News:

Willkommen im Notebookcheck.com Forum! Hier können sie über alle unsere Artikel und allgemein über Notebook relevante Dinge disuktieren. Viel Spass!

Main Menu

AMD Ryzen 9 5900X lags behind 12 Intel Rocket Lake and Comet Lake chips while a Coffee Lake CPU and a mobile processor outbench the Ryzen 9 5950X - UserBenchmark really doesn't like Zen 3

Started by Redaktion, May 12, 2021, 11:23:19

Previous topic - Next topic

Awesomeblackdude

Quote from: kek on May 13, 2021, 05:44:20
Quote from: Wereweeb on May 12, 2021, 22:17:44
You are all paid AMD shills, all know that the most important parameter for CPU performance are single-threaded tasks, like how quickly programs can check for updates, or how well games from 1998 perform.

Compiling? Video editing? Compression? Criptography? What's all that nonsense? I only use my computer to play Fallout 2.

I know this is a joke, but you actually stated a half true: single threaded tasks are still important for most application nowadays. While multicore has indeed caught up, it introduces several issues when coding a program, so it's still not taking advantage of 8 cores or so.

Compiling Java code, for example, is still single threaded so far for most IDEs.


For Java compilers, you can use several threads, such as Eclipse and Buck Build, but there are limitations.

When Java is compiled it can be run as multi-thread programming language which means that every thread can run in parallel.

The OS splits the processing time not only between different applications, but also between each thread in an application. 😬

Arne Bovarne

From a programmer's point of view: Many programs cannot scale with the number of cores. It is single-core performance that determines the performance of such programs. The scaling of the performance with the number of cores is determined by how much of the program can go in parallel and how much must go in series. Therefore, single core performance is very valuable. Furthermore, vectorization is better than parallelization. This is because parallelization has high latency. My experience with modern compilers is that Intel is best at vectorization and single-core performance.

Quick Reply

Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic.

Name:
Email:
Verification:
Please leave this box empty:

Shortcuts: ALT+S post or ALT+P preview