25 Sep
2004
25 Sep
'04
6:43 a.m.
25 Sep
25 Sep
7:11 a.m.
Brian Knittel wrote:
Only 2399 more to go and you can clone this. :)
http://www.epemag.com/zuse/part4a.htm
Imagine
implementing this trivial logic:
y1 = a & b;
y2 = a & c;
one relay 
8:53 a.m.
New subject: simple processor from 74154?
Does anyone remember a construction project from Popular Electronics (I think),
perhaps in the late '60's, that made a simple processor from a 74154 4 line
demultiplexer and a handful of diodes?
John Tinker
9:14 a.m.
New subject: QBUS boards...
I've got some QBUS boards that I'm trying to identify and place a value
on.
First up is a Parallax graphics board with the number 990-0001 on it.
It's got a small cab kit with four BNC connectors on it marked RGB and
Sync. There's a 75Ohm terminating resistor on the Sync connector.
Next is a board made by MDB Systems that has "MLSI-DR11-W" printed on it.
It's got a cab kit with it that has two 37 pin(?) d shell connectors on
it.
I've got a cab kit for a PDP-11 of some kind (I _think_). It's got two
25 pin d-shell connectors on it marked "A0 (Console)" and A1. Each has a
rotary knob underneath for setting a baud rate.
A D315 distribution panel. Has four 25 pin d-shell connectors on it.
A three board set joined with short IDC cables consisting of one M7169 and
two M7168 modules.
Any input on this will be greatly appriciated!
g.
--
"I'm not crazy, I'm plausibly off-nominal!"
Proud owner of 80-0007
http://www.f15sim.com - The only one of its kind.
11:59 a.m.
New subject: QBUS boards...
Gene Buckle declared on Saturday 25 September 2004 04:14 pm:
First up is a Parallax graphics board with the number
990-0001 on it.
It's got a small cab kit with four BNC connectors on it marked RGB and
Sync. There's a 75Ohm terminating resistor on the Sync connector.
It looks like Parallax produced at least a few different QBUS framebuffer
boards.
Next is a board made by MDB Systems that has
"MLSI-DR11-W" printed on
it. It's got a cab kit with it that has two 37 pin(?) d shell
connectors on it.
It's probably the MDB QBUS equivalent of a DEC DR11-W, which is a
"General purpose DMA parallel interface".
I've got a cab kit for a PDP-11 of some kind (I
_think_). It's got
two 25 pin d-shell connectors on it marked "A0 (Console)" and A1.
Each has a rotary knob underneath for setting a baud rate.
This sounds like the cab kit for an 11/23+, or possibly an 11/53.
A D315 distribution panel. Has four 25 pin d-shell
connectors on it.
A three board set joined with short IDC cables consisting of one M7169
and two M7168 modules.
That's a VCB02 framebuffer card for a QBUS VaxStation.
Nothing looks particularly valuable from the list. I'd say $5-10 per
item would be a fair value, unless you find someone who has a real need
for one of them.
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
26 Sep
26 Sep
7:08 a.m.
New subject: QBUS boards...
At 07:59 PM 9/25/2004, you wrote:
Nothing looks particularly valuable from the list.
I'd say $5-10 per
item would be a fair value, unless you find someone who has a real need
for one of them.
Ditto, except for the DR11-W.
People are still doing high speed data transfer to/from PDP11's and might
pay a premium for that board.
Ed K.
7:54 a.m.
New subject: QBUS boards...
Ed Kelleher declared on Sunday 26 September 2004 02:08 pm:
At 07:59 PM 9/25/2004, you wrote:
True, but it is an MDB part, not a DEC part, and you might have trouble
finding docs for setting the jumpers, etc.
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
Nothing looks particularly valuable from the list.
I'd say $5-10 per
item would be a fair value, unless you find someone who has a real
need for one of them.
Ditto, except for the DR11-W.
People are still doing high speed data transfer to/from PDP11's and
might pay a premium for that board. 
25 Sep
25 Sep
9:45 a.m.
New subject: simple processor from 74154?
On Sat, 25 Sep 2004, John Tinker wrote:
Does anyone remember a construction project from
Popular Electronics (I
think), perhaps in the late '60's, that made a simple processor from a
74154 4 line demultiplexer and a handful of diodes?
Did they even have 74154's in the late 1960s? At any rate, sounds like an
interesting project.
--
Sellam Ismail Vintage Computer Festival
------------------------------------------------------------------------------
International Man of Intrigue and Danger http://www.vintage.org
[ Old computing resources for business || Buy/Sell/Trade Vintage Computers ]
[ and academia at www.VintageTech.com || at http://marketplace.vintage.org ]
10:16 a.m.
New subject: simple processor from 74154?
At 05:45 PM 9/25/2004, Sellam wrote:
TI TTL Data Book for Design Engineers, 1st edition was copyrighted 1973.
It lists the 74154 and other parts up to the 74298.
Ed K.
Does anyone
remember a construction project from Popular Electronics (I
think), perhaps in the late '60's, that made a simple processor from a
74154 4 line demultiplexer and a handful of diodes?
Did they even have 74154's in the late 1960s? At any rate, sounds like an
interesting project.
10:55 a.m.
New subject: simple processor from 74154?
Ed Kelleher wrote:
At 05:45 PM 9/25/2004, Sellam wrote:
TI TTL Data Book for Design Engineers, 1st edition was copyrighted 1973.
It lists the 74154 and other parts up to the 74298.
How come the neat CMOS chips like BCD adders and stuff have never made it
to TTL?
Does
anyone remember a construction project from Popular Electronics (I
think), perhaps in the late '60's, that made a simple processor from a
74154 4 line demultiplexer and a handful of diodes?
Did they even have 74154's in the late 1960s? At any rate, sounds
like an
interesting project.
26 Sep
26 Sep
7:04 a.m.
New subject: simple processor from 74154?
At 06:55 PM 9/25/2004, Ben wrote:
TI TTL Data Book for Design Engineers, 1st edition was copyrighted 1973.
It lists the 74154 and other parts up to the 74298. How come the neat CMOS chips
like BCD adders and stuff have never made it
to TTL?
Probably because TTL was a step backwards.
I use 74HCxxx for my stuff now.
Does anyone
remember a construction project from Popular Electronics (I
think), perhaps in the late '60's, that made a simple processor from a
74154 4 line demultiplexer and a handful of diodes?
Did they even have 74154's in the late 1960s? At any rate, sounds like an
interesting project.
25 Sep
25 Sep
5:14 p.m.
New subject: simple processor from 74154?
9/25/2004 5:45:54 PM, Vintage Computer Festival <vcf(a)siconic.com> wrote:
On Sat, 25 Sep 2004, John Tinker wrote:
Maybe not the late 1960's. And maybe not PE, either. But I definitely saw the
article.
John Tinker
Does anyone remember a construction project from
Popular Electronics (I
think), perhaps in the late '60's, that made a simple processor from a
74154 4 line demultiplexer and a handful of diodes?
Did they even have 74154's in the late 1960s? At any rate, sounds like an
interesting project. 
26 Sep
26 Sep
1:05 a.m.
New subject: simple processor from 74154?
http://home.gwi.net/~jdebell/pe/PEIntro.htm
Above link is for someone assembling an online archive.
Geoff.
Does anyone remember a construction project from
Popular Electronics (I
think),
perhaps in the late '60's, that made a simple
processor from a 74154 4
line
demultiplexer and a handful of diodes?
John Tinker
25 Sep
25 Sep
6:15 p.m.
On Sat, 2004-09-25 at 12:11, ben franchuk wrote:
Only 2399 more to go and you can clone this. :)
http://www.epemag.com/zuse/part4a.htm
Are Zuse's machines stored-program? I think he had some odd ideas about
stored-program machines and self-modification, like von Neumann did
early on.
6:37 p.m.
Tom Jennings wrote:
Are Zuse's machines stored-program? I think he had
some odd ideas about
stored-program machines and self-modification, like von Neumann did
early on.
All machines of that era had too *small* a memory for internal storage
of programs, so external ( read manual ) input was needed like paper
tape or punched cards or punched film (Zuse).
7:10 p.m.
On Sat, 2004-09-25 at 23:37, ben franchuk wrote:
Tom Jennings wrote:
Eh, that's not a good enough excuse :-)
Are Zuse's machines stored-program? I think
he had some odd ideas about
stored-program machines and self-modification, like von Neumann did
early on.
All machines of that era had too *small* a memory for internal storage
of programs, so external ( read manual ) input was needed like paper
tape or punched cards or punched film (Zuse). 
8:37 p.m.
New subject: Early computers was Re: Relay computers
Tom Jennings wrote:
On Sat, 2004-09-25 at 23:37, ben franchuk wrote:
>Tom Jennings wrote:
>>Are Zuse's machines stored-program?
>All machines of that era had too *small* a memory
for internal storage
>of programs, so external ( read manual ) input was needed like paper
>tape or punched cards or punched film (Zuse).
Eh, that's not a good enough excuse :-)
Indeed, the Manchester Baby machine of 1948 had 32 words of memory, and
a 3 bit opcode field, despite this it was fully functional a stored
program computer, if somewhat limited in its application.
On the Zuse machine, there is a paper by Rojas which discusses how to
make it into a general purpose computer :
Rojas, Raul: How to make Zuse?s Z3 a Universal Computer.
IEEE Annals of Computing, Vol. 20, No. 3, July/Sept. 1998..
This web page also seems apropro :
http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/Neumann_vs_Zuse.html
-- HansP
26 Sep
26 Sep
5:12 a.m.
New subject: Early computers was Re: Relay computers
Hans B PUFAL wrote:
> Eh, that's not a good enough excuse :-)
umm... It was after midnight I did not want to try think much more.
Now re-reading Zuse again you find out he was a mechanical minded
person rather than pure science or math. To me his computing engine
was mechanical rather than electronic ( tubes ) and much a greater
advent guenus in that field compared to computing in general.
Also the fact that Germany lost the war, did not improve matters.
Indeed, the Manchester Baby machine of 1948 had 32
words of memory, and
a 3 bit opcode field, despite this it was fully functional a stored
program computer, if somewhat limited in its application.
So why does the Baby Emulator need Win/95 to run? :)
On the Zuse machine, there is a paper by Rojas which
discusses how to
make it into a general purpose computer :
I am more happy just to find out how you used them and what the
instruction set
was?
PS. I thinking of building a smaller decimal machine ( 6 digits ) and
wondering
where a good source of info on the web with the way to handle the math.
BCD? Execess 3, some other codeing?
Ben.
PS.
*Answer* You can have a bit mapped display to get the
Scope tube display.
6:33 a.m.
New subject: Early computers was Re: Relay computers
On Sep 26, 2004, at 10:12 AM, ben franchuk wrote:
PS. I thinking of building a smaller decimal machine (
6 digits ) and
wondering
where a good source of info on the web with the way to handle the math.
BCD? Execess 3, some other codeing?
Why not have it add ascii directly... "1" + "1" = "2"
:^)
7:49 a.m.
New subject: Early computers was Re: Relay computers
On Sep 26, 2004, at 11:51 AM, David V. Corbin wrote:
no one... just me just now... Didn't and older IBM machine store
numbers digit
by digit in consecutive words (allowing you to setup integers as
long/wide as you
needed) ?
>> Ron
wrote....
>> Why not have it add ascii directly... "1" + "1" =
"2" :^)
Who EVER implied a relation between decimal representation and ascii
representation??????????
David.
9:12 a.m.
New subject: Early computers was Re: Relay computers
Ron Hudson wrote:
no one... just me just now... Didn't and older
IBM machine store
numbers digit
by digit in consecutive words (allowing you to setup integers as
long/wide as you
needed) ?
The IBM 1620 did that. Mind you back then you used about
10 digits for most math. Simh will emulate that machine but I
don't think there is any software ( assembler / fortran II ) around still
for that machine.
9:51 a.m.
New subject: Early computers was Re: Relay computers
On Sun, 26 Sep 2004, ben franchuk wrote:
The IBM 1620 did that. Mind you back then you used
about
10 digits for most math. Simh will emulate that machine but I
don't think there is any software ( assembler / fortran II ) around still
for that machine.
The card cabinets that were dumpstered at my office two years
ago included PDQ FORTRAN for the 1620 and a 1401 emulator for the 1620.
12:01 p.m.
New subject: Early computers was Re: Relay computers
> The card cabinets that were dumpstered at my
office two years
> ago included PDQ FORTRAN for the 1620 and a 1401 emulator for the 1620.
On Sun, 26 Sep 2004, ben franchuk wrote:
Well get them read soon!
Will NEVER be able to read them.
I can provide you with the names of the college administrators
who were responsible for the destruction, which also included
more than a dozen office-supply-cabinets full of collectible
computer stuff and personal property, and bookcases full of
my old computer magazines. One of the other instructors got
in major trouble (they tried to arrest him!) for trying to
rescue some of the stuff.
--
Grumpy Ol' Fred cisin(a)xenosoft.com
27 Sep
27 Sep
2:03 a.m.
New subject: Early computers was Re: Relay computers
>>>> "Ron" == Ron Hudson
<ron.hudson(a)sbcglobal.net> writes:
Ron> On Sep 26, 2004, at 11:51 AM, David V. Corbin wrote:
>>>> Ron wrote.... Why not have it add
ascii directly... "1" + "1"
>>>> = "2" :^)
>
>
> Who EVER implied a relation between decimal representation and
> ascii representation??????????
>
> David.
>
>
Ron> no one... just me just now... Didn't and older IBM machine
Ron> store numbers digit by digit in consecutive words (allowing you
Ron> to setup integers as long/wide as you needed) ?
Yes, that exactly describes the IBM 1620. (It even had variable
length floating point -- any number of digits for the mantissa, though
always 2 digits for the exponent.)
paul
7:27 a.m.
New subject: Early computers was Re: Relay computers
On Sun, 2004-09-26 at 11:51, David V. Corbin wrote:
>>> Ron wrote....
>>> Why not have it add ascii directly... "1" + "1" =
"2" :^)
Who EVER implied a relation between decimal
representation and ascii
representation??????????
If you generalize "ASCII" to "character set values" then many
machines
did just that.
26 Sep
26 Sep
8:43 a.m.
New subject: Early computers was Re: Relay computers
Ron Hudson wrote:
On Sep 26, 2004, at 10:12 AM, ben franchuk wrote:
.-- .
.... .- ...- .
-- --- .-. ... .
-.-. --- -.. .
:)
PS. I thinking of building a smaller decimal
machine ( 6 digits ) and
wondering
where a good source of info on the web with the way to handle the math.
BCD? Execess 3, some other codeing?
Why not have it add ascii directly... "1" + "1" = "2"
:^)
27 Sep
27 Sep
4:06 a.m.
New subject: Early computers was Re: Relay computers
At 02:43 PM 9/26/2004 -0600, you wrote:
Ron Hudson wrote:
WE HAVE MORSE CODE WHAT?
(I believe that nanotech could change the world within my unnatural
lifetime.) --David M. Palmer palmer(a)alumni.caltech.edu
--... ...-- -.. . -. ----. --.- --.- -...
tpeters(a)nospam.mixcom.com (remove "nospam") N9QQB (amateur radio)
"HEY YOU" (loud shouting) WEB ADDRESS http//www.mixweb.com/tpeters
43? 7' 17.2" N by 88? 6' 28.9" W, Elevation 815', Grid Square
EN53wc
WAN/LAN/Telcom Analyst, Tech Writer, MCP, Cisco Certified CCNA
On Sep 26, 2004, at 10:12 AM, ben franchuk wrote:
.-- . W E
.... .- ...- . H A V E
-- --- .-. ... . M O R S E
-.-. --- -.. . C O D E
:) ? PS. I thinking of building a smaller decimal
machine ( 6 digits ) and
wondering
where a good source of info on the web with the way to handle the math.
BCD? Execess 3, some other codeing?
Why not have it add ascii directly... "1" + "1" =
"2" :^)
6:35 a.m.
New subject: Early computers was Re: Relay computers
Tom Peters wrote:
WE HAVE MORSE CODE WHAT?
that was a joke ...
forget ascii use morse.
PS. How will nanotech change the world to some thing better?
We have problems still keeping the 'BIG OLD TECH' tech
around.
(I believe that nanotech could change the world within my unnatural
lifetime.) --David M. Palmer palmer(a)alumni.caltech.edu
8:15 a.m.
New subject: Harvard vs. vonNeuman ... was Re: Early computers was Re: Relay computers
re: Zuse, and stored-program, etc...
On Sun, 2004-09-26 at 01:37, Hans B PUFAL wrote:
This web page also seems apropro :
http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/Neumann_vs_Zuse.html
First, I don't mean to disparage Zuse's accomplishments, especially
under the circumstances he lived in, it's pretty amazing.
The idea-of-the-century wasn't machines that calculate automatically,
but machines-that-modify-themselves-as-part-of-normal-operation. Other
than particular logical arrangement, I don't see how the Zuse machines
are an architectural improvement over the IBM and Harvard relay
calculators.
In many ways it's an unfair comparison, the IBM machines had a nearly
unlimited budget, and poor Konrad sat by himself in his living room
punching programs into used 35mm film stock, never mind the war going on
around him.
Less well-known but historically more interesting is his 'planKalculus',
his algorithmic nomenclature he worked up to describe computer
programs. He even had chess-playing code! It's described in fair detail
in A HIST. OF COMP. 20TH CENT. (1980, MIT press) by Knuth.
(For the record, PICs are "Harvard architecture"; data and program
memory are separate. "von Neumann architecture" is program and data
stored in the same memory.)
1:53 a.m.
New subject: Harvard vs. vonNeuman ... was Re: Early computers was Re: Relay computers
On Mon, 27 Sep 2004, Tom Jennings wrote:
Less well-known but historically more interesting is
his 'planKalculus',
his algorithmic nomenclature he worked up to describe computer
programs. He even had chess-playing code! It's described in fair detail
in A HIST. OF COMP. 20TH CENT. (1980, MIT press) by Knuth.
For a good history of Zuse's early work I highly recommend his
autobiography, _The Computer--My Life_:
http://www.quantumbooks.com/Merchant2/merchant.mvc?Screen=PROD&Product_…
--
Sellam Ismail Vintage Computer Festival
------------------------------------------------------------------------------
International Man of Intrigue and Danger http://www.vintage.org
[ Old computing resources for business || Buy/Sell/Trade Vintage Computers ]
[ and academia at www.VintageTech.com || at http://marketplace.vintage.org ]
10:57 a.m.
New subject: Harvard vs. vonNeuman ... was Re: Early computers was Re: Relay computers
On Mon, 27 Sep 2004, Tom Jennings wrote:
re: Zuse, and stored-program, etc...
Actually many modern CPU's with caches are Harvard Architecture, the
instruction streams and data streams only being mixed on the memory of the
caches.
I've also heard that machines that mix data and instruction streams are more
properly called "Princeton achitecture"...
Isn't self modifying code pretty much deprecated these days (aside from
trampolines and such)
Peter Wallace
On Sun, 2004-09-26 at 01:37, Hans B PUFAL wrote:
This web page also seems apropro :
http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/Neumann_vs_Zuse.html
First, I don't mean to disparage Zuse's accomplishments, especially
under the circumstances he lived in, it's pretty amazing.
The idea-of-the-century wasn't machines that calculate automatically,
but machines-that-modify-themselves-as-part-of-normal-operation. Other
than particular logical arrangement, I don't see how the Zuse machines
are an architectural improvement over the IBM and Harvard relay
calculators.
In many ways it's an unfair comparison, the IBM machines had a nearly
unlimited budget, and poor Konrad sat by himself in his living room
punching programs into used 35mm film stock, never mind the war going on
around him.
Less well-known but historically more interesting is his 'planKalculus',
his algorithmic nomenclature he worked up to describe computer
programs. He even had chess-playing code! It's described in fair detail
in A HIST. OF COMP. 20TH CENT. (1980, MIT press) by Knuth.
(For the record, PICs are "Harvard architecture"; data and program
memory are separate. "von Neumann architecture" is program and data
stored in the same memory.)
1:32 p.m.
New subject: Harvard vs. vonNeuman
Peter C. Wallace wrote:
Isn't self modifying code pretty much deprecated
these days (aside from
trampolines and such)
So what is a trampoline?
Peter Wallace
I have yet to see a good use for self-modifying code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Ben.
4:01 p.m.
New subject: Harvard vs. vonNeuman
It was thus said that the Great ben franchuk once stated:
I have yet to see a good use for self-modifying code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Back when I took Assembly langauge (8086) at college, I used
self-modifying code in one of the assignments (the teacher and I didn't get
along and this was my attempt at getting a "rise" out of the teacher). The
assignment was to sort a list of numbers in ascending or descending order
(depending upon user input) so I simply changed the jump instruction after
the compare in the sort routine. It saved either two copies of the routine
(differing in only a single instruction), a complicated logical nest in the
middle of a sort routine, or using a callback routine.
As far as I can recall, that was the *only* time I've ever done
self-modifying code.
-spc (And it seemed like a good reason at the time ... )
28 Sep
28 Sep
11:58 a.m.
New subject: Harvard vs. vonNeuman
I use a lot of self-modifying code: Perl scripts that generate HTML code!
Maybe that doesn't count...
At 11:43 PM 2/10/2006 -0500, you wrote:
It was thus said that the Great ben franchuk once
stated:
[Wickedness] Wickedness is a myth invented by good people to account
for the curious attraction of others. --Oscar Wilde
--... ...-- -.. . -. ----. --.- --.- -...
tpeters(a)nospam.mixcom.com (remove "nospam") N9QQB (amateur radio)
"HEY YOU" (loud shouting) WEB ADDRESS http//www.mixweb.com/tpeters
43? 7' 17.2" N by 88? 6' 28.9" W, Elevation 815', Grid Square
EN53wc
WAN/LAN/Telcom Analyst, Tech Writer, MCP, Cisco Certified CCNA
I have yet to see a good use for self-modifying code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Back when I took Assembly langauge (8086) at college, I used
self-modifying code in one of the assignments (the teacher and I didn't get
along and this was my attempt at getting a "rise" out of the teacher). The
assignment was to sort a list of numbers in ascending or descending order
(depending upon user input) so I simply changed the jump instruction after
the compare in the sort routine. It saved either two copies of the routine
(differing in only a single instruction), a complicated logical nest in the
middle of a sort routine, or using a callback routine.
As far as I can recall, that was the *only* time I've ever done
self-modifying code.
-spc (And it seemed like a good reason at the time ... ) 
30 Sep
30 Sep
7:38 p.m.
New subject: Harvard vs. vonNeuman
Did the use of your self-modifying code in the past propelled you to the
future from where you're sending us mail in the present? :-)
Sean 'Captain Napalm' Conner wrote:
It was thus said that the Great ben franchuk once
stated:
I have yet to see a good use for self-modifying
code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Back when I took Assembly langauge (8086) at college, I used
self-modifying code in one of the assignments (the teacher and I didn't get
along and this was my attempt at getting a "rise" out of the teacher). The
assignment was to sort a list of numbers in ascending or descending order
(depending upon user input) so I simply changed the jump instruction after
the compare in the sort routine. It saved either two copies of the routine
(differing in only a single instruction), a complicated logical nest in the
middle of a sort routine, or using a callback routine.
As far as I can recall, that was the *only* time I've ever done
self-modifying code.
-spc (And it seemed like a good reason at the time ... )
3 Oct
3 Oct
11:47 a.m.
New subject: Harvard vs. vonNeuman
It was thus said that the Great Wai-Sun Chia once stated:
Did the use of your self-modifying code in the past propelled you to the
future from where you're sending us mail in the present? :-)
No, that's a bad CMOS battery in the Linux server. Whenever it reboots I
have to check the time and date; only this time, I noticed the time was
okay, but didn't notice that the date was a month off ...
-spc (Server in question is not quite on topic yet ... but close)
27 Jan
27 Jan
4:22 a.m.
New subject: Harvard vs. vonNeuman
technically in the OS (unix model) when you exec a program you are
executing "self modifying code" because the program space and
the data of most machines are in shared spaces, or are assumed
to be.
There must be TLB and data and instruction cache adjustments
after reading any program text into memory and before trying to
execute it, or you have a problem. (at least allowances are made
for such at the end of reading and doing the load process)
At least from an architecural standpoint, you must deal with
code that gets loaded after the "program" runs in general on
any system, other than special standalone ones, perhaps.
Jim
Sean 'Captain Napalm' Conner wrote:
It was thus said that the Great ben franchuk once
stated:
>
> I have yet to see a good use for self-modifying code.
<snip>
As far as I can recall, that was the *only* time I've ever done
self-modifying code.
-spc (And it seemed like a good reason at the time ... )
27 Sep
27 Sep
4:39 p.m.
New subject: Harvard vs. vonNeuman
On Mon, 27 Sep 2004, ben franchuk wrote:
Peter Wallace
Peter C. Wallace wrote:
A piece of code used often used for signal handlers that pushes
instructions on the stack and then executes them. Newer systems are getting
away from this so that the stack can be flagged as non-executable to avoid
stack overun screws.
Isn't self modifying code pretty much
deprecated these days (aside from
trampolines and such)
So what is a trampoline? Peter Wallace
I have yet to see a good use for self-modifying code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Ben.
4:50 p.m.
New subject: Harvard vs. vonNeuman
On Mon, Sep 27, 2004 at 07:32:06PM -0600, ben franchuk wrote:
I have yet to see a good use for self-modifying code.
Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Self-modifying code was somewhat common in the 6502 world. I have seen
(and written) several examples for the PET, VIC-20, C-64, and Apple II.
