music5?

[Bill Schottstaedt]

I've added music5.f to the Snd tarball and CVS site.  It's the
version of music 5 we had here at Stanford in the late 60's,
typed in laboriously by yours truly from faded and yellowing
XGP output.  The "I" and "1" characters (which make up 50%
of the total) are nearly indistinguishable in this output, so
my eyes are blurry, my hands are palsied, my shoulders
hurt.  I need a beer.  Anyway, I would love to get this working,
but I can't remember much Fortran, and gfortran doesn't
like the dialect used in music5.f ("obsolete arithmetic IF" etc).
If anyone is interested, I'd be more than happy to run back
to the sources to check up on typos and so on.


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[M. Edward (Ed) Borasky]

Bill Schottstaedt wrote:

> I've added music5.f to the Snd tarball and CVS site.  It's the
> version of music 5 we had here at Stanford in the late 60's,
> typed in laboriously by yours truly from faded and yellowing
> XGP output.  The "I" and "1" characters (which make up 50%
> of the total) are nearly indistinguishable in this output, so
> my eyes are blurry, my hands are palsied, my shoulders
> hurt.  I need a beer.  Anyway, I would love to get this working,
> but I can't remember much Fortran, and gfortran doesn't
> like the dialect used in music5.f ("obsolete arithmetic IF" etc).
> If anyone is interested, I'd be more than happy to run back
> to the sources to check up on typos and so on.
>
>
> _______________________________________________
> Cmdist mailing list
> Cmdist@...
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
>

Well ... I started my career doing Fortran II and assembler programming
on the IBM 7094, which is where Max Matthews started the "Music" series.
What dialect is it -- Fortran II, Fortran IV or Fortran 66? What did you
run it on at Stanford, a PDP-6/10/Decsystem 10/20?

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[M. Edward (Ed) Borasky]

M. Edward (Ed) Borasky wrote:

> Bill Schottstaedt wrote:
>> I've added music5.f to the Snd tarball and CVS site.  It's the
>> version of music 5 we had here at Stanford in the late 60's,
>> typed in laboriously by yours truly from faded and yellowing
>> XGP output.  The "I" and "1" characters (which make up 50%
>> of the total) are nearly indistinguishable in this output, so
>> my eyes are blurry, my hands are palsied, my shoulders
>> hurt.  I need a beer.  Anyway, I would love to get this working,
>> but I can't remember much Fortran, and gfortran doesn't
>> like the dialect used in music5.f ("obsolete arithmetic IF" etc).
>> If anyone is interested, I'd be more than happy to run back
>> to the sources to check up on typos and so on.
>>
>>
>> _______________________________________________
>> Cmdist mailing list
>> Cmdist@...
>> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
>>
>
> Well ... I started my career doing Fortran II and assembler programming
> on the IBM 7094, which is where Max Matthews started the "Music" series.
> What dialect is it -- Fortran II, Fortran IV or Fortran 66? What did you
> run it on at Stanford, a PDP-6/10/Decsystem 10/20?
>
> _______________________________________________
> Cmdist mailing list
> Cmdist@...
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
>

P.S.: There's machine-level emulation available for a lot of systems, if
you have the compilers, etc. Check out http://simh.trailing-edge.com/


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

This is fantastic!  I got rid of some of the problems by splitting
it into 6 separate programs (one for each "page" in the output).
I wasn't around, or rather I wasn't at SAIL during those years
(I was an undergrad at Stanford, and actually took a course from
JC in 1969, but I wasn't interested in computer music at that
time -- I thought I wanted to be a mathematician).  SAIL had
a PDP-10 and a PDP-6.  My impression was that the musicians
used the PDP-6, but this code says it is for the PDP-10.
I like the "WHERE IS HARVEY" error message.  Snd needs
some of those.



_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[M. Edward (Ed) Borasky]

Bill Schottstaedt wrote:

> This is fantastic!  I got rid of some of the problems by splitting
> it into 6 separate programs (one for each "page" in the output).
> I wasn't around, or rather I wasn't at SAIL during those years
> (I was an undergrad at Stanford, and actually took a course from
> JC in 1969, but I wasn't interested in computer music at that
> time -- I thought I wanted to be a mathematician).  SAIL had
> a PDP-10 and a PDP-6.  My impression was that the musicians
> used the PDP-6, but this code says it is for the PDP-10.
> I like the "WHERE IS HARVEY" error message.  Snd needs
> some of those.
>
>
>
> _______________________________________________
> Cmdist mailing list
> Cmdist@...
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
>

Well ... there's a fundamental issue in PASS1 that may not be easily
overcome. On a PDP-6 or PDP-10, I believe the words were 36 bits and the
characters six bits, with six characters fitting into a word. It looks
like a lot of the DATA statements are expecting this. On Linux/i386,
however, an INTEGER is 32 bits and there are four eight-bit characters
in it.

PASS2 is relatively clean ... I think it had only one or two issues
preventing it from compiling. PASS3 is a mess :(.

