negative hex

I thought you mis-spelled steam. Steaming oneself did seem a bit odd. Oh, never mind...

Incidentally, a friends #2 son is a graduate of the self-esteem skool of worthless college education. He couldn't find an acceptable job after graduating college, so his father hired him to cut metal and shuffle paper in his machine shop. I've never seen so much ineptitude and sloth combined with supreme arrogance in one package. He can't do anything right, doesn't want to learn anything new, and is amazingly well versed in the art of manufacturing excuses. He spends considerable time and effort "finding himself". When his father get up enough nerve to throw him out and into the real world on his own, he'll probably figure out what he's doing wrong. Meanwhile, I'm fascinated by how many mistakes he can make without ever accepting any responsibility. Yep, teaching self-esteem is certainly useful.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558            jeffl@comix.santa-cruz.ca.us
# http://802.11junk.com               jeffl@cruzio.com
# http://www.LearnByDestroying.com               AE6KS

A career student?

"

Nope. His grades weren't good enough to consider going for his Masters. I forgot what he majored in, but it's possible that there is no masters program for his elected discipline. He didn't seem to like college anyway and decided to spend the summer "finding himself". Unfortunately, he's also broke, so "finding himself" will have to wait.

I would feel better if he suggested a get rich quick scheme, business deal, or engaged in some form of criminal activity. Those would all show initiative. Unfortunately, even those common activities are lacking.

What's apparent after even a few minutes of conversation is that he's supremely confident in anything he announces or pontificates. I've only seen that in royalty and the young children of the very rich. He has merely to say the magic words, and the entire world will stop, drop what they're doing, and engage in whatever activity he finds interesting. That's the result of self-esteem, where the emphasis is on the "self".

Anyway, I have more important things to do than to "straighten out" my friends kids. My crisis for today is whether and how much to invoice a customer for the 3 hours it took to recover from their totally disgusting office coffee.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558            jeffl@comix.santa-cruz.ca.us
# http://802.11junk.com               jeffl@cruzio.com
# http://www.LearnByDestroying.com               AE6KS

I used to be an SOB (son of boss), but I was acutely aware (without my dad having to make much point of it, that I can recall) that I had to prove myself twice over what anyone else did (since I couldn't rest on the assumption that my job meant that I was valuable). Knowing that really screwing up would negatively affect family finances kept me motivated, too.

Dunno yet if my kids will turn out that way, but if they do I hope I can bring myself to kick them out on the street. You serve no one by keeping them on.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

I had the same experience with my father. He owned a lingerie factory in Smog Angeles. I initially helped out on weekends and sometimes evenings. I started out as a shipping clerk and eventually became a fairly good industrial sewing machine mechanic. In between, I attempted literally every possible job function in the factory. This wasn't really a planned exercise in on the job training, but rather I filled in when someone was sick or on vacation. Nobody told me that I couldn't do anything, so I just did it. What little training there was, consisted of reading the machine manuals (in Italian and German), asking dumb questions, and making every possible mistake exactly once. I had no fear of tearing apart a complicated sewing machine although I did require some expert help getting a few back together. It's still my trademark as "Learn By Destroying".

When I turned 18, my father handed me the ritual presents, a rather large cheque, and informed me that I had 4 weeks to get out of the house and find a place to live. I was going to Jr College at the time. It was unexpected and quite a shock. Over a period of about 4 years, I came back about 5 times, usually to recover from my latest social or financial disaster. The last time, I didn't require an invitation to leave. My explosive experiments blew out one wall of the garage, destroyed a workbench, and blackened the side of my mothers car. It was over a year before my father would even speak to me.

There were also various useful adventures along the way. You might consider running your son through a few of these.

1. At age 16, a sit down lunch at the local homeless shelter (Union Rescue Mission in L.A.) where I got to rub elbows with society's losers. It gave me a clue as to what can happen if I screw up.
2. At age 14, a tour of the local hospital oncology ward after I was caught smoking. Getting the don't smoke lecture from someone with ephesima had the desired effect.
3. Getting stuck in the Lincoln Heights jail for an extra 2 weeks thanks to my father arranging for my bank account to be inaccessible so I couldn't make bail. Not my idea of fun especially after they couldn't figure out who I was or why I was there.
4. Getting lessons in high speed driving from a suicidal maniac driving instructor after rolling my mothers car.

