Oh I agree the iMac Pro is not going anywhere. Apple will be able to position the Mac Pro and Pro Display above it (especially if it is ~32" and 6K resolution). But I just can't help but believe the iMac Pro was meant to replace the Mac Pro at the top of the line-up. But the W-Series Xeons were the first that could actually work in an iMac chassis, so perhaps it always was a case of "we will do a new Mac Pro eventually" and the iMac Pro became a "hey, we can make a more powerful model that better meets the needs of some of our Pro Software customers than an Core-i7 iMac" as a complimentary model.
the "iMac Pro' was coming sooner or later anyway. Wheather in 2017 or 2019 it was coming. AMD getting back in the game versus Intel meant that something like the Rzyen 9 ( >10 cores in a ~100W TDP envelope ) was going to happen by 2020 in the x86 solution space.
It should have been clear in 2015-2016 if looped into the advanced R&D that the very high core count was going to shift to packaging multiple dies.
Apple did the iMac Pro because
i. It was easier. ( somewhat like slapping a A10 into a basically unchanged iPod Touch was easier. Avoids an "start from scratch" iteration through Industrial design choke point. )
ii. Because it was in demand from the iMac user base and that kind of "horsepower" was coming anyway.
[ that the RAM door disappeared an they could milk the RAM pricing was probably gravy on top. That made the product even less risky and more likely to meet Roi targets. ]
The core issue was that the 2021 iMac would probably be the iMac Pro 2017 in just a few more years.
All of that doesn't negative where the Mac Pro has traditionally sat in terms of performance coverage and expanded utility in configuration. The iMacs would be capped by the same basically sub 500W power lmit and internal volume constraints as the Mac Pro 2013 ( which didn't do as well as they wanted. It did OK to show that the iMac Pro could do well enough but a substantive number of folks refused for years to come into the space. )
So Apple covered the space they were already doing OK with. However, that doesn't mean it was the "replacement" for what they largely weren't covering well. So they laid out a less risky extention from the iMac to iMac Pro. Once they did that they would lay out another less risk extension from there incrementally farther. IMHO, probably same path of using some component and basaeline board/firmware work to 'share' some costs . (e.g., add power range to Mac Pro and cover that with more.. storage capacity , GPU dual , RAM capacity ( 4 more DIMMs) , and better CPU TDP flexibility.
This would be a longer redesign so it was queued up after
Yes, I could see Apple maintaining a very small "new products" team within the Mac Products Group that are only able to fully redesign a product every three to four years as they alternate amongst them. So each year they focus on one Mac family per year and that is the one that gets the redesign:
2013: Mac Pro
2015: MacBook
2016: MacBook Pro
2017: iMac Pro (which is perhaps why the iMac has not changed - the team spent their time on the iMac Pro, instead)
2018: MacBook Air and Mac Mini (they had the time to do both since the iMac Pro and Mac Mini were not full redesigns)
2019: Mac Pro
(2020 will probably be the MacBook ARM and then new MacBook Pro in 2021, followed by new iMac/iMac Pro in 2022)
But that is the primary point in that there was
two cases that needed executive buy in. One is to keep the Mac Pro in the line up. Second is that the Mac Product line needs more concurrent and pipelined development. Same choke point so any major screw up that slipped in has even bigger blowback for a longer period of time. Apple waiting unitl 2021 to et their MBP faux pas fully worked out is risky. MacBook ARM ( same thing ... would be much less risky if forked out a group to run that in parallel with keeping the x86 Mac stuff moving while ramp up the ARM).
That tension of spreading folks out too thin is exactly the tension of why there is usually a push to kill X if add Y. PCs are evolving slower now so not quite as "thin" now as compared to the pace 10 years ago.
2005-2006 Apple flipped the whole Mac line up in 12 months. ( some rumors that the Mac Pro was largely 'outsourced' to some bodies/resources that Intel provided. )
Or is indicative that they will announce it at WWDC so no need to do so two months prior.
From the prospective of buyers who started their clocks in 2013 ( devoted to skipping the 2013 iteration) the product is 6 years late. If Apple was pragmatically almost completely done with Product verification there is zero rational reason o add 'extra' months to the reveal. None. It is grossly late. At that point they are not "building anticipation to delight". It is far closer to managing disappointment. They were 4 years late when they first started talking about it. It was going to be minimally 2 years they should have say "not this year" , but " at least 2 years out". The plan should have been to release just as prudently soon as they could. Not wait for some relatively arbitrary date.
Delaying to WWDC only makes any remote sense in that context that they were not done months ago and have perhaps barely squeaked close enough to either run a rigged demo or a static "not pure vaporware" demo.
If they were "enough for a demo" stage done 2-3 months ago what rational reason would they have to sit on it?????? They are over a 1,000 days late ( triple digits ). Running it up to (or over) 2,000 is a bigger 'win'?
If they are not at the "enough for a demo" stage now what would they gain by showing some mostly incomplete system at this point? That would aid in their competence reputation? Or diminish it. ( especially after the AirPower SNAFU and 4 iterations to a working keyboard drama. ).
Apple waited until the "last" moment to cancel the AirPower. If the new Mac Pro schedule is cluster-screwed up then they'll probably wait as long as possible to reset the expectations for an even later 2020 product. That wouldn't be done with fanfare. It is about equally likely they are going to serve up yet another "dog ate my homework" moment as do any grand demo. Apple ducking that April anniversary meeting time probably means that they were hiding "bad news" .... not triumphant.
They could have come out and said " No problems with the roll out schedule at all so far. Most definitely this is a 2019 product and you won't have to wait too much longer to see. Allocate your budgets at some sane point points because you'll want to spend money on a Mac pro this year. We are dropping the Mac Pro 2012 onto the Vindtage list because were are moving on. " that would acutally build hype for grand WWDC extraganza.
Apple has been doing 100% the opposite. Almost radio silence even less than a week before to build hype for WWDC grand reveal. If it is somewhat bad news that would line up that approach. Burying somewhat bad news would help protect the WWDC build up.
It is more or less what we got with the 2013 Mac Pro and 2017 iMac Pro: they showed a deck of slides with CG renderings during the keynote with a physical unit in the Hands On Area afterwards. I believe the 2013 Mac Pro was just an unpowered display unit under glass and while the iMac Pro was powered on, I do not believe anybody was allowed to do anything with it.
The gross disconnect is that the iMac Pro was not 4-6 years late. Even the MP 2013 wasn't 4 years late. ( Intel had no Xeon E5 in 2011 ( 2011 was a gap year for just about everyone in the higher end workstation space) so missing June 2012 was were the 'late' clock reasonably started for the Mac Pro in that timeframe. ). The MP 2013 was in the range of being only about a 12-18 late in debut. [ it was 18 months late in terms of shipping to a competitive market ]. Saddling the MP 2013 with TBv2 made it more so a 2014 product.
Apple didn't do a "repeat" of glass exhibit with the iMac Pro. That glass case wasn't a "good thing" in 2013. It was 'necessary' more than good. It isn't something to explicitly target for revival.
Well then the 2013 Mac Pro and 2017 iMac Pro as shown at their respective WWDCs were also "vaporware" as they were not shown on stage as a physical, working product and no one was allowed to actually touch them to benchmark performance after the keynote. I'm not sure they even showed actual working units to the media before the models shipped their respective Decembers.
Technically yes, vapourware ..... which why Apple would be slavishly trying to repeat that is dubious. AirPower blew up in their face. They grossly stumbled HomePod out the door after doing a big dog and pony for it.
One o the major things that Apple has done in the Mac Pro space is fumble away user confidence in their ability to execute. As much as folks rag on Intel's stumbles on 10nm this is actually worse ( Intel isn't 6 years late). This might work ... just don't blow on it too hard ( it might fall down) isn't going to build much confidence outside the 'cult zone' folks. Apple has screwed up enough stuff lately that some half-done demo is just that half-done.
Minimally, it needs to be at the point where the iMac Pro was. If they don't have that, then it probably isn't worth the trade-off of showing it.
