Skip to content

README: parallel between operations, not inside them - #4

Merged
hokb merged 3 commits into
mainfrom
readme-hero
Nov 13, 2026
Merged

hokb merged 3 commits into
mainfrom
readme-hero

Conversation

@hokb

@hokb hokb commented Nov 11, 2026

Copy link
Copy Markdown
Member

The README now states up front that numpy2x does not split single operations (an FFT, a matrix product) across threads. It runs different operations at the same time.

  • Hero line: "Parallel between NumPy operations, not inside them. One switch, two more cores: your NumPy program about 2x faster, with bit-identical results."
  • New paragraph after the command: contrasts numpy2x with multithreaded backends. Those split one large operation; numpy2x runs independent operations across lines, loop iterations and function calls, which also pays off for programs made of many small and medium operations.

Example 2 is not changed yet.


Generated by Claude Code

… not one operation on several threads

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JT2yHm85oFjgf1AcgGLUux
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JT2yHm85oFjgf1AcgGLUux
@hokb
hokb merged commit f031bd0 into main Nov 13, 2026
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants