Guy Macon, do you have anything to contribute?

[Jim Thompson]

On Sat, 22 May 2004 17:55:36 -0700, Guy Macon
<http://www.guymacon.com> wrote:

Jim Thompson <thegreatone@example.com> says...

RPN is for dweebs. My mind works Algebraically.

...

Analog Innovations, Inc.
Analog/Mixed-Signal ASIC's and Discrete Systems

Ever put a FORTH core in an ASIC?

B16: a FORTH processor in an FPGA
http://www.jwdt.com/~paysan/b16-eng.pdf

VLSI Design Tools In 500 lines of colorForth these tools
provide everything required to design a chip...
http://www.colorforth.com/vlsi.html

I have a subcontractor who does digital cores for me when I need them.
The same fellow I recommended for VHDL consulting.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice480)460-2350 | |
| E-mail Address at Website Fax480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

 

[Jim Thompson]

On Sat, 22 May 2004 18:08:49 -0700, Guy Macon
<http://www.guymacon.com> wrote:

Terry Given <the_domes@xtra.co.nz> says...

"Jim Thompson" <thegreatone@example.com> wrote...

RPN is for dweebs. My mind works Algebraically.

Well put. So does my HP28S.

Of course it does. Only the Nazis over at TI give you no choice...

My theory is that once Jim Thompson dicovers the power and elegance
of putting a FORTH core into an ASIC, he will switch to RPN.

Barf !-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice480)460-2350 | |
| E-mail Address at Website Fax480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

 

[Al Borowski]

(I can't see Jim's post, so I'm replying to this one)

RPN is for dweebs.

Oh, how mature (rolls eyes)

My mind works Algebraically.

Really?

So, if I asked you to calculate

(5+4)/(3+7) * 9

by hand, how would you do it?


Personally, I'd add 5 and 4, kepping the result in my head. Then add 3
and 7, divide the results, then multiply the result by 9.

The same thing on an RPN calc is

5 [enter] 4 +
3 [enter] 7 +
/
9 *

Of course thats a dead simple example, theres no real advantage here.
Symbolic stuff is a different story.

cheers,

Al

...Jim Thompson


Well put. So does my HP28S.

Cheers,
Terry

 

[KR Williams]

In article <X5ydnXeCeooCZzLdRVn-hA@speakeasy.net>, Guy Macon
<http://www.guymacon.com> says...

KR Williams <krw@att.biz> says...

RPN is for dweebs. My mind works Algebraically.

Well put. So does my HP28S.

What a waste.

Think of it as training wheels...

I didn't learn to ride a bike until my dad took the training
wheels off. It only took 20 minutes after.

--
Keith

 

[KR Williams]

In article <X5ydnXWCeoovZDLdRVn-hA@speakeasy.net>, Guy Macon
<http://www.guymacon.com> says...

Terry Given <the_domes@xtra.co.nz> says...

"Jim Thompson" <thegreatone@example.com> wrote...

RPN is for dweebs. My mind works Algebraically.

Well put. So does my HP28S.

Of course it does. Only the Nazis over at TI give you no choice...

My theory is that once Jim Thompson dicovers the power and elegance
of putting a FORTH core into an ASIC, he will switch to RPN.

So, Guy... Since this has now turned into a political argument,
why haven't you put your patented [OT] (or is that [OT:]) tag on
the subject? ...and declare all that don't agree with you
"heathens"? ;-) Hey, I'm not going to "change" threads by
changing the subject line. DIYDDIYD (I'm not going there .

--
Keith

 

[Terry Given]

"Guy Macon" <http://www.guymacon.com> wrote in message
news:69OdnXv7I-yDajLdRVn-ig@speakeasy.net...

KR Williams <krw@att.biz> says...

The newer HPs are both arithmetic and RPN (or so they state), but
why would you want arithmetic? RPN rulz!

It's a mystery! I am still trying to figure out why so many people
use (algebraic) C/C++ when (RPN) FORTH is clearly superior....

I havent programmed in FORTH since I was a teenager. I have seen an
industrial control system implemented in FORTH though. Come to think of it,
other than DSP algorithms I havent programmed anything for about 10 years.

Terry

 

[Terry Given]

"KR Williams" <krw@att.biz> wrote in message
news:MPG.1b19d27581ec14c59898e8@news1.news.adelphia.net...

In article <X5ydnXeCeooCZzLdRVn-hA@speakeasy.net>, Guy Macon
http://www.guymacon.com> says...

KR Williams <krw@att.biz> says...

RPN is for dweebs. My mind works Algebraically.

Well put. So does my HP28S.

What a waste.

Think of it as training wheels...

I didn't learn to ride a bike until my dad took the training
wheels off. It only took 20 minutes after.

--
Keith

Gentlemen,
please dont assume I actually USE my HP28S for algebra - I dont. I do almost
all of my algebra by hand (calcs too, at least until I set up detailed
mathematical models), only occassionally resorting to MAPLE (well mathcad,
but its the maple engine) - usually to prove my hand-derived results. Alas,
Maple does a great job of mangling algebraic expressions, to minimise their
readibility/usability, BUT it is real good at re-arranging my expression to
look as nasty as its own, at which point elemntary sesame-street theory (one
of these things is not like the other) can be used to determine veracity.

Terry

 

[Guy Macon]

Al Borowski <aj.borowski@erasethis.student.qut.edu.au> says...

(I can't see Jim's post, so I'm replying to this one)

RPN is for dweebs.

Oh, how mature (rolls eyes)

My mind works Algebraically.

Really?

So, if I asked you to calculate

(5+4)/(3+7) * 9

by hand, how would you do it?

Personally, I'd add 5 and 4, kepping the result in my head. Then add 3
and 7, divide the results, then multiply the result by 9.

The same thing on an RPN calc is

5 [enter] 4 +
3 [enter] 7 +
/
9 *

Of course thats a dead simple example, theres no real advantage here.
Symbolic stuff is a different story.

IRRC, studies have shown that RPN users calculate faster with fewer errors.

 

[Guy Macon]

KR Williams <krw@att.biz> says...

So, Guy... Since this has now turned into a political argument,

....a good-natured one, I hope. If anyone thinks that RPN vs. Algebraic
is worth having a *real* fight over instead of the present joking around,
well what else would you expect from RPN-rejecting heretics?

....Uuuuh...because it's already there?

 

[Guy Macon]

Jim Thompson <thegreatone@example.com> says...

Guy Macon <http://www.guymacon.com> wrote:

My theory is that once Jim Thompson dicovers the power and elegance
of putting a FORTH core into an ASIC, he will switch to RPN.

Barf !-)

That's O.K. Jim... Someday you will see the light. <big grin>

 

[Terry Given]

"Al Borowski" <aj.borowski@erasethis.student.qut.edu.au> wrote in message
news:40b00d76$0$16578$5a62ac22@freenews.iinet.net.au...

(I can't see Jim's post, so I'm replying to this one)

RPN is for dweebs.

Oh, how mature (rolls eyes)

My mind works Algebraically.

Really?

So, if I asked you to calculate

(5+4)/(3+7) * 9

by hand, how would you do it?


Personally, I'd add 5 and 4, kepping the result in my head. Then add 3
and 7, divide the results, then multiply the result by 9.

The same thing on an RPN calc is

5 [enter] 4 +
3 [enter] 7 +
/
9 *

Of course thats a dead simple example, theres no real advantage here.
Symbolic stuff is a different story.

cheers,

Al

Nope, I do it "algebraically" - actually, by pattern matching. I recognise
5+4 as 9, likewise 3+7 as 10, and I know dividing by 10 is easy, so I do it
last. I also know 9*9 is 81, so 8.1 is the answer. I do similar things to
calculate answers in my head, as 2,5,10 are easy to work with, and addition
is easy too. pick a harder example.

RPN is a nice, systematic algorithm. there are others.

Funny Story:
doing a contract compliance test for our newly designed product, we had to
image a screen with a fancy camera. We loacted the camera at the
contractually specified distance and height, then had to work the angle. My
boss (who is studying for a PhD in EE) asked me to calculate the angle from
the 2 measurements as he didnt have a calculator ie tan(theta)=y/x =
sin(theta)/cos(theta). Using the series approximation sin(m)=m and
cos(m)=1-m^2/2, along with a pen and the flap of a cardboard box, I
calculated theta for him. It took far longer to find a pen than do the
maths. He thanked me, and proceeded to turn on his laptop then enter the
number into Excel. When I pointed out to him that he could just calculate
the exact answer he looked pretty sheepish. I was within 1% of the correct
angle.

Terry

 

[Al Borowski]

So, if I asked you to calculate

(5+4)/(3+7) * 9

by hand, how would you do it?


Nope, I do it "algebraically" - actually, by pattern matching. I recognise
5+4 as 9, likewise 3+7 as 10, and I know dividing by 10 is easy, so I do it
last. I also know 9*9 is 81, so 8.1 is the answer. I do similar things to
calculate answers in my head, as 2,5,10 are easy to work with, and addition
is easy too.

I made the numbers too easy. Repeat with 5.436, 4.225 etc

People's minds work different ways.

pick a harder example.

Seeing this is supposedly an electronics group (at least in name), how
about an equiv resistor problem? I'll keep the numbers simple again just
for easy typing.

Suppose you have 4,5 and 6 ohm resistors in parallel. In series with
them you have a 7 ohm resistor. In parallel with *that* arrangement you
have an 8 ohm resistor. Whats the equiv resistance?


0----------------------------------
| |
| |
.-. |
| | |
|7| |
'-' |
| |
| .-.
| | |
-------|------- |8|
| | | '-'
.-. .-. .-. |
| | | | | | |
|4| |5| |6| |
'-' '-' '-' |
| | | |
-------|------- |
| |
| |
0-------------|-------------------|



created by Andy´s ASCII-Circuit v1.22.310103 Beta www.tech-chat.de

Note: # means the '1/X' key.

first find the equiv resistance of the 4,5,6 ohm part.

[ 4 # 5 # 6 # + + # ] (find the inverses of 4,5 and 6, adds them and
inverts the resul)

then add the 7 ohm resistor in series

[ 7 + ]

then place the 8 ohm resistor in parallel with the resistance so far

[ # 8 # + #] (again, find the inverse of the answer so far, find the
inverse of 8, add them, find the inverse of that)


This look terrible to read on a PC screen, but is shown graphically as
you type it - you can follow the working easily if you do it on a
graphics calc. After the first part the screen would show (1/
(1/4+1/5+1/6) in textbook form for instance.

RPN is a nice, systematic algorithm. there are others.

Yep. Its just personal preference. Thats enough procrastination for now,
I have an assignment due tomorrow I really need to finish :-(

cheers,

Al

 

[Terry Given]

"Al Borowski" <aj.borowski@erasethis.student.qut.edu.au> wrote in message
news:40b07de8$0$16574$5a62ac22@freenews.iinet.net.au...

Seeing this is supposedly an electronics group (at least in name), how
about an equiv resistor problem? I'll keep the numbers simple again just
for easy typing.

Suppose you have 4,5 and 6 ohm resistors in parallel. In series with
them you have a 7 ohm resistor. In parallel with *that* arrangement you
have an 8 ohm resistor. Whats the equiv resistance?


0----------------------------------
| |
| |
.-. |
| | |
|7| |
'-' |
| |
| .-.
| | |
-------|------- |8|
| | | '-'
.-. .-. .-. |
| | | | | | |
|4| |5| |6| |
'-' '-' '-' |
| | | |
-------|------- |
| |
| |
0-------------|-------------------|



created by Andy´s ASCII-Circuit v1.22.310103 Beta www.tech-chat.de

OK, 4,5,6 are all common factors of 60 - 10x6, 12x5, 15x4. therefore the
parallel combination is 60/37 ohms. Add 7=259/37 to get 319/37 ohms = 8 +
23/37 ~ 8.7 ohms so the answer will be R//1.1R where R=8 = 1.1/2.1 * R =
8.8/2.1 = 4.2Ohms. look ma, no calculator.

check with HP28S: 4||5||6 ohms = 1.6216 = 60/37. +7 = 8.6216 ohms || 8 oms =
4.1496 ohms. my "guess" is out by (1-4.2/4.1496) = 1.2%.

pattern matching wins again.

actually thats how i always do parallel/series calcs, its easy (well it is
for me). N resistors in parallel is always one Nth. so I set 6R = 10 60R in
parallel....

QED
Terry

 

[Al Borowski]

OK, 4,5,6 are all common factors of 60 - 10x6, 12x5, 15x4. therefore the
parallel combination is 60/37 ohms. Add 7=259/37 to get 319/37 ohms = 8 +
23/37 ~ 8.7 ohms so the answer will be R//1.1R where R=8 = 1.1/2.1 * R =
8.8/2.1 = 4.2Ohms. look ma, no calculator.

show off


Al

 

[Richard Henry]

"Terry Given" <the_domes@xtra.co.nz> wrote in message
news:Xa0sc.8952$XI4.322171@news.xtra.co.nz...

"Al Borowski" <aj.borowski@erasethis.student.qut.edu.au> wrote in message
news:40b07de8$0$16574$5a62ac22@freenews.iinet.net.au...

Seeing this is supposedly an electronics group (at least in name), how
about an equiv resistor problem? I'll keep the numbers simple again just
for easy typing.

Suppose you have 4,5 and 6 ohm resistors in parallel. In series with
them you have a 7 ohm resistor. In parallel with *that* arrangement you
have an 8 ohm resistor. Whats the equiv resistance?


0----------------------------------
| |
| |
.-. |
| | |
|7| |
'-' |
| |
| .-.
| | |
-------|------- |8|
| | | '-'
.-. .-. .-. |
| | | | | | |
|4| |5| |6| |
'-' '-' '-' |
| | | |
-------|------- |
| |
| |
0-------------|-------------------|



created by Andy´s ASCII-Circuit v1.22.310103 Beta www.tech-chat.de


OK, 4,5,6 are all common factors of 60 - 10x6, 12x5, 15x4. therefore the
parallel combination is 60/37 ohms. Add 7=259/37 to get 319/37 ohms = 8 +
23/37 ~ 8.7 ohms so the answer will be R//1.1R where R=8 = 1.1/2.1 * R =
8.8/2.1 = 4.2Ohms. look ma, no calculator.

check with HP28S: 4||5||6 ohms = 1.6216 = 60/37. +7 = 8.6216 ohms || 8 oms
=
4.1496 ohms. my "guess" is out by (1-4.2/4.1496) = 1.2%.

pattern matching wins again.

actually thats how i always do parallel/series calcs, its easy (well it is
for me). N resistors in parallel is always one Nth. so I set 6R = 10 60R
in
parallel....

Hmm...well...

I first looked at the drawing holistically: 8 ohms in parallel with (7 ohms
plus about 1/3 of 5 ohms), so the combination will be a little over 4 ohms,
say 4 and a quarter as a right-brain guess. That's within 2.5% of the
left-brain calculation.

What's the tolerance on those resistors again?

 

[KR Williams]

In article <PNudnTLJmptAki3dRVn-iQ@speakeasy.net>, Guy Macon
<http://www.guymacon.com> says...

Al Borowski <aj.borowski@erasethis.student.qut.edu.au> says...

(I can't see Jim's post, so I'm replying to this one)

RPN is for dweebs.

Oh, how mature (rolls eyes)

My mind works Algebraically.

Really?

So, if I asked you to calculate

(5+4)/(3+7) * 9

by hand, how would you do it?

Personally, I'd add 5 and 4, kepping the result in my head. Then add 3
and 7, divide the results, then multiply the result by 9.

The same thing on an RPN calc is

5 [enter] 4 +
3 [enter] 7 +
/
9 *

Of course thats a dead simple example, theres no real advantage here.
Symbolic stuff is a different story.

IRRC, studies have shown that RPN users calculate faster with fewer errors.

That's only because people who use RPN are smarter. ;-)

--
Keith

 

[Guy Macon]

Al Borowski <aj.borowski@erasethis.student.qut.edu.au> says...

Richard Henry <rphenry@home.com> says...

Terry Given <the_domes@xtra.co.nz> says...

Al, Richard and Terry, would you be son kind as to edit the subject
line so as to remove my name? I would very much appreciate it.

 

[Guy Macon]

KR Williams <krw@att.biz> says...

Guy Macon <http://www.guymacon.com> says...

IRRC, studies have shown that RPN users calculate faster with fewer errors.

That's only because people who use RPN are smarter. ;-)

Ah, but does using RPN make you smarter, or do the smart people naturally
gravitate towards RPN?

 

[John Larkin]

On Sun, 23 May 2004 10:50:25 -0700, Guy Macon
<http://www.guymacon.com> wrote:

KR Williams <krw@att.biz> says...

Guy Macon <http://www.guymacon.com> says...

IRRC, studies have shown that RPN users calculate faster with fewer errors.

That's only because people who use RPN are smarter. ;-)

Ah, but does using RPN make you smarter, or do the smart people naturally
gravitate towards RPN?

Both. Of course, HP was dumb enough to quit making RPN calculators, so
there may be some flaw in the theory. Maybe all the smart people split
off to Agilent.

John

 

[Rich Grise]

"Terry Given" <the_domes@xtra.co.nz> wrote in message
news:1_Frc.8404$XI4.308502@news.xtra.co.nz...

"did you know, that by not even looking in my general direction, let alone
at my face, throughout the entire transaction, you have made me feel both
insignificant and worthless. you cant be bothered to even acknowledge my
presence"

Personally, I have a strong opinion about the text of what you're saying
here, and the undercurrent/attitude it puts forth.

The shithead couldn't possibly have made you feel insignificant and
worthless unless you already have insignificant and worthless feelings
waiting in you for any excuse to boil to the surface. Learning how
to deal with them is a whole nother topic, but for this instance,
The correction I'd use would be along the lines of:

"Ignoring a customer during a transaction is inconsiderate and rude.
When you don't acknowledge a customer it's impossible for him to know
if you're even hearing the order, let alone getting it right. And it's
bad business practce. Keep the <product>, I won't be doing any business
here as long as customers get treated with such disrespect, and please
let the manager know I said so and why."

I was in a local Wendy's recently, and while I was ordering my burger,
the drone interrupted to say, "Do you want everything on that?" This
is supremely annoying. I stopped, and said, "If you'll stop interrupting,
I'll tell you what I want." He was dumbfounded. He sort of backed away
from the register, looking like he didn't know whether to threaten me
or cry. The manager had to take over. In cases like this, I often remark:
"It's so hard to get polysyllabic help these days."

Nobody gets it.

Cheers!
Rich