It's of limited use in cached architectures like the 68030, but it would
work fine with, say, the 68000, 68008, or 68010; but by that time, self-
modifying code was "out of fashion" (plus the 680x0 instruction set is
so much richer than the 6502 that there's much less need to try it).
-ethan
--
Ethan Dicks, A-130-S Current South Pole Weather at 28-Sep-2004 04:20 Z
South Pole Station
PSC 468 Box 400 Temp -48.7 F (-44.9 C) Windchill -90.5 F (-68.09 C)
APO AP 96598 Wind 23.5 kts Grid 344 Barometer 683.5 mb (10496. ft)
Ethan.Dicks(a)amanda.spole.gov http://penguincentral.com/penguincentral.html
28 Sep
28 Sep
8:16 a.m.
New subject: Harvard vs. vonNeuman
I have yet to
see a good use for self-modifying code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Self-modifying code was somewhat common in the 6502 world. I have seen
(and written) several examples for the PET, VIC-20, C-64, and Apple II. 
27 Sep
27 Sep
8:22 p.m.
New subject: Harvard vs. vonNeuman
ben franchuk wrote:
Peter C. Wallace wrote:
ISTR I saw it in two area's:
- delay loop calibration -> delay routine contains a number of nops,
jump instruction moves 'upward' depending on amount of delay required in
the loop.
- one program contained some jump instructions that were replaced by
nops (or vice-versa) when a dongle was detected.
Found by reverse engineering some DOS software.
Bert
Isn't self modifying code pretty much
deprecated these days (aside from
trampolines and such)
So what is a trampoline?
Peter Wallace
I have yet to see a good use for self-modifying code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088. 
28 Sep
28 Sep
1:58 a.m.
New subject: Harvard vs. vonNeuman
ben franchuk wrote:
I have yet to see a good use for self-modifying code.
Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
I saw it actually pretty often, in computer graphics/imaging.
A lot of times you got very fast blitter routines by writing a
small function (at runtime) which performs what you like to do,
and branch to it. Or just changing the actual pixel-op at runtime,
or design the filter the "right" way (meaning that the routine knows
already that it starts on odd byte adressess, but can transfer
longwords, deals with overlaps, etc)
cheers
6:46 a.m.
New subject: Harvard vs. vonNeuman
I have yet to see a good use for self-modifying
code. Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Ben.
On the CDC6600 - and, for that matter most machine architectures that
preceded the IBM /360 - the subroutine call instruction writes a jump to the
return address as the first word of the subroutine and then branches to the
second. Before Manchester invented "B-lines" (index registers)
self-modifying code was needed even for indexing.
It is almost a definition of a modern architecture that code need not be
(and often cannot be) self-modifying.
Andy
8:02 a.m.
New subject: Harvard vs. vonNeuman
Andy Holt wrote:
On the CDC6600 - and, for that matter most machine architectures that
preceded the IBM /360 - the subroutine call instruction writes a jump to the
return address as the first word of the subroutine and then branches to the
second. Before Manchester invented "B-lines" (index registers)
self-modifying code was needed even for indexing.
What about later machines like the PDP-8, that I have used.
The next machine I build I don't plan to have index registers
just simple indirect addressing. I have done one or two designs
in a FPGA but this time I want to use real TTL.
Anybody got a good web link on decimal floating point,I am not sure
how to handle the sign bit with decimal math for adc operations.
It is almost a definition of a modern architecture
that code need not be
(and often cannot be) self-modifying.
ROM able code does say a lot about how you
program your machine.
Andy
Ben.
1 Nov
1 Nov
2:29 a.m.
New subject: Harvard vs. vonNeuman
On Mon, Sep 27, 2004 at 07:32:06PM -0600, ben franchuk wrote:
Peter C. Wallace wrote:
I'm not sure if this fits his usage of the word here, but in the Redboot
boot loader (part of the eCos operating system), they use something they
call a trampoline to start a program loaded into ram.
Isn't self modifying code pretty much
deprecated these days (aside from
trampolines and such)
So what is a trampoline? I have yet to see a good use for self-modifying code.
Other than
subroutines and interupts placing data into a code segement the only
self modifying code I have seen was for the 8088.
Wouldn't JIT compilers be essentially self modifying code?
--
Joshua D. Boyd
jdboyd(a)jdboyd.net
http://www.jdboyd.net/
http://www.joshuaboyd.org/
27 Sep
27 Sep
5:35 p.m.
New subject: Harvard vs. vonNeuman ... was Re: Early computers was Re: Relay computers
On Mon, Sep 27, 2004 at 03:57:20PM -0700, Peter C. Wallace wrote:
On Mon, 27 Sep 2004, Tom Jennings wrote:
Self modifying code is _very_ strongly frowned upon when working with
halfway modern machines, because it messes up the I-caches (which
operate best when the memory theuy cache is read-only and there almost
guaranteed not to change under them) and therefore _hurts_ performance.
Regards,
Alex.
--
"Opportunity is missed by most people because it is dressed in overalls and
looks like work." -- Thomas A. Edison
re: Zuse, and stored-program, etc...
Actually many modern CPU's with caches are Harvard Architecture, the
instruction streams and data streams only being mixed on the memory of the
caches.
I've also heard that machines that mix data and instruction streams are more
properly called "Princeton achitecture"...
Isn't self modifying code pretty much deprecated these days (aside from
trampolines and such) On Sun, 2004-09-26 at 01:37, Hans B PUFAL wrote:
This web page also seems apropro :
http://irb.cs.tu-berlin.de/~zuse/Konrad_Zuse/Neumann_vs_Zuse.html
First, I don't mean to disparage Zuse's accomplishments, especially
under the circumstances he lived in, it's pretty amazing.
The idea-of-the-century wasn't machines that calculate automatically,
but machines-that-modify-themselves-as-part-of-normal-operation. Other
than particular logical arrangement, I don't see how the Zuse machines
are an architectural improvement over the IBM and Harvard relay
calculators.
In many ways it's an unfair comparison, the IBM machines had a nearly
unlimited budget, and poor Konrad sat by himself in his living room
punching programs into used 35mm film stock, never mind the war going on
around him.
Less well-known but historically more interesting is his 'planKalculus',
his algorithmic nomenclature he worked up to describe computer
programs. He even had chess-playing code! It's described in fair detail
in A HIST. OF COMP. 20TH CENT. (1980, MIT press) by Knuth.
(For the record, PICs are "Harvard architecture"; data and program
memory are separate. "von Neumann architecture" is program and data
stored in the same memory.)
8:18 p.m.
New subject: Harvard vs. vonNeuman ... was Re: Early computers was Re: Relaycomputers
"Peter C. Wallace" wrote:
I've also heard that machines that mix data and instruction streams are more
properly called "Princeton achitecture"...
Aggreed, Mauchly and Eckert invented it (for the Eniac), von Neumann
only documented it.
Bert
28 Sep
28 Sep
1:25 a.m.
New subject: Harvard vs. vonNeuman ... was Re: Early computers was Re: Relay computers
>>>> "Peter" == Peter C Wallace
<pcw(a)mesanet.com> writes:
Peter> Actually many modern CPU's with caches are Harvard
Peter> Architecture, the instruction streams and data streams only
Peter> being mixed on the memory of the caches.
You could look at it that way, but I don't think that is the normal
usage of the terms.
Peter> I've also heard that machines that mix data and instruction
Peter> streams are more properly called "Princeton achitecture"...
Says who? That's the first time I've heard that term in 25+ years.
Peter> Isn't self modifying code pretty much deprecated these days
Peter> (aside from trampolines and such)
Self modifying code in the early sense of the word has been frowned on
for decades. But program loaders, pagers, and OS components like that
all modify program memory -- something that's only possible in von
Neumann machines.
paul
7033
days inactive
7157
days old
51 comments
27 participants
participants (27)
-
als＠thangorodrim.de -
andyh＠andyh-rayleigh.freeserve.co.uk -
bert＠brothom.nl -
bfranchuk＠jetnet.ab.ca -
brian＠quarterbyte.com -
cisin＠xenosoft.com -
dickset＠amanda.spole.gov -
dvcorbin＠optonline.net -
emu＠ecubics.com -
geneb＠deltasoft.com -
geoffreythomas＠onetel.com -
hansp＠citem.org -
jdboyd＠jdboyd.net -
jtinker＠coin.org -
jwstephens＠msm.umr.edu -
pat＠computer-refuge.org -
pcw＠mesanet.com -
pkoning＠equallogic.com -
Pres＠macro-inc.com -
ron.hudson＠sbcglobal.net -
spc＠conman.org -
spectre＠floodgap.com -
tomj＠wps.com -
tpeters＠mixcom.com -
vcf＠siconic.com -
waisun.chia＠hp.com -
waltje＠pdp11.nl