I’m : a programmer, writer, podcaster, geek, and coffee enthusiast.

A Mac Pro Mini

For years, geeks have wanted Apple to make an “xMac”: an expandable desktop tower like the Mac Pro, but much cheaper, generally achieved in theory by using consumer-class CPUs and motherboards instead of Intel’s expensive, server-grade Xeon line. Apple’s refusal to release such a product is almost single-handedly responsible for the Hackintosh community.1

Apple has shown that they don’t want to address this market, presumably because the margins are thin. And the demand probably isn’t as strong as geeks like to think: most businesses buy Windows PCs for their employees, and most consumers buy laptops for themselves. The relatively small group of people who still want desktop Macs seems to be served adequately by the iMac and Mac Mini.

With Apple now letting the Mac Pro stagnate without an update since August 2010, while the rest of the Mac line has seen multiple new generations and huge improvements since then, it seems like this just isn’t a segment that they care about anymore. The Mac Pro stagnation and the improvements in the other lines have pushed many former Mac Pro owners to the new iMacs and Retina MacBook Pros, which are so fast that they can compete with 2010’s Mac Pros.

The Mac Pro is all but forgotten now, but Dan Frakes restarted the discussion of the xMac this week, arguing for the next Mac Pro to be a consumer minitower:

Put all this together—Apple’s relentless efforts over the past few years to make everything smaller, cooler, and less power-hungry; the fact that you don’t need massive components to get good performance; and an apparent trend towards conceding the highest-end market—and it seems like a Mac minitower is a logical next step for the Mac Pro line.

I don’t think it’ll happen like this.

The Mac Pro has three problems:

Back in 2011, I wrote Scaling down the Mac Pro. Most of it still applies: most ways to significantly reduce the Mac Pro’s cost or size would make it far less attractive to the people who do still buy it.

My conclusion in 2011:

It’s impossible to significantly change the Mac Pro without removing most of its need to exist.

But I think it’s clear, especially looking at Thunderbolt’s development recently, that Apple is in the middle of a transition away from needing the Mac Pro.

Fewer customers will choose Mac Pros as time goes on. Once that level drops below Apple’s threshold for viability or needing to care, the line will be discontinued.

I bet that time will be about two years from now: enough time for Apple to release one more generation with Thunderbolt and the new Sandy Bridge-based Xeon E5 CPUs in early 2012, giving the Mac Pro a full lifecycle to become even more irrelevant before they’re quietly removed from sale.

A few power users will complain, but most won’t care: by that time, most former Mac Pro customers will have already switched away.

It’s turning out to be even more brutal than I expected: Apple didn’t even bother with a Xeon E5 update.2 It looks like they’re accelerating the Mac Pro’s demise by neglect, not preparing for its recovery.

There’s even less of a reason to buy the Mac Pro today than I expected. With Fusion Drive, an iMac can have 3 TB of storage that’s as fast as an SSD most of the time (for most use, including mine3) without any external drives. And the recent CPUs in the iMac and MacBook Pro are extremely competitive with the Mac Pro at a much lower cost.4

Coincidentally, the Mac Pro discontinuation date I predicted as two years from that post will be this November: that’s roughly when Tim Cook hinted that we should expect to see “something really great” for “pro customers” to address Mac Pro demand.

I really want that to be a great new Mac Pro, but I can’t deny that both Apple and the market are sending strong signals suggesting otherwise.

Desktop Retina displays are a wildcard. We can already observe from the Retina MacBook Pros that high-density LCD panels appear to be nearly ready for desktop sizes, but GPUs and video-cable interfaces (e.g. Thunderbolt) will be severely limiting factors. It’s possible that bringing Retina density to desktop-size displays without sucking will require the Mac Pro’s margins, thermal and power capacities, and horsepower for the first few years.

But Apple is patient, and building a Retina iMac might be significantly easier since it avoids the high-speed-video-cable problem. With Fusion Drive significantly improving disk speeds, high-end Thunderbolt A/V peripherals finally starting to appear, and so many pros already using Retina MacBook Pros and iMacs today, a Retina iMac would be a strong temptation for current Mac Pro users.

Maybe Cook’s “something really great for pro customers” is the Retina iMac. If so, as much as I hate to say it as a fan of expandable towers, that’s probably enough.

  1. A lot of people, my past self included, also just want to build their own towers from individually purchased parts, which Apple will probably never support. But I bet Apple could win a lot of them over by just shipping an “xMac” for a reasonable price. ↩︎

  2. A few high-end users have built Xeon E5 Hackintoshes with strong results compared to current Mac Pros↩︎

  3. I’ve been using Fusion Drive in my Mac Pro for almost a month and I’m very pleased. In practice, it actually feels faster overall than the separate SSD and HDD setup did, because the SSD is finally being intelligently used for my working files even when they’re so large that I previously kept them on the HDD.

    I now have 4 TB of space performing at SSD speeds most of the time, at a very reasonable cost, using only two drives. ↩︎

  4. Today, you can get a 27” iMac that’s competitive with or faster than most sensible 2010 Mac Pro configurations (3.4 GHz quad-core Ivy Bridge, 16 GB RAM, 3 TB Fusion Drive, GTX 680MX) for $2,949.

    The most sensible, comparable Mac Pro to get is probably the 3.33 GHz 6-core with 16 GB RAM, 2 TB hard drive (no small SSDs are available to fuse, but you could get an aftermarket one), the ancient Radeon HD 5870, and a 27” LED Cinema Display (not even the Thunderbolt Display, which isn’t compatible because there are still no Thunderbolt ports on the Mac Pro). It would be slightly faster than the iMac in the most CPU-bound roles, but substantially slower in many other tasks. It costs $4,673. ↩︎