personal note on ACS and NG-SPICE (fwd)
Al sent me this message and I would like to hear some commetns from the
ng-spice developers community.
Paolo
---------- Forwarded message ----------
Date: Sat, 9 Sep 2000 21:52:13 -0700
From: Al Davis <aldavis@ieee.org>
To: pnenzi@ieee.ing.uniroma1.it
Subject: personal note on ACS and NG-SPICE
I just thought I would drop you a note on where I stand on ACS and
NG-SPICE.
I consider myself to be a researcher, although I have had a hard time
convincing universities of that. I have been working in simulation
for almost 20 years, including for some of the EDA companies. Before
that I was an analog circuit designer. Some of my dealings with
corporate IP conflicts are very convincing about the importance of
publishing your own work under GPL, and what Richard Stallman really
means when he says "the GPL is about freedom".
My interest, and goal with ACS is to make the most advanced, highest
performance general purpose mixed signal simulator engine, and then
hope somebody else makes it into a complete product. I am
emphasizing the new methods, rather than re-implementation of the
traditional. I find the product oriented part of the work to be a
nuisance.
I have the impression that the primary goal of the NG-SPICE group is
to bring the features of the commercial Spice tools into free
software. In other words, to make a complete product.
I do not see a conflict here. I see some interesting opportunities.
By lurking on the NG-SPICE mail lists, and perusing the
archives, I noticed some trends.
One is the politics of Berkeley and licensing.
Another is the constant mention of "we are rewriting to make it GPL",
"we need to redo this", "I don't understand this", "we need to
redo the parser", .....
I have even seen comments like "we might be able to copy this part
from ACS".
There are good reasons why I did not base my research on Spice.
There are also good reasons for the move to C++. I only wish I had
done it sooner. The design of Spice is such that if you really want
to move ahead, it really will require a full rewrite. Spice-3 is as
far behind today as Spice-2 was when Tom Quarles undertook the job of
converting it to C.
This all makes me wonder why it is so important to base it on Spice,
especially considering that none of the group has any real
personal attachment to it.
With all this in mind, I would like to offer you ACS.
The biggest down side I see with this is that there are some in the
group that will not touch C++. This seems strange to me, because I
found C++ to be a big productivity improvement over C. Much more
than the improvement of C over Fortran. But, to each his own. If
the system is modular enough it doesn't matter what language the
pieces use, and C++ modules can be C inside.
So, what do you think of this?
al.
Partial thread listing: