Congrats BrownOut! sounds like a good opportunity.
As far as the "simulator as a search-space" discussion... I am a grad student in analog design. Ive taken a 3 classes in IC design, even designed a DAC that passed DRC and LVS checks and is being fabbed on AMI 0.5u... using the simulator as a "search" for the right design happened at a few points in the circuit. when? well, when the stinking square law equation for mosfets didn't accurately predict circuit behavior (what a surprise). I often would get results from hand calculation that would only be within +/- 50% of the simulated result, because the equations dont work very well when you get devices that are "on the edge." Of course, I'm sure I failed to account for certain things in the calculations that I would have gotten if I'd had more experience and more knowledge, but the fact is, systems are infinitely more complex these days. The best you can do is include as many second order effects as you can in maybe a 2-4 transistor circuit. after 4 devices, incorporating secondary effects often leads to solutions that aren't closed form, requiring numerical solution... but at that point you might as well just ditch your hand-equations based on a faulty equation and hop on the simulator, since its a numerical solver.
The most significant advances in the design was when something "clicked" in my head (i.e., i finally figured out exactly how and why it worked), because then I could actually change parameters and add circuitry in ways that made sense... after that, the simulator just allowed for optimization. I think the solution is somewhere between the simulator and hand calculation - but hand calculation is even inferior to just gut instinct. Once you get an intrinsic understanding of the way devices work in a given process it starts to come much faster, and hand calculations only start to be good for "order of magnitude" starting points.
EDIT: again sorry to derail the post, we could separate these posts out to another thread and continue the discussion there, because it is an interesting one. I am personally interested in the pedagogy of engineering...
As far as the "simulator as a search-space" discussion... I am a grad student in analog design. Ive taken a 3 classes in IC design, even designed a DAC that passed DRC and LVS checks and is being fabbed on AMI 0.5u... using the simulator as a "search" for the right design happened at a few points in the circuit. when? well, when the stinking square law equation for mosfets didn't accurately predict circuit behavior (what a surprise). I often would get results from hand calculation that would only be within +/- 50% of the simulated result, because the equations dont work very well when you get devices that are "on the edge." Of course, I'm sure I failed to account for certain things in the calculations that I would have gotten if I'd had more experience and more knowledge, but the fact is, systems are infinitely more complex these days. The best you can do is include as many second order effects as you can in maybe a 2-4 transistor circuit. after 4 devices, incorporating secondary effects often leads to solutions that aren't closed form, requiring numerical solution... but at that point you might as well just ditch your hand-equations based on a faulty equation and hop on the simulator, since its a numerical solver.
The most significant advances in the design was when something "clicked" in my head (i.e., i finally figured out exactly how and why it worked), because then I could actually change parameters and add circuitry in ways that made sense... after that, the simulator just allowed for optimization. I think the solution is somewhere between the simulator and hand calculation - but hand calculation is even inferior to just gut instinct. Once you get an intrinsic understanding of the way devices work in a given process it starts to come much faster, and hand calculations only start to be good for "order of magnitude" starting points.
EDIT: again sorry to derail the post, we could separate these posts out to another thread and continue the discussion there, because it is an interesting one. I am personally interested in the pedagogy of engineering...
Last edited:
