Nope, alumina is what I meant-- substrates are dielectric. Amorphous alumina has very good thermal properties. There's also sapphire (crystalline alumina), aluminum nitride, etc.
Last edited:
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.
M1X Mac mini's Leaked Schematics. Fake or?
- Thread starter sunny5
- Start date
- Sort by reaction score
The paragraph after he talks about it being “bad” to put SoC on inside he mentions complaints of the cell modem throttling . In the ifixit taerdown the modem is on the inside . The NAND storage chip seems like it would cover modem to some extend . Heat from to SoC can so leak through bottom to modem and RF chip area .
If go back to iPhone 11
iPhone 11 Teardown
For the first time ever, Apple unleashed three new iPhones all at once, making this officially our busiest teardown day ever. We focused most of...
www.ifixit.com
suggests that he is alluding to general practice by Apple ….
If the cellular modem is throttling next to the NAND , then pretty doubtful it would work well with the A15 on top of it .
PoP is generally used to describe stacking a memory package on a SoC package. It isn't related to PCBs. (Though, to make things extra confusing (imagine me with an evil grin), most of the packages in question are based on organic substrates, AKA advanced types of PCB.)
I think that's just a sticker Apple puts on the outside so you can see something marked as A15. Other pictures from the iFixit teardown show the actual A15 SoC package on one of the inner sides of that PCB sandwich.
It's two PCBs stacked and interconnected using a third PCB as a combination interconnect and mechanical spacer. Here's an image showing one PCB plus the spacer PCB the from the iFixit teardown of the iPhone X, the first model where Apple used this design:
Ah that makes more sense. Step 7 is both bottoms . Step 8 both tops ( the NAND chip is blurred to mask the confusing “A15” label. Kind of cuurio had the decipher the vendor if Apple painted over chip )
Very cool! Er… cozy. Well, plenty of heat will dissipate through the vias. I wonder what substrate they’re using.
Good point on the vias. Even if Apple didn't bury the SoC in this kind of board sandwich, they might be the best thermal path. After all, A13/14/15 etc are space-optimized phone PoPs with an entire DRAM package piggybacked on top of the SoC package. If you attached a heatsink to the top of the SoC package, it would have a poor heat conduction path to the SoC die - there's DRAM package encapsulation material, substrate, and DRAM die to get through.
That is exactly my thought pattern. Any thoughts about what substrate they’re using?
Thinking about thermals, I looked at ceramics at first, but it’s dark brown; reminds me of phenolic composites, like old school impedance controlled dielectrics. I wish more youtubers had X-ray spectrometers— maybe someone can airdrop an iPhone to Cody Reeder.
I had 2 minis that went kaput due to heating issues , why on earth are apple obsessed with form over function making everything thin ?
I can understand it in portable items like laptops and iPads , but a desktop ?
I can understand it in portable items like laptops and iPads , but a desktop ?
I am not sure whether results from Nexus 6 translate to the latest iPhones. Neither do the authors offer the definition of "thermal throttling" or any source for their claim (they mention the PhoneLab paper but I don't find any discussion of throttling there)
One could easily test this: like do a long sustained performance benchmark (e.g. Wild Life Ultimate) on an iPhone in a warm and in a cold environment and see whether there are significant differences in the reached equilibrium levels.
Well, wait a minute, that is not the same at all. Thermals of the trashcan Mac Pro were excellent and their cooling system nothing but beautiful. The "thermal corner" referred to the generations of the CPUs/GPUs that came afterwards: Apple apparently didn't anticipate the average TDP of high end chips to go up as it did. If you want, the problem of the cylinder MP was the inability to scale, it was a strategic mistake. For its contemporary hardware it was perfectly adequate. No engineering mistakes, managerial ones.
Edit: the cylinder MP had no problem cooling down a 130W CPU and a 250+W GPU. But later higher-core Xeons have brought that up to over 200W and the W6800X Duo has a whopping 400W TDP...
Last edited:
The last board design I did was probably about 15 years ago, so I'll bow out of speculating about substrate material. Wasn't too involved with substrate choice when I did it, and it's been so long I've forgotten what little I knew.
I’ve been going down the RF “Signal Path” recently, so substrates, cavities, and physics in general have become a recent passion (resulting in many “Noise Paths”). So, full disclosure, I identify as an enthusiastic engineer at heart.The last board design I did was probably about 15 years ago, so I'll bow out of speculating about substrate material. Wasn't too involved with substrate choice when I did it, and it's been so long I've forgotten what little I knew.
Last edited:
Here are the results they show for performance on an unspecified machine-learning workload as a function of ambient temperature. Yes, it's unacceptably lacking in explanatory details (the IEEE peer reviewers were asleep at the wheel here), and the paper should certainly have been better written—but I don't think we need a precise defintion of "thermal throttling" to understand what's being shown here: With the stock Nexus phone, there's a drop in performance when ambient temperature exceeds 25C, due to thermal effects (and the overclocked variant is even more sensitive to ambient temp, as one would expect):
And yes, a Nexus isn't an iPhone, and it would be good to have iPhone-specific data, which I don't. But let's be reasonable here: Isn't it more plausible than not that, if an iPhone's performance on a task is thermally limited, this performance is going to be more reduced at 32C than at 4C (to use the example ambient temps I gave in my last post)?
