| > This is a little mini-moan. I only have an old-fashioned user-unfriendly
> UNIX mail system available to read postings, so I cannot read the fancy non-text
> postings some subscribers make. In some cases the mail system cannot even display an encoded
> version, so I can't even copy it somewhere else to view.
> Is it possible that those making such postings can send a standard text version
> instead of, or even as well as, the deluxe version?
>
> Richard Dobson.
i would prefer to receive both text and "fancy", as well.
les
---------------------------------------------------------------------
Les Rackl rackl@math.utah.edu
---------------------------------------------------------------------
Received: from stork2.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa04821;
18 May 97 0:19 BST
Received: from goggins.bath.ac.uk by stork.maths.Bath.AC.UK id aa18387;
18 May 97 0:23 BST
Received: from hermes.ex.ac.uk by goggins.bath.ac.uk with SMTP (PP);
Sun, 18 May 1997 00:22:49 +0100
Received: from noether [144.173.8.10] by hermes via SMTP (AAA02239);
Sun, 18 May 1997 00:17:41 +0100
Received: from hermes.ex.ac.uk by maths.exeter.ac.uk;
Sun, 18 May 97 00:17:35 +0100
Received: from ulysses.Stanford.EDU [36.49.0.124] by hermes
via ESMTP (AAA02236); Sun, 18 May 1997 00:17:34 +0100
Received: (from tkunze@localhost)
by ulysses.stanford.edu (950413.SGI.8.6.12/950213.SGI.AUTOCF)
id QAA11931 for csound@maths.exeter.ac.uk;
Sat, 17 May 1997 16:18:08 -0700
From: Tobias Kunze
Message-Id: <9705171618.ZM11930@ulysses.stanford.edu>
Date: Sat, 17 May 1997 16:18:07 -0700
Reply-To: t@ulysses.stanford.edu
X-Url: http://www.stanford.edu/~tkunze
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: Csound mailing list
Subject: SGI benchmarks
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk
I redid the benchmarks with the new 7.1 compilers. Here they are:
Test Bach-d Bach-m Riss-g Riss-m Guit-d Guit-m Jame-g pvanal lpanal
Length 115.73s 115.73s 63.75s 63.75s 87.75s 87.75s 6.75s 6.75s 5.00s
------------------------------------------------------------------------------
O2 r5 4.13 6.43 5.00 3.13 18.78 19.31 1.63 - -
Indy r5 4.70 4.55 2.96 2.92 21.24 21.23 0.50 - -
Indy r46 6.81 6.69 4.53 4.42 36.46 36.54 0.71 7.39 4.77
Notes:
O2 r5 -- SGI O2 r5000sc, 64MB, IRIX 6.3
recompiled bath version 3.44 using SGI's 7.1 compilers
Indy r5 -- SGI Indy r5000sc, 96MB, IRIX 6.2
recompiled bath version 3.44 using SGI's 7.1 compilers
Indy r46 -- SGI Indy r4600sc, 160MB, IRIX 6.2
recompiled bath version 3.44 using SGI's 7.1 compilers
Received: from stork2.maths.bath.ac.uk by omphalos.maths.Bath.AC.UK id aa04985;
18 May 97 2:11 BST
Received: from goggins.bath.ac.uk by stork.maths.Bath.AC.UK id aa21782;
18 May 97 2:15 BST
Received: from hermes.ex.ac.uk by goggins.bath.ac.uk with SMTP (PP);
Sun, 18 May 1997 02:15:03 +0100
Received: from noether [144.173.8.10] by hermes via SMTP (CAA04254);
Sun, 18 May 1997 02:11:46 +0100
Received: from hermes.ex.ac.uk by maths.exeter.ac.uk;
Sun, 18 May 97 02:11:40 +0100
Received: from root@big.fishnet.net [205.216.133.3] by hermes
via ESMTP (CAA04248); Sun, 18 May 1997 02:11:35 +0100
Received: from sinewave (port002.vta.fishnet.net [205.216.133.151])
by big.fishnet.net (8.7.5/8.6.9) with SMTP id RAA02115
for ; Sat, 17 May 1997 17:15:41 -0700
Message-Id: <337E49F0.5FE59691@rcsreg.com>
Date: Sat, 17 May 1997 17:14:40 -0700
From: Toby
X-Mailer: Mozilla 3.01Gold (X11; I; Linux 2.0.27 i486)
Mime-Version: 1.0
To: Csound Mailing List
Subject: A drawback in the new organization of CSCORE
Content-Type: text/plain; charset=us-ascii; name="letter"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline; filename="letter"
Sender: owner-csound-outgoing@maths.ex.ac.uk
Precedence: bulk
I have been using a very old version of Cscore, because I
couldn't get the new one to compile under Linux for some
time. I was successful however, with version 3.46.
It looks as though the new Cscore is far more robust.
Two things bother me about it though.
First, the library has been rearranged so that the function
main() is defined in the library rather than in the user
program. One is forced to give a single argument to the
resultant program which is the name of the input score file
which will source functions like lget() and getev().
This causes some problems for me. For example, if my
program generates notes by itself and does not use
an input score, cscore complains and exits. The only
way to get around this is to create a dummy score
(which has to be a valid score) and point to it.
This is inelegant at best.
Naming the file with an argument also prevents the
use of stdin, which is a great drawback. I might
like to do something like:
amplify 2.3 < template | chop 3 4 | randomize 1.1 > my.sco
Furthermore, the present arangement prevents my
programs from receiving command line arguments, so
the above would be impossible even with the use of
tempfiles and a shell script!
Now, it seems to me as though I have fixed the
problem by changing main() in cscormai.c thusly:
extern FILE *scfp /* I added this line */
main(int argc, char **argv)
{
dribble = NULL;
scfp = stdin; /* I added this line */
// scfopen(argc,argv); /* This is the old file opener */
cscorinit();
cscore(argc, argv); /* Pass args to user function */
}
This seems to cure all of the problems on my system,
and the resulting library can be used without any
modification to my old programs and makefiles (save lcount(), see below).
The only reasons that I can think of that cscore was changed this
way in the first place are:
1) Mac's don't have stdin? (Is this true?)
2) cscore uses fseek() or something on the file and
therefore cannot use pipes?
If (1) is true, then it would be easy to look for a flag like -i
to allow the option of reading from a file.
If (2) is true, then perhaps -istdin could be supported, with the stipulation
that certain functions will not be usable.
Perhaps there is an unavoidable circumstance of which I am
unaware that causes stdin to be unusable. I would be happy to make
the necessary adjustments to the code once I am sure that I understand
the situation fully.
The second item which concerns me is that the current documentation
for CSCORE is a light year behind the code. I noticed that 'warping'
of the score is now available. Is there some documentation available
on how to use this?
Also, lcount() has been dropped, presumably because list->nevents
suffices nicely now. It would be nice to have it back however, for
backward compatibility.
Toby
-There otta be a law-
|