From: Peter Seibel
Subject: Another way to get a Lisp Machine
Date: 
Message-ID: <m3acuhv28w.fsf@javamonkey.com>
I don't think I've seen anything about this here or on the Lisp 
blogs.  Someone's got a CADR emulator up and running: 
<http://www.heeltoe.com/retro/cadr/>

-Peter

-- 
Peter Seibel                                      ·····@javamonkey.com

         Lisp is the red pill. -- John Fraser, comp.lang.lisp

From: Joe Marshall
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <hdoptlm6.fsf@ccs.neu.edu>
Peter Seibel <·····@javamonkey.com> writes:

> I don't think I've seen anything about this here or on the Lisp blogs.
> Someone's got a CADR emulator up and running:
> <http://www.heeltoe.com/retro/cadr/>

Wow!
From: Philip Haddad
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <ba57c4f9.0410201621.283d31f7@posting.google.com>
Peter Seibel <·····@javamonkey.com> wrote in message news:<··············@javamonkey.com>...
> I don't think I've seen anything about this here or on the Lisp 
> blogs.  Someone's got a CADR emulator up and running: 
> <http://www.heeltoe.com/retro/cadr/>
> 
> -Peter

That's awesome!

-- 
Certum quod factum.
Philip Haddad
From: Fred Gilham
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <u7mzyg4qcc.fsf@snapdragon.csl.sri.com>
> I don't think I've seen anything about this here or on the Lisp
> blogs.  Someone's got a CADR emulator up and running:
> <http://www.heeltoe.com/retro/cadr/>

I tried it out --- it's actually pretty neat in a kind of painful
way.  I think I even managed to get into Zmacs and I got some neat
looking menus to pop up.

I had to fiddle with a couple things in the code to get it to run
under FreeBSD, and the disk.img.gz file is actually a tar file not
just a gzipped file so you have to use tar and not just gzip -d to
extract the file.

-- 
Fred Gilham                                       ······@csl.sri.com
The reason the left prefers the neocon "right" to a paleo alternative
is, quite simply, that the neocons are essentially of the left
themselves and, thus, provide a fake opposition against which the rest
of the left can shadowbox and thereby perpetuate its own political and
cultural hegemony unchallenged by any authentic right. --Sam Francis
From: William Bland
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <pan.2004.10.21.05.36.10.800827@abstractnonsense.com>
On Wed, 20 Oct 2004 22:01:23 -0700, Fred Gilham wrote:

> 
>> I don't think I've seen anything about this here or on the Lisp
>> blogs.  Someone's got a CADR emulator up and running:
>> <http://www.heeltoe.com/retro/cadr/>
> 
> I tried it out --- it's actually pretty neat in a kind of painful
> way.  I think I even managed to get into Zmacs and I got some neat
> looking menus to pop up.
> 
> I had to fiddle with a couple things in the code to get it to run
> under FreeBSD, and the disk.img.gz file is actually a tar file not
> just a gzipped file so you have to use tar and not just gzip -d to
> extract the file.

I just got it working, but it seems to be... um, insane:

http://www.abstractnonsense.com/cadr.png

;-)

It is fun though - I got into Zmacs and knew my way around a bit from
using Emacs (well, until it dropped me into a debugger - then I hadn't a
clue).

Cheers,
	Bill.
-- 
"If you give someone Fortran, he has Fortran. If you give someone Lisp,
he has any language he pleases." -- Guy Steele
From: Robert Strandh
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <6wr7ns38y7.fsf@serveur5.labri.fr>
William Bland <·······@abstractnonsense.com> writes:

> I just got it working, but it seems to be... um, insane:
> 
> http://www.abstractnonsense.com/cadr.png

It looks like the base is set to 8 by default.

-- 
Robert Strandh

---------------------------------------------------------------------
Greenspun's Tenth Rule of Programming: any sufficiently complicated C
or Fortran program contains an ad hoc informally-specified bug-ridden
slow implementation of half of Common Lisp.
---------------------------------------------------------------------
From: William Bland
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <pan.2004.10.21.06.24.45.936993@abstractnonsense.com>
On Thu, 21 Oct 2004 08:02:24 +0200, Robert Strandh wrote:

> William Bland <·······@abstractnonsense.com> writes:
> 
>> I just got it working, but it seems to be... um, insane:
>> 
>> http://www.abstractnonsense.com/cadr.png
> 
> It looks like the base is set to 8 by default.

Ah yes, of course.  And I guess it doesn't have bignums, which is why it
gets (fact 10) wrong.

Cheers,
	Bill.
-- 
"If you give someone Fortran, he has Fortran. If you give someone Lisp,
he has any language he pleases." -- Guy Steele
From: William Bland
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <pan.2004.10.21.06.34.48.684734@abstractnonsense.com>
On Thu, 21 Oct 2004 06:21:41 +0000, William Bland wrote:

> On Thu, 21 Oct 2004 08:02:24 +0200, Robert Strandh wrote:
> 
>> William Bland <·······@abstractnonsense.com> writes:
>> 
>>> I just got it working, but it seems to be... um, insane:
>>> 
>>> http://www.abstractnonsense.com/cadr.png
>> 
>> It looks like the base is set to 8 by default.
> 
> Ah yes, of course.  And I guess it doesn't have bignums, which is why it
> gets (fact 10) wrong.

Oops, no (sorry, it's late) both *print-base* and *read-base* must be 8...
but then why does it let me ask for (fact 8) and (fact 9)? Surely the
digits 8 and 9 should cause an error?  Hmm...

Cheers,
	Bill.
-- 
"If you give someone Fortran, he has Fortran. If you give someone Lisp,
he has any language he pleases." -- Guy Steele
From: David Golden
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <S7Udd.39740$Z14.13854@news.indigo.ie>
William Bland wrote:

> 
> Oops, no (sorry, it's late) both *print-base* and *read-base* must be 8...
> but then why does it let me ask for (fact 8) and (fact 9)? Surely the
> digits 8 and 9 should cause an error?  Hmm...
> 

This might have been a deliberate "convenience" - I seem to recall
(but note "seem", maybe I hallucinated the whole thing) digits 8 and 9 being supported
in octal numbers, deliberately, by several unix tools I used in days of yore,
at least for single-digit \ escapes in strings, which are \ooo where ooo was
supposed to be 1-3 octal digits - so \9 worked as "tab" even though it really
shouldn't.

I think some (I have a nagging feeling TCL was involved, but it definitely throws
an error these days) even went as far as e.g.

0999 => 657

i.e. (+ (* 8 8 9) (* 8 9) 9)

as you'd expect... maybe...
From: David Golden
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <SrUdd.39741$Z14.14326@news.indigo.ie>
David Golden wrote:

> This might have been a deliberate "convenience" - I seem to recall
> (but note "seem", maybe I hallucinated the whole thing) digits 8 and 9
> being supported in octal numbers, deliberately, by several unix tools I
> used in days of yore, at least for single-digit \ escapes in strings,
> which are \ooo where ooo was supposed to be 1-3 octal digits - so \9
> worked as "tab" even though it really shouldn't.
> 

After a bit of a search, self-replying with a reference - see e.g.
http://www.gnu.org/software/binutils/manual/gas-2.9.1/html_chapter/as_3.html

"""
An octal character code. The numeric code is 3 octal digits. For compatibility
with other UNIX systems, 8 and 9 are accepted as digits: for example, \008 
has the value 010, and \009 the value 011
"""
From: William Bland
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <pan.2004.10.21.20.20.01.805498@abstractnonsense.com>
On Thu, 21 Oct 2004 20:52:17 +0100, David Golden wrote:

> After a bit of a search, self-replying with a reference - see e.g.
> http://www.gnu.org/software/binutils/manual/gas-2.9.1/html_chapter/as_3.html
> 
> """
> An octal character code. The numeric code is 3 octal digits. For compatibility
> with other UNIX systems, 8 and 9 are accepted as digits: for example, \008 
> has the value 010, and \009 the value 011
> """

Presumably I'm not the only one who thinks this "convenience" is actually
horribly ugly and could easily lead to hard-to-find bugs?  Or is that just
a hangover from my overly pedantic mathematical background?

Cheers,
	Bill.
-- 
"If you give someone Fortran, he has Fortran. If you give someone Lisp,
he has any language he pleases." -- Guy Steele
From: David Golden
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <8nVdd.39742$Z14.14106@news.indigo.ie>
William Bland wrote:

> 
> Presumably I'm not the only one who thinks this "convenience" is actually
> horribly ugly and could easily lead to hard-to-find bugs?