I think throwing me out was the "right" thing to do. I could easily have become too comfortable and complacent living at home, and not derived the necessary lessons from my subequent mistakes. One doesn't learn basic urban survival skills out of a book. Throw the kid out, be prepared for him to return a few times, and he'll probably thank you for it later. I did.

Incidentally, self-esteem was never a problem or even a question. I was far too busy causing trouble, or wiggling out of trouble, to even think about it. I do run on pride, but that's the pride of accomplishment, not the pride of just being there (i.e. self-esteem).

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558            jeffl@comix.santa-cruz.ca.us
# http://802.11junk.com               jeffl@cruzio.com
# http://www.LearnByDestroying.com               AE6KS

In defence of sign magnitude, two's compliment gets ugly fast when dealing with bignums, it works admirably when dealing with fixed size numbers however.

yes, one good reason not to use assembler.

some things are just easier to do efficiently in assembler, especially when the processor offers features that the compiler doesn't understand.

things like carry, and overflow, flags, opcodes that enable efficient n-word wide bit-shifts...

Bye.

You mean excess-3? That confuses some, but radix negative three is more fun, it can express signed integers of any mangnitude without need for a sign bit...

eg; 17 decimal is "212" and -5 is "21"

Not using assembler on critical embedded systems presupposes two conditions:

1. The guy who wrote the compiler is smarter than you are.
2. You understand exactly how the compiler works.

John

Or just that he had more time than you do and was "reasonably" smart.

Just looking at the assembly output and making sure it's "reasonable" and running it through your own validation routines & profiler works for this.

In general I agree with you, John, although these days I find it's often easier to pay a few bucks more for a faster CPU or a "hardware assist" in the form of an FPGA/PLD than to spend a week optimizing low-level assembly code to squeeze the last ounce of performance out of a cheap CPU.

You're in the rather unique position of having been programming 68k CPU cores in assembly for well over a decade now, right?

How often do you nead to port a critical embedded system to other CPUs?

Thanks, Rich

I'd rather write correct, carefully commented assembly code than check some compiler's cryptic output line-by-line.

Yup. But I wouldn't program some machines in assembly, like some RISC chips, or any Intel horror, and I wouldn't do high-level stuff like TCP/IP stacks or GUIs in assembly. For deep-embedded stuff, assembly is as coding-efficient as anything else, and usually a lot more hardware-efficient.

We are lately doing a lot of the heavy lifting in an FPGA, and letting the uP do a more supervisory role. Doing 64 multiplies and a hundred adds every 8 ns is hard to beat.

John

I don't know, but I sure do see a lot of wingeing here about FPGAs and such going out of support after a few years. . .

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

On a sunny day (Thu, 03 Jul 2008 14:07:27 -0700) it happened John Larkin wrote in :

It may be of interest to you, this link:

They want to use artificial intelligence to optimise GCC (the C compiler) for embedded. I went trough the links, and am very sceptical about this, did not download the soft though. But then the compiler would be smart or smarter then you (given a fixed development time frame).

It seems that your standard is made out of tin rather than gold.

2s compliment is quite easy and easily generalizable if you understand the hardware.

Incorrect approach. Implement the basal functions is asm. Then go to more regular languages like c/c++.

Not to mention direct arithmetic on BCD.

Interesting idea of more than word wide shifts being implemented in asm opcodes, i have yet to see it implemented.

If anything, a serious understatement.

When the base device becomes unobtainium. About a deacadely occurrence from what i read. This is often somewhat reduced by moving to a similar device, which accounts for much of the 8*51 class devices longevity.

More closely, they encode conditions that may be interesting or useful to the user, perhaps joined with an (optional) interrupt. So far they are sans will, AFAIK.

The creator of the method created the "XS-3" moniker, sorry about mistyping it. Would you explain the encoding rules for Radix (-3)? I may find it mathematically interesting.