Which HP calculators DO NOT use the Saturn processor?
jrla...@shell.golden.net
don't. For instance what do the HP-1x series use in them? Are they Saturn
based?
--
john R. Latala
jrla...@golden.net
Jonathan Busby
>Subject line says it all. Many calculators I know what they use but some I
>don't. For instance what do the HP-1x series use in them? Are they Saturn
>based?
The HP-71B was the first calculator to use the Saturn. Everything
after it made use of the Saturn except the HP-1x series and some of
the newest algebraic models. ( HP-10BII I believe employs some type of
microcontroller to emulate the old Saturn chip that the 10B used to
use. I don't what the 30S and the 6S use )
The HP-1x series used a derivative of the HP-41 "NUT" processor. See
http://www.hpmuseum.org/tech10.htm .
-------------------------------------------------------------------------------
Jonathan Busby - <j...@SNMAPOhouston.rr.com>
Remove the random permutation of "NOSPAM" from my e-mail address
before replying.
Christoph Giesselink
news:scd4pukq5k5jmnj0q...@4ax.com...
> On 25 Sep 2002 17:22:14 -0400, jrla...@shell.golden.net wrote:
>
> >Subject line says it all. Many calculators I know what they use but some
I
> >don't. For instance what do the HP-1x series use in them? Are they Saturn
> >based?
>
>
> The HP-71B was the first calculator to use the Saturn. Everything
> after it made use of the Saturn except the HP-1x series and some of
> the newest algebraic models. ( HP-10BII I believe employs some type of
> microcontroller to emulate the old Saturn chip that the 10B used to
> use. I don't what the 30S and the 6S use )
The Bert, Lewis, Clarke and Yorke are special chips having a Saturn kernel.
The following list may be incompleted, because I made it from my mind.
- Saturn
HP71B, HP18C, HP28C
- 1LU7 Bert
HP10B, HP20S
- Lewis
HP17B, 17BII, 19B, 19BII, 27S, 28S, 42S
- Clarke
HP48S, HP48SX
- Yorke
HP38G, HP39G, HP40G, HP48G, HP48G+, HP48GX, HP49G
All other HP calculators don't use a Saturn processor kernel and that are of
course much more that I listed above.
Christoph
John H Meyers
http://www.hpmuseum.org/saturn.htm
http://www.hpmuseum.org/techcpu.htm
((o))
.
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Bruce Horrocks
<j...@SNMAPOhouston.rr.com> writes
>The HP-71B was the first calculator to use the Saturn. Everything after
>it made use of the Saturn except the HP-1x series and some of the
>newest algebraic models. ( HP-10BII I believe employs some type of
>microcontroller to emulate the old Saturn chip that the 10B used to
>use. I don't what the 30S and the 6S use )
The HP10BII uses a 6502 core. No emulation - the math routines were
re-written in C just for it.
This fact brought to you courtesy of the HPCC Conference in London,
21-22 September 2002.
Regards,
--
Bruce Horrocks
Hampshire
England
b...@granby.demon.co.uk
Jonathan Busby
Hmmm... I guess I was remembering wrong. (that's a surprise ;) So the
correct term should be "simulate"? (at least for the math routines)
For the rest, I don't know if the 10BII resembles the 10B, just with
"prettier" packaging. Maybe someone that has used it can chime in.
Eric Smith
> Subject line says it all. Many calculators I know what they use but some I
> don't. For instance what do the HP-1x series use in them? Are they Saturn
> based?
The Voyager series (HP-10C, -11C, -12C, -15C, and -16C) use the "Nut"
processor that was first used in the HP-41C ("Coconut"). For many years
the Voyagers used the same actual processor die (but in a QFP package
rather than a DIP). Later cost-reduced versions of the 12C have integrated
more functions onto fewer chips, resulting in a single-chip version, but
the CPU still has the same architecture.
The HP-71B, clamshell, Pioneer, and graphing calculators use variants
of Saturn.
The HP 10BII uses a Mitsubishi processor that is a 6502-derivative.
I have no idea what sort of processor is used in the HP 6S, 6S Solar,
and 30S. They were not designed by HP.
Eric Smith
> Hmmm... I guess I was remembering wrong. (that's a surprise ;) So the
> correct term should be "simulate"? (at least for the math routines)
No, it's not simulating anything either. The code has been carefully
written from scratch to produce the same numeric results. In other
words, it's new code implementing the same algorithms.
Jonathan Busby
<eric-no-s...@brouhaha.com> wrote:
But this is the definition of "simulate". ;)
To rephrase the second part of my question I was wondering if the user
interface has also been rewritten to produce similar behavior.
Jonathan Busby
<cgi...@swol.de> wrote:
>The Bert, Lewis, Clarke and Yorke are special chips having a Saturn kernel.
>The following list may be incompleted, because I made it from my mind.
>
>- Saturn
>HP71B, HP18C, HP28C
>
>- 1LU7 Bert
>HP10B, HP20S
>
>- Lewis
>HP17B, 17BII, 19B, 19BII, 27S, 28S, 42S
>
>- Clarke
>HP48S, HP48SX
>
>- Yorke
>HP38G, HP39G, HP40G, HP48G, HP48G+, HP48GX, HP49G
>
>
>All other HP calculators don't use a Saturn processor kernel and that are of
>course much more that I listed above.
>
>Christoph
>
There's also the Saca (Sacajawea) chip which is used by the 14B, 22S,
32S, and 32SII . (if my sources are correct) According to the same
sources, the Saca is basically just a Bert but with more RAM and ROM (
16 KB and 512 KB respectively ) and with a one line alphanumeric
display controller instead of a 7 segment one.
Jonathan Busby
<j...@SNMAPOhouston.rr.com> wrote:
>There's also the Saca (Sacajawea) chip which is used by the 14B, 22S,
>32S, and 32SII . (if my sources are correct) According to the same
>sources, the Saca is basically just a Bert but with more RAM and ROM (
>16 KB and 512 KB respectively ) and with a one line alphanumeric
>display controller instead of a 7 segment one.
Well, I tried to cancel that message but no luck. My fingers are
getting ahead of me again. ;) That should be 512 *bytes* of RAM and 16
KB of ROM.
Eric Smith
> No, it's not simulating anything either. The code has been carefully
> written from scratch to produce the same numeric results. In other
> words, it's new code implementing the same algorithms.
Jonathan Busby <j...@SNMAPOhouston.rr.com> writes:
> But this is the definition of "simulate". ;)
simulate (v.t. [imp. & p.p. Simulated; p. pr. & vb. n. Simulating] To
assume the mere appearance of, without the reality; to assume the signs
or indications of, flasely; to counterfeit; to feign.
-- Webster's Revised Unabridged Dictionary
If I buy a 2002 Mazda to replace my 1992 model, it will still get me
to and from work just fine. But I don't claim that the 2002 model
simulates the 1992 model.
The 10BII has genuine math routines. They don't simulate anything.
Jonathan Busby
<eric-no-s...@brouhaha.com> wrote:
I think I know the definition of a common English word - I've been
speaking the language for 22 years. Perhaps you need to work out some
personal issues since you feel the need to patronize people over
extremely trivial matters.
If you want to nitpick, then consulting
http://www.dictionary.com/search?q=simulate gives a definition which
reads :
"
1. a. To have or take on the appearance, form, or sound of;
imitate.
b. To make in imitation of or as a substitute for.
2. To make a pretense of; feign: simulate interest.
3. To create a representation or model of (a physical system or
particular situation, for example).
"
So, you are saying the following sentence is semantically
incorrect ? : "The new math functions take on the behavior of the old
ones while being different internally." Replace "take on" with
"simulate". I think if you browse google you'll find thousands of
sentences that fit with this usage.
Wow. I thought I had finished English class a long time ago. And of
all places, I'm relearning the joy of this great language right here.
Let me ask you a question. Do you point out subtle differences in your
interpretation of word semantics with *everyone* you talk to, or do
you just do it here?
Eric Smith
> Do you point out subtle differences in your interpretation of word
> semantics with *everyone* you talk to, or do you just do it here?
No.
Having published five simulators as free software (GPL), and having
contributed code to a simulator that tens of thousands of people use
(maybe more), I think I have some idea as to what the term means. I'll
concede that other people might be more expert in this area than I am.
You can obviously choose to use whatever definition for "simulate" you
like, but I don't know anyone else who would view a piece of code
rewritten to run natively on a different processor to be a simulation.
That's certainly not common usage in the industry. More typically that
might be referred to as "porting," although that term can also be used
in reference to making relatively small changes to source code written
in a high level language for use on a different platform.
Another sure way to start an argument is to bring up the difference
between "simulate" and "emulate"; nearly everyone uses the latter when
they mean the former.
Jonathan Busby
<eric-no-s...@brouhaha.com> wrote:
>Another sure way to start an argument is to bring up the difference
>between "simulate" and "emulate"; nearly everyone uses the latter when
>they mean the former.
So if I said "The new math functions emulate the behavior of the old
ones" then would you be happy?
Eric Smith
> Another sure way to start an argument is to bring up the difference
> between "simulate" and "emulate"; nearly everyone uses the latter when
> they mean the former.
Jonathan Busby <j...@SNMAPOhouston.rr.com> writes:
> So if I said "The new math functions emulate the behavior of the old
> ones" then would you be happy?
Not especially. The term "emulate" as applied to computers was first used
for publication by IBM in 1964. They defined emulation as the use of one
processor or system to simulate another through the use of hardware and/or
microcode assistance. Specifically, various IBM 360 models offered optional
emulation features which included extended microcode to allow the processor
to run code for older IBM computers, such as the 1401 and 7090.
In more recent years, the most common form of emulation is the In Circuit
Emulator (ICE), a tool commonly used for embedded system development.
I think the most appropriate terms to describe what the software engineers did
to produce the HP 10BII firmware are "port" (as mentioned in my last posting)
or "reimplement".
Jonathan Busby
<eric-no-s...@brouhaha.com> wrote:
Since I don't want this to draw on forever I'll just state that I
don't agree and leave it at that. But perhaps if we use a more neutral
synonym we can satisfy everyone's need for verbal perfection.
"The new math functions [blank] the behavior of the old ones."
[blank] = imitate|copy|mirror|reflect|reproduce|match|parallel
Take your pick ;)
Raymond Hellstern
before we're getting too off-topic:
Consider the 'HP-41CV Emulator Card' for the HP-48 as a simulation,
in contrast to the program from HrastProgrammer which is an emulation.
The HP/Zengrange card just simulates to be something like an HP-41,
HrastProgrammer's software emulates the HP-41 hardware in software,
so the original ROM images can be used.
Regards,
Raymond
"Jonathan Busby" <j...@SNMAPOhouston.rr.com> schrieb im Newsbeitrag
news:bav9puo6gdfn6ufbp...@4ax.com...
Eric Smith
> Consider the 'HP-41CV Emulator Card' for the HP-48 as a simulation,
> in contrast to the program from HrastProgrammer which is an emulation.
>
> The HP/Zengrange card just simulates to be something like an HP-41,
> HrastProgrammer's software emulates the HP-41 hardware in software,
> so the original ROM images can be used.
Except that "emulation" was defined by its inventors at IBM to mean
simulation with hardware and/or microcode assistance. HrastProgrammer's
software simulates the 41 hardware at a low level, but it doesn't have
any special hardware or microcode assistance, so it is not emulation.
References:
_IBM's Early Computers_, Charles J. Bashe et al
_IBM's 360 and Early 370 Systems_, Emerson W. Pugh et al
_Building IBM: Shaping an Industry and its Technology, Emerson W. Pugh
_A History of Computing in the 20th Century_, Nicholas Metropolis
_A History of Modern Computing_, Paul E. Ceruzzi
_IBM Dictionary of Computing_, George McDaniel
Jonathan Busby
<Raymond....@T-Online.de> wrote:
>Hi,
>
>before we're getting too off-topic:
>
>Consider the 'HP-41CV Emulator Card' for the HP-48 as a simulation,
>in contrast to the program from HrastProgrammer which is an emulation.
>
>The HP/Zengrange card just simulates to be something like an HP-41,
>HrastProgrammer's software emulates the HP-41 hardware in software,
>so the original ROM images can be used.
>
>Regards,
>
>Raymond
These are exactly the definitions I was thinking of when I used
"emulate" and "simulate" .
There are a multitude of software projects that describe themselves as
"emulators" and conform to the definition of something that replicates
the external and *internal* behavior of some architecture by accepting
the same data/programs and producing the same results. ( see
http://www.dictionary.com/search?q=emulate )
I can also think of many instances where the word "simulation" is
applied to something that replicates the external ( ie. interface )
behavior of something to a close degree but operates differently
internally. Some examples are a flight simulator, a mechanical arcade
machine simulator, a HDL test bench program with behavioral simulation
of say memory chips , and the list goes on. This would fit with my
usage of "simulate" to describe the 10BII math functions because of
the fact that they have the same external behavior as the 10B
functions but on a low level they operate differently internally.
Of course, there are also many counterexamples to the above. I think
it's just a matter of personal preference and is also dependent on
what context you're working in.
-------------------------------------------------------------------------------
John H Meyers
Prior competing definitions and historical references omitted,
but just to add some more chaos to the above:
http://www.dictionary.com/search?q=simulate
http://www.dictionary.com/search?q=emulate
http://www.dictionary.com/search?q=presently
"When I use a word," Humpty-Dumpty said,
it means just what I choose it to mean -- neither more nor less"
- Lewis Carroll, "Through the Looking Glass..." (1872)
"And I am right, and you are right, and all is right as right can be"
- W. S. Gilbert ("The Mikado")
John H Meyers
Raymond Hellstern
"Eric Smith" <eric-no-s...@brouhaha.com> schrieb im Newsbeitrag
news:qhk7l5x...@ruckus.brouhaha.com...
> "Raymond Hellstern" <Raymond....@T-Online.de> writes:
> > Consider the 'HP-41CV Emulator Card' for the HP-48 as a simulation,
> > in contrast to the program from HrastProgrammer which is an emulation.
> >
> > The HP/Zengrange card just simulates to be something like an HP-41,
> > HrastProgrammer's software emulates the HP-41 hardware in software,
> > so the original ROM images can be used.
>
> Except that "emulation" was defined by its inventors at IBM to mean
> simulation with hardware and/or microcode assistance. HrastProgrammer's
> software simulates the 41 hardware at a low level, but it doesn't have
> any special hardware or microcode assistance, so it is not emulation.
>
It was defined by people at IBM for their type of emulation,
where is the problem that other people may have redefined the meaning?
Take, for example, any terminal emulator program for a PC.
None of those can rely on a real terminal hardware,
but I never heard someone talk about those programs as simulators.
Take Emu48: It uses the original ROM images,
so it can be considered as emulation.
Even the HP/Zengrange product I mentioned in my earlier post
used 'Emulator' as part of the official HP product name,
which I consider wrong.
From my (and many other people's) point of view this is a simulation,
whereas Emu48 is an emulation.
If I write some software for the HP-48, I can test the *actual* binary for
the real machine in Emu48.
Emu48 runs exactly the same binary as the real hardware.
If it were a simulation, I would have to write maybe similar, but
*different* code for it.
That's exactly what distinguishes an emulation from a simulation.
Software for a specific platform can be run on a totally different platform
using special mechanics.
It doesn't matter if it's tied to hardware,
like the 8086 emulation in a Pentium processor,
or in software, like Emu48 (or nearly any other emulator out there)
It's just an other abstraction layer.
It seems that the words 'simulation' and 'emulation' and their meaning can
be interpreted somehow;-)
If there is enough interest, you could ask the author of Emu48 and the other
guys about their understanding regarding this topic.
However, I don't care too much about the difference.
Raymond
Eric Smith
> Except that "emulation" was defined by its inventors at IBM to mean
> simulation with hardware and/or microcode assistance. HrastProgrammer's
> software simulates the 41 hardware at a low level, but it doesn't have
> any special hardware or microcode assistance, so it is not emulation.
"Raymond Hellstern" <Raymond....@T-Online.de> writes:
> It was defined by people at IBM for their type of emulation,
> where is the problem that other people may have redefined the meaning?
It was defined by people at IBM for a new form of computer simulation
that involved hardware or microcode assistance. As the inventors of this
technique, they certainly have some priority in terms of defining what
the word means in this field. Certainly the word "emulate" existed before
1964, but not with any meaning particularly significant to computers.
If the "other people" you refer to have actually invented something new,
then by all means they should get to name it. But they haven't. They're
just trying to change the definition of a word for no good reason.
Given that there is a perfectly good word ("simulate") to describe
software that makes one computer execute software for another *without*
special microcode and hardware, using the word "emulate" to mean the
same thing does not provide any benefit.
It is useful to have two words that describe the same process with or
without the hardware/microcode assistance; making one of those words
refer to both just confuses the issue.
> Take, for example, any terminal emulator program for a PC.
> None of those can rely on a real terminal hardware,
> but I never heard someone talk about those programs as simulators.
True. I'm not sure how that terminology came into place. But it is
a different situation than making one computer execute instructions
written for another.
> Take Emu48: It uses the original ROM images,
> so it can be considered as emulation.
Emu48 does a darned good job of simulating the HP 48 hardware, and running
the ROMs. But calling it an emulator does not change its nature, nor does
it clarify anything.
There is a distinction between instruction-level simulation and high-level
functional simulation, but this is completely orthogonal to the
distinction between simulation and emulation.
If we use your definitions of simulate and emulate, it would suggest
that a terminal emulator should really be called a terminal simulator,
since it doesn't run the ROM code of the original terminal.
> If I write some software for the HP-48, I can test the *actual* binary for
> the real machine in Emu48.
> Emu48 runs exactly the same binary as the real hardware.
>
> If it were a simulation, I would have to write maybe similar, but
> *different* code for it.
> That's exactly what distinguishes an emulation from a simulation.
No, it's not. I cited some authoritative references for the definition
of emulation. Can you cite some for your definition?
Eric Smith
> I think
> it's just a matter of personal preference and is also dependent on
> what context you're working in.
Yes. And in the context of computers, the engineers who developed the
System/360 at IBM *invented* emulation and defined what it means. I
have never seen any historical record of the word emulate being used
in any reference to computers prior to 1964, though certainly the word
existed with non-computer-specific meaning before that.
Unless there's a very good reason for it, why should we abuse the term
to mean something else? That can only lead to confusion.
Veli-Pekka Nousiainen
news:qhadm1b...@ruckus.brouhaha.com...
X
> > Take Emu48: It uses the original ROM images,
> > so it can be considered as emulation.
>
> Emu48 does a darned good job of simulating the HP 48 hardware, and running
> the ROMs. But calling it an emulator does not change its nature, nor does
> it clarify anything.
I still prefer to use the word "emulation" here, since the hardware
is also simulated, not just the software eg. userlanguage & interface.
So I disagree with Eric (for the first time :)
It's time to change the old IBM definition!!
Veli-Pekka
PS:
Is firmware the ROM of a calc or the microcode in the CPU
??? ~( :-o)
Raymond Hellstern
Even IBM talks of 'Terminal Emulation' when they talk about software for a
specific machine (e.g. PC) to
do as if it were another.
http://www.networking.ibm.com/525/525files.html
and many many other places.
So if even the inventors use the word 'emulation' for software alone w/o
hardware/microcode support,
Emu48 is an emulator.
The text behind the following link
http://www.dictionary.com/search?q=emulate
is printed below (after the '>'):
>Computer Science.
>To imitate the function of (another system),
>
Yes
>as by modifications to hardware or software
>
Yes. Note *or software*
>that allow the imitating system to accept the same data,
>
Yes
>execute the same programs,
>
Yes
>and achieve the same results as the imitated system.
>
Yes
According to the above definition Emu48 is an emulator. Period.
Raymond
Christoph Giesselink
>
> There's also the Saca (Sacajawea) chip which is used by the 14B, 22S,
> 32S, and 32SII . (if my sources are correct) According to the same
> sources, the Saca is basically just a Bert but with more RAM and ROM (
> 16 KB and 512 KB respectively ) and with a one line alphanumeric
> display controller instead of a 7 segment one.
Are you sure that this is correct? The Lewis doc say in chapter 6:
"The chip has a 5th configuration register that is unused in the LEWIS chip.
It will be used for RAM configuration on the SACJAWEA chip."
The Bert doc says nothing about the Saca chip.
But you're absolutely right in the case that there's a chip missing between
the Bert and Clarke "powered" machines.
Christoph
Jonathan Busby
The RAM/ROM sizes are wrong due to a typo. I corrected myself here :
http://groups.google.com/groups?selm=ba2apu89lrv46v4tm8ifpees1162pl7fcb%404ax.com
It should be 512 bytes of RAM and 16 KB of ROM. These numbers are from
Craig A. Finseth's HPDATABase and Marc Staps book "Codenames of HP
Handheld Calculators and PDAs: facts and speculations" (in this
particular book he states that this information was obtained from
Dennis York, former manager of the Handheld Products Division in
Corvallis, 1989)
I'm sure there is a Sacajawae (1LE2) chip from the fact that there
would be a missing Saturn chip otherwise and various trusted sources
mention it. (including ones other than those mentioned)
The statement "the Saca is basically just a Bert" may be incorrect if
the Bert uses a different version of the processor core or is
substantially different from the Lewis in other respects. (other than
the display controller and RAM/ROM sizes) If that's the case then
maybe it should read "the Saca is basically just a Lewis but with a 1
line alphanumeric display controller instead of a 2x16 character dot
matrix display one" .
Jonathan Busby
<j...@SNMAPOhouston.rr.com> wrote:
>The statement "the Saca is basically just a Bert" may be incorrect if
>the Bert uses a different version of the processor core or is
>substantially different from the Lewis in other respects. (other than
>the display controller and RAM/ROM sizes) If that's the case then
>maybe it should read "the Saca is basically just a Lewis but with a 1
>line alphanumeric display controller instead of a 2x16 character dot
>matrix display one" .
Argg! That should be "2x22 character" . ;)