Re: saner shootout programs

Bulat Ziganshin wrote:
> Hello Richard,
> 
>> Yes it was just a plausible guess, but not contradicted by the experts.
>> And that was using the Windows version of GHC so other versions may
>> have better optimisation. I don't know how to check, maybe the experts
>> can illuminate the subject?
> 
> note that my letter was not contradicted too LOL
> 
> 1. many experts just don't read this list due to its high traffic,
> write into ghc-users instead

Good point. I have started a thread here

http://www.haskell.org/pipermail/glasgow-haskell-users/2008-May/014742.html

> 
> 2. ghc is too large system and no one know all its details. this
> particular option may be set up 10 years ago and never
> touched/studied by anyone since then. for example, i'm ghc performance
> expert [to some degree], but i don't know anything about its build
> system. all that i can tell is that ghc base library build times don't
> prohibit use of -O2
> 
>