3 ms·
Haven't PCs stopped getting more powerful quite a few years ago? The only exceptions I see is swapping spinning disk with ssd and ever-growing gaming GPUs. If y
by lrem 10y ago
Haven't PCs stopped getting more powerful quite a few years ago? The only exceptions I see is swapping spinning disk with ssd and ever-growing gaming GPUs. If you're not gaming, in a couple years you might be hard pressed to spot the difference in performance.
- zokier 10y agoThe past couple of years (maybe since Haswell, or 3 years) have been slower progress, but on the other hand I'd say the low hanging fruit on mobile chips have been picked too by now and the progress there is also winding down. People have been chatting and surfing and whatnot just fine with their 200MHz Pentiums at the time, so its not that surprising that it might be bit difficult for many people to truly appreciate just the amount of power available these days. Of course Wirth's and Parkinson's laws do not help here either. Just for fun, here is a small comparison with 9 year old CPU and 3 year old CPU: http://www.anandtech.com/bench/product/55?vs=837 http://www.anandtech.com/bench/product/55?vs=837
- lrem 10y agoYou're actually consistent with what I had in mind. A 3 year old quad core is 2-3 times faster than a 9 year old dual core, in most of the embarrassingly parallel benchmarks. I.e. despite all the process improvements, computing power per core stagnated. Personal computers mostly remain at two cores, four accessible at a premium. Personal computers are simply fast enough and the progress is better used by moving more pieces off the motherboard onto the CPU. If someone needs more power, then he's considered to be building a workstation or a server and charged way more than for personal stuff. I have no clue if you're right about arm progress saturating, I haven't been following that.