RC4 in x86 assembly – 77 bytes

odzhan.wordpress.com

45 points·odzhan··13 comments

[EDIT:] For those who didn't read the article, this is a cute parlor trick, but RC4 is broken and should not be used. If you really want a small cipher, use TEA or one of its more recent variants.

https://en.wikipedia.org/wiki/Tiny_Encryption_Algorithm

To be fair, the introduction states that it is broken and links to two papers as to why.

Good point. I edited my comment to reflect this.

>but RC4 is broken and should not be used.

I think the author is pretty clear about that in the first chapter. And it's not because it's "broken" that it's not interesting to study it.

Good point. I edited my comment to reflect this.

Is TEA less broken than RC4? It's also bigger. RC4 is remarkable because it's so small. TEA does somewhere around ~8 times more work.

Yes, TEA is less broken than RC4. This is not to say that it is inherently more secure, just that no effective attacks on it are currently widely known the way they are with RC4. But this could just be because no one uses TEA and so it has received comparatively little cryptanalytical attention.

RC4 is remarkable because it's so small.

With AESNI, you could probably make a pretty small AES implementation too...

https://en.wikipedia.org/wiki/Treyfer seems exceptionally short, even more so than RC4, and simple enough to memorize. Googling around suggests it's still okay to use, too (but don't use it as a hash).

I wrote implementation of Treyfer today out of curiosity but haven't completely optimized it. https://odzhan.wordpress.com/2015/11/13/asmcodes-treyfer/

You should edit your decrypter into Wikipedia.

Brilliant! Great work!

Can we stop promoting the use of a broken cipher?

Reply on news.ycombinator.com