When I thought about this, my initial first thought was the entire Mac lineup (once fully moved over to Apple Silicon) would get on a predictable yearly upgrade schedule, M1 to M2, M1X to M2X, etc. I mean, there's precedent for this when you look at iPhones.
That has not happened for the iPad Pro's A1_X SoCs that are larger than the the iPhone SoC die size. The the A-X variants have been about 120mm^2 large and Apple has only updated them when a major process shrink was available. 10nm -> 7nm -> (likely 5nm (A14X coming). Major shrinks used to be on a Moore's Law cadence of 18-24 months and that is about where the iPad Pro has been.
More than pretty good chance that this M1 is really going to be the same exact die that they use for the A14X with some features turned off/on for the iPad. ( would explain the port shrinkage on Mini and display limitation on Mini also. Shrinkage on max RAM capacity. etc. ). For example will get an iPad Pro with same core count but probably just USB 3.2 2x2 support (flip off several of the features of the USB4 engine inside to save power and better driver compatibility). Magic Keyboard might also get a boost on port support. [ If not exactly the same , then just some subsystem or two redacted out.. ( thunderbolt controller ) for a relatively minor reduction in size. But pretty good chance the same because it is just easier inventory and R&D wise (and just pass along the cost increment. ) ]
If volume of iPad Pro was not large enough for Apple to be interesting in allocating the resources to be on a yearly cycle then merging the "bottom" of the Mac line up with iPad Pro could 'uncork' that limitation. if volume wasn't the primary issue and Apple has high preference to keeping more resources allocated on long , deeper, more concurrent design pipelines for the iPhone SoC then adding Mac probably won't cure that resource allocation "problem" (preference).
IMHO, I doubt Apple is going get on a yearly time frame for whole line up. Apple may run the part of the Mac line up with iPad Pro overlap faster than the rest, but has go up in die size ( more "stuff" , ARM cores , GPU core , bigger caches, large 'backside' cache for GPU , etc. ) there is likely going to be a slower cycle because both the volume is lower and the upfront costs are higher and harder to get extremely high yields on larger dies. Apple probably won't get into the monstrously high size zone but going from 120mm^2 to 240-300mm^2 is going to substantially reduce yields on bleeding edge process. 300-500mm even more; that is in part why don't see AMD/Nvidia etc on those now for their GPUs / big AI/ML dies .
The farther away the M-series variants gets ( e.g, something for " Mac Pro / iMac Pro " zone as opposed to the MBP 13" four port zone ) the more likely it will get decoupled from the iPhone design cycle. Those systems are updating on slower cycle and Apple probably won't be looking to "speed that up". More users on longer cycle paying for a more "tangent" design.
part of Apple's 12 cycle on the iPhone SoC is limiting them to sub 100mm^2 sizes. The smaller the die , the quicker can jump onto new bleeding edge process and not have to worry as much about yields and costs. ( incrementally more expensive rather than being more substantively more expensive. ) Macs with multiple ports , far more sophisticated I/O requirements basically run counter to that "small size" constraint if Apple is looking to put a performance gap in there also over the mainstream PC SoC market.
After further thought, I'm not so sure Apple is going to want to refresh their entire Mac lineup (even if it is just a processor bump) on a yearly basis. The more I think about it, I think the M1 chips may be on an 18 month upgrade cycle. I obviously have no factual basis for this, but for whatever reason, I just don't think they are going to update all their models on a yearly basis.
Intel dropped 3 Xeon E5/W updates while Apple was in Rip van Winkle slumber on the Mac Pro updates. That wasn't/isn't a CPU availability problem. That probably isn't going to "go away" with M-series either. ( Rumors of a half sized one (probably taking away vast majority of slots being indicative that won't be getting a full sized one again for long time. )
As go "up" the Mac product line up the volumes drop. Apple is generally less interested in lower volume stuff than high volume stuff. The performance gap Apple has at the lower end of the line up where there is high volume is also likely to make them less paranoid about competition at the lower end ( hence slower cycles).
If Apple did a there SoC line up:
M1
M1X - mid range ( MBP 13" four , MBP 16" , iMac 21-24" , upper Mini )
M1Z - high range ( iMac 27" , iMac Pro , Mac Pro (half sized) )
Then pretty good chance they would do a multiple staged roll out all three before going to M2. If they do more.
M1
M1X - mid range ( MBP 13" four , MBP 16" , upper Mini )
M1Y - upper mid ( iMac 21-24" , iMac 27" , Mac Pro "half" )
M1Z - high ( iMac Pro ' > 27" ' , Mac Pro full )
Then might get into some overlap between appearance of M2 ( at the bottom) and M1Z (at the very top ). [ IMHO, the M1Z in that second sequence might not even make the 2 year window they gave themselves. The iMac 27 and "half" Mac Pro would be good enough for claiming completion. ]
The more M-series designs they have to validate and being to market the less concurrent pipeline design they are likely to do. More variants --> slower iterations . Faster iterations --> less variants. In the latter case , the question will be how much can the shoehorn into Mac products without chopping off substantive features.
The likelihood that Apple is doing more variants and faster iterations combination is very slim. That is exactly opposite of their policy they have followed so far of maximum "hand me down" usage to choke off variantions per product. (all mainstream iPads get "hand me down" , All watches $300-6000 get exact same SoC., All next gen, mid-high range iPhones get exact same SoC. etc. ) . The historic Mac product line up breadth runs opposite of what they have done on other Apple Silicon products.
If Apple sells substantially more lower end laptops than during the Intel era and treads water on the desktop sales then I think they would ecstatically happy. Which means they have a lead and probably not too worried about their lead at all. ( AMD , Intel and most of the ARM vendors are focused elsewhere. )
