Re:Pessimism about parallelism
by rossdavidh on Wednesday December 26, 2018 @10:12AM
(#18762495)
All interesting, but if we're talking about typical laptop/desktop users, beside the point, I think. The reality is that the entire computer's performance is not the bottleneck, for most of what a typical user is doing, most of the time. The limitation on performance is either:
1) in the user's head (e.g. a word processor and an author)
2) in the network the computer is attached to (much of what is done on the web)
Getting faster/better performance out of your laptop/desktop won't help any, for the reason that your computer's performance is already not the limiting factor, for most of what a typical user is doing.
Although, this does reinforce the point that adding even more cores to a typical user's laptop/desktop, isn't really accomplishing much, but for much simpler reasons than the issues discussed in the article.
[ Parent
| on HN ]
3) in the storage
4) in the antivirus insisting on chewing each piece of IO very slowly in case it might have a virus
Re:Pessimism about parallelism
by mattnewport on Wednesday December 26, 2018 @10:43AM
(#18762653)
This is not the case in practice for most of what I'm doing, most of the time. I'm more of a power use than a typical user perhaps but I also have much higher end hardware than a typical user and I'm continually frustrated by performance issues that are primarily CPU, GPU or local storage related, although the network being a limiting factor is also common.