If there are still a lot of people looking for desktop, I don't think we are in the "post-PC era" quite yet.
This isn't the "post-PC" era.
It may be the post-buying-a-new-PC-every-18-months era, the post every-kid-in-a-family-needing-their-own-PC era or the post-this-years-model-twice-as-fast-and-10%-cheaper era, but the PC is still an essential tool for many.
I see lots of people using mobile phones and tablets... but outside of purely domestic users, they're using them
in addition to a laptop or desktop. While mobile is great for communication, taking notes in meetings, entertainment or a handful of niche applications where touch/stylus works well, as soon as people need to write a 10000-word report, edit a video, write some code or do their accounts, they reach for a proper computer.
Other industries have managed to cope with the maturation of the market without abandoning product lines. Fridges, washing machines, cookers last for ever, but they still get made. A new car from a decent maker will last 10 years, but they still release new models every couple of years.
Unfortunately, the PC industry has, in the past, only known an expanding market, exponential growth in power and capacity and ever-reducing component costs - so the "Post PC era" meme was invented as an excuse to neglect the PC and go looking for the next boom product that might generate windfall profits.
Some of it is self-fulfilling: Intel has been focussing on low-power, cool-running mobile chips rather than raw power (OK, part of that is because, if they don't, ARM will clean their clock) and a major cause of the Windows 8 debacle was Microsoft trying to force a mobile-style user interface on desktop users. Unfortunately, the next big push in CPU power probably involves improving
compiler technology to automatically optimise software for lots of CPU cores and GPU-accelerated systems... which is
hard and doesn't pay off this quarter.
In a future where PC sales - particularly "power user" PCs as opposed to the tablet/laptop crossover market - are declining maybe Apple needs to stop trying to pass off (very nice) "ultrabooks" as "pro" machines and re-visit some past assumptions.
What if - alongside their 'prestige' ultrabooks and all-in-ones, they built a Mini-ITX-sized, expandable, versatile, desktop that was
nicely built but largely used (good quality) generic parts, for an affordable price? Or a slightly chunky "desktop replacement" laptop In the past, the answer was simple: it would decimate iMac and Mac Pro sales. I've used that argument before. In the alleged "post-PC" era, though, would it really - or would it help keep "pros" and "power users" on the MacOS platform, and counter the lure of being able to go out and order a Windows or Linux PC with
exactly the components you wanted? Is the option of a $1000 xMac
really going to dissuade someone prepared to spend $3-4k on an ultrabook with an emoji bar and thermally-throttled-to-death CPU/GPU? If that's too risky, at least build a new Mac Mini around the Skull Canyon chipset.
I'll tell you one thing that
will decimate the sales of MacPro/iMac*/Mac Mini - not updating them for years because of the huge design and re-tooling costs of changing all those bespoke parts to work with this years chips. Another sales-killer: the office/family nerd that everybody asks about what computer they should buy gets fed up and switches to Windows.
A lot of the wailing and gnashing of teeth about the new MacBook Pros might have been avoided if Apple had, at least, offered an
alternative by updating the Mini, Pro and iMac.
*To be fair, the 27" iMac is relatively up-to-date and has a Skylake CPU - I don't know that there
is a suitable Kaby Lake CPU for it yet - however, it needs USB C/TB3 (in place of the TB2 ports - for pity's sake keep the USBA and ethernet!) and maybe a graphics bump. I seriously hope that's what we're waiting for, and not a 20% thinner case.