Which CPU is faster, TI-85's Z80 6 Mhz or HP's GX Saturn 4 Mhz?
Lil Thumpr
i can safely say that the ti blows the hp out of the water as far as
speed. i hate the hp48g, i wish i never bought the damn thing. anybody
want to trade a ti-86 for it? it responds so so so so so slowly, i press
a button, and a minute later it executes the command. the hp sucks.
Per Blomqvist
I have a HP48 GX and a TI-82, and the graphic plot on the TI is muth more
faster than the HP, I don't know what procesor the TI-82 have.
But on my TI-82 i can't whange the X-res.
Dave Arnett
Lil Thumpr wrote:
>
> i can safely say that the ti blows the hp out of the water as far as
> speed. i hate the hp48g, i wish i never bought the damn thing. anybody
> want to trade a ti-86 for it? it responds so so so so so slowly, i press
> a button, and a minute later it executes the command. the hp sucks.
Hey, great testimonial!
I didn't know the HP48 would work under water. And you've used it
as part of a dam, too?
Now, folks, we get stories here about how various HP products
survive abusive environments and keep right on performing. Let
me remind everbody that an anectdote posted here does not mean you
should go out and try this kind of thing intentionally. I know that
my four-year-old got hold of an HP48G prototype and got hooked on mine
hunt. (His goal was to FIND the mines and blow them up.) At bath time,
he snuck it into the tub and ...well, it's good I had extra protoypes
because that one never worked again. It may have been the Mr. Bubble[1]
that caused the problem, but the point is that I prefer to keep water
OUT of the HP48.
I am sorry, Lil' Thumper, that the TI didn't fare as well. I hope
nobody was hurt during its explosion.
Good Day!
Dave.
PS: You claim you press a single button and within 60 seconds the HP
sucks as well? How many gallons-per-minute will it pump? It this with
an add-on attachment, or is it an effect of osmosis across the membrane
keyboard? DA.
PPS: I hope nobody is offended that I had a little fun with this.
Clearly, different products match different people. If Lil Thumper
doesn't like the '48, it doesn't mean the '48 is universally useless,
nor does it mean that Lil Thumper is unenlightened. It may well be that
another product meets his needs better. There's no dishonor in that. I
hope we all wish him well in his endeavors, even though some may
wholeheartedly disagree with his analysis.
------
I don't speak for HP, or for Proctor & Gamble, when I post here.
[1] Mr. Bubble may well be a trade name protected under law. For
those who live in more civilised countries, it's a brand of soap used
to make mountains of bubbles in the bath, usually for children.
Luka Crnkovic
Lil Thumpr wrote:
>
> i can safely say that the ti blows the hp out of the water as far as
> speed. i hate the hp48g, i wish i never bought the damn thing. anybody
> want to trade a ti-86 for it? it responds so so so so so slowly, i press
> a button, and a minute later it executes the command. the hp sucks.
Oh, come on, get serious. You couldn't *possibly* relate a 48 to a
ti-86.
This could have two reasons
1. You work for TI
2. You are to stupid to use your 48.
I somehow suspect that the later is the case. I suggest that you RTFM
and
learn something about your machine. I doubt that you would question
the 48's capabilities again.
Stephane Jantzen
> Oh, come on, get serious. You couldn't *possibly* relate a 48 to a
> ti-86.
> This could have two reasons
>
> 1. You work for TI
> 2. You are to stupid to use your 48.
Very smart answer, dude. If you were not so blinded by your devotion to
your calculator (how pathetic), you could make the following answer :
- Saturn kicks z80's ass in FP numbers
- z80 kicks Saturn's ass in int numbers, and in asm (unless you use FP
in asm, but I seriously doubt it)
That's it, whether you like it or not. You can easily verify this by
computing the same loop on both machines, with ints and then floats. But
referring to some of your previous posts, I suspect you don't know much
about TI. You guy seem to make a *serious* misunderstanding between a
calc features (in this case the HP's much better) and its instrinsic
speed.
PS : as a former TI user, I'm probably stupid. But I'm still ready to
compare my diplomas with yours.
--
| Stephane J. Jantzen (mailto:step...@jriver.com) |
| J. River, Inc., 125 N First St, Minneapolis (http://www.jriver.com) |
JEEjohn
In article <342190...@jriver.com>, Stephane Jantzen
<step...@jriver.com> writes:
>
>PS : as a former TI user, I'm probably stupid. But I'm still ready to
>compare my diplomas with yours.
>
>--
>
>| Stephane J. Jantzen (mailto:step...@jriver.com) |
>| J. River, Inc., 125 N First St, Minneapolis (http://www.jriver.com) |
Looks like now we have somebody who thinks "He or she who dies
with the most diplomas - wins" !
Sgt. Pepper
That just proved your point that you are stupid, by wanted to compare your
diplomas with those of others. Inferiority complex?
In article <342190...@jriver.com>, step...@jriver.com wrote:
> PS : as a former TI user, I'm probably stupid. But I'm still ready to
> compare my diplomas with yours.
>
> --
>
> | Stephane J. Jantzen (mailto:step...@jriver.com) |
> | J. River, Inc., 125 N First St, Minneapolis (http://www.jriver.com) |
--
The road to this paradise was not as comfortable and alluring as the
road to the religious paradise; but it has shown itself reliable, and
I have never regretted having chosen it. --Albert Einstein
My email address is purposely modified to avoid spam.
Please send all replies to happ...@pipeline.com
Thanks.
Stephane Jantzen
> That just proved your point that you are stupid, by wanted to compare your
> diplomas with those of others.
I just couldn't figure out how I could convince him I wasn't ... but
maybe am I :) Anyway, I don't insult people asking for a question, only
because the question seems irrelevant to me ...
Jarno Peschier
happy4me@#pipeline#.com (Sgt. Pepper) wrote the following:
>That just proved your point that you are stupid, by wanted to compare your
>diplomas with those of others. Inferiority complex?
Ad hominem attacks are so... childish... <sigh>
Jarno Peschier, computer science student, Utrecht University
mailto:jpes...@cs.ruu.nl http://jarno.home.ml.org/
____________________________________________________________
'avwI' nejDI' narghta'bogh qama' reH 'avwI' Sambej
Josephine
Inferiority complex or not, he is right. I have compared my 48 GX with
the TI 86, and the TI is faster in some instances. However, I have not
tried it with the meta kernel.
On Thu, 18 Sep 1997 19:51:09 -0400, happy4me@#pipeline#.com (Sgt.
Pepper) wrote:
>That just proved your point that you are stupid, by wanted to compare your
>diplomas with those of others. Inferiority complex?
>
Luka Crnkovic
Stephane Jantzen wrote:
>
> > Oh, come on, get serious. You couldn't *possibly* relate a 48 to a
> > ti-86.
> > This could have two reasons
> >
> > 1. You work for TI
> > 2. You are to stupid to use your 48.
>
> Very smart answer, dude. If you were not so blinded by your devotion to
> your calculator (how pathetic), you could make the following answer
> But referring to some of your previous posts, I suspect you don't know much
> about TI. You guy seem to make a *serious* misunderstanding between a
> calc features (in this case the HP's much better) and its instrinsic
> speed.
I have never denied that the TI's processor was faster. I just reacted
against a person who said that a ti-86 was better. I said that
he couldn't possibly RELATE a 48 to a 86. I didn't say anything
about speed. I still think it is pointless to compare the 48 to
any non-symbolic, non-RPN machine.
>
> PS : as a former TI user, I'm probably stupid. But I'm still ready to
> compare my diplomas with yours.
MY DAD HAS GOT A BIGGER CAR THAN YOUR DAD!!!!!
Jean-Yves Avenard
Hello..
It is the first time I here that the Saturn CPU can manage FP numbers ?? Is
it a new Saturn ? where did you find here ??
The Saturn is very good for BCD calculation and have a lot of BIG register
(64bits), the range adress is a bit short (512kb) but remember me the adress
range of a Z80 ???
If you compare the Saturn with a "real" processor like the 68K or Risc , ok,
the Saturn is not so good afterall. But no comparison can be made with a Z80
(20 years old).
I don't know what you learn with your "diplomas", but surely not enough to
talk about this subject.
Jean-Yves Avenard
Stephane Jantzen wrote in article <342190...@jriver.com>...
>> Oh, come on, get serious. You couldn't *possibly* relate a 48 to a
>> ti-86.
>> This could have two reasons
>>
>> 1. You work for TI
>> 2. You are to stupid to use your 48.
>
>Very smart answer, dude. If you were not so blinded by your devotion to
>your calculator (how pathetic), you could make the following answer :
>
>- Saturn kicks z80's ass in FP numbers
>- z80 kicks Saturn's ass in int numbers, and in asm (unless you use FP
>in asm, but I seriously doubt it)
>
>That's it, whether you like it or not. You can easily verify this by
>computing the same loop on both machines, with ints and then floats. But
>referring to some of your previous posts, I suspect you don't know much
>about TI. You guy seem to make a *serious* misunderstanding between a
>calc features (in this case the HP's much better) and its instrinsic
>speed.
>
>PS : as a former TI user, I'm probably stupid. But I'm still ready to
>compare my diplomas with yours.
>
Stephane Jantzen
> It is the first time I here that the Saturn CPU can manage FP numbers ?? Is
> it a new Saturn ? where did you find here ??
There is a big difference between "managing FP numbers" and "having a
FPU". Any CPU can handle FP numbers ... it's either microcoded (x86's
FDIV) or must be done in another way. For instance, the z80 has no
built-in instruction for integer multiplication, this doesn't mean you
can't multiply. I don't know how the Saturn handles FP numbers (I
suspect it's microcoded in a way or the other, considering the speed),
but the fact is that FP operations on a Saturn are much faster than on a
z80. This is not surprising when you know that z80 has only integer
instructions.
> The Saturn is very good for BCD calculation and have a lot of BIG register
> (64bits), the range adress is a bit short (512kb) but remember me the adress
> range of a Z80 ???
Once more, the topic was the SPEED of the CPU. Nobody said that the z80
address space was larger. But 64bits registers are undoubtly a factor of
speed, especially considering the pseudo-16bits registers of the z80.
> If you compare the Saturn with a "real" processor like the 68K or Risc , ok,
> the Saturn is not so good afterall. But no comparison can be made with a Z80
> (20 years old).
Well, using the AGE of a CPU is quite irrelevant when talking about its
intrinsic performance ... even if this is a good indication most of the
time, it's not always the case. Moreover, even 20 years after its
design, the z80 is still used when high-speed is not an issue
David Maurice Cooke
Stephane Jantzen <step...@jriver.com> writes:
>
> > It is the first time I here that the Saturn CPU can manage FP numbers ?? Is
> > it a new Saturn ? where did you find here ??
>
> There is a big difference between "managing FP numbers" and "having a
> FPU". Any CPU can handle FP numbers ... it's either microcoded (x86's
> FDIV) or must be done in another way.
> For instance, the z80 has no
> built-in instruction for integer multiplication, this doesn't mean you
> can't multiply.
The Saturn can't multiply directly, either (except by 2).
> I don't know how the Saturn handles FP numbers (I
> suspect it's microcoded in a way or the other, considering the speed),
> but the fact is that FP operations on a Saturn are much faster than on a
> z80. This is not surprising when you know that z80 has only integer
> instructions.
IIRC, the Saturn can hold a (12-digit) FP in one of its 64-bit
registers (as a BCD number), and can manipulate the various parts,
like the exponent and mantissa, directly. On a Z80 you would have to
do most of the manipulations in memory.
I believe the difference that people most see in terms of speed is
graphing: Yes, the TI's are faster in this. Is there any other reason
besides processor speed? Could the plotting routine on the HP be
written in assembly to speed it up? What's the HP doing that makes it
slow?
|>|\/|<
----------------------------------------------------------------------------
David M. Cooke
dmc...@sfu.ca
Richard M. Smith
<snip all>
Most people would buy the H.P. calculator because of the *rich* programming
language designed by creative people. I almost never used the plotting
features in the SX or GX because the P.C. does a much better job in this
area. In school, Civil Engineering, I relied on the H.P. to store programs
personally written to execute intensive algorithms. I also purchased the
P.C. program development link software which allows programming on a real
screen.
If you want to save money, then go with the other guys. Their keyboards are
*poor*, and the displays are not crisp. Most of the people I knew at school
always wished they owned H.P. calculators. Save yourselves the anxiety. .
buy what you really want!
