• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI

Bend 2 launches; pixel-fixer port reportedly runs about 4× slower than Rust

Bend 2's announcement claims it blocks AI mistakes through proof checking. A user testing a Claude-converted pixel-fixer reports matching Rust's grid decisions on all 18 test images, despite a difficult port.

2 Sources, 21d ago, first seen 21d ago

TLDR

Bend 2's announcement describes a new programming language that runs on GPUs and claims to block AI mistakes through proof checking.

A user who had Claude convert a Python/Rust pixel-fixer to Bend reports identical grids, sub-pixel steps and consensus paths across all 18 test images compared with the Rust core. Bend ran about four times slower than Rust on the same machine, they say.

The user describes a difficult conversion, though replacing lists with Bend Arrays cut processing time from 3.6 minutes to 4.5 seconds per image. They report no memory or concurrency bugs and praise the proof checks that gate the build. But they caution that floating-point math is opaque to the checker, so numerical errors could remain.

Combined views

—

2 Sources, first seen 21d ago

— likes— comments— saves— reposts

Combined views

—

2 Sources, first seen 21d ago

— likes— comments— saves— reposts

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

2 Sources

Astropulse@RealAstropulseI was curious about the advantages/speed claims here so I put it to the test. I had claude convert my entire python/rust pixel-fixer tool to bend. Same algorithm: three grid detectors (autocorrelation + cepstrum, run-length lattice fit, self-similarity tiebreak) and the two-stage k-means repack that turns mushy AI "pixel art" into real 1:1 pixels. Result: 18/18 test images give the identical grid, identical sub-pixel steps and the same consensus path as the Rust core. F32 only, no f64 in Bend, and it still didn't flip a single decision. The clean test image reconstructs pixel-for-pixel. Speed: about 4x slower than Rust on the same box. Rust has rayon, rustfft and f64; Bend got ~2x from its fork-join on 6 threads. Honest number, not a blowout either way. The conversion was a real fight though. Bend is affine, lazy and termination-checked: no loops, no mutation, no if, no mutual recursion. Every array read hands the array back, so a 3x3 median is nine chained continuation functions. Fable ended up writing a kernel generator and a topological sorter just to keep compiling. First port used lists everywhere: correct, but 3.6 minutes per image. Switching every pixel stage to Bend Arrays (~1 ns reads) made that 4.5 s. Two of the three real bugs were silent U32 wraparounds. Zero memory or concurrency bugs, which is the point of the language. PROOF.bend gating the build is genuinely cool. Float math is opaque to the checker though, so the numerics could be wrong even if the 'laws' are correct. What would make Bend great for this kind of work: f64, a shareable read-only array, loop/if sugar over what it already compiles to, checked integer arithmetic, and an FFT in the stdlib. With those it'd be a few hundred lines shorter and read like the Rust. Highly recommend checking out Bend, its genuinely interesting and it will only get better I'm sure. Taelin is getting a lot of pushback now but I genuinely couldn't find a single false claim. It does what he says. It's missing some things, sure, but it literally just got released. I didn't try the cuda path, since that would be a pain for the systems I use to run the pixel-fixer. I'm really excited to see where this goes. The proofs and laws genuinely made this conversion easy. Just pointed fable pixel-fixer it and said to convert it from rust to bend.21d
Taelin@VictorTaelinRT @RealAstropulse: I was curious about the advantages/speed claims here so I put it to the test. I had claude convert my entire python/ru…21d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    2 Sources

    Astropulse@RealAstropulseI was curious about the advantages/speed claims here so I put it to the test. I had claude convert my entire python/rust pixel-fixer tool to bend. Same algorithm: three grid detectors (autocorrelation + cepstrum, run-length lattice fit, self-similarity tiebreak) and the two-stage k-means repack that turns mushy AI "pixel art" into real 1:1 pixels. Result: 18/18 test images give the identical grid, identical sub-pixel steps and the same consensus path as the Rust core. F32 only, no f64 in Bend, and it still didn't flip a single decision. The clean test image reconstructs pixel-for-pixel. Speed: about 4x slower than Rust on the same box. Rust has rayon, rustfft and f64; Bend got ~2x from its fork-join on 6 threads. Honest number, not a blowout either way. The conversion was a real fight though. Bend is affine, lazy and termination-checked: no loops, no mutation, no if, no mutual recursion. Every array read hands the array back, so a 3x3 median is nine chained continuation functions. Fable ended up writing a kernel generator and a topological sorter just to keep compiling. First port used lists everywhere: correct, but 3.6 minutes per image. Switching every pixel stage to Bend Arrays (~1 ns reads) made that 4.5 s. Two of the three real bugs were silent U32 wraparounds. Zero memory or concurrency bugs, which is the point of the language. PROOF.bend gating the build is genuinely cool. Float math is opaque to the checker though, so the numerics could be wrong even if the 'laws' are correct. What would make Bend great for this kind of work: f64, a shareable read-only array, loop/if sugar over what it already compiles to, checked integer arithmetic, and an FFT in the stdlib. With those it'd be a few hundred lines shorter and read like the Rust. Highly recommend checking out Bend, its genuinely interesting and it will only get better I'm sure. Taelin is getting a lot of pushback now but I genuinely couldn't find a single false claim. It does what he says. It's missing some things, sure, but it literally just got released. I didn't try the cuda path, since that would be a pain for the systems I use to run the pixel-fixer. I'm really excited to see where this goes. The proofs and laws genuinely made this conversion easy. Just pointed fable pixel-fixer it and said to convert it from rust to bend.21d
    Taelin@VictorTaelinRT @RealAstropulse: I was curious about the advantages/speed claims here so I put it to the test. I had claude convert my entire python/ru…21d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet