Re: Overview Of New Intel Core i7(Nehalem) Processor

On Sat, 13 Jun 2009 10:48:16 GMT, Jan Panteltje
<pNaonStpealmtje@xxxxxxxxx> wrote:

On a sunny day (Fri, 12 Jun 2009 16:43:52 -0700) it happened John Larkin
<jjlarkin@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote in
<q5p5351phu2ga739k97qku73f5rh99egip@xxxxxxx>:

When is teh last time yo uwrote a big program for something
tha tdid not yet exist YOURSELF?

This, every line of it:

http://www.highlandtechnology.com/DSS/T346DS.html

I am impressed, but was not that in ASM, where the poor customer
has to create funny RS232 commands to have it do anything?



What difference does it make what it's programmed in? What matters is
that it works, that it has no bugs, and that it took less than two
weeks to code, test, and release.

And the serial commands are not "funny." They are clean and easy to
use and they include a nice set of HElp commands. People buy this
stuff and nobody has complained about the syntax. SCPI is, on the
other hand, a nightmare both to compose and to parse.

To set the channel 2 frequency to 31.552 MHz, 2.5 volts peak, just
send

2Freq 31.552M; 2Ampl 2.5 <cr>

and it replies

OK;OK <cr>

where only the first two chars of a token are needed. What's wrong
with that?

Send STatus <cr> to get a nice system summary screen. It's very
friendly.

This thing has over 250 distinct serial commands. In under 2 weeks.
The parser is pretty elegant.


And I did the product definition, the schematic design, the manual,
the test procedure, the PC-resident test/cal program, and the embedded
code. I negotiated the FPGA architecture but didn't drive the Xilinx
software myself. Zero bugs in the code. No prototype, and we shipped
the first-etch Rev A boards.


Because I think if you ever did, you would not talk like this.
Am I right?

No.

hehe
Yes, boards first time right is good.
But it seems you have never done any of those higher level software
program... especially ones that depend on a hundred other libs.

Exactly. By choice. That would take longer, run slower, be mostly out
of my control, require a bunch of external RAM, and probably ship with
bugs. Why would I want to do that?

One joy of being an electronics designer is that you don't have to get
tangled up in insane crap like that.


Basically what you wrote is just a driver ;-)
After that the real thing starts,

It manages a really cool DDS synthesizer/ARB. It even includes a set
of demo waveforms, like the one that generated the mpeg movie. Yes,
that's built-in. If a user calls us and says something like "channel 2
is broken" we can just say "send it the 'DEMO 1' command and check the
output" or "send it the 'BIST' command and it will completely test
itself." Yes, it has full BIST; that was the worst part of the
programming. Composing the HELP screens was a pain, too.




It seems so silly to time and time again insult programmers and programs
you can get the same 'expert' advice in any bar close before cloing time.

I don't like bad products, and there's tons of really bad software
around. As a science and a practice, software is immature and erratic.
Big software projects fail as often as they succeed, and seldom come
in on schedule or within budget.

Google "software project failure." If you find the results offensive,
don't blame me.


Yes, I know that, but same for hardware, from space missions where the first
3 multi-million rockets fail for reasons that a non-expert would think
are mistakes he or she would never make, like a booster
running into the next stage (next stage engine firing too late), leaky
fuel leads as now with the shuttle launch delay, faulty speedometers (pitot tubes) on
airplanes, tunnels made for underground trains that are too low for the trains..
the list is endless.
On the other side of the spectrum there are more good, say successful, hardware project, AND software projects.
those just do not get press.
You say : 'All my programs work', I can say the same, *** I also know the limitations
of mine, and neither get a big press.
Although I have designed projects that did get nation wide TV coverage, I have always
preferred the place behind the camera, never in front.
There is nothing that is forgotten so fast as 'news'.


I've probably written a million lines of code in my life.

John

Yes a million is a lot, these days it fits on one USB stick.
One cosmic ray and there goes your million.

I haven't lost a source file since I started my company. We have a
formal release procedure and over-the-top backups.

You seem to object to the tools I use to design products. That makes
no sense unless you can show that something is wrong with the products
themselves.

Show us some electronic products that you've designed.

John

.