from the community.

 

[Argon_]

Do you have any examples in mind? I think Apple thermal designs are generally top-notch, but they are focused on overall utility and not delivering lowest possible temperature at the expense of other properties.

I9-8 core 15MBP, rMBA. On a less severe level, all 6 core touch bar 15MBPs, iMac Pro, 6-8 core Intel Mac Mini. One often has to choose between burning hot temperatures or high fan noise. In the case of the first two machines, usually both. Of course, a lot of this has to do with Intel first failing to reduce their TDPs, then raising them, then outright lying about them.

I'd be all for a 'Mac SE' in a couple years, packing an M1 Mac Mini into a form factor the size of a piece of toast, with passive cooling. For a high end machine though, no. It's a desktop; keep the old form factor and enjoy the thermal headroom. Particularly for audio and video professionals for whom a quiet machine improves their quality of work.

The X suffix ASi might range from 30-45W of real TDP.

 

[sunny5]

 

[CWallace]

I9-8 core 15MBP, rMBA. On a less severe level, all 6 core touch bar 15MBPs, iMac Pro, 6-8 core Intel Mac Mini. One often has to choose between burning hot temperatures or high fan noise. In the case of the first two machines, usually both. Of course, a lot of this has to do with Intel first failing to reduce their TDPs, then raising them, then outright lying about them.

As you noted, most of this blame lies on Intel and in my first-hand experience with Dell XPS, Lenovo Carbon and HP Spectre laptops (all similar "thin form factor" models like the MacBook Pros) with similar configurations, they are all nuclear furnaces, as well (I actually cannot keep my left hand near the exhaust port of my Carbon as it physically hurts from the heat).

The MBPs are actually the coolest since they can use their aluminum chassis to help dissipate heat, something the CFRP and plastic cases on the PCs are not nearly as effective as.

That being said, going thicker is not necessarily a solution. Our older Dell Precision portable workstations that are an inch thick roast from their Xeons and nVidia GPUs. I shudder to think how much worse it is on the newer, thinner models.

 

[leman]

I9-8 core 15MBP, rMBA. On a less severe level, all 6 core touch bar 15MBPs, iMac Pro, 6-8 core Intel Mac Mini. One often has to choose between burning hot temperatures or high fan noise. In the case of the first two machines, usually both.

I agree that these products could be described as "underwhelming" in regards to thermals/noise under load, but I wouldn't say that any of them was a bad thermal design. Each of these were more than capable of providing in excess of the official advertised platform TDP, thus matching the spec. It's a different issue that the spec was sometimes at odds with reality. But than again, what is one supposed to do? A mobile Intel i9 can draw over 100 watts under turbo, does this mean that laptop manufacturers have to provide over 100W of thermal dissipation? Why even have laptops than if one demands desktop-like cooling capabilities?

Of course, a lot of this has to do with Intel first failing to reduce their TDPs, then raising them, then outright lying about them.

I had a look at the leaked Adler Lake power limits... their 15W CPUs have PL2 of 55watts... in contrast, an M1 when used in a 15W chassis has a top power draw of... 15W. Talk about expectations and marketing.

 

[anshuvorty]

These don’t look very official TBH. The ports for example aren’t detailed enough (HDMI port is just a square) so these to me look like poorly created fan-made “CAD” drawings based off of prossers leaks. That’s why the USB-C ports are way too close together.

It still could very much look similar or close to this design, but based on the history of the leaker I’m skeptical of the legitimacy of this.

^. This. The post just took Prosser's claim and claimed it as his own, it seems. We shall know soon enough..

 

[thingstoponder]

Leaks Apple pro is a fraud who no serious leaker takes seriously.

This “schematic” looks like someone just made a diagram of the Prosser mockup which probably isn’t real either.

 

[Argon_]

I agree that these products could be described as "underwhelming" in regards to thermals/noise under load, but I wouldn't say that any of them was a bad thermal design. Each of these were more than capable of providing in excess of the official advertised platform TDP, thus matching the spec. It's a different issue that the spec was sometimes at odds with reality. But than again, what is one supposed to do? A mobile Intel i9 can draw over 100 watts under turbo, does this mean that laptop manufacturers have to provide over 100W of thermal dissipation? Why even have laptops than if one demands desktop-like cooling capabilities?



I had a look at the leaked Adler Lake power limits... their 15W CPUs have PL2 of 55watts... in contrast, an M1 when used in a 15W chassis has a top power draw of... 15W. Talk about expectations and marketing.

