Proposed Assembler Commands

[keith]

[...]
Wasn't it the 6800 (or perhaps 6502??) that had the LSEX (Load with Sign
EXtended)?

RCA 1802 had a SEX instruction.

68K sports a BRA.

Oh, my. I bet that sports-BRA supports more than a bit. ...a nybble,
perhaps.

So that's why they're extinct!

Can we figure out how to do this to Windows?

 

[Larry Brasfield]

[...]
I always liked the Moto instruction sets. The Intel stuff has always
seemed klugey and disorganized to me. So, we all wound up with the
worst instruction set and the worst OS the industry could invent.

I sometimes suspected IBM wanted to help ensure
their PC would never compete with real computers.

I often suspect that you haven't a clue what you're talking about.

You often strike me as prone to boorishness.

The 8088, with its segmented addressing scheme,
was commonly recognized as a bitch to program.
IBM's choice of that uP hampered development
of software for the PC for years, until the '386
with its flat addressing and VM support appeared.

Ok, what is *your* definiton of RISC? I suspect we're going to be
"enlightened" here.

There you go again. An examination of the 1802 registers
and instruction set shows many of the characteristics that
later uP's claimed to be RISC had. I'm not going to attempt
your enlightenment or worry about your boorish suspicions.

 

[Ken Smith]

The 1802 did RISC before its heyday. I suspect it was
motivated by transistor count rather than speed. At the
time, CMOS had the lowest gate density.

No, the PDP-8 was RISC. The 1802 was MISC, a Moronic Instruction Set
Computer.

"when stupidity will serve as an explaination you need look no further" I
think applies quite well to the 1802. Any programmer using one would have
given up half of the 16 registers to have those transistors used for
making instuctions that were more useful.

 

[Ken Smith]

Sounds tedious.

680x:

LDAA A
ADAA B
STAA C

PDP-11:

MOVB A, C
ADDB B, C

PDP-8:
CLR ; Make the accumulator zero
TAD A ; Make it the first value
TAD B ; Add in the second
STC ; Store and clear accumulator

Less transistors than a 1802 (I'd bet) and 4 times less instructions. The
folks at DEC obviously studied the question of getting the most out of
every transistor.

 

[Ken Smith]

I do believe the 1802 is outta this world! IIRC is't on the Pioneers and
Voyagers. Wonderful accomplishments!

Yes, I think you are right. They may have been SOS ones. I know RCA made
them for some purpose way be then. They cost like the devil.

 

[Ken Smith]

But sadly lacked the necessary hardware for multiplication.

I think I should go look for my 64Bit/32Bit = 32,32Bit divide code for the
1802. It was highly optimized for speed so it could do many per second.

 

[Spehro Pefhany]

On Mon, 2 May 2005 03:18:05 +0000 (UTC), the renowned

[...]
Wasn't it the 6800 (or perhaps 6502??) that had the LSEX (Load with Sign
EXtended)?

RCA 1802 had a SEX instruction.

But sadly lacked the necessary hardware for multiplication.

I think I should go look for my 64Bit/32Bit = 32,32Bit divide code for the
1802. It was highly optimized for speed so it could do many per second.

A sophisticated algorithm, or just the rudimentary shift and subtract
algorithm, but tightly coded?



Best regards,
Spehro Pefhany

 

[John Larkin]

PDP-8:
CLR ; Make the accumulator zero
TAD A ; Make it the first value
TAD B ; Add in the second
STC ; Store and clear accumulator

(That last one was actually DCA, deposit and clear accumulator.)

The PDP-8 had interesting micro-operations. In one instruction, you
could do

CLA CLL CMA ROL clear acc; clear link; complement accumulator;
rotate left

which loaded 4094 into the accumulator. So there were all sorts of
constants you could cobble up in one instruction.

The original PDP-8 was made from discrete-transistor modules.

Less transistors than a 1802 (I'd bet) and 4 times less instructions.

The original PDP-8 was made from discrete-transistor modules, probably
a few hundred transistors.

John

 

[Mike Monett]

keith wrote:

[...]

My fav editor gave up the ghost about then. I loved EPM, but it went the
way of 16bit addressing. I still try to find E-like editors.

Keith

You can program SPRINT to emulate just about any kind of editor, from Wordstar
to EMACS and everything in between.

The SPRINT engine uses 32-bit addressing, so the file size is limited only by
disk space. I put the swap file in a 30 meg ramdisk, so it's plenty fast enough.
It can open 26 files simultaneously, but I rarely hit this limit.

I wrote my editor to automatically detect all the common file types: ASM, C,
Pascal, ascii, HTML, etc. So when I press a function key, it accomplishes the
same task regardless of the type file I'm editing. This is very convenient.

It means I don't have to switch editors for different types of files, and I
don't need to memorize a bunch of contradictory commands, each with their own
quirks depending on the programmer's philosphy.

For example, after a copy and paste, where do you put the cursor - at the start
or the end of the string? Actually, it depends on what you are doing, so I added
a function to switch the cursor location. So I can have it either way

And whenever some new requirement pops up, I can quickly add or change a command
to make the editor do anything I need. This is impossible with closed-source
programs, so I don't think I will ever need any of the Microsoft bloatware that
everyone complains about.

If you have had any experience writing your own editor, you will find it very
difficult to use a "store-bought" one with canned commands that may not do what
you need.

Mike Monett

 

[Mike Monett]

Dave Holford wrote:

[...]

We also had a list which included things like:

RWT Rewind and wipe tape
IAC Ignore all commands
DPL Destroy progam and loop
etc.

Regretfully I don't have a list and have long forgotten most of them; but HCF
is one that I remember.

Dave

Heh - very good. Thanks - I added them to the list.

Mike Monett

 

[Ken Smith]

I think I should go look for my 64Bit/32Bit = 32,32Bit divide code for the
1802. It was highly optimized for speed so it could do many per second.

A sophisticated algorithm, or just the rudimentary shift and subtract
algorithm, but tightly coded?[/QUOTE]

At its core was the usual shift and subtract/add method. The method is
sometimes called "Booth's method". It avoids the cost of having to do a
compare and then conditionally a subtract. You just go ahead and
subtract. If the subtract borrows, from then on you add in each loop
until you get a carry. I did this by having two copies of the main loop
that go jumped between when you swithc from adding to subtracting and
back.

On the way in, the numbers were examined. Leading and trailing zero bytes
on the values allowed it to shorten up the loops. For cases that must
return zero results, the code just jumped to a routine that copied the
numerator into the remainder and zeros to the result.

 

[James Beck]

o All,

Microprocessors have become so simple that it seems anyone can program
them nowadays. But think of what it was like when the first processors
came out and you had to program around noisy circuits and unreliable
memory.

For example, here is a list of assembler commands that emulate the
performance of the first computers. (They were found in old Deja View
files.)

You forgot the most important one of all

BOC Branch On Conundrum

That way Spock couldn't tie up all of your computers resources just by
telling it to calculate PI to the last decimal place.

 

[John Woodgate]

I read in sci.electronics.design that James Beck

You forgot the most important one of all

BOC Branch On Conundrum

That way Spock couldn't tie up all of your computers resources just by
telling it to calculate PI to the last decimal place.

That's not a conundrum. "Verify 'This statement is false.'" is a
conundrum.

 

[John Larkin]

[...]
I always liked the Moto instruction sets. The Intel stuff has always
seemed klugey and disorganized to me. So, we all wound up with the
worst instruction set and the worst OS the industry could invent.

I sometimes suspected IBM wanted to help ensure
their PC would never compete with real computers.

I often suspect that you haven't a clue what you're talking about.

You often strike me as prone to boorishness.

Now *that* is truly funny! I suppose opposite pedants repel.

John

 

[John Woodgate]

I read in sci.electronics.design that John Larkin

Now *that* is truly funny! I suppose opposite pedants repel.

Law of Nature, that. Furthermore, pedants occupy an N-dimensional
universe, where N is equal to the number of pedants. Ergo, all pedants
are orthogonal.

 

[Ben Bradley]

Yes. Some time back I was following a bickup truck down I89 with the
license plate EIEIO. Thinking it was a PPC type with strange sense of
humor, I pulled up along side. Nope, McDonald's Tree Farms. ;-)

Wasn't it the 6800 (or perhaps 6502??) that had the LSEX (Load with Sign
EXtended)?

Not the 6502, I still have many opcodes memorized from my KIM-1
days. I'm a lot less familiar with the 6800, but I did write some code
for it and don't recall that instruction on it either.

 

[John Larkin]

I read in sci.electronics.design that John Larkin


Law of Nature, that. Furthermore, pedants occupy an N-dimensional
universe, where N is equal to the number of pedants. Ergo, all pedants
are orthogonal.

Do three pedants define a plain obtuse triangle?

John

 

[John Woodgate]

I read in sci.electronics.design that John Larkin

Do three pedants define a plain obtuse triangle?

John

No. Being all orthogonal, they define the surface of an octant of a
sphere; a triangle with three right angles. (Think Lat and Long of three
points, 0,0; 90,0 and 0,90.)

 

[keith]

keith wrote:

[...]

My fav editor gave up the ghost about then. I loved EPM, but it went the
way of 16bit addressing. I still try to find E-like editors.

Keith

You can program SPRINT to emulate just about any kind of editor, from Wordstar
to EMACS and everything in between.

If you have had any experience writing your own editor, you will find it
very difficult to use a "store-bought" one with canned commands that may
not do what you need.

The *last* thing I want to do is program an editor. COnfiguring one is
bad enough. I don't much like making hammers and straightening nails
either. I've been using MED on Windows, and VE (though I don't like it) on
*IX for some time.

 

[keith]

[...]
I always liked the Moto instruction sets. The Intel stuff has always
seemed klugey and disorganized to me. So, we all wound up with the
worst instruction set and the worst OS the industry could invent.

I sometimes suspected IBM wanted to help ensure
their PC would never compete with real computers.

I often suspect that you haven't a clue what you're talking about.

You often strike me as prone to boorishness.

Ah, you're too kind.

The 8088, with its segmented addressing scheme,
was commonly recognized as a bitch to program.
IBM's choice of that uP hampered development
of software for the PC for years, until the '386
with its flat addressing and VM support appeared.

Commonly recognized by you? It was not a "bitch" to program, though that
has nothing to do with your professed ignorance of the subject.

There you go again. An examination of the 1802 registers
and instruction set shows many of the characteristics that
later uP's claimed to be RISC had. I'm not going to attempt
your enlightenment or worry about your boorish suspicions.

It is noted that you still haven't answered the question.