Indeed. I only know about it because it was horribly ugly and tended
to lead to hard-to-find bugs... 
From: Karl A. Krueger
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <clc14a$n0p$1@baldur.whoi.edu>
Ingvar <······@hexapodia.net> wrote:
> William Bland <·······@abstractnonsense.com> writes:
>> 
>> Presumably I'm not the only one who thinks this "convenience" is actually
>> horribly ugly and could easily lead to hard-to-find bugs?  Or is that just
>> a hangover from my overly pedantic mathematical background?
> 
> Not as bad as C's sscanf() function that thinks everything with a
> leading 0 is an octal number. Depending on how C code parses the IP
> address 10.010.020.030, this is either 10.10.20.30 or 10.8.16.24, for
> whole-family fun.

I've had that.  The network monitoring tool "tcpflow" prints out each
byte of an IP address zero-padded, so as to make them line up vertically.
This of course means that they cannot be cut-and-pasted into a command
line to a tool such as nmap or even ping.

Just Say No to Octal on modern systems ... :)

-- 
Karl A. Krueger <········@example.edu> { s/example/whoi/ }

Every program has at least one bug and can be shortened by at least one line.
By induction, every program can be reduced to one line which does not work.
From: Daniel Barlow
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <876552473p.fsf@noetbook.telent.net>
Ingvar <······@hexapodia.net> writes:

> Not as bad as C's sscanf() function that thinks everything with a
> leading 0 is an octal number. Depending on how C code parses the IP
[...]

javascript:alert(010+1)

I don't know what IE will do with tht, but Mozilla-based browsers (and
most likely also netscape 4, if anyone still has it) pop up a window
containing the text '9'



-dan

-- 
"please make sure that the person is your friend before you confirm"
From: Julian Stecklina
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <86hdol5xvg.fsf@goldenaxe.localnet>
Daniel Barlow <···@telent.net> writes:

> javascript:alert(010+1)
>
> I don't know what IE will do with tht, but Mozilla-based browsers (and
> most likely also netscape 4, if anyone still has it) pop up a window
> containing the text '9'

Just checked IE6: '9' ...

Regards,
-- 
                    ____________________________
 Julian Stecklina  /  _________________________/
  ________________/  /
  \_________________/  LISP - truly beautiful
From: Lukas Mai
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <cldqbn$g1i$1@wsc10.lrz-muenchen.de>
Ingvar schrob:
[...]

> Not as bad as C's sscanf() function that thinks everything with a
> leading 0 is an octal number.

Minor nit: It's not scanf itself but its %i (general integer) format
specifier. %d (decimal) doesn't have this problem.

-----8<-----
#include <stdio.h>
#include <stdlib.h>

int main(int argc, char **argv) {
	int n, i;
	if (argc < 2) {
		fprintf(stderr, "Usage: %s INTEGER [...]\n", argv[0]);
		return EXIT_FAILURE;
	}
	for (i = 1; i < argc; ++i) {
		printf("string: \"%s\"; ", argv[i]);

		if (sscanf(argv[i], "%i", &n) == 1) {
			printf("%%i: %d", n);
		} else {
			printf("%%i: conversion failed");
		}
		fputs("; ", stdout);

		if (sscanf(argv[i], "%d", &n) == 1) {
			printf("%%d: %d", n);
		} else {
			printf("%%d: conversion failed");
		}
		putchar('\n');
	}
	return 0;
}
----->8-----

$ ./a.out 2 010 019
string: "2"; %i: 2; %d: 2
string: "010"; %i: 8; %d: 10
string: "019"; %i: 1; %d: 19

> Depending on how C code parses the IP
> address 10.010.020.030, this is either 10.10.20.30 or 10.8.16.24, for
> whole-family fun.

Agreed.
Lukas
-- 
exit sub{&{$_[0]}}->(sub{defined$_[1]&&$_[0]->($_[0],getc,print chr sub{$_[1]
&&$_[1]<27?96&$_[0]|1+(12+$_[1])%26:$_[0]}->(ord$_[1],64^ord$_[1]&223))},getc)
From: Bruce O'Neel
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <w9loeipssz0.fsf@obs.unige.ch>
For those of use who don't know about the CADRs, how does one get
Zmacs to start?

Thanks!

cheers

bruce

Fred Gilham <······@snapdragon.csl.sri.com> writes:

> > I don't think I've seen anything about this here or on the Lisp
> > blogs.  Someone's got a CADR emulator up and running:
> > <http://www.heeltoe.com/retro/cadr/>
> 
> I tried it out --- it's actually pretty neat in a kind of painful
> way.  I think I even managed to get into Zmacs and I got some neat
> looking menus to pop up.
> 
> I had to fiddle with a couple things in the code to get it to run
> under FreeBSD, and the disk.img.gz file is actually a tar file not
> just a gzipped file so you have to use tar and not just gzip -d to
> extract the file.
> 
> -- 
> Fred Gilham                                       ······@csl.sri.com
> The reason the left prefers the neocon "right" to a paleo alternative
> is, quite simply, that the neocons are essentially of the left
> themselves and, thus, provide a fake opposition against which the rest
> of the left can shadowbox and thereby perpetuate its own political and
> cultural hegemony unchallenged by any authentic right. --Sam Francis

-- 
Reality is 80m polygons - Alvy Ray Smith
Bruce O'Neel - ·······@mindspring.com 
http://homepage.iprolink.ch/~bioneel/beo/beo.html - daily stuff
From: Christopher C. Stacy
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <uoeipailk.fsf@news.dtpq.com>
Bruce O'Neel <···········@obs.unige.ch> writes:

> For those of use who don't know about the CADRs, how does
> one get Zmacs to start?

Type these two characters: [SYSTEM] and the letter "e",
or call the function ED.

You can also use APROPOS to help guess at things like this.
From: Vladimir Sedach
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <87hdopuybg.fsf@shawnews.cg.shawcable.net>
Peter Seibel <·····@javamonkey.com> writes:

> I don't think I've seen anything about this here or on the Lisp blogs.
> Someone's got a CADR emulator up and running:
> <http://www.heeltoe.com/retro/cadr/>

The fun doesn't stop there!
http://www.heeltoe.com/retro/lispm/l-machine.html

This is the most _awesome_ news I've heard in a while!

> -Peter
> 
> -- 
> Peter Seibel                                      ·····@javamonkey.com
> 
>          Lisp is the red pill. -- John Fraser, comp.lang.lisp

Vladimir
From: William Bland
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <pan.2004.10.21.01.22.15.921265@abstractnonsense.com>
On Wed, 20 Oct 2004 22:48:36 +0000,  wrote:

> Peter Seibel <·····@javamonkey.com> writes:
> 
>> I don't think I've seen anything about this here or on the Lisp blogs.
>> Someone's got a CADR emulator up and running:
>> <http://www.heeltoe.com/retro/cadr/>
> 
> The fun doesn't stop there!
> http://www.heeltoe.com/retro/lispm/l-machine.html
> 
> This is the most _awesome_ news I've heard in a while!
> 

That is indeed absolutely fantastic news!  Can't wait to play with it. 
Does anyone know anything about the legal status of the stuff being
used? (I'm not trying to imply anything btw - I just don't know much
about legal stuff and am interested).  Is code in patents, for example,
automatically in the public domain?  Or has a sufficient amount of time
passed that it is now?

Cheers,
	Bill.
-- 
"If you give someone Fortran, he has Fortran. If you give someone Lisp,
he has any language he pleases." -- Guy Steele
From: Christopher C. Stacy
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <ulldzlbsq.fsf@news.dtpq.com>
William Bland <·······@abstractnonsense.com> writes:

> On Wed, 20 Oct 2004 22:48:36 +0000,  wrote:
> 
> > Peter Seibel <·····@javamonkey.com> writes:
> > 
> >> I don't think I've seen anything about this here or on the Lisp blogs.
> >> Someone's got a CADR emulator up and running:
> >> <http://www.heeltoe.com/retro/cadr/>
> > 
> > The fun doesn't stop there!
> > http://www.heeltoe.com/retro/lispm/l-machine.html
> > 
> > This is the most _awesome_ news I've heard in a while!
> > 
> 
> That is indeed absolutely fantastic news!  Can't wait to play with it. 

It is very cool, and should be lots of fun to play with.

Not to throw cold water or anything, but I'd like to point out that
this is the original version of the MIT Lisp Machine.  That puts it
about 1,000 man-years behind the Symbolics version of the Lisp Machine
operating system, so it does not have many of the nice things that you
hear people talk about all the time when they discuss "Lisp Machines".
(It would be a major undertaking to go into a good explanation all the
differences.  I don't think it has ANSI Common Lisp, for one small example.)
I just wanted to mention this in order to set expectations properly, 
rather than having people come along complaining, "_That's it_?!?"

(And before people start asking if there is a version for the Symbolics...
The Symbolics version is still a commercial product that you can buy,
either as obsolete hardware, or in its software-emulator form.
The software version runs on DEC Alphas.  Last I heard, Symbolics
was interested in funding for porting it to more popular architectures.
It seems unlikely that any of this would become available for free.)
From: William Bland
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <pan.2004.10.22.03.34.50.650349@abstractnonsense.com>
On Fri, 22 Oct 2004 02:38:28 +0000, Christopher C. Stacy wrote:

> It is very cool, and should be lots of fun to play with.
> 
> Not to throw cold water or anything, but I'd like to point out that
> this is the original version of the MIT Lisp Machine.  That puts it
> about 1,000 man-years behind the Symbolics version of the Lisp Machine
> operating system, so it does not have many of the nice things that you
> hear people talk about all the time when they discuss "Lisp Machines".
> (It would be a major undertaking to go into a good explanation all the
> differences.  I don't think it has ANSI Common Lisp, for one small example.)
> I just wanted to mention this in order to set expectations properly, 
> rather than having people come along complaining, "_That's it_?!?"

Understood, but taken in it's historical context I still see much to
be impressed with.  For a trivial example, I was surprised that a piece of
hardware built in the 70s had a high enough resolution display that the
emulator window looks big on a display built in 2004.

Cheers,
	Bill.
-- 
"If you give someone Fortran, he has Fortran. If you give someone Lisp,
he has any language he pleases." -- Guy Steele
From: Tim Bradshaw
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <1098437449.455455.233140@c13g2000cwb.googlegroups.com>
William Bland wrote:
> Understood, but taken in it's historical context I still see much to
> be impressed with.  For a trivial example, I was surprised that a
piece of
> hardware built in the 70s had a high enough resolution display that
the
> emulator window looks big on a display built in 2004.
>

I'm not sure that display resolutions have gone up very much for a
long time. The screen I'm using now (on some random PC) is 1280x1024.
I think the standard Sun3 screen resoluton was 1152x900, and that's
probably the same for Sun2s I think (so that resolution was common at
least back to 1984 I think).  Hires Sun3s were 1600x1280 and were
certainly available in 1986 or so I think: monitors not being sharp
enough was often a problem for them.

So I think it's fair to say that screen resolutions haven't gone up
that much since the early 80s, anyway.  Of course colour depths have
gone up a bit...

It may be that I'm stuck in the past and everyone is using 3000x2000
displays now.  If so I want one!  I suspect I'm not because a
3000x2000 CRT would have to be fairly well-aligned, especially in
colour, and a 3000x2000 LCD would be rather expensive I think.  So may
be the answer is that resolutions fairly rapidly got high enough that
the physical construction of a display device became an issue, and
display technology (although it's changed completely from CRT to LCD
for most people) hasn't actually got that much better in terms of
resolution for 20 years or something.

--tim
From: Christopher C. Stacy
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <uacuesyew.fsf@news.dtpq.com>
"Tim Bradshaw" <··········@tfeb.org> writes:

> William Bland wrote:
> > Understood, but taken in it's historical context I still see much to
> > be impressed with.  For a trivial example, I was surprised that a
> > piece of hardware built in the 70s had a high enough resolution display 
> > that the emulator window looks big on a display built in 2004.
> >
> So I think it's fair to say that screen resolutions haven't gone up
> that much since the early 80s, anyway.  Of course colour depths have
> gone up a bit...

But given that these Lisp Machines with their cutting-edge bitmap
displays cost about $75,000 each, it's fair to say that the costs
_have_ gone down significantly!  (Also, the CRTs used on the CADRs
were rather unreliable and very fussy!  I don't remember what fraction
of the cost was the display memory and CRTs, but it was noticable.)

CADRs also had optional high-res color displays.  I don't remember
what "high res" was or the depth.  Only a few were built; mostly
just used for Mandlebrot demos and VLSI design applications.

Later Liap Machines were much fancier, of course, 
From: Tim Bradshaw
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <1098827771.684845.10530@c13g2000cwb.googlegroups.com>
Christopher C. Stacy wrote:

> But given that these Lisp Machines with their cutting-edge bitmap
> displays cost about $75,000 each, it's fair to say that the costs
> _have_ gone down significantly!  (Also, the CRTs used on the CADRs
> were rather unreliable and very fussy!  I don't remember what
fraction
> of the cost was the display memory and CRTs, but it was noticable.)

yes, of course, the costs have come down enormously - I'd not thought
of that.  The workstations I used were 10s of K but now everyone has
high-resolution (and many-bit colour) for hardly anything.  And, as you
say, you don't make support calls on screens very often now...

Interestingly, I suspect the fractional cost might not have gone down
at all: some of the graphics cards you see for games applications (I
guess) are pretty expensive!

--tim
From: Espen Vestre
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <kw1xfqappu.fsf@merced.netfonds.no>
"Tim Bradshaw" <··········@tfeb.org> writes:

> It may be that I'm stuck in the past and everyone is using 3000x2000
> displays now.  If so I want one!  I suspect I'm not because a

Not quite. Even the 30" Apple LCD is "just" 2560 x 1600.

(But I think the screen resolution revolution I have been waiting for
 since the first time I used a laser printer about 20 years ago is
 finally on the horizon - in a few years we might be using 300dpi
 lcd or oled screens)
-- 
  (espen)
From: Bulent Murtezaoglu
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <87u0sm8zwh.fsf@p4.internal>
>>>>> "TB" == Tim Bradshaw <··········@tfeb.org> writes:
[...]
    TB> It may be that I'm stuck in the past and everyone is using
    TB> 3000x2000 displays now.  If so I want one!  

If you can justify $3k and are willing to locate a suitable card for 
high refresh rates on eBay:

http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=36083&Sku=I74-1544&CatId=170

    TB> I suspect I'm not
    TB> because a 3000x2000 CRT would have to be fairly well-aligned,
    TB> especially in colour, and a 3000x2000 LCD would be rather
    TB> expensive I think.  

It seems you cannot carry that kind of signal around at high enough
refresh by analog means with commodity hardware.  I got high-end
sony CRT, and as the dot clock goes over 300MHz, I start to get enough
ringing and ghosting artifacts that even when the color alignment is
OK, a decent reproduction of the signal as seen by the monitor doesn't
yield great results.

    TB> So may be the answer is that resolutions
    TB> fairly rapidly got high enough that the physical construction
    TB> of a display device became an issue, and display technology
    TB> (although it's changed completely from CRT to LCD for most
    TB> people) hasn't actually got that much better in terms of
    TB> resolution for 20 years or something.

Yes.  I'd trade a gigahertz or so from the CPU speed for an equivalent 
progress in displays.

BM
From: Pascal Bourguignon
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <87ekjqxzuk.fsf@thalassa.informatimago.com>
"Tim Bradshaw" <··········@tfeb.org> writes:
> It may be that I'm stuck in the past and everyone is using 3000x2000
> displays now.  If so I want one!  I suspect I'm not because a
> 3000x2000 CRT would have to be fairly well-aligned, especially in
> colour, and a 3000x2000 LCD would be rather expensive I think.  

IBM has a 200 dpi LCD that must be close.  About $8000.

-- 
__Pascal Bourguignon__                     http://www.informatimago.com/

Voting Democrat or Republican is like choosing a cabin in the Titanic.
From: Paul Khuong
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <a828a711.0410222359.6e13f0f4@posting.google.com>
Pascal Bourguignon <····@mouse-potato.com> wrote in message news:<··············@thalassa.informatimago.com>...
> "Tim Bradshaw" <··········@tfeb.org> writes:
> > It may be that I'm stuck in the past and everyone is using 3000x2000
> > displays now.  If so I want one!  I suspect I'm not because a
> > 3000x2000 CRT would have to be fairly well-aligned, especially in
> > colour, and a 3000x2000 LCD would be rather expensive I think.  
> 
> IBM has a 200 dpi LCD that must be close.  About $8000.
The Toshiba M200 has a 1400x1050 12.1" screen. That's (iirc, i don't
feel like doing the calculations) 144 dpi, and 432 horizontal dpi with
subpixel rendering. It's probably much less than 8k too.

Paul Khuong
From: Paul Khuong
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <a828a711.0410230913.4ab6a679@posting.google.com>
·······@gmail.com (Paul Khuong) wrote in message news:<····························@posting.google.com>...
> Pascal Bourguignon <····@mouse-potato.com> wrote in message news:<··············@thalassa.informatimago.com>...
> > "Tim Bradshaw" <··········@tfeb.org> writes:
> > > It may be that I'm stuck in the past and everyone is using 3000x2000
> > > displays now.  If so I want one!  I suspect I'm not because a
> > > 3000x2000 CRT would have to be fairly well-aligned, especially in
> > > colour, and a 3000x2000 LCD would be rather expensive I think.  
> > 
> > IBM has a 200 dpi LCD that must be close.  About $8000.
> The Toshiba M200 has a 1400x1050 12.1" screen. That's (iirc, i don't
> feel like doing the calculations) 144 dpi, and 432 horizontal dpi with
> subpixel rendering. It's probably much less than 8k too.
Argh. I'm replying to myself: This was meant to go under Espen
Vestre's post (http://groups.google.com/groups?hl=en&lr=&selm=kw1xfqappu.fsf%40merced.netfonds.no)
From: Steven M. Haflich
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <J0yed.17560$nj.15059@newssvr13.news.prodigy.com>
Tim Bradshaw wrote:

> I'm not sure that display resolutions have gone up very much for a
> long time. The screen I'm using now (on some random PC) is 1280x1024.
> I think the standard Sun3 screen resoluton was 1152x900, and that's

(For many years I used a particular Sun3 that had a 1600x1200 display
as my X server, until the hardware died...)

No one has remembered that 1200 or 1600 linear pixel displays were not
unknown in the 1980's on normal, stock systems, and certainly not in
the late 1980's.  However, color displays _were_ rare in the 1980's
since both the monitor and display hardware was much more expensive.
The 64MB display cards that are common today have more memory than
the entirety of any period Lisp Machine I've ever seen.

When Nicholas Negroponte was opening the MIT Media Lab around 1984,
be commented (only partially in jest) that he didn't want any B&W
monitors in the building.  But even many years after I still
preferred a B&W monitor since most of what I did involved program
text, and at the time affordable color monitors were not as good as
B&W when displaying text.  Nowadays, of course, technology has
improved, there is no reason not to use a color display, and there
are more applications that exploit color effectively.

Color monitors were available on Symbolics, but they were expensive
and correspondingly rare.
From: Rainer Joswig
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <joswig-DE8228.21382623102004@news-50.dca.giganews.com>
In article <····················@newssvr13.news.prodigy.com>,
 "Steven M. Haflich" <·················@alum.mit.edu> wrote:

> Tim Bradshaw wrote:
> 
> > I'm not sure that display resolutions have gone up very much for a
> > long time. The screen I'm using now (on some random PC) is 1280x1024.
> > I think the standard Sun3 screen resoluton was 1152x900, and that's
> 
> (For many years I used a particular Sun3 that had a 1600x1200 display
> as my X server, until the hardware died...)
> 
> No one has remembered that 1200 or 1600 linear pixel displays were not
> unknown in the 1980's on normal, stock systems, and certainly not in
> the late 1980's.  However, color displays _were_ rare in the 1980's
> since both the monitor and display hardware was much more expensive.
> The 64MB display cards that are common today have more memory than
> the entirety of any period Lisp Machine I've ever seen.
> 
> When Nicholas Negroponte was opening the MIT Media Lab around 1984,
> be commented (only partially in jest) that he didn't want any B&W
> monitors in the building.  But even many years after I still
> preferred a B&W monitor since most of what I did involved program
> text, and at the time affordable color monitors were not as good as
> B&W when displaying text.  Nowadays, of course, technology has
> improved, there is no reason not to use a color display, and there
> are more applications that exploit color effectively.
> 
> Color monitors were available on Symbolics, but they were expensive
> and correspondingly rare.

They weren't that rare, because many Symbolics Lisp Machines
were sold as graphics workstation in the area of
scientific imaging, knowledge based image processing,
modelling and animation. In the AI lab here in Hamburg
we had a 3640 with a 1280x1024 24bit color monitor.
It has for example been used in a work on temporal
filters of 3d image sequences. A bit later Symbolics tried
to leverage the less expensive Macintosh hardware.
They supported some of their graphics software on
the MacIvory platform (for example using the Nuvista
framebuffer).
From: Tim Bradshaw
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <1098828087.614403.62120@z14g2000cwz.googlegroups.com>
Steven M. Haflich wrote:
> Tim Bradshaw wrote:
>
> > I'm not sure that display resolutions have gone up very much for a
> > long time. The screen I'm using now (on some random PC) is
1280x1024.
> > I think the standard Sun3 screen resoluton was 1152x900, and that's
>
> (For many years I used a particular Sun3 that had a 1600x1200 display
> as my X server, until the hardware died...)
>
> No one has remembered that 1200 or 1600 linear pixel displays were
not
> unknown in the 1980's on normal, stock systems, and certainly not in
> the late 1980's.

