On Wed, Dec 3, 2008 at 6:49 PM, John Doty <[EMAIL PROTECTED]> wrote: > > On Dec 3, 2008, at 7:40 AM, Svenn Are Bjerkem wrote: > >> >> Ken Kundert knows what he is talking about > > I don't think so. He thinks mixed-signal design is like computer > programming. But it's not: to do it well you have to start from the > *physics* of what you're trying to accomplish.
I'll rephrase myself: "Ken Kundert *should* know what he is talking about." At one point, sooner or later, mixed-signal *is* about programming. You don't start with the circuit simulator when you're down at physics, unless you are simulating physics. Verilog-A or VHDL-A is very much about simulating models of physics. Where spice only offer you the opportunity to either compile a new bsim* model or tie existing sources and components together, -A tools give you the possibility to write the node and branch behaviour inside an element, and you can set the accuracy of your model to tune simulation time. Or maybe you had something else in mind when you say *physics*? To test your "concepts" you use Matlab and SimuLink (another expensive pack of tools) If you *know* everything up front, then just go directly to schematics and pcb. When Ken worked for Cadence he developed their Spectre simulator, at least he was a part of the development team. The first versions of spectre used spectreHDL for analog modelling, and that was later replaced by Verilog-A. If he is not able to *sell* his work that is maybe because he is an *engineer* and most engineers have a hard time selling. (Unless they work in sales) > > I wonder what tools Paul Horowitz uses these days. *There's* a guy > who understands mixed signal disign and can write clearly about it. Are you thinking of "Art of electronics"? or any other book by him? I don't think I have access to the papers he has published on IEEE. Have to pay to download pdf. > >> and he offers enough >> information on http://www.designers-guide.org/index.html for anybody >> interested in AMS to get up to date on the advantages of doing things >> the AMS way. > > Kundert's book is what Al got me to buy. It's a sales pitch, not a > textbook: the signal to noise ratio is very low. Someone who has a > problem to solve doesn't want a bunch of hype on the advantages of a > tool: they want a roadmap that will allow them to find a solution. I won't defend a person I don't know and a book I haven't read, but I would sadly expect that a lot of AMS books spend a lot of time promoting AMS as there is a lot of work to do convincing engineers to use AMS. As long as there are no open-source tools (yet) that give real engineers a real "teaser" and those tools that exist either cost a fortune or are crippled beyond practical use, evangelism will be a part of the game. I think some of the articles on the website gave me enough inspiration to try out the AMS features of said Spectre-AMS and found that Cadence had solved the problem rather poorly. The tutorials were good, but when we tried to implement AMS on our technology with our requirements, things got tricky. AMS is a lot like SystemVerilog in the sense that it is really difficult to see an advantage before it has been tried out on a live project. -- Svenn _______________________________________________ geda-dev mailing list geda-dev@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev
