18 Sep
2004
18 Sep
'04
9:17 a.m.
As I stated earlier, I have cannibalized three HP laserjet series II
printers. In looking over the parts recovered I find that I have three
(3) totally different memory boards.
The first is populated with small chips on BOTH sides of the board with
a total of 128 soldered in chips that appear from their part numbers to
be 256K memory chips, That is the number 256 appears within the part
number. Now I don't know how wide the buss is but I am assuming 8 bits
because of its age. If so this board is a 4 Megabyte memory board.
The second is only half populated with 16 soldered in chips on one side
that have the number 1000 in their part number so I am assuming they are
1meg chips and using the afore mentioned logic I come up with a
2Megabyte memory board.
The third is fully populated with 32 socketed chips with the number 1000
in their part number and that works out to a 4 Megabyte memory board.
Is my logic sound?
Do I have two 4 meg boards and a 2 meg board?
What is the largest memory board the Series 2 could hold?
I am assuming that the socketed memory is best because it is easily
repaired, but are there any problems with the socketed memory having bad
contacts after time?
18 Sep
18 Sep
1:52 p.m.
At 02:17 PM 9/18/2004, you wrote:
As I stated earlier, I have cannibalized three HP
laserjet series II
printers. In looking over the parts recovered I find that I have three
(3) totally different memory boards.
The first is populated with small chips on BOTH sides of the board with a
total of 128 soldered in chips that appear from their part numbers to be
256K memory chips, That is the number 256 appears within the part
number. Now I don't know how wide the buss is but I am assuming 8 bits
because of its age. If so this board is a 4 Megabyte memory board.
um, that sounds probable...
The second is only half populated with 16 soldered in
chips on one side
that have the number 1000 in their part number so I am assuming they are
1meg chips and using the afore mentioned logic I come up with a
2Megabyte memory board.
yes, that's a 2meg board using 1 megabit chips
The third is fully populated with 32 socketed chips
with the number 1000
in their part number and that works out to a 4 Megabyte memory board.
yes, that is a 4 meg board.
Is my logic sound?
Do I have two 4 meg boards and a 2 meg board?
What is the largest memory board the Series 2 could hold?
4 megabyte
I am assuming that the socketed memory is best because
it is easily
repaired, but are there any problems with the socketed memory having bad
contacts after time?
the biggest problem I've had with laserjet boards is the PALs and GALs they
used on some boards (HP and aftermarket) for addressing failing over time :(
4:44 p.m.
As I stated earlier, I have cannibalized three HP laserjet series II
printers. In looking over the parts recovered I find that I have three
(3) totally different memory boards.
The first is populated with small chips on BOTH sides of the board with
a total of 128 soldered in chips that appear from their part numbers to
be 256K memory chips, That is the number 256 appears within the part
number. Now I don't know how wide the buss is but I am assuming 8 bits
because of its age. If so this board is a 4 Megabyte memory board.
The bus width doesn't matter in determining the memory size. If you had
16 1 megabit memory chips arranged as 2 megs * 8 bits, that a 2 megabyte
board. If you had them arrangeded as 1 meg * 16 bits, that's still a 2
megabyte board.
The Series 2 formatters use a 68000 processor (the 64 pin chip on the
formatter board) and thus have a 16 bit wide databus.
-tony
19 Sep
19 Sep
1:35 a.m.
Sure the buss width matters if you are measuring memory as it is NORMALY
measured, in Bytes, not Bits.
Tony Duell wrote:
As I stated
earlier, I have cannibalized three HP laserjet series II
printers. In looking over the parts recovered I find that I have three
(3) totally different memory boards.
The first is populated with small chips on BOTH sides of the board with
a total of 128 soldered in chips that appear from their part numbers to
be 256K memory chips, That is the number 256 appears within the part
number. Now I don't know how wide the buss is but I am assuming 8 bits
because of its age. If so this board is a 4 Megabyte memory board.
The bus width doesn't matter in determining the memory size. If you had
16 1 megabit memory chips arranged as 2 megs * 8 bits, that a 2 megabyte
board. If you had them arrangeded as 1 meg * 16 bits, that's still a 2
megabyte board.
The Series 2 formatters use a 68000 processor (the 64 pin chip on the
formatter board) and thus have a 16 bit wide databus.
-tony
6:39 a.m.
Sure the buss width matters if you are measuring memory as it is NORMALY
measured, in Bytes, not Bits.
No it doesn't.
Suppose I take 16 of the 64K * 1 DRAM chips (4164s). I wire them up to an
8 bit databus with the apporporate control electronics so I get 128K 8
bit locations. That stores 128K bytes.
I now wire them up differently to a 16 bit data bus system. I haev 64K 16
bit locations. That's _still_ 128K bytes of storage.
-tony
9:03 a.m.
Tony Duell declared on Sunday 19 September 2004 01:39 pm:
Yes, but it does matter if the memory is ECC or something. 9 x 1Mbit
chips doesn't (typically) give you 1152KB of ram, it give you 1024KB.
The same goes for ECC, and the memory used in old (on-topic) IBM
RS/6000s which is 40 bits wide (ECC + a spare bit). a 2M x 40bit module
has 8MB of ram, not 9MB.
Also, I guess, it's not proper (at least in my opinion) to say a PDP-8
with 32kW of 12bit wide memory has 48KB of ram.
However for multiple of 8 (or 9 with byte-parity or ...) bus widths, the
bus width doesn't really matter with how many bytes the memory contains.
Pat
--
Purdue University ITAP/RCS --- http://www.itap.purdue.edu/rcs/
The Computer Refuge --- http://computer-refuge.org
Sure the buss
width matters if you are measuring memory as it is
NORMALY measured, in Bytes, not Bits.
No it doesn't.
Suppose I take 16 of the 64K * 1 DRAM chips (4164s). I wire them up to
an 8 bit databus with the apporporate control electronics so I get
128K 8 bit locations. That stores 128K bytes.
I now wire them up differently to a 16 bit data bus system. I haev 64K
16 bit locations. That's _still_ 128K bytes of storage.
9:14 a.m.
I didn't understand that you were holding the word width constant. If
you hold the word width constant, yes, you are right. But that is not
what I was talking about. In many early computers, the data buss and
the word width were the same. It was not till they began to make
processors with larger word widths that there was a difference.
Actually, not till they started making processors. The early computers
didn't have processors in the same sense as we now know them as a single
component.. The buss normally matched the word width.
In this particular case the memory I was counting, on the board in the
printer HP printer, I was counting the memory AS IT IS NORMALLY COUNTED
for an HP laserjet Series II, and I was right, to count it as 8 bits
wide regardless of the word width, I have two 4 meg boards and one 2 meg
board. That has been confirmed. My reason for saying the meg count was
dependent on buss width was because in this case, IT IS. The buss is 8
bits wide and that is how you count the memory on a HP laserjet card.
Tony Duell wrote:
Sure the buss
width matters if you are measuring memory as it is NORMALY
measured, in Bytes, not Bits.
No it doesn't.
Suppose I take 16 of the 64K * 1 DRAM chips (4164s). I wire them up to an
8 bit databus with the apporporate control electronics so I get 128K 8
bit locations. That stores 128K bytes.
I now wire them up differently to a 16 bit data bus system. I haev 64K 16
bit locations. That's _still_ 128K bytes of storage.
-tony
10:01 a.m.
I didn't understand that you were holding the word width constant. If
I'm not. But I am keeping the size of a 'byte' constant at 8 bits (this
is actually not always the case, but anyway).
you hold the word width constant, yes, you are right.
But that is not
what I was talking about. In many early computers, the data buss and
the word width were the same. It was not till they began to make
OK....
Let's think of those 16 64K*1 DRAMs again.
If I wire them up as a 16-bit wide memory (to link to a 68000, say), it's
correct that I have 64K _words_ of storage (where one word == 16 bits ==
data bus width). But it's also conventional to call that 128K bytes (the
origianl Mac, for example, had exactly that configuration of memory, it
was called a 128K machine, even though it's really got 64K words of
memory (each being 16 bits or 2 bytes).
My PDP11/45 has 124K words of memory. I have seen this quoted as 248K
bytes (IMHO that is also correct).
processors with larger word widths that there was a
difference.
Actually, not till they started making processors. The early computers
didn't have processors in the same sense as we now know them as a single
component.. The buss normally matched the word width.
Usually, but not always The Philips P850 looks to be a 16 bit machine to
the programmer, the front panel has 16 toggles, the registers look to be
16 bits wide, as do the memory locations, etc. At the hardware level it's
an 8 bit machine, 8 bit ALU (a pair of 74181s) 8 bit data bus to memory,
etc.
In this particular case the memory I was counting, on
the board in the
printer HP printer, I was counting the memory AS IT IS NORMALLY COUNTED
for an HP laserjet Series II, and I was right, to count it as 8 bits
wide regardless of the word width, I have two 4 meg boards and one 2 meg
YEs, becasue a byte is 8 bits, and because the memory chips all store
data (not parity or ECC bits, for example).
It really doesn't matter if the processor accesses that memory 8 bits at
a time or 16 bits at a time. The total number of bits that can be stored
is the same. So, therefore, is the total number of 8 bit bytes.
board. That has been confirmed. My reason for saying
the meg count was
dependent on buss width was because in this case, IT IS. The buss is 8
bits wide and that is how you count the memory on a HP laserjet card.
Actually, on the Laserjet 2, the bus is 16 bits wide. It's a 68000
processor.
-tony
22 Sep
22 Sep
7:11 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
If
you hold the word width constant, yes, you are right. But that is not
what I was talking about. In many early computers, the data buss and
the word width were the same.
... and many did not. The 'byte' as a convention for talking about
memory is just that, a convention, and fails miserably on machines whose
major casual metric is not a multiple of 8 bits. Many, many machines
were built on a multiple of 6 bits because that's how many it took to
define a character.
For machines which have some architectural feature > 8 but modulo 8 ==
0, 32- and 64-bit wide memory and paths could be byte-addressed. I don't
know for sure, but I would imagine there are 6-bit-character-addressable
instruction sets too.
Until more or less when CPUs fit entirely within silicon, there was no
hard and true correlation between the bit-widths of busses, registers
and paths; this was because constructing those things cost actual money
and scaling of silicon didn't exist. Lots of machines have different
width regs/accumulator, memory, index regs, program counters, arithmetic
units, etc.
(My LGP-21 is a good example: 32-bit accumulator, 31-bit memory, 12-bit
program counter, double-32 product reg, 4- or 6-bit I/O.)
(Nothing in a Microchip Inc PICxxxx except the register files is
byte-width!)
For non-multiple-of-8 machines, the 'byte' is not relevant generally.
Boy with hammer: everything looks like a nail.
7:26 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
>>>> "Tom" == Tom Jennings
<tomj(a)wps.com> writes:
> If you hold the word width constant, yes, you are
right. But that
> is not what I was talking about. In many early computers, the
> data buss and the word width were the same.
Tom> ... and many did not. The 'byte' as a convention for talking
Tom> about memory is just that, a convention, and fails miserably on
Tom> machines whose major casual metric is not a multiple of 8
Tom> bits. Many, many machines were built on a multiple of 6 bits
Tom> because that's how many it took to define a character.
Tom> For machines which have some architectural feature > 8 but
Tom> modulo 8 == 0, 32- and 64-bit wide memory and paths could be
Tom> byte-addressed. I don't know for sure, but I would imagine there
Tom> are 6-bit-character-addressable instruction sets too.
In some sense, you could say that of the Burroughs 5000/6000 series.
Those had character pointers, though most of the memory addressing was
to 48 (or 51, if you count the flag field) bit words.
How about the IBM 1620, with digit-addressable memory, 5 bits plus
parity per addressable entity?
Tom> Until more or less when CPUs fit entirely within silicon, there
Tom> was no hard and true correlation between the bit-widths of
Tom> busses, registers and paths; this was because constructing those
Tom> things cost actual money and scaling of silicon didn't
Tom> exist. Lots of machines have different width regs/accumulator,
Tom> memory, index regs, program counters, arithmetic units, etc.
Tom> (My LGP-21 is a good example: 32-bit accumulator, 31-bit memory,
Tom> 12-bit program counter, double-32 product reg, 4- or 6-bit I/O.)
Or the CDC 6600 CPU: 60 bit memory, some 60 and some 18 bit
registers. In the PPU: 12 bit memory, 18 bit accumulator.
If you count the link bit, the PDP-8 has a 13 bit accumulator I
believe.
Tom> For non-multiple-of-8 machines, the 'byte' is not relevant
Tom> generally.
...though the term might be used to describe a 6 bit chunk, or a 12
bit chunk (CDC), or a variable width chunk (PDP-10).
paul
7:41 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Wed, 22 Sep 2004, Tom Jennings wrote:
I always thought that the technical definition of a "byte" is
"8-bits".
--
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 ]
If
you hold the word width constant, yes, you are right. But that is not
what I was talking about. In many early computers, the data buss and
the word width were the same.
... and many did not. The 'byte' as a convention for talking about
memory is just that, a convention, and fails miserably on machines whose
major casual metric is not a multiple of 8 bits. Many, many machines
were built on a multiple of 6 bits because that's how many it took to
define a character.
8 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
>>>> "Vintage" == Vintage
Computer Festival <vcf(a)siconic.com> writes:
Vintage> On Wed, 22 Sep 2004, Tom Jennings wrote:
> > If > you hold the word width constant,
yes, you are right. But
> that is not > what I was talking about. In many early computers,
> the data buss and > the word width were the same.
>
> ... and many did not. The 'byte' as a convention for talking about
> memory is just that, a convention, and fails miserably on machines
> whose major casual metric is not a multiple of 8 bits. Many, many
> machines were built on a multiple of 6 bits because that's how
> many it took to define a character.
Vintage> I always thought that the technical definition of a "byte"
Vintage> is "8-bits".
Not in this community! Newfangled usage, yes. If you want a word for
8-bit chunk, nothing smaller, nothing larger, say "octet". (That
makes you sound like a network geek, admittedly -- it's where the term
originated, I believe.)
paul
9:07 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
>>>> "William" == William
Donzelli <aw288(a)osfn.org> writes:
> Not in this community! Newfangled usage, yes.
William> Well, if newfangled means after 7 April 1964.
Is that when IBM did the 360?
But at that time they were merely one of the users of the term "byte",
and arguably the odd man out by applying it to 8 bits rather than the
more common 6. It wasn't until all the other machines had disappeared
that you could argue "of course byte means 8 bits". So, perhaps,
around 1985 or so (when the PDP-10 pretty much disappeared).
Or perhaps not yet. There are still machines in production use where
byte does not mean 8 bits. Cyber mainframes, for example.
paul
9:24 a.m.
New subject: bit-widths
And I always though the exact definition of a byte was based on how many
teeth your 'puter had.....
11:14 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Wed, 2004-09-22 at 14:07, Paul Koning wrote:
I've seen references to 'byte' meaning lots of numbers 6 - 9, but was it
ever really weighted towards 6? Most references I've seen to the
6-bit-unit was simply, 'character'. But I spent no time around IBM
(except one brief stint on TSO onna 360 writing 8x300 code).
>>>> "William" == William Donzelli <aw288(a)osfn.org> writes:
> Not in this community! Newfangled usage,
yes.
William> Well, if newfangled means after 7 April 1964.
Is that when IBM did the 360?
But at that time they were merely one of the users of the term "byte",
and arguably the odd man out by applying it to 8 bits rather than the
more common 6. Or perhaps not yet. There are still machines in
production use where
byte does not mean 8 bits. Cyber mainframes, for example.
What's a cyber byte size?
23 Sep
23 Sep
4:13 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
>>>> "Tom" == Tom Jennings
<tomj(a)wps.com> writes:
Tom> On Wed, 2004-09-22 at 14:07, Paul Koning wrote:
> Or perhaps not yet. There are still machines in
production use
> where byte does not mean 8 bits. Cyber mainframes, for example.
Tom> What's a cyber byte size?
12 bits, the PPU word size.
paul
11:10 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Wed, 2004-09-22 at 13:55, William Donzelli wrote:
Now dearest William, I know your birthday is important to you, but it
many not necessarily be cause for general celebration...
Not in this
community! Newfangled usage, yes.
Well, if newfangled means after 7 April 1964.
10:51 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
Vintage> I always thought that the technical
definition of a "byte"
Vintage> is "8-bits".
On Wed, 2004-09-22 at 13:00, Paul Koning wrote:
Not in this community! Newfangled usage, yes. If you
want a word for
8-bit chunk, nothing smaller, nothing larger, say "octet". (That
makes you sound like a network geek, admittedly -- it's where the term
originated, I believe.)
Well, in general today (there's that dangerous phrase again...) a byte
is 8 bits, and rarely will you go wrong. I believe that most uses of
'byte' to mean 'smallest multi-bit unit' .NE. 8 are IBMish, 1960-ish,
and obsolete (there's another...) and its adherents were not only killed
in the Wars but their bodies stomped into the mud and gravestones turned
over. But I've been wrong before.
Paul is absolutely correct -- octet is the UNambiguous word for a
grouping of 8 bits. I also believe came from the telecom world, it's
certainly in heavy use there (also most internet RFCs).
11:49 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Wed, Sep 22, 2004 at 03:51:09PM -0700, Tom Jennings wrote:
Paul is absolutely correct -- octet is the UNambiguous
word for a
grouping of 8 bits. I also believe came from the telecom world, it's
certainly in heavy use there (also most internet RFCs).
I believe that is because many of the guys making early contributions
had their feet firmly planted in the 36-bit world (with 6 6-bit chars
per machine word). An 'octet' is unambiguously 8 bits.
-ethan
--
Ethan Dicks, A-130-S Current South Pole Weather at 22-Sep-2004 23:40 Z
South Pole Station
PSC 468 Box 400 Temp -94.0 F (-70.0 C) Windchill -128.3 F (-89.09 C)
APO AP 96598 Wind 6.7 kts Grid 084 Barometer 672 mb (10929. ft)
Ethan.Dicks(a)amanda.spole.gov http://penguincentral.com/penguincentral.html
10:20 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
The IBM 7070, The first computer I worked on, had 10 bits per byte and
9K of memory. The 10 bits must have been a hold over from the decimal
system. I have no idea why there was 9K of memory.
Tom Jennings wrote:
If
you hold the word width constant, yes, you are right. But that is not
what I was talking about. In many early computers, the data buss and
the word width were the same.
... and many did not. The 'byte' as a convention for talking about
memory is just that, a convention, and fails miserably on machines whose
major casual metric is not a multiple of 8 bits. Many, many machines
were built on a multiple of 6 bits because that's how many it took to
define a character.
For machines which have some architectural feature > 8 but modulo 8 ==
0, 32- and 64-bit wide memory and paths could be byte-addressed. I don't
know for sure, but I would imagine there are 6-bit-character-addressable
instruction sets too.
Until more or less when CPUs fit entirely within silicon, there was no
hard and true correlation between the bit-widths of busses, registers
and paths; this was because constructing those things cost actual money
and scaling of silicon didn't exist. Lots of machines have different
width regs/accumulator, memory, index regs, program counters, arithmetic
units, etc.
(My LGP-21 is a good example: 32-bit accumulator, 31-bit memory, 12-bit
program counter, double-32 product reg, 4- or 6-bit I/O.)
(Nothing in a Microchip Inc PICxxxx except the register files is
byte-width!)
For non-multiple-of-8 machines, the 'byte' is not relevant generally.
Boy with hammer: everything looks like a nail.
10:49 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
The IBM 7090 was the first machine I worked on - but I am familiar with the
IBM 7070.
The IBM 7070 was modeled after the IBM 650 - so the 7070 was a bi-quinary (2
out of 5) encoded machine. There were 3 accumulators and 99 index registers.
Memory locations were used for the accumulators (9991, 9992 and 9993), the
program counter was 9995. The index registers were memory locations 1-99.
Technically memory ran from 0-9999, but maximum memory was specified as 9900
words - as that was approximately what was available as "standard" (non
register) memory.
Memory words were 10 decimal digits plus sign
Lyle
On Wednesday 22 September 2004 15:20, Jim Isbell wrote:
The IBM 7070, The first computer I worked on, had 10
bits per byte and
9K of memory. The 10 bits must have been a hold over from the decimal
system. I have no idea why there was 9K of memory.
Tom Jennings wrote:
>>If
>>you hold the word width constant, yes, you are right. But that is not
>>what I was talking about. In many early computers, the data buss and
>>the word width were the same.
>
>... and many did not. The 'byte' as a convention for talking about
>memory is just that, a convention, and fails miserably on machines whose
>major casual metric is not a multiple of 8 bits. Many, many machines
>were built on a multiple of 6 bits because that's how many it took to
>define a character.
>
>For machines which have some architectural feature > 8 but modulo 8 ==
>0, 32- and 64-bit wide memory and paths could be byte-addressed. I don't
>know for sure, but I would imagine there are 6-bit-character-addressable
>instruction sets too.
>
>Until more or less when CPUs fit entirely within silicon, there was no
>hard and true correlation between the bit-widths of busses, registers
>and paths; this was because constructing those things cost actual money
>and scaling of silicon didn't exist. Lots of machines have different
>width regs/accumulator, memory, index regs, program counters, arithmetic
>units, etc.
>
>(My LGP-21 is a good example: 32-bit accumulator, 31-bit memory, 12-bit
>program counter, double-32 product reg, 4- or 6-bit I/O.)
>
>(Nothing in a Microchip Inc PICxxxx except the register files is
>byte-width!)
>
>For non-multiple-of-8 machines, the 'byte' is not relevant generally.
>
>Boy with hammer: everything looks like a nail.
--
Lyle Bickley
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
11:13 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
--- Lyle Bickley <lbickley(a)bickleywest.com> wrote:
The IBM 7070 was modeled after the IBM 650 - so the
7070 was a bi-quinary (2
out of 5) encoded machine.
Do you know why a bi-quinary encoding was used instead
of the more obvious 8-4-2-1 BCD encoding scheme? I've
been a bit curious about that. More generally, I've
been curious why decimal architectures held on for so
long, given the higher cost of the CPU logic. In the
older semi-electronic punch-card calculators, it
clearly simplified the logic overall, as they were
running simple hard-wired (plug-board) programs
interfacing with decimal-oriented electromechanical
input and output. I can also see where it made sense
on smaller digit-serial designs. In a large-scale
parallel CPU, however, it's not clear to me why
a binary architecture (doing binary to decimal
conversion in software) wasn't preferred right from
the start.
--Bill
12:54 p.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Wednesday 22 September 2004 16:13, William Maddox wrote:
--- Lyle Bickley <lbickley(a)bickleywest.com>
wrote:
--snip--
To the best of my knowledge, it was a leftover from punched cards. On a 650,
10 digits per word were used for storing data - and a punched card could hold
8 words of data (8x10=80 column punched card). "Instructions" had 2 digits
OP code, 4 digits data address, 4 digits next address and sign.
Given that most programmers were "graduating" from unit record machines (IBM
407, etc.) to program the 650 - they were used to "decimal everything". IBM
didn't want to change the paradyme - at least not in early commercial
systems. (Note that all of IBM's scientific computers [704, 709, Stretch,
709x, 704x, etc.] were pure binary).
Internally, the 650 used bi-quinary which only requires 6 bits to represent
one of 10 decimal digits. One bit (the "bi" in bi-quinary) represented
"0"
or "5" and what the other five bits represented was dependent on the state of
the "bi" bit; i.e.,
"bi" bit
---------
0 1
---------
0 5
1 6
2 7
3 8
4 9
Kinda shrewd - and helped early unit record folks in deal with simple binary
and "mostly" decimal ;-)
Lyle
--
Lyle Bickley
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
The IBM 7070 was modeled after the IBM 650 - so
the
7070 was a bi-quinary (2
out of 5) encoded machine.
Do you know why a bi-quinary encoding was used instead
of the more obvious 8-4-2-1 BCD encoding scheme? I've
been a bit curious about that.
1:04 p.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
>>
>> Do you know why a bi-quinary encoding was used instead of
>> the more obvious 8-4-2-1 BCD encoding scheme?
Possible answers....
A) It matches an Abacus...
B) It matches (some?) nixie style tubes?
C) It is just a really cool word?
D) It is a great stumper for Geek-Jeopardy?
[actually I blieve the answer to be "B" with a touch of "A"]
23 Sep
23 Sep
2:26 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
>>>> "David" == David V Corbin
<dvcorbin(a)optonline.net> writes:
>>> Do you know why a bi-quinary encoding was
used instead of the
>>> more obvious 8-4-2-1 BCD encoding scheme?
David> Possible answers....
David> A) It matches an Abacus...
David> B) It matches (some?) nixie style tubes?
David> C) It is just a really cool word?
David> D) It is a great stumper for Geek-Jeopardy?
David> [actually I blieve the answer to be "B" with a touch of
"A"]
That doesn't sound right; a Nixie has 10 wires, one per digit, plus a
common wire.
A possible answer: because the logic to do add or multiply is simpler
in biquinary than in BCD? You'd have to do the equations and
synthesize the logic to check that.
paul
9:11 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Thu, 2004-09-23 at 07:26, Paul Koning wrote:
That doesn't sound right; a Nixie has 10 wires,
one per digit, plus a
common wire.
See below, scroll down to the bottom!
http://wps.com/archives/tube-datasheets/Datasheets/Burroughs-catalog-918A/8…
10:19 p.m.
New subject: dual anode nixies-was- all sorts of other stuff
The AN/USM-207 counter used these dual anode nixies.
I've put a cct here of one of the counter/display sections-
http://homepages.ihug.co.nz/~tractorb/BQ%20nixie%20cct.pdf
Cheers
Dave Brown, NZ
----- Original Message -----
From: "Tom Jennings" <tomj(a)wps.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk(a)classiccmp.org>
Sent: Friday, September 24, 2004 9:11 AM
Subject: RE: bit-widths, was Re: HP Laserjet ..again....
On Thu, 2004-09-23 at 07:26, Paul Koning wrote:
That doesn't sound right; a Nixie has 10
wires, one per digit, plus
a
common wire.
See below, scroll down to the bottom!
http://wps.com/archives/tube-datasheets/Datasheets/Burroughs-catalog-918A/8…
22 Sep
22 Sep
1:17 p.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Wed, 2004-09-22 at 16:13, William Maddox wrote:
Do you know why a bi-quinary encoding was used
instead
of the more obvious 8-4-2-1 BCD encoding scheme? I've
been a bit curious about that. More generally, I've
been curious why decimal architectures held on for so
long, given the higher cost of the CPU logic. In the
older semi-electronic punch-card calculators, it
clearly simplified the logic overall, as they were
running simple hard-wired (plug-board) programs
interfacing with decimal-oriented electromechanical
input and output. I can also see where it made sense
on smaller digit-serial designs. In a large-scale
parallel CPU, however, it's not clear to me why
a binary architecture (doing binary to decimal
conversion in software) wasn't preferred right from
the start.
Biquinary has some advantages when constructed from vacuum tubes,
because tubes are so awful as logic elements (every element in a tube
has a fairly large capacitor connected between each element, in effect,
and electrons have to travel a *fraction of an inch* within the device).
(You can even get a single tube to count to N, however unreliably, using
a phantastron circuit.)
Don't take the existence of B-to-D software conversions for granted --
on slow memory (drums for instance) that alone could double or more
runtimes already measured in hours, on machines with MTBF of hundreds of
hours (early on). Besides, what else is there to do with an electronic
automatic computer but to calculate numbers? And when you are doing so
in machine code, with no what we'd call developmental software about,
it's a substantial advantage to do so in human systems. There were many
hardware solutions to B-to-D and D-to-B, some quite involved.
For the first half of the 1950's, it wasn't universally agreed that
compilers and interpreters were even a *good idea*, since so much
performance and control would be lost, etc. FORTRAN removed most of
those objections.
At some point it was obvious that decimal wasn't the right approach from
a density and facilities view, and it Went Away.
(There exist biquinary NIXIE tubes; two anodes for the 10 digits. I have
a few I think.)
There are plenty of electronic numbering systems; things you'd think are
obvious, like Binary Coded Decimal bits weighted 8, 4, 2, 1, are not
obvious, there were/are a lot of other schemes.
I'm sure actual old timers could fill in a lot here.
1:30 p.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
There are plenty of electronic numbering systems;
things you'd think are
obvious, like Binary Coded Decimal bits weighted 8, 4, 2, 1, are not
obvious, there were/are a lot of other schemes.
I'm sure actual old timers could fill in a lot here.
Well, I was born before 7 April 1964, but I do know that some of the other
schemes were used in telemetry because they were 2 of 5 - sort of like
built-in stupid parity.
William Donzelli
aw288(a)osfn.org
1:52 p.m.
New subject: Decimal V Binary??
There are SOME people (I am NOT one) who actually feel that
DECIMAL is the right way to go in computers!
For example, check Mike Cowlishaw's pages at IBM:
http://www2.hursley.ibm.com/decimal/decifaq.html
http://www2.hursley.ibm.com/decimal/decifaq3.html#bingood
etc.
--
Grumpy Ol' Fred cisin(a)xenosoft.com
How would society have developed if the more dominant
of our ancestors didn't have stinky feet?
3:47 p.m.
New subject: Decimal V Binary??
On Wed, 2004-09-22 at 19:08, William Donzelli wrote:
I don't see this page as a need to 'return to decimal hardware' so much
as people really need to consider accuracy and precision et al in
arithemetic systems.
These are turing-derived machines, after all, and can do decimal
arithmetic when we want them to.
There are SOME
people (I am NOT one) who actually feel that
DECIMAL is the right way to go in computers!
If the computers are doing my bank statements, then yes, I am one of them.
5:12 p.m.
New subject: Decimal V Binary??
Having spent four and a half years running IT at the Philadelphia Federal
Reserve Bank and almost ten years responsible for Investment Quantitative
Analysis, Trust Systems and Automation Planning for a major regional bank
(now Wachovia) - I can assure you that banks do all their accounting in
decimal.
Investment analysis, manket and economic models, arbitrage analysis,
regression analysis, etc. are done in floating point.
They really know the difference and when it's important do use decimal v.
"binary"...
Lyle
On Wednesday 22 September 2004 20:47, Tom Jennings wrote:
On Wed, 2004-09-22 at 19:08, William Donzelli wrote:
I don't see this page as a need to 'return to decimal hardware' so much
as people really need to consider accuracy and precision et al in
arithemetic systems.
These are turing-derived machines, after all, and can do decimal
arithmetic when we want them to.
--
Lyle Bickley
Bickley Consulting West Inc.
http://bickleywest.com
"Black holes are where God is dividing by zero"
There are
SOME people (I am NOT one) who actually feel that
DECIMAL is the right way to go in computers!
If the computers are doing my bank statements, then yes, I am one of
them.
23 Sep
23 Sep
3:30 a.m.
New subject: Decimal V Binary??
>>>> "Tom" == Tom Jennings
<tomj(a)wps.com> writes:
Tom> On Wed, 2004-09-22 at 19:08, William Donzelli wrote:
> > There are SOME people (I am NOT one) who
actually feel that >
> DECIMAL is the right way to go in computers!
>
> If the computers are doing my bank statements, then yes, I am one
> of them.
Tom> I don't see this page as a need to 'return to decimal hardware'
Tom> so much as people really need to consider accuracy and precision
Tom> et al in arithemetic systems.
Tom> These are turing-derived machines, after all, and can do decimal
Tom> arithmetic when we want them to.
And indeed they usually do, when you write in Cobol.
paul
22 Sep
22 Sep
11:40 p.m.
New subject: Decimal V Binary??
On Wed, 2004-09-22 at 22:08, William Donzelli wrote:
I've always wondered what systems/processors (other than IBM mainframes)
that had hardware decimal arithmetic support. I don't mean 'decimal
adjust' instructions, I'm talking about full
add/subtract/multiply/divide/etc. The only other one I know of is the
Wang VS, which came standard with support for 'packed decimal' data, and
even floating-point decimal. The Wang VS instruction set was very
similar to the IBM 360/370, even the packed decimal data format was the
same.
Jim
There are SOME
people (I am NOT one) who actually feel that
DECIMAL is the right way to go in computers!
If the computers are doing my bank statements, then yes, I am one of them. 
23 Sep
23 Sep
12:05 a.m.
New subject: Decimal V Binary??
At 07:40 AM 9/23/2004, you wrote:
I've always wondered what systems/processors (other
than IBM mainframes)
that had hardware decimal arithmetic support. I don't mean 'decimal
adjust' instructions, I'm talking about full
add/subtract/multiply/divide/etc. The only other one I know of is the
Wang VS, which came standard with support for 'packed decimal' data, and
even floating-point decimal. The Wang VS instruction set was very
similar to the IBM 360/370, even the packed decimal data format was the
same.
DEC PDP11/23 CPU has sockets for 2 extra chips (40 pin DIPS).
One is hardware floating point. This is fairly common.
The other for a CIS - Commercial Instruction Set with full set of decimal
instructions as you list.
Never used one though.
Ed K.
2:37 a.m.
New subject: Decimal V Binary??
Ed Kelleher wrote:
At 07:40 AM 9/23/2004, you wrote:
The dual-height 11/03 processor allows a CIS as well. I have a
couple that have that chip. As far as I can determine, both those
machines were used to sort, analyse, and archive incoming raw data.
Honestly, though, I don't think I've ever run a job on the /03 that
touched the CIS.
Doc
I've always wondered what systems/processors
(other than IBM mainframes)
that had hardware decimal arithmetic support. I don't mean 'decimal
adjust' instructions, I'm talking about full
add/subtract/multiply/divide/etc. The only other one I know of is the
Wang VS, which came standard with support for 'packed decimal' data, and
even floating-point decimal. The Wang VS instruction set was very
similar to the IBM 360/370, even the packed decimal data format was the
same.
DEC PDP11/23 CPU has sockets for 2 extra chips (40 pin DIPS).
One is hardware floating point. This is fairly common.
The other for a CIS - Commercial Instruction Set with full set of
decimal instructions as you list.
3:56 a.m.
New subject: Decimal V Binary??
>>>> "Doc" == Doc Shipley
<doc(a)mdrconsult.com> writes:
Doc> Ed Kelleher wrote:
> DEC PDP11/23 CPU has sockets for 2 extra chips (40
pin DIPS).
>
> One is hardware floating point. This is fairly common.
>
> The other for a CIS - Commercial Instruction Set with full set of
> decimal instructions as you list.
Doc> The dual-height 11/03 processor allows a CIS as well. I have a
Doc> couple that have that chip. As far as I can determine, both
Doc> those machines were used to sort, analyse, and archive incoming
Doc> raw data.
Doc> Honestly, though, I don't think I've ever run a job on the /03
Doc> that touched the CIS.
I would assume that Cobol and (possibly) RPG would have an option to
use CIS if available, but no other language is likely to do so.
There was a DEC internal prototype of an 11/70 with CIS added; it was
called 11/74 I believe (not to be confused with the SMP machine also
referred to as 11/74 that the RSX development team used). Apparently
it was canceled shortly before it was ready to ship. Nasty rumors
claimed that this was because it was outperforming the 11/780 in every
way; I don't know how valid that might be.
paul
7:50 a.m.
New subject: Decimal V Binary??
On Thu, 23 Sep 2004, Paul Koning wrote:
I would assume that Cobol and (possibly) RPG would
have an option to
use CIS if available, but no other language is likely to do so.
IIRC, Only if explicitly specified in the ENVIRONMENT division.
COBOL had extensive [but not often used] cross-platform
capabilities by specifying in the source code both what
kind of system was to compile it, AND what the target system
was.
27 Sep
27 Sep
1:27 a.m.
New subject: Decimal V Binary??
The dual-height 11/03 processor allows a CIS as
well. I have a
couple that have that chip. As far as I can determine, both those
machines were used to sort, analyse, and archive incoming raw data.
Are you sure... I don't remember a CIS option for the 11/03. There
was an EIS/FIS option chip which could be used on 11/03, 11/2 and
PDT-11/1x0 systems. But the only CIS option chip I know of works
only on 11/23[+] (and possibly the 11/24?)
Megan Gentry
Former RT-11 Developer
+--------------------------------+-------------------------------------+
| Megan Gentry, EMT/B, PP-ASEL,ST| email: mbg at world.std.com |
| Member of Technical Staff | megan at savaje.com |
| SavaJe Technologies, Inc. | (s/ at /@/) |
| 100 Apollo Drive | URL: http://world.std.com/~mbg/ |
| Chelmsford, MA 01824 | "pdp-11 programmer - some assembler |
| (978) 256 6521 (DEC '77-'98) | required." - mbg KB1FCA |
+--------------------------------+-------------------------------------+
23 Sep
23 Sep
3:43 a.m.
New subject: Decimal V Binary??
>>>> "Jim" == Jim Donoghue
<jim(a)smithy.com> writes:
Jim> On Wed, 2004-09-22 at 22:08, William Donzelli wrote:
> > There are SOME people (I am NOT one) who
actually feel that >
> DECIMAL is the right way to go in computers!
>
> If the computers are doing my bank statements, then yes, I am one
> of them.
Jim> I've always wondered what systems/processors (other than IBM
Jim> mainframes) that had hardware decimal arithmetic support. I
Jim> don't mean 'decimal adjust' instructions, I'm talking about full
Jim> add/subtract/multiply/divide/etc. The only other one I know of
Jim> is the Wang VS, which came standard with support for 'packed
Jim> decimal' data, and even floating-point decimal. The Wang VS
Jim> instruction set was very similar to the IBM 360/370, even the
Jim> packed decimal data format was the same.
VAX and PDP11 (with the CIS option) for starters.
paul
10:03 a.m.
New subject: Decimal V Binary??
I've always wondered what systems/processors
(other than IBM mainframes)
that had hardware decimal arithmetic support. I don't mean 'decimal
adjust' instructions, I'm talking about full
add/subtract/multiply/divide/etc. The only other one I know of is the
PDP11/44 CIS option supported packed decimal 'strings'. Didn't some VAXen
do so as well?
-tony
10:09 a.m.
New subject: Decimal V Binary??
I've always wondered what systems/processors
(other than IBM mainframes)
that had hardware decimal arithmetic support. I don't mean 'decimal
adjust' instructions, I'm talking about full
Not really want you're asking for, but I have a machine here (actually 2
diffeerent machines using the same processor) where the hardware is
bit-serial for binary operatiosn, but does BCD operations 1 nybble at a
time.....
A simple puzzle : What machine is it?
-tony
27 Sep
27 Sep
5:12 a.m.
New subject: Decimal V Binary??
On Thu, 23 Sep 2004 07:40:20 -0400, Jim Donoghue <jim(a)smithy.com> wrote:
On Wed, 2004-09-22 at 22:08, William Donzelli wrote:
I've always wondered what systems/processors (other than IBM mainframes)
that had hardware decimal arithmetic support. I don't mean 'decimal
adjust' instructions, I'm talking about full
add/subtract/multiply/divide/etc. The only other one I know of is the
Wang VS, which came standard with support for 'packed decimal' data, and
even floating-point decimal. The Wang VS instruction set was very
similar to the IBM 360/370, even the packed decimal data format was the
same.
If I remember right, the NCR315 had full BCD arithmetic. At least the
arithmetic instructions it had were decimal, I do not remember if it
actually had a divide instruction in hardware.
It was even so decimal-oriented that memory was addressed in decimal.
A rather nice feature was that all internal registers were ferrite core.
If it stopped because of a power falure, it could be restarted in the
exact instruction step where it had stopped.
When I was working for NCR Norway in the early 70's, the bulk of input to
the 315 came from OCR - reading cash register rolls printed in National
Optical Font. Not bad for a cumputer which was shipping earlier than the
IBM 360.
We also had dial-up data transfer over modem from England, but that seemed
a bit trivial when Norway was connected to ARPANET the year after. That,
however, is a totally different story.
--
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/
There are
SOME people (I am NOT one) who actually feel that
DECIMAL is the right way to go in computers!
If the computers are doing my bank statements, then yes, I am one of
them. 
9 Oct
9 Oct
4:07 p.m.
New subject: Decimal V Binary??
> I've always wondered what systems/processors
(other than IBM
> mainframes) that had hardware decimal arithmetic support.
The VAX has packed decimal arithmetic instructions. Depending on which
VAX, they may be microcoded or they may just trap to the OS for
"macrocoded" emulation (of course the OS may choose to error out
instead of bothering).
I don't know of any VAX that does them in non-microcoded hardware, and
am inclined to doubt such a beast exists.
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse(a)rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
5:02 p.m.
New subject: Decimal V Binary??
On Sun, 10 Oct 2004, der Mouse wrote:
The lowly 8080 had DAA, Decimal Adjust for Addition. If I recall,
it handled "carry" from bit5 (and Cy?) after adding a BCD pair
in A.
Man, I wrote a lot of 8080 code, so much I still remember
opcodes and could puzzle out binary 20 years later, but it's
possible I never once used DAA. Hell, I used MOV A, A even.
(An inline pseudo-disassembler; a config file could specify
the IO port a tape drive interface was located and the program
modified itself to change all the port numbers for IN and
OUT. IO instructions following MOV A, A were modified.)
> I've
always wondered what systems/processors (other than IBM
> mainframes) that had hardware decimal arithmetic support.
The VAX has packed decimal arithmetic instructions. Depending on which
VAX, they may be microcoded or they may just trap to the OS for
"macrocoded" emulation (of course the OS may choose to error out
instead of bothering). 
8:09 p.m.
New subject: Decimal V Binary??
Hi,
Tom Jennings <tomj(a)wps.com> said:
The lowly 8080 had DAA, Decimal Adjust for Addition. If I recall,
it handled "carry" from bit5 (and Cy?) after adding a BCD pair
in A.
The even lowlier SC/MP had DAD (Decimal Add), DAI (Decimal Add Immediate)
and DAE (Decimal Add Extension) instructions. They all add 2-digit
(8-bit) decimal numbers.
--
Cheers,
Stan Barr stanb(a)dial.pipex.com
The future was never like this!
27 Sep
27 Sep
1:22 a.m.
New subject: Decimal V Binary??
Actually, the CIS option chip for the 11/23 spans two sockets. It
plugs into the outer edge of one socket and the outer edge of the
adjacent socket...
Megan Gentry
Former RT-11 Developer
+--------------------------------+-------------------------------------+
| Megan Gentry, EMT/B, PP-ASEL,ST| email: mbg at world.std.com |
| Member of Technical Staff | megan at savaje.com |
| SavaJe Technologies, Inc. | (s/ at /@/) |
| 100 Apollo Drive | URL: http://world.std.com/~mbg/ |
| Chelmsford, MA 01824 | "pdp-11 programmer - some assembler |
| (978) 256 6521 (DEC '77-'98) | required." - mbg KB1FCA |
+--------------------------------+-------------------------------------+
22 Sep
22 Sep
11:21 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
I'm honored to be around so many 2nd-gen machine users!
Clearly the 'byte' discussion applies to binary machines. I assume the
unit for decimal machines is simply a 'digit'?
On Wed, 2004-09-22 at 15:49, Lyle Bickley wrote:
The IBM 7090 was the first machine I worked on - but I
am familiar with the
IBM 7070.
The IBM 7070 was modeled after the IBM 650 - so the 7070 was a bi-quinary (2
out of 5) encoded machine. There were 3 accumulators and 99 index registers.
Memory locations were used for the accumulators (9991, 9992 and 9993), the
program counter was 9995. The index registers were memory locations 1-99.
Technically memory ran from 0-9999, but maximum memory was specified as 9900
words - as that was approximately what was available as "standard" (non
register) memory.
Memory words were 10 decimal digits plus sign
Lyle
On Wednesday 22 September 2004 15:20, Jim Isbell wrote:
> The IBM 7070, The first computer I worked on, had 10 bits per byte and
> 9K of memory. The 10 bits must have been a hold over from the decimal
> system. I have no idea why there was 9K of memory.
>
> Tom Jennings wrote:
> >>If
> >>you hold the word width constant, yes, you are right. But that is not
> >>what I was talking about. In many early computers, the data buss and
> >>the word width were the same.
> >
> >... and many did not. The 'byte' as a convention for talking about
> >memory is just that, a convention, and fails miserably on machines whose
> >major casual metric is not a multiple of 8 bits. Many, many machines
> >were built on a multiple of 6 bits because that's how many it took to
> >define a character.
> >
> >For machines which have some architectural feature > 8 but modulo 8 ==
> >0, 32- and 64-bit wide memory and paths could be byte-addressed. I don't
> >know for sure, but I would imagine there are 6-bit-character-addressable
> >instruction sets too.
> >
> >Until more or less when CPUs fit entirely within silicon, there was no
> >hard and true correlation between the bit-widths of busses, registers
> >and paths; this was because constructing those things cost actual money
> >and scaling of silicon didn't exist. Lots of machines have different
> >width regs/accumulator, memory, index regs, program counters, arithmetic
> >units, etc.
> >
> >(My LGP-21 is a good example: 32-bit accumulator, 31-bit memory, 12-bit
> >program counter, double-32 product reg, 4- or 6-bit I/O.)
> >
> >(Nothing in a Microchip Inc PICxxxx except the register files is
> >byte-width!)
> >
> >For non-multiple-of-8 machines, the 'byte' is not relevant generally.
> >
> >Boy with hammer: everything looks like a nail.
11:52 a.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
I worked on the 7070 in 1959/60 time frame. It was a wonder of science
at the time. I remember looking at the new frame of core memory that
we had just added and imagining that if I stuck a screw driver into it
and twisted it around, that the computer might scream.
Lyle Bickley wrote:
The IBM 7090 was the first machine I worked on - but I
am familiar with the
IBM 7070.
The IBM 7070 was modeled after the IBM 650 - so the 7070 was a bi-quinary (2
out of 5) encoded machine. There were 3 accumulators and 99 index registers.
Memory locations were used for the accumulators (9991, 9992 and 9993), the
program counter was 9995. The index registers were memory locations 1-99.
Technically memory ran from 0-9999, but maximum memory was specified as 9900
words - as that was approximately what was available as "standard" (non
register) memory.
Memory words were 10 decimal digits plus sign
Lyle
On Wednesday 22 September 2004 15:20, Jim Isbell wrote:
The IBM 7070, The first computer I worked on, had
10 bits per byte and
9K of memory. The 10 bits must have been a hold over from the decimal
system. I have no idea why there was 9K of memory.
Tom Jennings wrote:
>>If
>>you hold the word width constant, yes, you are right. But that is not
>>what I was talking about. In many early computers, the data buss and
>>the word width were the same.
>>
>>
>... and many did not. The 'byte' as a convention for talking about
>memory is just that, a convention, and fails miserably on machines whose
>major casual metric is not a multiple of 8 bits. Many, many machines
>were built on a multiple of 6 bits because that's how many it took to
>define a character.
>
>For machines which have some architectural feature > 8 but modulo 8 ==
>0, 32- and 64-bit wide memory and paths could be byte-addressed. I don't
>know for sure, but I would imagine there are 6-bit-character-addressable
>instruction sets too.
>
>Until more or less when CPUs fit entirely within silicon, there was no
>hard and true correlation between the bit-widths of busses, registers
>and paths; this was because constructing those things cost actual money
>and scaling of silicon didn't exist. Lots of machines have different
>width regs/accumulator, memory, index regs, program counters, arithmetic
>units, etc.
>
>(My LGP-21 is a good example: 32-bit accumulator, 31-bit memory, 12-bit
>program counter, double-32 product reg, 4- or 6-bit I/O.)
>
>(Nothing in a Microchip Inc PICxxxx except the register files is
>byte-width!)
>
>For non-multiple-of-8 machines, the 'byte' is not relevant generally.
>
>Boy with hammer: everything looks like a nail.
>
>
12:16 p.m.
New subject: bit-widths, was Re: HP Laserjet ..again....
On Wed, 22 Sep 2004, Lyle Bickley wrote:
The IBM 7070 was modeled after the IBM 650 - so the
7070 was a
bi-quinary (2 out of 5) encoded machine. There were 3 accumulators and
Here's a good description of bi-quinary:
http://en.wikipedia.org/wiki/Bi-quinary_coded_decimal
--
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 ]
7378
days inactive
7400
days old
52 comments
21 participants
participants (21)
-
ard＠p850ug1.demon.co.uk -
aw288＠osfn.org -
bv＠norbionics.com -
cisin＠xenosoft.com -
dickset＠amanda.spole.gov -
doc＠mdrconsult.com -
dvcorbin＠optonline.net -
geoffr＠zipcon.net -
jim＠smithy.com -
lbickley＠bickleywest.com -
mbg＠TheWorld.com -
millenniumfalcon＠cableone.net -
mouse＠Rodents.Montreal.QC.CA -
pat＠computer-refuge.org -
pkoning＠equallogic.com -
Pres＠macro-inc.com -
stanb＠dial.pipex.com -
tomj＠wps.com -
tractorb＠ihug.co.nz -
vcf＠siconic.com -
wmaddox＠pacbell.net