Stanford doesn't throw *anything* away ... you might be able to find a
PDP-6 compiler somewhere and get "simh" to work on it. IIRC "simh" is
"mostly" old DEC boxes and the folks who hang out on that web site
probably have some of the PDP-6 software. The PDP-6 was a much simpler
beast than the 10, so if you can get this to work on an emulated PDP-6,
you might be better off than trying to port it.


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[M. Edward (Ed) Borasky]

M. Edward (Ed) Borasky wrote:

> Well ... there's a fundamental issue in PASS1 that may not be easily
> overcome. On a PDP-6 or PDP-10, I believe the words were 36 bits and the
> characters six bits, with six characters fitting into a word. It looks
> like a lot of the DATA statements are expecting this. On Linux/i386,
> however, an INTEGER is 32 bits and there are four eight-bit characters
> in it.
>
> PASS2 is relatively clean ... I think it had only one or two issues
> preventing it from compiling. PASS3 is a mess :(.
>
> Stanford doesn't throw *anything* away ... you might be able to find a
> PDP-6 compiler somewhere and get "simh" to work on it. IIRC "simh" is
> "mostly" old DEC boxes and the folks who hang out on that web site
> probably have some of the PDP-6 software. The PDP-6 was a much simpler
> beast than the 10, so if you can get this to work on an emulated PDP-6,
> you might be better off than trying to port it.

Well ... simh doesn't do PDP-6 but does do PDP-10. *And* there are
copies of TOPS-10 and quite a few Fortran compilers. So ... that might
be your best bet.

http://pdp-10.trailing-edge.com/


>
>
> _______________________________________________
> Cmdist mailing list
> Cmdist@...
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
>

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

I think at least the big DATA statement is defining the words
like ENV and AD4 that were music 5's generator names; they
then walk though the array one character at a time later to
recognize the names, so there the 36-bit business shouldn't
matter.  In PASS3, they're packing 3 12-bit samples in a word,
but that's easy to undo -- I'd just output the raw samples.


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[errordeveloper]

hm ..
i'd like to have a go with that ..
can you give some basic instruction how to get it running?
i have compiled the simulator on x86_64 linux and downloaded some other
stuff from the site ..but don't really have any clue how to stick it all
together ..i also have no any experience with that hardware due to my age ..

but i really wish to get a feeling of how it was !

cheers,
--
ilya .d
On Sat, Apr 26, 2008 at 05:25:08PM -0700, M. Edward (Ed) Borasky wrote:

> M. Edward (Ed) Borasky wrote:
> > Well ... there's a fundamental issue in PASS1 that may not be easily
> > overcome. On a PDP-6 or PDP-10, I believe the words were 36 bits and the
> > characters six bits, with six characters fitting into a word. It looks
> > like a lot of the DATA statements are expecting this. On Linux/i386,
> > however, an INTEGER is 32 bits and there are four eight-bit characters
> > in it.
> >
> > PASS2 is relatively clean ... I think it had only one or two issues
> > preventing it from compiling. PASS3 is a mess :(.
> >
> > Stanford doesn't throw *anything* away ... you might be able to find a
> > PDP-6 compiler somewhere and get "simh" to work on it. IIRC "simh" is
> > "mostly" old DEC boxes and the folks who hang out on that web site
> > probably have some of the PDP-6 software. The PDP-6 was a much simpler
> > beast than the 10, so if you can get this to work on an emulated PDP-6,
> > you might be better off than trying to port it.
>
> Well ... simh doesn't do PDP-6 but does do PDP-10. *And* there are
> copies of TOPS-10 and quite a few Fortran compilers. So ... that might
> be your best bet.
>
> http://pdp-10.trailing-edge.com/
>
>
> >
> >
> > _______________________________________________
> > Cmdist mailing list
> > Cmdist@...
> > http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
> >
>
> _______________________________________________
> Cmdist mailing list
> Cmdist@...
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

I kept digging around where I found that PDP-10 printout, and
found a listing of Music V for the PDP-11 (it's calling itself
music11), a complete copy of the Music IV programmer's
manual (Mathews and Miller), Technical Information for the
Baldwin Model 5 Electronic Organ; not to mention innumerable
old language manuals, the best of which is Spitbol.
I feel like an archaeologist.


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[M. Edward (Ed) Borasky]

Bill Schottstaedt wrote:

> I kept digging around where I found that PDP-10 printout, and
> found a listing of Music V for the PDP-11 (it's calling itself
> music11), a complete copy of the Music IV programmer's
> manual (Mathews and Miller), Technical Information for the
> Baldwin Model 5 Electronic Organ; not to mention innumerable
> old language manuals, the best of which is Spitbol.
> I feel like an archaeologist.
>
>
> _______________________________________________
> Cmdist mailing list
> Cmdist@...
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
>

I thought I had a copy of Max Matthews' book on the MUSIC series, but I
haven't been able to find it. Regarding the PDP-11 version, you might
check with Barry Vercoe to see if he has any insight. IIRC the first
version of CSound was on  the PDP-11 and was a more or less direct port
of Music-11. Laurie Spiegel might also know something about it.

PDP-11 Fortran would be a lot easier to port to Linux. :)

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

I'll ask Max if I can scan in the book -- it's not "Technology
of Computer Music", but an explicitly uncopyrighted memo
from Bell Labs -- not sure how the law treats that, but surely
if Max says it's ok, nobody else will mind.  I also found
class notes and handouts from a course on Music IV at
UCLA in 1969 -- I think they come ultimately from Gilbert
Strang -- is he still alive?

Once I recover from the previous binge of typing, I might
type in the PDP-11 version -- it's actually close to the
PDP-10 version.  My copy has many comments from
Leland Smith who was doing the local port.


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

oops -- a typo -- I meant Gerald Strang -- Gilbert Strang
is a very well known math guy -- anyway, I came upon this
interesting paper while stumbling around:

http://www-math.mit.edu/~gs/papers/newsigproc.pdf


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bret Battey]

I just been transferring my CLM4 operations to an Intel Mac Pro, OSX.5, using SBCL, and with a new MOTU 828 Mk3 interface. All has gone well until I tried playing back a sound with CLM -- it immediately forces my MOTU into 192kHz mode and plays back the sound exceedingly fast. True also if I use (dac) to playback an AIF file.

It doesn't do this if I use built in sound -- stays in 44.1 like it should.

Hard to know where the problem lies here. Any clues or anyone who has overcome this?

Thanks,
-=Bret

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Heinrich Taube]

hi bret i have a mkII in my office and the lab and i dont have this  
problem with clm3+sbcl. i know thats not what you have but maybe that  
data point will help.

On Jul 20, 2008, at 2:23 PM, Bret Battey wrote:

> I just been transferring my CLM4 operations to an Intel Mac Pro, OSX.
> 5, using SBCL, and with a new MOTU 828 Mk3 interface. All has gone  
> well until I tried playing back a sound with CLM -- it immediately  
> forces my MOTU into 192kHz mode and plays back the sound exceedingly  
> fast. True also if I use (dac) to playback an AIF file.
>
> It doesn't do this if I use built in sound -- stays in 44.1 like it  
> should.
>
> Hard to know where the problem lies here. Any clues or anyone who  
> has overcome this?
>
> Thanks,
> -=Bret
> _______________________________________________
> Cmdist mailing list
> Cmdist@...
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

This has come up before -- I think in the previous cases, it was
fixable by disabling the error check around line 7345 of audio.c --
either don't set "err" via AudioDeviceSetProperty or afterwards
set it to noErr.  



_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bret Battey]

If I remove that entire error block, the MOTU srate remains unchanged, but playback is exceedingly fast.

If I force err = NoErr at line 7358, MOTU srate remains unchanged, but playback is exceedingly slow (!)

If I force err = NoErr at line 7368, MOTU is forced to 192k

Puzzled,
-=Bret

-----Original Message-----
From: Bill Schottstaedt [bil@...]
Sent: Sun 7/20/2008 9:54 PM
To: Bret Battey; cmdist@...
Subject: Re: [CM] CLM 4 Mac, DAC forces MOTU to 192k

This has come up before -- I think in the previous cases, it was
fixable by disabling the error check around line 7345 of audio.c --
either don't set "err" via AudioDeviceSetProperty or afterwards
set it to noErr. 

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

Is there a programmatic way to recognize this device?  If so, and
if you can figure out what it wants, we could add that case to
audio.c.  If there's some open-source program that works, I'd
be happy (well, to be honest, I'd be annoyed as hell, but it's
the Code of The Developer that any bug is a shining delight)  to
try to tease out how it differs from my code.


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bret Battey]

Hi Bill,

I just gave Audacity a try. Audacity preferences has an option "Do not modify audio device settings (such as sample rate)". With that checked, things work fine (Audacity plays back a 44.1kHz audio file at the correct speed with my MOTU set at 48kHz) -- and that is my preferred way of working.

If I uncheck that, however, Audacity hangs up. So it doesn't have a perfect solution either ;)

-=Bret


-----Original Message-----
From: Bill Schottstaedt [bil@...]
Sent: Mon 7/21/2008 2:31 PM
To: Bret Battey; cmdist@...
Subject: Re: [CM] CLM 4 Mac, DAC forces MOTU to 192k

Is there a programmatic way to recognize this device?  If so, and
if you can figure out what it wants, we could add that case to
audio.c.  If there's some open-source program that works, I'd
be happy (well, to be honest, I'd be annoyed as hell, but it's
the Code of The Developer that any bug is a shining delight)  to
try to tease out how it differs from my code.

_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

[Bill Schottstaedt]

I added mus-audio-output-properties-mutable in the Mac case.
I wonder what happens if you

   (mus-audio-output-properties-mutable #f)

On my Mac, it hangs, so I have high hopes.


_______________________________________________
Cmdist mailing list
Cmdist@...
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist