A very good point that I also wondered about. I think the original leaked files had more info and were maybe even dated but they've all been pulled/deleted, so I cannot confirm.
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 worst case scenario
- Thread starter JMacHack
- Start date
- Sort by reaction score
I dont think we know the exact timeline of validation testing. I could be wrong though.
Edit: Sorry just saw your most recent reply.
Maybe someone has already mentioned it, but Apple has a history of shipping bigger devices than iPhones with older cores than the current iPhone. They either never designed or never shipped A7X, A11X, and A13X. In each case, that led to windows of time where high end iPads were shipping with older CPU and GPU core designs, eg A12X iPad models being introduced with fanfare at a time when A13 iPhones were already shipping.
In practice, of course, these high end iPads were still more powerful. They had more cores, more memory, and better cooling.
Just as with iPads, I don't think it's likely Apple will maintain a 1-year cadence on M-series chip updates. Everything else can and will be delayed as necessary to keep iPhone on track for a new chip every fall.
In practice, of course, these high end iPads were still more powerful. They had more cores, more memory, and better cooling.
Just as with iPads, I don't think it's likely Apple will maintain a 1-year cadence on M-series chip updates. Everything else can and will be delayed as necessary to keep iPhone on track for a new chip every fall.
Very likely.
No way.
No way.
Almost certainly. Memory speed doesn’t matter that much.Bandwidth does for GPU though. It will have a wider bus.
Per core? I don’t expect that at all.
I think it they double the cores it would be at least 80% more performance. Rumor is 16 and 32 core variants.
My worst case scenario is only 32GB of RAM max.
I think it will have 4 ports and more video outputs.
Will not happen because then high-end PC laptops are better than the high-end MacBook Pro’s. I doubt Apple will release a product like this.
The 24" iMac chassis redesign points against the "M1X'. It is relatively clear Apple was out to gut as much thermal cooling capacity as possible with the new case design. ( Ethernet chucked to the power adatper. Can't even get audio jack on the back because it is so thin. ). It is down to about the thickness of the LCD screen panel and see if there is some left over space for the computer part. It is a 24" ipad in terms of primary chassis design.
The other problem for the 24" is price.
i. the bigger die 'M1X' will cost more than the smaller M1 die. The SoC will cost different.
ii. What Apple hasn't replaced from the old 21.5" Mac line up yet is the "edu" , non-Retina iMac with the ancient MBA processor.
When the M2 comes out the "full port" iMac 24" could be moved to the new chip. When "higher end" boils down to a couple more ports there isn't much of a differentiation happening. The more cripple port model could just keep the M1 chip. Wouldn't be surprising if Apple gutted the motherboard of Ethernet support and maybe crank down the screen to pay for a price cut back to "edu" iMac prices. Once Apple has a M(n-1) or M(n-2) part to use they can use that trailing end part to sell a cheaper iMac. To jumpstart the path they are just using the only single one they have now ( and keeping the Intel alterantive around. It is over on the 27" iMac "buy" page, but it is still around. )
This would be very similar to what Apple has down with the iPad. "regular" iPad versus iPad Air. the former is often used as tool to sell to the "edu" market.
In short, gimping the ports gives Apple a path to cut costs later.
As far as "high priority"... 16" before iMac was extremely likley lower priority to growing the M-series powered systems ecosystem as fast as possible. That is way Apple lead with the lower cost M1 systems. The objective was probably to stop buying as many Intel CPUs as fast as possible. That meant taking everything that were the highest volume sellers in the lower "30%" of the line up that could be powered by an M1 and flipping them over almost all at once. MBA and MBP 13' two port were probably highest volume. The Mini was just "too easy" since the DTK had also just used it win an Apple chip also. It probably wasn't doing the volume that the MBA and MBP 13" were doing but it was just move volume to throw onto the pile. ( Plus they could push discounts at the lower end through Amazon, BH , etc. to soak up any production volume estimation adjustments they needed to do. Push a decent discount on the lowest end configuration and drawn in those looking in from the outside the lowest entry price. ) . Between the 21.5" and 27" the smaller one was very likely the highest volume seller. Putting all the volume sellers onto one chip makes it cheaper for Apple. Much cheaper.
Putting a variant of the "M1X" into the "larger screen" ( 27" - 32" whatever Apple picks) will help differentiate the two iMacs ( just like before when separated by CPU+GPU combos. ) That 27' model can better soak up the higher price of the "M1X". And if Apple doesn't go crazy thin on the 27" it would have better thermals too. [ go back more to the box with thickness shape of the XDR and pre "thinnest edge" iMac design shift. so not constrained to stuff the electronics in the chin because there is no room behind the screen for anything substantive. ]
The M2 will be more "powerful" than the M1. The M3 will likely supercede the M2. The iMac 24" can get to "more powerful" models by just following that update track. It really doesn't need the 'M_X' to do that. Apple has been doing lots of spinning over last several years about how the iPad Pro chip ( 12Z , 12X ) were 'desktop' like .
The system that is far more likely to be split M1 and "M1X" is the Mini. There is gobs more Thermal headroom in the current Mini chassis. There is also gobs of money already invested in lots of data centers and racks in custom holders to put lots of Minis into the rack , shelf, custom cabinet , etc. Selling something that just slots right into where it fits makes far more sense that Apple making those folks go back and spend another round of money for new holders and enclosures. Money spent on that is just less money that could be spent on a new Mini.
The more common low end use desktop customers could get a new "thinner" enclosure with better Wi-Fi ( which is next to useless in a rack/enclosure deployment in vast majority of cases ) and perhaps an incrementally smaller desktop footprint. That thinner Mini case goes on the M1 , M2 , M3 track and the classic Mini base chassis goes on the M1X , M2X , M3X track.
That would fail TBv4 certification. A host system has to be able to do video out concurrently on all of the TBv4 ports to certify for TBv4. ( Most of TBv4 is about dropping optional features. One of TBv3 options feature is shorting on provisioning the TB controller on PCI-e and DisplayPort (DP) inputs . TBv4 removes that. )
The TB and HDMI 2.1 ports could share a DP stream and probably still pass. But if the "independance" of the HDMI port was having its own private stream then would need a four stream SoC. Apple probably isn't going to provision HDMI 2.1 either because it is going to be a conversion of a DisplayPort 1.4 streams coming of the TB controller.
The Apple SoC doesn't provide Ethernet. I'd be surprised if they were up for scaraficing larger logic board space and power to a discrete Ethernet controller. Yes, Apple probably got some space back by absorbing other stuff into the "M1X" SoC, but they are probably more interested on "spending" that on bigger battery than logic board space. The other thing is Apple is probably not looking to put a bigger MagSafe connector on the side.
MagSafe is likely going to do two things. First, charge the bigger battery faster ( which also might take up logic board space that a discrete Ethernet controller would occupy). Second, it gets around the fact that Apple gimped the number of TB ports. Dropping one of the TB ports on the right side for a HDMI port means down one TB port. That's the port that could have held the USB-C power supply cable. If also need to use a USB-C Power port now down to just two TB ports. Two ports is "low end" MBP 13 territory, not $2.5K laptop territory. MagSafe solves that problem that the HDMI port creates.
The large battery in the MBP expands out to the outer edges which is the primary driver of the highly limited edge space for ports. The Apple SoC probably reduces overall power consumption but super, ultra bring mode microLEDs are probably a net increase from that subsystem. The increased "bump" in battery life the M1 systems got is probably going to be lower than the bump the "M1X + microLED" systems get.
Extremely unlikely. On Macs, Apple has consistently used the Type-C connector for Thunderbolt. Probably not going to change paths now. The four port MBP 13" and MBP 16" all have TB ports. Going to USB-C only for one of those ports is huge backslide fo very likely zero decrease in the overall system price. Enough folks have grumpbled about SD card and HDMI that dropping one TB ports and still charging the same price would work. Bringing back just regular MagSafe makes that even more likely. That is also a source of complaints and as outlined almost necessary since nuked a TB port.
There are tons of deployed high quality monitors out there with HDMI 2.0. The notion that all of those were just "bottom dwelling PowerPoint only worhty" monitors is goofy.
Apple has a "warmied over, incrementally bumped" M1 coming in the M2 (in 5-10 months. Get outside the initial demand bubble for the A15 and then shift some wafers over to making the M2. ). The "M1X" absolutely does not need to be the "early arrival of the M2" SoC. It is a bigger die that does more.
A TB or USB hub on the inside of the system would use more power than if the TB and/or USB was integrated on the SoC. Apple's SoC program is all about performance/power. There is little likelihood they would pass up a power saving solution that they already have a solution for in the SoC. All they need to do is use a bigger die to Thunderbolt and/or USB-C which they already have.
For the number of display they need some more display controllers and more bandwidth. The display controllers they already have with the limit of 2. Just need more. The bandwidth would come with using one or two more RAM stacks ( and associated memory controllers).
The worst case would be that Apple kneecaps the "M1X" with some artificially low die space increase. The A14 is about 85mm^2 and the M1 about 120mm^2 ( as were previous iPad Pro chips.) . Worse case would be some arbritrary rule that said "can only add 40mm^2 becuase that is gap between A-series and old A_X series. ( can only go to 160-180mm^2 because trying to save a couple bucks. )
Given that these were stolen from the contractor means they are likely real. This wasn't a leaker out to get ad page views or reputation. It was a ransom threat. Can't ask for lots of money for "fake" stuff. Quanta and Apple knew the hackers were coming so set up a super elaborate , fake honeypot to draw them in...... probably not.
If the 14" has a miniLED screen there is no "low end" (as in much more affordable 14" ). The high price of the screen is going to drive the system price higher. It will be similar to when Apple did the original transition to "Retina" screens and kept the "non Retina" MBP 13" around to be the lower price point.
The bottom end of the 14" is probably starting where the 4 port currently starts. $1,799 . Decent chance it backslides a bit on RAM ( back to 12GB instead of 16GB). So that to get back to equal specs on RAM have to move higher to around $2,099.
If there is a "chopped down die" version of "M1X" that will also go into the base 14" model. (e.g. 16 GPU cores versus 32 ). But that is still likely going to land with the $1,799 starting point. ( drop in GPU cores would probably be accompanied with drop in RAM stacks. drop to 3 from 4 ( 3 x 4GB = 12 GB ). )
The "bezels are evil" crowd and/or "gotta have very high HDR" crowd will just have to pay. ( just like back when Retina was new). The dramatic uptick in response to the iPad Pro's mniLED screen update means Apple probably thinks that will work well.
Edit: miniLED for microLED. microLED will be much later.
Last edited:
Looks like low-end and high-end. I'd be happy with M1 performance with more ports and more RAM and the bigger screens. I suspect that they just sell the M1 MacBook Air/Pro 13 as the low-end. It's still a very capable machine at a great price. I do not need the new screen technology myself and probably won't be able to tell the difference.
Apple can do whatever they want to. They're the only game in town with Apple Silicon.
I think one can reasonably expect the base model will be "M1X" with 16 GPU cores, 16GB of RAM and 512GB of SSD.
It would be great if Apple matched the $1799 of the entry-level 13.3" 4-port Intel Core-i5 MacBook Pro, but one rumor says it will be $1899 or $1999. And when you consider the "M1X" will handily outrun the Core-i7 (which is a $200 upgrade) and utterly crush the Intel iGP and the screen is larger and it's MiniLED, I'm fine with the $1999 price (even though many have noted the loss of the Intel CPU and the TouchBar would have lowered the BOM to help offset the BOM increases from the new display and the "M1X" over the M1).
I also think that we will have an entry price of $1999 for the smaller model and $2499 (or higher) for the larger model. Which will be a good price point given that an 8-core Firestorm will easily outperform all but fastest enthusiast-class desktop processors.
