Re: [Haskell-cafe] haskell crypto is reaaaaaaaaaally slow

On Wed, 2007-06-20 at 15:23 -0700, Jason Dagit wrote:

On 6/20/07, Andrew Coppin wrote:

...

Donald Bruce Stewart wrote:

...

Finally, to actually get C speed, use a C md5.

I always feel worried when people say this... It's almost like admitting "hey, Haskell is beautiful, but it can never be fast". I always find myself wanting that statement to be false...

I agree with you, but at the same time, if Don says something about the performance of Haskell I tend to trust him on it. I find that his ability to optimize Haskell tends to set the bar. Otoh, I'd love to see someone demonstrate otherwise here :)

Well this is something else Don said,

I'd suspect a pure haskell md5 over bytestrings would be competitive with a C implemetation though. Easier to just call OpenSSL.