The 10-core Comet Lake-S doesn't obviate the need for up to an 18-core iMac Pro...Apple still needs to update the iMac Pro to the Xeon W-22xx-Series CPUs among other updates. I do wish there were some benchmarks to tell us how much the potential increase in horsepower we could expect, but I have found none at this point.
Intel has slashed CPU prices, contract NAND is cheaper, but prices are expected to go up, contract DRAM is cheaper, but prices are expected to go up on that this year as well. The possible GPU replacements for the iMac Pro update is the biggest unknown to me.
If Apple choose to button up the iMac, I think it would be a mistake. I am hoping that Apple introduces a redesigned chassis that allows memory access for both models, takes XDR Display design cues, etc.
Apple might lower the price of entry of the iMac Pro, but my bet is that they will start at 10-cores (W-2255) at $4999 as the iMac now maxed out goes to $4849, and a future 10-core CL-S model should cost the same.
A de-specced iMac Pro is a non-starter at this point.
Let's hope Apple gets on the stick with new ID for the iMac and iMac Pro. The Pro is an important bridge between the iMac and the Mac Pro and the iMac 21.5" and 27" are important to Apple, but really long in the tooth.
10-core Comet Lake in a regular iMac could enable Apple to quietly kill off the iMac Pro. The reason why I think it should be kept is because of that status as a bridge machine for professionals who can't stretch to the Mac Pro.
10 core is probably going to remain the sweet spot for iMac Pro but Apple still need an entry level device, it's more logical that 8 cores would continue to be that entry level - especially as the Mac Pro also starts with 8 cores. A $4k 16Gb, 512Gb SSD 8 core iMac Pro with an AMD 5500 Pro GPU should surely be of interest for people who want a bigger screen for example. Leave the $5k machine for users who want to continue with 32Gb, 1Tb SSD, 10 cores, and AMD 5700 Pro.
Apple probably hedge NAND and DRAM prices with their massive bulk orders - their pricing of the 16" MacBook Pro certainly exhibits their buying power after they effectively doubled the storage on both of their SKUs without increasing prices and could in theory allow Apple to ride increases in NAND prices.
Let's also not forget that Intel have slashed CPU prices thanks to AMD's Ryzen although the price of 4k and 5k panels must be coming down too.
This is my thinking behind Apple having the budget to do something interesting within the current named price points of the iMac and iMac Pro because they have to add a T2 CPU onto the motherboard which will cost extra. The released budget from DRAM, NAND, and CPU savings could add up to an all-SSD design for the regular iMac.
The issue with the W-22xx series preventing it from being a slot in replacement for the W-21xx CPUs in the current iMac Pro is the increased TDPs on the higher end CPUs. Now Apple could mitigate by under clocking but they'd be putting the package into an ageing enclosure with inevitable remarks about 'massive bezels' on the 27" 5k panel.
So, in conclusion, if iMac Pro remains a 27" screen with workstation Xeon parts they may have to redesign for the extra heat produced by the W-22xx CPUs and powerful AMD GPUs going into it.
I'm still thinking a more practical and popular normal iMac would be a 24-25" with the same resolution as the 4k iMac (just a bigger screen) which would contain the Comet Lake parts - perhaps even high end Comet Lake H parts if Apple were perhaps thinking towards a design that let people more easily mount on a VESA monitor arm rather than a full-on space heater.
I think there’s a market, but I don’t think there is any will within Apple. A double-wide mini with the 5300M/5500M would sell. A 2-3 PCIe slot Mac is even less likely. I have yet to understand why Apple has treated the Core X-Series like it has the plague, but I cannot for the life of me understand why. Again, I just don’t think anyone within Apple can get either one of these models past MarCom and the executive suite.
Attacking things from the mobile end would be likely. Apple still needs a 15” MacBook Air, 14” MacBook Pro and quad-core standard on the Air. This dual-core nonsense is pitiful in 2020.
Intel's X-series processors have to be in serious danger from AMD's higher end Ryzen CPUs. The heat (and subsequent cooling noise) they put off is very high and they would present competition to the Mac Pro and iMac Pro in horse power terms.
Surely more interesting would be a Cube Mac with Comet Lake H, S, or even T parts that can be cooled to silence (a third of the TDP budget of some X series CPUs) but come with an onboard GPU which would preclude the need for expensive external eGPU. I'm thinking a headless MacBook Pro 16" with GPU on board and upgradeable RAM in a bigger case that optimises silence over performance.
Top SKU would be 8 cores, 16 threads, 3.1GHz base clock (4.98Ghz turbo) (i9-10980HK) - pair that with AMD 5500 Pro or better in a Cubic case with big fan cooling it and plan to refresh it every 2 years.
Most of Intel's 10th generation mobile products are quad core now, right down to the 15w parts so there would be plenty of threads even at entry level.
Also, I think the real proof in the pudding, so to speak, is going to be next quarter if the Mac share stays stagnant and the iPad share goes up...or vice versa. They basically tied, but I still don’t see enough workflows that can actually migrate to the iPad at this point. Apple is at a bit of a crossroads right now, and hopefully the Mac and the iPad will get more attention this year despite the iPhone and the Watch being the dominant selling hardware at this point.
What if Apple want to reduce costs on the Mac range by reducing overall number of SKUs and maybe also trying to go for economy of scale? If Comet Lake S parts are only going into a consumer desktop products such as iMac and Mac mini (making up less than 20% of Mac sales) they could redesign the lines around mobile Comet Lake Y, U, H, and use Xeon with 'Pro' products.
If they wanted to move things to iPad they would have to introduce changes to iPad OS that properly allow mouse pointers. By forking iPad OS away from iOS they have made a step in the right direction but we may be some distance away from challenging MacOS especially with the state of Catalina at the moment. I'd like to see how Final Cut Pro X could make a transition to an iPad but the user interface without proper mouse support could be challenging although I guess pencil support could replace that.
And so-called full Photoshop on an iPad would be interesting.
