News for hackers. Stuff that matters.
Re:Our Plan for Python 3.13 by hospitalJail on Thursday June 15, 2023 @09:45AM (#36340421)
>I still don't get why today people care so much about single thread performance.

For about 10 minutes a few years ago, when the M1 had the best single threaded performance per buck, people cared.

Now that the M1 isnt the leader in single threaded, we are back to the 'multithread is most important'.

Which has always been true. If your program needs an improvement in speed, you can multithread it. The opposite isnt true.

Re:Our Plan for Python 3.13 by Joker_vD on Thursday June 15, 2023 @10:32AM (#36341102)
> If your program needs an improvement in speed, you can multithread it. The opposite isnt true.

What do you mean by "the opposite"? "If your program doesn't need an improvement in speed, you can't multithread it"? "If you can multithread your program, then it doesn't need an improvement in speed"? Well, yeah, obviously both of those statements are false but they're also quite useless, so who cares?

Re:Our Plan for Python 3.13 by hoosieree on Thursday June 15, 2023 @01:17PM (#36343709)
Add negative threads to fix a program that runs too fast?
Re:Our Plan for Python 3.13 by Joker_vD on Thursday June 15, 2023 @02:29PM (#36344651)
Well, having too many threads can slow down a program as well (extra context switches, extra synchronization) so... no idea.
Re:Our Plan for Python 3.13 by holoduke on Thursday June 15, 2023 @10:16AM (#36340857)
Not all algorithms can be chunked up. Single thread performance is and will always be important.
Re:Our Plan for Python 3.13 by insanitybit on Thursday June 15, 2023 @11:22AM (#36341834)
You can improve performance by moving to a single thread. Pinning work to a single core will improve cache performance, avoid overhead of flushing TLBs and other process specific kernel structures, and more.