--
Sincerely,
Richard M. Smith
(805) 528-1282
rms...@pobox.com
For all you automated e-mail spammers,
the Federal Communications Commission:
Chairman Reed Hundt: rhu...@fcc.gov
Commissioner James Quello: jqu...@fcc.gov
Commissioner Susan Ness: sn...@fcc.gov
Commissioner Rachelle Chong: rch...@fcc.gov
Jean-Yves Avenard
My question now is :
Why do you continue to compare the Saturn and the Z80, since you prove that
you don't know at all the Saturn processor.
First, like the Z80, the Saturn as no instruction to multiply or divide two
registers.
2nd: There is no instruction to manage FP numbers (for example add or
substract FP numbers)
Now you said that 64bits register is a factor of speed, when in your
previous message you said that the Saturn is slower than the Z80 : you do
you think now ?
Is it slow or fast ?
There is no "microcoded" instruction in the Saturn, because no instruction
require it..
The z80 can manage only 64kb, it is not enough to use it in "good"
calculators like the HP48GX. Maybe it's enough with the TI-86, but you can't
compare this one with a 48GX: it's a nonsense.
The TI-86 is a toy for young student. If it's enough for High School, for
engineering school you will need a more sophisticated tool like the TI-92.
Jean-Yves
Stephane Jantzen wrote in article <34269D...@jriver.com>...
>> It is the first time I here that the Saturn CPU can manage FP numbers ??
Is
>> it a new Saturn ? where did you find here ??
>
>There is a big difference between "managing FP numbers" and "having a
>FPU". Any CPU can handle FP numbers ... it's either microcoded (x86's
>FDIV) or must be done in another way. For instance, the z80 has no
>built-in instruction for integer multiplication, this doesn't mean you
>can't multiply. I don't know how the Saturn handles FP numbers (I
>suspect it's microcoded in a way or the other, considering the speed),
>but the fact is that FP operations on a Saturn are much faster than on a
>z80. This is not surprising when you know that z80 has only integer
>instructions.
>
Stephane Jantzen
> Why do you continue to compare the Saturn and the Z80, since you prove that
> you don't know at all the Saturn processor.
Well ... I admit I'm not a Saturn guru, but I know enough, I think. And
moreover, even without knowing everything about the Saturn, it's easy to
see that in some cases it *seems* faster than a z80, in other cases not.
In all cases, what is interesting is the speed the CPU seems to provide
to the end user. And to do so, all you have to know is how to perform
the same kind of operations on both calcs. Hardware considerations just
come to explain why one is better than the other
> Now you said that 64bits register is a factor of speed, when in your
> previous message you said that the Saturn is slower than the Z80 : you do
> you think now ?
> Is it slow or fast ?
same answer : Saturn for FP, z80 for ints. Having a fast feature doesn't
mean the whole unit is faster. You can have a faster ram on your 386
than on your Pentium, this won't necessarily mean the 386 will always be
faster, because the factor we consider is not predominant.
> The z80 can manage only 64kb, it is not enough to use it in "good"
I agree with you, 64k is too less.
> The TI-86 is a toy for young student. If it's enough for High School, for
> engineering school you will need a more sophisticated tool like the TI-92.
I finished with school without having *more* than a TI-85 ... I guess
your rule isn't a strict one :)
Jonathan duSaint
Stephane Jantzen wrote:
<snip>
> > Is it slow or fast ?
>
> same answer : Saturn for FP, z80 for ints.
Actually, the Saturn handles both quite remarkably. Example:
LA(5) 2 the integer 2 in A[A]
C=A A save a copy in C[A]
ASL multiply by 16
A=A+C A A[A] now = 17 * 2 (int)
LAHEX 0000000000000020 A[W] = 2 (fp)
LCHEX 0000000000000020 C[W] = 2 (fp)
GOSBVL =SPLTAC convert both to long real
GOSBVL =ADDF add them
GOSBVL =PACKSB convert to short real, A[W] = 2 + 2 = 4
--Jon
--
==========================================
| Jonathan duSaint |
| mailto:jdus...@mathworks.com |
| |
| The Mathworks, Inc. |
| http://www.mathworks.com |
==========================================
Stephane Jantzen
Jonathan duSaint wrote:
> Actually, the Saturn handles both quite remarkably. Example:
Yep, but somewhere between the CPU and the user there must be a slow
layer, because a simple loop like : "for i between 0 and 1000, sum all
i's" is faster on a TI (I think twice). If you replace "sum all i's" by
"sum the square roots of all i's", then the HP is *far* better.
The apparent speed ratio cannot be directly translated into a CPU power
ratio, but it is a good approximation, if you suppose that both OS'es
have the same efficiency.
Dan Kirkland
In article <yzk1d8m0...@fraser.sfu.ca>,
David Maurice Cooke <dmc...@fraser.sfu.ca> writes:
>I believe the difference that people most see in terms of speed is
>graphing: Yes, the TI's are faster in this. Is there any other reason
>besides processor speed? Could the plotting routine on the HP be
>written in assembly to speed it up? What's the HP doing that makes it
>slow?
It has been said many times that the HP48 uses finer resolution
at default settings. This means the HP48 does more calculations
for the plot.
And while I have never used a TI, I am pretty sure the HP48 has
many more options with it's plotter. (Can a TI plot for any
named variable? Or just X?) Having more options can also slow
things down.
dan
Balazs Fischer
JA> The TI-86 is a toy for young student. If it's enough for High
JA> School, for engineering school you will need a more sophisticated
JA> tool like the TI-92.
This might seem like heresy, but you don't need more than a basic calc that
can do +, -, *, /, sin, cos, ln, exp and inverse functions. JEHOVA!!
cu
Balazs Fischer
Balazs....@studbox.uni-stuttgart.de
PGP Key: http://pgp5.ai.mit.edu:11371/pks/lookup?op=get&search=0xED7447E5
... Help stamp out and abolish redundancy!
tomas muehlhoff
[Stephane schrieb am Dienstag folgendes:]
>in other cases not. In all cases, what is interesting is the speed
>the CPU seems to provide to the end user. And to do so, all you
you can't compare processor speeds if you want to know the provided
end-user-speed.
f.e. the 48 has to work out stuff besides the real calcing stuff. think
about the clock and stuff. if you use the 48 you will know, that it
pauses sometimes, so you can't just compare that. you know thath the 48
is slower or faster, depending on how much ram is installed, because
the 48 has to 'handle' so much stuff besides calcing your problem.
>have to know is how to perform the same kind of operations on both
>calcs. Hardware considerations just come to explain why one is
>better than the other
work out some usual operations, think how you would realize them. look
into the data book of the processor, count the clock cycles needed for
your operations and divide them by the clock speed.
>same answer : Saturn for FP, z80 for ints. Having a fast feature
btw. in what comps IS the saturn ?
>doesn't mean the whole unit is faster. You can have a faster ram on
>your 386 than on your Pentium, this won't necessarily mean the 386
what does all this have to do with the speed of the ram ? (which means
nothing else than, when do i have to refresh the memory contents)
>> The z80 can manage only 64kb, it is not enough to use it in
>> "good"
>
>I agree with you, 64k is too less.
too less for what ?
>> The TI-86 is a toy for young student. If it's enough for High
>> School, for engineering school you will need a more sophisticated
>> tool like the TI-92.
>
>I finished with school without having *more* than a TI-85 ... I
>guess your rule isn't a strict one :)
i finished school with a TI-33 galaxy, then i had a Casio FX4500P and a
FX5500G, after all that i now own a 48G, all love RPN :) didn't program
that 48 yet, but 8085s and PDP11s and stuff.
sometimes i prefer the 4500P :)
tom
## "Ich wuenschte, ich wuerde mich fuer Tennis interessieren, [...]
## Es ist besser vor dem Stumpfsinn zu kapitulieren."
## (Tocotronic)
Richard M. Smith
Balazs Fischer <Balazs....@studbox.uni-stuttgart.de> wrote in article
<858d...@studbox.uni-stuttgart.de>...
<snip>
> This might seem like heresy, but you don't need more than a basic calc
that
> can do +, -, *, /, sin, cos, ln, exp and inverse functions. JEHOVA!!
<snip>
Yes, your correct about minimum requirements. However, the rich programming
language of the HP's is the fruit for the creative mind.