Well, in the bit of my message you deleted I said that Hires suns were
1600x1280, so I remembered.  We had two, in fact, and very nice they
were too.

> However, color displays _were_ rare in the 1980's
> since both the monitor and display hardware was much more expensive.
> The 64MB display cards that are common today have more memory than
> the entirety of any period Lisp Machine I've ever seen.

Yes, and as you said in the bit I elided (see, I get to do it too (:-),
they were rotten for text.  Even good colour monitors were marginal,
especially if you moved / dropped them or they got old.

--tim
From: Brad Parker
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <f570f7ac.0410210733.4b4aa49f@posting.google.com>
Peter Seibel <·····@javamonkey.com> wrote in message news:<··············@javamonkey.com>...
> I don't think I've seen anything about this here or on the Lisp 
> blogs.  Someone's got a CADR emulator up and running: 
> <http://www.heeltoe.com/retro/cadr/>

I just fixed the bug which made floating point print out wrong
and caused the dates be wrong.

Basically the divide logic didn't take into account that
the Q and ALU rotate occur at the same moment in time.
(ooops, von Neumann versus hardware :-)

-brad
From: Thibault Langlois
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <1fff13b3.0410220118.6b48f873@posting.google.com>
····@heeltoe.com (Brad Parker) wrote in message news:<····························@posting.google.com>...
> Peter Seibel <·····@javamonkey.com> wrote in message news:<··············@javamonkey.com>...
> > I don't think I've seen anything about this here or on the Lisp 
> > blogs.  Someone's got a CADR emulator up and running: 
> > <http://www.heeltoe.com/retro/cadr/>
> 
> I just fixed the bug which made floating point print out wrong
> and caused the dates be wrong.
> 
> Basically the divide logic didn't take into account that
> the Q and ALU rotate occur at the same moment in time.
> (ooops, von Neumann versus hardware :-)
> 
> -brad

Hello,

It would be nice to have a qwerty keyboard mapping. On my portuguese
(slightly modified qwerty keyboard) I do not have the / key for
example.

When the systems starts, it asks for date and time, what is the
expected format ? From a previously posted bitmap it seems to be
mm/dd/yy hh:mm:ss . It seems strange to me.

Thibault
From: Joe Marshall
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <oeiug8vf.fsf@ccs.neu.edu>
····@heeltoe.com (Brad Parker) writes:

> Peter Seibel <·····@javamonkey.com> wrote in message news:<··············@javamonkey.com>...
>> I don't think I've seen anything about this here or on the Lisp 
>> blogs.  Someone's got a CADR emulator up and running: 
>> <http://www.heeltoe.com/retro/cadr/>

How's the performance compare to the original?
From: Dave
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <d4aa8e8a.0411011705.1cf51f79@posting.google.com>
Peter Seibel <·····@javamonkey.com> wrote in message news:<··············@javamonkey.com>...
> I don't think I've seen anything about this here or on the Lisp 
> blogs.  Someone's got a CADR emulator up and running: 
> <http://www.heeltoe.com/retro/cadr/>
> 
> -Peter

If anyone has this built for Mac OS-X would you please send it to me
or post somewhere?  I've gotten hopelessly lost building the SDL
library (./configure can't find Makefile.in?).  Thanks.

-Dave
From: Rahul Jain
Subject: Re: Another way to get a Lisp Machine
Date: 
Message-ID: <87k6t241kd.fsf@nyct.net>
············@yahoo.com (Dave) writes:

> Peter Seibel <·····@javamonkey.com> wrote in message news:<··············@javamonkey.com>...
>> I don't think I've seen anything about this here or on the Lisp 
>> blogs.  Someone's got a CADR emulator up and running: 
>> <http://www.heeltoe.com/retro/cadr/>
>> 
>> -Peter
>
> If anyone has this built for Mac OS-X would you please send it to me
> or post somewhere?  I've gotten hopelessly lost building the SDL
> library (./configure can't find Makefile.in?).  Thanks.

You need to run automake first.

-- 
Rahul Jain
·····@nyct.net
Professional Software Developer, Amateur Quantum Mechanicist