78"OsbQ["56/!nisvyB/uA5Q.{)gm6?q.j71ww.>b9b]-sG8zNt%KkIa>xWg&1VcjZk[hBQ>]j~`Wq
Xl,y1a!(>6`UM{~'X[Y_,Bv+}=L\SS*mA8=s;!=O`ja|@PEzb&i0}Qp,`Z\:6:OmRi*
X-Hashcash: 1:25:050824:[EMAIL
PROTECTED]::K7WVp5LtQH+6JUFQ:00000000000000000000000000000000000000hE5o
Mail-Followup-To: Eduard Bloch <[EMAIL PROTECTED]>, [EMAIL PROTECTED],
[EMAIL PROTECTED]
Date: Wed, 24 Aug 2005 18:22:01 -0500
In-Reply-To: <[EMAIL PROTECTED]> (Eduard Bloch's message of
"Wed, 24 Aug 2005 23:44:40 +0200")
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
severity 324929 wishlist
thanks
Hi,
Err, this is not a very robust benchmark. With input of any
size whatsoever, results would depend on what your CPU was doing --
but the fact that the Perl code is faster is undisputable. In
any case, the design goal of mime-codecs is to provide people with a
simple, very portable tool set that can be used to read mail -- and,
as such, is rarely going to matter for the common case -- I seem to
be getting about 11MB/s, which should be enough for the use case
mime-codecs was designed for.
If you feel that we need faster tools, perhaps mime-codecs can
be split out from VM and reimplemented in Perl.
On my tests,
Perl:
thru: 0.007ms at 0B/s ( 18.6MB/s avg) 4.0GB
thru: 0.008ms at 0B/s ( 20.5MB/s avg) 4.0GB
base64-encode:
thru: 0.011ms at 0B/s ( 11.8MB/s avg) 4.0GB
thru: 0.009ms at 0B/s ( 11.9MB/s avg) 4.0GB
manoj
--
The full impact of parenthood doesn't hit you until you multiply the
number of your kids by thirty-two teeth.
Manoj Srivastava <[EMAIL PROTECTED]> <http://www.debian.org/%7Esrivasta/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
