The conversation in another thread about chip shortages and their impact on PC and PC upgrade prices got me thinking about my 9yo PC.
I built it 9 years ago to be a bit of a "do all" machine. Over the years I put a decent video card in it (since replaced with a better one, but still "outdated" by today's standards), upgraded the ram to 16gb, and replaced the main drive with an SSD (but kept the spinny drive for user home directories and storage). Despite its age, it still works well and does everything I need to do. It's a tad sluggish here and there and I can see the CPU maxing out at these times, but memory is never over 50% utilization, nor is the graphics card ever taxed. Basically, only the CPU is underperforming.
The CPU is an AMD FX-4100 (4 cores, 4 threads, 3.8ghz max speed), which was pretty low end at the time. My mobo will support an 8-core/8-thread CPU from the same generation. I think the extra thread capacity would resolve my CPU bottleneck and only cost $100 or so. I'd also drop in a new 2tb SSD and dump the spindle drive, which should give another small performance boost. Grand total should be under $300.
If I built a new system, sticking with the traditional PC model, it would cost me twice that or more. Or, I could go with one of the new "micro" systems like the Intel NUC, but the total cost after adding memory and drives would be about the same.
I use this system for basic computing tasks, web surfing, streaming video services (Netflix, etc), light gaming (mainly TF2), occasional video editing, and as a centralized storage nexus for my family (all systems back their files up to this one, which then dumps to our server in the basement). It's the video streaming and the browser continuously showing the view from our Nest cameras that tend to use the most CPU resource. Everything else is low.
I'm really not looking forward to building, tweaking, and testing a new build, which has me leaning towards a NUC if I replace this system entirely.
So, upgrade what I have or build a new one?
Chris