I think we agree here. I said in my last post that these aren't engineering mistakes--the engineers are smart enought to know what's right. I'm saying these are design mistakes--not implementing their best engineering.
[I do grant you that the Mac Pro example didn't correspond to what we discussed in specifics (though it did in a general sense, which is why I supplied it--but never mind that; arguing that would send us down a rabbit hole
).]But I did provide another example that did correspond specificially to a poor thermal design decision (not for the future, but for the current product): The large iMac. Why didn't you comment on that, since it is more directly germane? The iMac's high noise levels under load obviously compromised user experience, and could have been addressed by an easy, low-cost engineering modification--which Apple didn't implement b/c of a design decision, rather than for engineering reasons. And if Apple made a suboptimal thermal design decision for the iMac, couldn't they have also have made one for the iPhone?
Just to clarify: I acknowledge we don't know enough to conclude that what Apple did with the iPhone is necessarily a poor thermal design decision, in spite of what Anandtech said. [As you rightly noted, Apple knows far more than we do, so what may not make sense to us may have been done for a good reason that we can't see with our current fact set.] [Though I regard the Anandtech writers as experienced industry professionals, so they wouldn't have written that without good reason.] What I'm saying that one can't simply conclude the opposite, i.e., that "It's Apple, so it must have been a good decision." Apple's decision are made by humans, and humans can make mistakes and misjudgements.
Last edited:
I've seen dogs that, when unleashed, promptly sat down.
If you see a "promise" in a marketing graphic, well...... I guess, for marketing folks, it's great that someone does.
With all due respect, I don't need a snarky lecture presuming to instruct me on what marketing means. I thought you were better than that (in the past you've been friendly and collegial; did someone just piss in your cereal?). And I've probably given more detailed, substantive criticisms on here about Apple's marketing BS than you have, including Job's cherry-picking of Mathematica benchmark specs when he introduced the first Mac Pro (https://forums.macrumors.com/thread...1.2240395/page-31?post=28570835#post-28570835), and their ridiculous comparison of the Pro Display XDR to a Sony HDR Mastering Monitor (https://forums.macrumors.com/thread...e.2258421/page-14?post=29053436#post-29053436).
Having said that, you can infer what sorts of claims Apple plans to make in their marketing from how they initially present it. And thus far, unlike in other areas, their marketing claims for the performance of the M1 have generally been conservative relative to actual performance. Thus if their marketing suggests they're planning to make strong claims, about an area in which their claims have thus far been conservative, that's intriguing.
I.e., you could say there are three levels of understanding of marketing:
Level 1: Those that simply believe it (i.e., not knowing the "rules" that you can't trust marketing)
Level 2: Those that simply disbelieve it (i.e., knowing the rules well enough to follow them)
Level 3: Those that understand the marketing games well enough to know when information can be gleaned in spite of the games (i.e., understanding the rules well enough to know when they can be broken).
I'm focusing on Level 3 here, while you're stuck at Level 2.
Last edited:
Oh, I agree that this is very likely. The question is what kind of performance delta one will get. There is still the hypothesis that the thermal constrain lies in the propagation of heat through the PCB rather than heat dissipation from the phone (in which case the sustained performance should not change that much with the ambient temperature). This is really something that should be tested, otherwise we are just throwing stones into water here.
Ah, sorry, I have overlooked it. I mean, I can't speak for Apple, but the simplest explanations that they didn't feel like doing a major internal redesign for a model they knew is approaching end of life anyway. The 27" iMac was released around 2015 if I remember correctly and it's thermals very absolutely ok. I can definitely imagine that they are getting more constrained with the hot running Comet Lake CPUs, but frankly, should Apple do a redesign every year because Intel can't keep it in their pants... and I mean thermal profile?
Promising ≠ promise lol
But their marketing historically often hints at something related to the event. It’s not always just random graphics and marketing copy. Looking at some past events:
- September 12, 2018 - "Gather round" Round = Apple Park
- March 25, 2019 - "It's show time" Show = Apple Tv+
- September 15, 2020 - "Time flies." Time = Apple Watch. (Graphic was a nod to iPad Air)
- October 13, 2020 - "Hi, Speed." High Speed = 5G iPhone (Graphic also invoked homepod + 5G waves)
- November 10, 2020 "One more thing." Graphic was a nod to Macbooks opening. Phrase is only used for very special announcements - like Apple's first AS macs.
- April 20, 2021 - "Spring Loaded." Loaded = Apple announced 5 new products. (Graphic was a nod to iPad Pro and the new iMac colors)
- September 14, 2021 - "California streaming" Streaming = Apple TV+ shows streaming in the fall. Event used locations around California.
So it's not too odd to try and figure out any sort of clues or hidden meanings in their marketing. The fact that Apple is using an obvious Star Wars hyperspace image means "something" is going to be fast/unleashed.
Great take… but I think you meant Star Trek. Either way, thanks for your analysis. Many 👍
LoL, yeah.
Of course they are going to release something faster than the one year old low end Mac chip. I wouldn’t try to convert jumping-into-hyperspace imagery to GeekBench 5 scores or predictions about memory types though. 😀
Exactly. That graphic suggests this event is going to be heavily performance-focused (as opposed to, say, feature-focused).
Last edited: