That last "only with Apple's modules" is extremely unlikely. Apple's representative example of modularity was with a "modular display". That was explicitly mentioned in their talk. Aside of their brief foray into the Apple Display Connector , they have used standard connections. ( yes mini-DisplayPort is a standards connection that Apple favored. Thunderbolt is defacto a standard. etc. ) So inferencing that they are going to deeply commit to exactly 180 opposite of one of their few explicit examples is a huge leap.
The iMac Pro has standard so-DIMMs even though you can't easily get to them. The Mac Pro 2013 has DIMMs. Standards which don't get in the way of their requirements/objectives aren't necessarily thrown out willy nilly.
Apple has a modern mantra now of security and privacy so yeah the boot SSD controller will probably be soldered onto the board. If the storage is not the primary boot device then the more mainstream standards have a shot.
Apple has talked about Thunderbolt being high throughput versus its other standard port competitors ( Std Serial Port, USB , Firewire, PCMCIA/ExpresCard/PCCard , etc. ). However, they haven't couched it as being the high end of throughput.
In the pow-wow session
"... so you look at that architecture of that Mac Pro, it had great Thunderbolt external I/O and we said: ‘This is a great opportunity to change what had been a conventional build a big card rack and slot a bunch of cards in there.’ We said: ‘a lot of this storage can be achieved with very high performance with Thunderbolt. So we built a design in part around that assumption, as well. Some of the pro community has been sort of moving that direction, but we had certainly in mind the need for expandability. If you wanted a great RAID solution in there, it probably made a lot more sense to put it outside the box than actually be constrained within the physical enclosure that contained the CPU. So, I think we went into it with some interesting ideas, and not all of them paid off. ..."
Some bets they made on Thunderbolt paid off in the 2013 design and some did not. Thunderbolt is neither completely a looser ( as some on this forum keep trying to characterize it as ) nor is is 100% covering all of the highest end solutions ( which is another far-out fringe keep characterizing it as also). Thunderbolt is more than "good enough" to make it into the next Mac Pro. It has a bigger impact on the rest of the Mac line up, but that role doesn't have to go down to zero on the next Mac Pro.
The one, and only one, storage drive would hopefully be in that scope of "not all of the paid off" camp for the next Mac Pro. That doesn't necessarily mean empty SATA bays and dangling power cables, but some additional drives inside would help ( some M.2 slots , some 2.5" sleds , etc.) isn't spelled out. Given how higher end M.2 SSD drives can saturate x4 PCI-e v3 lanes something outside of Thunderbolts scope would likely have a bigger pay-off than slapping yet another (past 4 TB ) sockets on a New Mac Pro.
They are very focused on eGPUs because probably 85+ % of the Macs they sell can use them. (a decent chance that's in the 95+ % range ). Mac Pro are a single digit now and extremely likely single digits back in 2009-2010. A revised Mac Pro isn't going to change that ( probably would be doing well to get into the 1-3% range. )
So yes. Display docking stations and eGPUs will probably be in the Apple store because most of the Macs they sell can use them. ( the next Mac Pro would only be an incremental addition to that group. The Mac Pro probably won't be the main driver of 'display' sales. That's a tail wags dog notion. Apple does display so 'need' Mac Pro to be the main driver. )
Although I can see how you might get to that position from what I wrote, you are reading far more into that. There is a difference between the full range of "race to the bottom" commodity stuff and the space than the space a substantive number of "Pros" in the Mac space have cards for. For example, Sonnet has a list of PCI-e cards that work with their Thunderbolt expansion enclosures. The first column on that list is "Mac". ( some of those don't have Windows check boxes... only as a point that there is more than the lowest commodity in cards is not the whole market. ) Pragmatically that is the older Mac Pro 2009-2010 (models and a bit earlier in some cases ). Once those all lapse into Obsolete status at Apple that column pragmatically disappears. That's wouldn't be a good thing for Apple in trying to sell a future Mac Pro.
More than likely Apple already has a "slot usage" survey that is at least a good (and probably better) than the "What's in your slots " survey on this forum. Some things folks used slots for would be better covered with better internal hardware ( modern SATA 6G bus , M.2 slots , USB 3 Type-A sockets (by the PCH that wasn't from 'stone ages' ) , etc. ). But an open slot or two would cover a wide sprectrum of those folks without having to jump into an Express III D ( some MBP users will 'get by' with that, so just pure slot count 'war' isn't necessarily a big winner. )
The six years don't mean a whole lot in "PCI-e slot" ware rhetoric. They didn't build any non PCI-e slot having Mac Pro in that span either ! The bulk of that six year span is doing a whole lot of nothing. That doesn't build presence to non-slots in Mac Pro space.
It would be entirely different had Apple churned out 2-3 non PCI-e slot Mac Pro systems in that six year span. That would be setting precedence and groundwork. Do nothing means little more than needing to fix their "Rip van Winkle" problem. Not a PCI-e slot war mandate problem.
If the uncouple the GPU and CPU thermal solutions ( likely since they said that was a problem) then the CPU and primary GPU solutions can be distinct and they could more readily do updates. That leaves the resources for a 3rd x16 slot at least as decoupled from the other two. That should open more flexitlibity in the configuration options, not less.
The impediment constraint that may lead them to leave the '2nd' x16 lane resources unused if they are still fixated on a literally desktop solution with a relatively small footprint. If they are willing to go deskside again that blocking out some power ( ~2800W ) and volume ( 2-3 slow widths ) for a slot wouldn't be a problem.
All of that wasn't a mistake. Some of it worked and some of it didn't. Apple explicitly said so. Trying to huff and puff until Apple admits that every aspect was a mistake .... Yeah that is going to be a very long wait because they have already blown that notion up. It wasn't. ( Apple's Mac sales didn't materially get impacted. They sold Mac Pro. They have sold iMac Pros and iMacs to former Mac pro class system buyers. )
The Pro Workflow exercise is about "pro" Macs. not the Mac Pro solely. Surprise, surprise, surprise Apple cares about the Mac ecosystem as a whole more than just only one specific Mac product. Apple clearly outline that their pro sales are e MBP , iMac , and sngle digit Mac Pro in that order of magnitude of sales.
Second, they described next to no details about the new Mac Pro at all. The vast majority of what they talk about over these two sessions was what they had done and are doing. Besides dispelling the notion that they have completely walked away from the Mac Pro space, they didn't violate the long term corporate practice of talking in detail about upcoming products. They didn't. Most of the folks yelping about that is just projection; often more so of what they wanted to hear rather than what was actually said.
Yanking out the default boot drive and display probably. However, with Thunderbolt PCI-e expansion enclosures and Thunderbolt making those PCI-e equipment pragmatically transparent to the software, macOS has to handle 3rd party cards anyway.
It really isn't about tight integration. It could be about saving volume (and power). But that more so could be limited to lopping of the 5.25" drive bays ( Opticals extremely likely aren't going to make the cut) and perhaps spinning metal all together.
About 3:42 ( where starts talking about transitions and being abrupt ... )
'they call us crazy. ... picking horses to ride .... "
[ Resources in the Scrooge McDuck money pit and resources assigned to the Mac Product are not the same thing. ]
" we are going to leave it out .... some folks are going to call us names. ... they pay us to make those choices '
then about 11:30 -13:40
There is huge conflict in Apple worldview in that folks are paying Apple to choose nothing about a system ( i.e., to sell an empty as possible shell ). That isn't Apple's objective.
I can't find the video with the quote from Jobs about IT folks 'being often wrong" ( it is all things digital interview by Mossberg ). Similar with the notion that when people buy then great (get to come to work next day) and don't buy then (don't if longer term). That people who indirectly buy for others ( or repair/service for others ) aren't necessarily 100% aligned with end users.
The other vendors will copy it because they are trying to be everything to everybody. That is not what Apple's trying to do.
[doublepost=1534200085][/doublepost]
Apple has a metric of measuring what people buy (and don't buy ) in repsonse. So their "worked OK" has a customer component in it. The farce is that the MP 2013 and iMac Pro have utterly and completely failed int he first year or so. Not really true.
Got numbers to back that up? As a percentage of Macs sold it was probably trending down even in the 2009-2010 era. If Mac Pros were growing at a equal rate to laptops, Apple wouldn't have had a huge lull from 2010-2013.
Folks "protested" the new Mac Pro by not buying and Mac sales never saw a big drop in sales.
Chuckle... a fair amount of the yelping is about old stuff. 5.25" drive bays and SATA power/data cables. a phypsical specification that dates from almost two decades ago. etc.
