On Fri, 28 Dec 2012 15:27:12 -0600, John Fields
wrote:

On Thu, 27 Dec 2012 11:46:10 -0800, John Larkin
wrote:


Synchronous programming isn't a religion or a rule.

---
Synchronous programming???

What is that?
---

It's just a fact that even trivial async designs are wont to have bugs,
and serious async designs are so bug-prone that they are rarely done.

You have posted several async designs that had low-probability (ie, long-period
intermittent) bugs. You either do that to prove that async design is
safe (which you have proved the opposite) or because you don't
understand how to do reliable synchronous logic design.

---
On the contrary, I've posted innumerable times, in response to your
never-ending hostile challenges, that those designs were used to
illustrate concepts, and weren't intended to represent fully
fleshed-out circuits.

Errors?

We all make them, and most of us own up to them, yet you persist in
your dogged determination to ignore facts and to manufacture fantasies
which make your errors seem trivial - even humorous - and easily
glossed over, and everyone else's, Mortal Sins.
---

Safe synchronous logic design can be taught in about 15 minutes on a
whiteboard. I got straightened out on this when I was still a
teenager.

---
And, as a consequence, your champion vanquished your asynchronous
skills to the point where they've atrophied and are now useless.
---

When logic was expensive and problems were simple, like up to around
1970, some people still tended to do async design. DEC's PDP-8,
PDP-8/I, and PDP-11/20 were async computers, and they were ugly and
buggy, full of one-shots, RCs, and delay lines.

---
"Ugly" is unkind, since, at the time, they were miracles of ingenuity.

Being a Monday morning quarterback, you, of course, declare how it
should have been done then with what you know now.
---

A synchronous state machine can be documented and analyzed. Every
state and every state transition can be known and controlled, and
there are conventions for doing so.

---
Rather rigid from someone who claims to shoot from the hip, and then
there's metastability...
---

|Async systems really don't have
distinct states, and transitions happen all over space and time,
creating endless hazards.

---
Utter nonsense, and whoever gave you that 15 minute crash course on
"Synchronous Good, Asynchronous Bad" during your tender years seems to
have really done a number on your head.
---

No serious enterprise would hire someone if they knew that they
preferred ad-hoc hairballs to serious state-machine design.

---
Well, neither of us is employable, so it's not a question of bias.

Instead, it's a question of who's not afraid to use what's most
efficacious for the task at hand.

Since you - apparently - refuse to even consider asynchronous logic
acceptable for at least some tasks, I think that leaves you in the
lurch.

Quoting that great sage, Ron White (and appropriately applicable to
Larkin), "You can't cure STUPID... STUPID is FOREVER!" :-}

...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

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