Funny, because aside from the os, the only thing that differentiates the iMac from a plugged-in iPad Pro 24 is the lack of a touch screen. I know that’s a controversial subject, but after living with my iPad Pro for a year, I’d love a touch screen iMac.
Oops that should have been miniLED.
If the new miniLED systems have a SD Card slot that probably 60+ % of the buyers will never use it either. The MagSafe port... a large number of folks have uSB-C chargers that can do 85-90W which will probably work (just take longer to charge). There is things like EU standard push for common chargers that will only make that pool grow bigger over time.
More RAM will come with M2 (or M3) pick up LDDR5. ( the RAM dies will get denser. So the RAM stack capacity will go up. ) . Instead of 4GB and 8GB stacks it would be 8 and 16GB stacks. Apple's prices wouldn't change much for BTO upgrades. More ports probably won't come with time (although may get MagSafe tickle down to free up a port with a USB-C charger on it. )
Short term they could do a 'current tech' screen upgrade to 14", but I doubt it. Apple really wants to do microLED in a couple (or several) years. There will probably be some "hand me down" process when that transition happens at the "high" end to the "lower" end of the then "old" tech. The miniLED tech will be cheaper and able to hit the lower system price points without Apple losing margins. Also will need something new to justify the still higher prices that have never moved.
Apple generally doesn't decrease prices , they just throw in more expensive components. ( or swap out new high cost ones for eventually lower costs ones. )
[ Forgot also that the TouchBar screen is probably being swapped out in part for more expensive mainscreen. If miniLED were cheap Apple might have kept the TouchBar. ]
Even with Intel CPUs they were still effectively the only game in town with Macs. Lots of folks grumbled and bought the TouchBar MBPs even though didn't want the Touchbar.
Apple isn't out to sell everything to everybody. They cluster customers onto a just a handful of products. The notion is that they are suppose to do those fewer products better rather than too distracted chasing every individual niche with something different.
Intel has been other there trying to do everything ( dozens and dozens of CPU SKUs ) for everybody. If Intel was more focused they may not have left the door open for Apple to go through. Or at least it would have taken a couple more years for Apple to transition off.
Thermals -- via Fans and hence long term sustain performance on larger workloads. The iPad Pro will likely sag hours before an iMac will. If all have is short run workloads then won't really see much of a difference.
Sound -- Six speakers versus 4 speakers. More ports to attach a external DSP if that is not enough withough using up the single port on the iPad Pro. ( if need to walk and chew gum on ports at the same time).
Displays -- drive multiple high resolution displays in non-mirror mode.
RAM max capacity. Once you buy you are stuck with the RAM you picked , but can pick a larger number with iMac 24". if want to run multiple virtual machines, that comes in handy. if just running on "big" app at a time then won't matter as much.
But yes .... for lots of regular, highly mainstream , everyday stuff I think the long term Apple solution for "Mac with touchscreen" is future version of iPadOS. If Apple 'walks' Thunderbolt ports down to iPad Air then would be a big transition point for iPadOS. Probably not happening soon , but looks like minimally going USB-C.
if the lower model runs up $200 it isn't likely that the 16" increase will be lower. Unless they have keeping the zones constant for the two screens , the 16" screen is a bigger 'pain' to construct and run. It won't be cheap.
And if the 13" has a different chopped down die then it is a different "M1X" SoC. BOM cost on that is going to be higher on that also (and Apple will apply a high mark up to a larger number which will be an even larger number. ). Perhaps the entry 16" model will have same smaller die, but there won't be a performance gap between the two models.
I suspect Apple will be more willing to shift the higher priced model higher rather the lower one. The lower priced model already have a target market that is price sensitive. That is wrong place to apply a relatively larger price increase. If the 16's price increase is lower it seems more likely there is some bigger user value drop off also.
If your workload doesn't need lots of RAM. Or actually need all 16+ threads. Apple cores are faster, but 2x faster? Or if throw a 32 core Threadripper 5000 at the workload; 4x faster? A Ryzen 5950X and desktop Nvidia 3070 is probably going to win on a wide variety of workloads for pretty similar price range. It won't be mobile but it won't get "crushed" in a workload contest either.
Go around 'Kicking sand' in Gen 11 ( Rocket Lake) CPU's 'face'. Yeah sure. But otherwise Apple is bring half the cores to the 'fight'. Intel Gen 11 isn't the only competition in desktop land.
I'm sure that only point to their own relatively stale Desktop stack and to $600 BestBuy discount special boxes that sell in volume in Windows PC world.
A 3970X 32-core runs $2,900 and TDP of 280 Watts. I personally don't know anyone with this or similar class chips. 5900X and 5950X is more common. I've considered a 5950X myself but securing a suitable GPU is a pain and Apple will make life a lot easier with a good iGPU.
I don't think that the M1X is going to beat the high-core AMD CPUs but the M1 and M1X are Apple's first-gen CPUs. That we are comparing high-end AMD CPUs at all to their first gen products is quite remarkable.
I think that it's going to be very difficult to best the M1X in CPU horsepower in mobile chips. I think that M1X will get to mid-range and maybe even the bottom of the high-end in GPU. All while sipping power. We will have to wait sometime before seeing what they do with the Mac Pro. I expect that they will want to exceed the performance of the 18 Core Xeon. There is some speculation that they will use a couple of M1X chips tied together by some fabric but I don't know of the Apple Silicon architecture is designed to accommodate this.
Both of the TR 3960X and TR 3970X processors have a list box TDP of 280 W, which is a new ‘record’ for high power consumption in a consumer CPU. In the enterprise space we see some specialist processors break the 400W mark, but those CPUs exist in environments with a variety of cooling methods and sound isn’t much of a concern. Conversely, these AMD processors will have to live in a box under someone’s desk, so there has to be a point where the TDP is too much. Last AMD generation was 250W, this one is 280W: if we’re not there already, then this should be a practical limit. AMD of course recommends liquid cooling with a good pump and a big radiator, so anyone buying one of these processors should look into spending at least another $120+ on a good liquid cooling system.
The same rumor that said the 14" would be $100-200 more expensive did state that the 16" would hold the line on price and therefore would remain $2399. I presume the BOM savings from losing the Intel CPU and AMD GPU along with the Touchbar either fully alleviate the price differential of "M1X" and MiniLED.
And if the rumors that the 14" will also have the 32-core GPU as BTO, then the only difference between the two models will likely be the screen size and battery life (from a larger pack) so $400 is likely the maximum delta the majority people would accept and therefore Apple would not try and squeeze another $100-200 more like they (evidently) feel they can with the 14".
And if the rumors that the 14" will also have the 32-core GPU as BTO, then the only difference between the two models will likely be the screen size and battery life (from a larger pack) so $400 is likely the maximum delta the majority people would accept and therefore Apple would not try and squeeze another $100-200 more like they (evidently) feel they can with the 14".
Last edited:
Not just the GPU, but have given up another 4GB of RAM also. So go from an aggregate RAM footprint of 20GB down to 16GB. Holding the line on price. Backsliding on capacity. If crank the capacities down then easier to hold the line. In general if you get less you should pay less. [ Similar issue with the entry M1 Mini mini dropping in price at first glance. Up until look at features lost. ]
'Spin' is likely going to be that the VRAM data was all duplicated in the main system RAM so it doesn't really count as a capacity loss. There will be some overlap, but saying it is 100% is probably too far. If gave it a 60% overlap, then it is still a 1.6GB backslide. At the prices Apple charges end users for RAM, that could "pay" for a lot.
The reasoning is that the 14" and 16" have the same hardware, the only difference being the larger display. And while you are absolutely right that the larger screen is significantly more expensive to construct, Apple will also be saving some money on the third-party dGPU. Anyway, it's all just speculation.
Oh, sure, but an 8-core Firestorm should hold it's own agains the likes of AMD 5900X or Intel 11900K. For a compact laptop with more than a full day battery life, that is going to be a tremendous amount of performance.