Re: [Gcl-devel] Speed

[Camm Maguire]

Greetings!

I have no comprehensive results at the moment but am always interested
in fixing known slowness in GCL generated code, in case you have a
favorite benchmark or two.  I would love to update the suite in the
source tree with more modern realistic tests that could be used as a
quick parity check before release.

I'm aware of at least one interesting item in this regard -- 2.7.0
automatically transmutes the takr benchmark into a mutually recursive
equivalent state machine.

Take care,

"Adrienne Thompson" <address@hidden> writes:

> The release notes for GCL 2.6.2 indicated that GCL was about as fast as CMUCL.
> I'm curious to know if we have surpassed that benchmark with the 2.7.x 
> unstable
> series. Is there any new data comparing GCL's speed with say that of SBCL?
> 
> Adrienne
> 
> 
> 
> _______________________________________________
> Gcl-devel mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/gcl-devel
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah