Reply 20 of 100, by XCVG
- Rank
- Newbie
I think anything rare, high-end, or unusual will be what's desirable/valuable in the future.
Generally, I don't think today's hardware- or anything past maybe 2005- will be as big of a deal as older hardware at all. It'll be outdated, but it won't be retro, and the appeal will be even more fringe than stuff from the 1980s, 1990s and early 2000s.
There are a few reasons I can think of why today's hardware will simply be less of a big deal. We hit the point where even cheap PCs were not just good enough but quite good for everyday tasks years ago, and for more intensive tasks like gaming we're hitting increasingly diminishing returns. The rate of advance has slowed dramatically; Skylake was all of 40% faster than Sandy Bridge five years later. And finally, there's a lot more abstraction between the hardware and the software, so there's no need for older hardware to run older programs (the inverse is also largely true). In fact, I found that my main machine running Windows 10 ran some XP-era games better than my XP rig.
That, and today's hardware is kind of boring. Very pretty, but at the end of the day there isn't a ton of variety. Practical, but boring.
My Sandy Bridge rig was my mainstay for five years, and is still in service as a server. Barring hardware failure- and keep in mind that 2010s hardware is more reliable than it was in the past- it will probably continue in that role into the 2020s. I'm planning on replacing my laptop as it just doesn't have enough grunt for me, but with a fresh battery it'll probably serve someone else well for years. My dad's Haswell machine will serve him well into the foreseeable future.
Coming back to my original point, I doubt there will be a ton of interest in standard-issue, boring machines from the late 2000s to the present. There's nothing special about a Core 2 Duo. It's a cheap option that will run modern software surprisingly well, but I struggle to think of anything that a brand new i7 won't do better. There are a few legacy roles, but all the ones I can think of either have to do with older retrocomputing (circular reasoning) or legacy industrial systems (which is actually a major problem in IT and always has been). Moreover, I don't see a system built five years from now being that much different compatibility-wise than that i7.
The only reason to build a machine from this era in the future would be for the cool factor of it. It's not going to be the dime-a-dozen Optiplexes that we're going to want to return to, nor the $100 bang-for-the-buck AM3 motherboard. It's going to be weird stuff like the 775DUAL-VSTA (I threw mine out 🙁 ), expensive stuff like the GTX Titan, and rare stuff like Thunderbolt controllers. I mean, this is true for any era, but I think it's going to be especially true going forward.
Of course, that's just my humble opinion, and I don't have the best record of predicting trends. I dismissed both VR and cryptocurrency, and look where those ended up.