I'd say that the latter set of products I mentioned were underwhelming, while the first two were legitimately bad design. The I5 2020 Air shoved a quad core chip into a dual core thermal design, assuming similar heat production per core based on product line, and proceeded to roast itself under normal loads like streaming music. The I9 15MBP was the choice of professionals, yet it throttled badly under continuous loads, a PRO machine should be designed with extended, heavy workloads in mind.

As for the Intel TDPs, they are shady marketing at best, and blatant lies at worst.

 

[jdb8167]

I had a look at the leaked Adler Lake power limits... their 15W CPUs have PL2 of 55watts... in contrast, an M1 when used in a 15W chassis has a top power draw of... 15W. Talk about expectations and marketing.

That's Alder Lake-P I presume? There is a lower power version with only 2 performance cores and 8 efficiency cores that has a PL2 of 30W. It is rated by Intel as TDP of 5-7W according to Tom's Hardware but listed at 9W so I'm not sure what that means. Downclock to get to 5W? The PL4 for Alder Lake-P (6p8e) can be as high as 215W. That one is supposedly a 45W TDP.

Alder Lake Specs. The nomenclature is x+y+z where x is the # of performance cores, y efficiency cores, and z graphics level which are all GT2 level with 96 execution units.

	Alder Lake-P	Alder Lake-M
	(2 + 8 + 2): <= 15W
PL1 (TDP)	(4 + 8 + 2): <= 28W	(2 + 8 + 2): <= 9W
	(6 + 8 + 2): <= 45W
	(2 + 8 + 2): <= 55W
PL2	(4 + 8 + 2): <= 64W	(2 + 8 + 2): <= 30W
	(6 + 8 + 2): <= 115W
	(2 + 8 + 2): <= 123W
PL4	(4 + 8 + 2): <= 140W	(2 + 8 + 2): <= 68W
	(6 + 8 + 2): <= 215W

I don't think any laptop is going to reach PL4 on the Alder Lake-P (6+8+2) no matter how good their cooling solution is.

 

[CWallace]

I'd say that the latter set of products I mentioned were underwhelming, while the first two were legitimately bad design. The I5 2020 Air shoved a quad core chip into a dual core thermal design, assuming similar heat production per core based on product line, and proceeded to roast itself under normal loads like streaming music. The I9 15MBP was the choice of professionals, yet it throttled badly under continuous loads, a PRO machine should be designed with extended, heavy workloads in mind.

I still place the blame firmly on Intel for the MBP. When Apple were designing the 2016 MBP, Intel's road-maps were claiming they'd be at 10nm very soon with much lower TDPs. Instead, they ran hard aground on the 14nm shoals and the only way to get more performance (to get people to buy new generations) was to crank the speeds and TDP.

And again, Apple was not alone in getting bit by this. Our company has Tier One Windows OEM laptops of similar design and they roast worse than the MacBook Pros. And even the inch-plus-thick portable workstation slabs crank their fans to keep the plastic from softening under sustained full loads.

I will spot you the 2020 MacBook Air since that was a new design, but even there, Apple was between a rock and a hard-place as the whole point of the Air line since it launched was to be thin and light and the only CPU options they had from Intel were furnaces.

Honestly, if any Mac model drove Apple to creating M1, it was the Air, because they either had to put "crippled" chips in it (like the i3s) to keep the thermals reasonable or choose the "performance heaters" in the i5 and i7 line and hammer the active cooling.

 

[LeeW]

All could be true, but then that does not mean it is 'coming soon' just because they have a design in mind.

 

[Argon_]

I still place the blame firmly on Intel for the MBP. When Apple were designing the 2016 MBP, Intel's road-maps were claiming they'd be at 10nm very soon with much lower TDPs. Instead, they ran hard aground on the 14nm shoals and the only way to get more performance (to get people to buy new generations) was to crank the speeds and TDP.

And again, Apple was not alone in getting bit by this. Our company has Tier One Windows OEM laptops of similar design and they roast worse than the MacBook Pros. And even the inch-plus-thick portable workstation slabs crank their fans to keep the plastic from softening under sustained full loads.

I will spot you the 2020 MacBook Air since that was a new design, but even there, Apple was between a rock and a hard-place as the whole point of the Air line since it launched was to be thin and light and the only CPU options they had from Intel were furnaces.

Honestly, if any Mac model drove Apple to creating M1, it was the Air, because they either had to put "crippled" chips in it (like the i3s) to keep the thermals reasonable or choose the "performance heaters" in the i5 and i7 line and hammer the active cooling.

The four core '16 pros had tolerable thermals. Then Intel added two more cores. Extra bilge pumps to keep up with the damage wrought by the 14nm shoal. The six core processors were warm, but controllable with a custom fan curve. Then came the toasty 8 core I9. I'm sure that a company like Apple could design a heatsink with five percent better capacity, sourced fans that spin a few hundred RPM faster, and minimized the problem.

What's ridiculous about the Air is, one, the use of a thermal pad rather than thermal paste, two, the lack of a spring loaded mounting mechanism. You'd think that Apple would eat the $.50 for a few tiny springs, double the fin density on the Intel Air's heatsink, and make it less of a battery powered stove.

I'll concede your point about the Pro, but the Air's thermals made it a laughingstock until ASi launched.

 

[jdb8167]

The four core '16 pros had tolerable thermals. Then Intel added two more cores. Extra bilge pumps to keep up with the damage wrought by the 14nm shoal. The six core processors were warm, but controllable with a custom fan curve. Then came the toasty 8 core I9. I'm sure that a company like Apple could design a heatsink with five percent better capacity, sourced fans that spin a few hundred RPM faster, and minimized the problem.

What's ridiculous about the Air is, one, the use of a thermal pad rather than thermal paste, two, the lack of a spring loaded mounting mechanism. You'd think that Apple would eat the $.50 for a few tiny springs, double the fin density on the Intel Air's heatsink, and make it less of a battery powered stove.

I'll concede your point about the Pro, but the Air's thermals made it a laughingstock until ASi launched.

They also have to keep the bottom case within legal limits. Hard to do without a fan.

Watch this Linus Tech Tips video.

 

[theorist9]

Yea, but It can still achieve all of that with a lower thermal envelope. the M1 doesn’t need a 65W thermal envelope to sustain max load. That’s my point.

Nobody has said they are compromising on thermals. We don’t know what the thermal envelope is of the new design. All I was saying is the current mini is designed to handle a 65W processor and the M1 is only 15W. Even if they cut the thermal envelope in half, it would still be more than enough to sustain max load on the M1.

I think everyone agrees with you that the Mini has far more thermal capacity than it needs to handle max load with no thermal throttling, which is why they're arguing with you .

I.e., it's precisely because of this that we can move on from questions of thermal throttling (which AS mostly puts to bed) to the other key attribute associated with thermals, which is quietness. A great result of the Mini having so much thermal headroom is that, even at max load, the fan only needs to spin relatively slowly, making it a a very quiet machine.

Being able to stay quiet under max load is a rare and highly desirable feature that obtains only because the Mini's thermal dissipation capacity is multiples beyond the max TDP. A smaller Mini with an M1X would not have as much thermal headroom as the current M1 Mini, and thus would be louder at max load. Unless you're installing the Mini in a cramped place, I think being extraordinarily quiet is a much more desirable feature than being even smaller.

 

[Jorbanead]

I think everyone agrees with you that the Mini has far more thermal capacity than it needs to handle max load with no thermal throttling, which is why they're arguing with you .

I.e., it's precisely because of this that we can move on from questions of thermal throttling (which AS mostly puts to bed) to the other key attribute associated with thermals, which is quietness. A great result of the Mini having so much thermal headroom is that, even at max load, the fan only needs to spin relatively slowly, making it a a very quiet machine.

Being able to stay quiet under max load is a rare and highly desirable feature that obtains only because the Mini's thermal dissipation capacity is multiples beyond the max TDP. A smaller Mini with an M1X would not have as much thermal headroom as the current M1 Mini, and thus would be louder at max load. Unless you're installing the Mini in a cramped place, I think being extraordinarily quiet is a much more desirable feature than being even smaller.

Correct. But even at half the TDP the mini could sustain loads of the M1 with very low volume. I am an audio engineer so fan noise is something I am very privy too. My comments are merely stating the mini doesn’t necessitate that much cooling, even if you want low fan noise.

The reason I’m saying this is because people are worried Apple is going to compromise on thermals, but there’s nothing that really shows one way or the other. Apple has been known to prioritize low fan noise in the past, but Intel over the last decade proved to not deliver the thermals apple expected. Intel instead ramped up TDP to achieve better performance, but Apple was already locked into a smaller form factor.

By having Apple create their own chips, industrial design and hardware teams can optimize their designs specifically for Apple silicon as they know what’s coming down the pipeline. Before we all worry about thermals, I’d rather stay optimistic and just wait to see what Apple can do with their own silicon.