2016 development laptop prognostications

2016/09/02 Tags: hardware development

What makes a great development laptop for me?

In 2013 it was, by order of personal value:

  1. Portability (weight, battery life, yet “decent” (13"+) screen size) - I develop away from my office regularly
  2. Performance (CPU, SSD) - Vim/Emacs scripting, Visual Studio (via VM), Java dev with IntelliJ, VM’s via VMware; not too heavy most of the time.

Based on what I felt was available at the time, I purchased a mid-2013 MacBook Air 13" with 8GB and the Core i7.

Happy choice.

Only whine is about the 1440x900 resolution and its fat bezels. A higher-res screen (perhaps just HD 1920x1080) in the same overall dimensions would have made it a perfect choice.

(Why not the MacBook Pro 13 retina? I wasn’t convinced that the new range was without any reliability issues, that the high PPI was necessary (2560x1600 for the 13"), that the GPU was powerful enough for the increased workload from the resolution, and that the reduced battery life made it worthwhile for the additional cost.)

(Why not a 15"? In my mind laptops of this class were big and clunky. But times change.)

In 2016 my requirements are much as before, except that I’m finding Android Studio’s Gradle build times rather irritating. (“Instant Run” doesn’t help as much as I might have hoped, and it can be annoying when one changes an asset and has to clean and rebuild all to see it (for example).)

So, I’ve been thinking about how best I can improve this.

Obvious thought: stop looking for ways to spend money and appreciate what I’ve got. I’m chasing after a pipe-dream as compiling and packaging is slow and will continue to be so for years to come.

Maybe so… but I like this stuff. If I admit it’s also a treat would that help? Besides, this is how I earn more money, and if I can make the process a little less painful and a little more productive, isn’t that worth it?

OK! It’s because I like having nice computers to play with. Fine. Let’s move on.

First option: a more powerful laptop.

Main candidates in 2016:

Notice how I couldn’t care less about touchscreen? Why does everyone seem to want it so much? Maybe I’m showing my age here. “Get your fingers off my screen!”

A MacBook Pro update is now so overdue, there must be an update to the product line this year. (Current estimate is for an October announcement, Q4 shipping.) I feel it’s prudent to wait to see how it turns out. But I have my doubts that it’ll be my final choice.

What makes me pessimistic are the rumours:

If, as I’m inclined to imagine it will be, the big October reveal is all about “thinner”, “butterfly keys”, “dynamic function bar”, “touch ID”, “metal injection mold-made hinges”, “awesome gaming”… I’m likely to be underwhelmed. I’ll be interested, in the same way as I am with the new MacBook, Razor Blade Stealth and many other machines, but I won’t be in a hurry to buy.

So, I’ve been contemplating an alternative approach.

Second option: The return of the desktop.

Network connectivity at my typical working locations has improved over the years and is already much at the point where remoting from my laptop into a desktop appears to be a viable option. It also appears to be feasible to build a quiet but powerful machine (Intel Core i7-6700K, 32GB DDR4, 512GB M.2 SSD, Noctua CPU cooler, Corsair RMx 650W power supply, etc) for less than £1,500.

With this strategy my laptop become solely about portability. (A thin client if you will.) So it can actually be relatively inexpensive. (Perhaps even Chromebook cheap?!) And, perhaps I can finally have the portable Linux desktop I’ve always wanted (battery life will need to be unaffected though).

It also happens that I still think this is where technology is headed in our post-“Moore’s Law” dystopia (cheap, ubiquitous, fashion-conscious, disposable - basically, like clothes now). As William Gibson noted: “The future is already here - it’s just not very evenly distributed”. Imagine being able to buy something as capable as the Lenovo Yoga 910 for around $100 (in 2016 value). (Instead you’ll be spending your money on lots of other things, like subscriptions to various clouds (content, storage, computing, etc.).) Let’s check back in 2025 to see how close this is to how things turned out.

Being portable, laptops are more likely to get stolen, damaged or lost. (Although it’s relative - I’ve been lucky and had no losses.) In 2016, they are also quite unmodifiable (e.g. soldered RAM). (Much like smartphones really.) On the other hand, desktops can be upgraded. They are cheaper per processing-unit than laptops. There’s a lot going for it - provided there’s sufficient network access available to remote at a decent refresh rate. (As they obviously score zero on my portability scale, which is why I last built a desktop in 2010. (It’s currently our home’s media server.))

The laptop still needs to be capable if there is no connectivity. As my Air already is, that’s a reasonable benchmark met by plenty of 2016 ultrabooks. (So not really that thin of a client.) I think the Air has aged very well, and if this solves my performance irritations, I can see it being my portable machine for the time being. (Until there’s a cheap ultrabook with a big screen and long battery life that tempts me away. And even then the Air will still have a home in our household.)

At this point, I think the second path will be my choice. Let’s meet back here in November and see how it’s all turned out.