Hi,
First of all I start with Genesys ones of the advantages of this simulator that it's have very simple interface!
I am will give a few examples why not using Genesys (of course only if u have other option):
1) in currently job i work in Xband (up to 6GHz) my friend make simulation (linear and EM) for small signal the results are very close, but when i simulated the same schematic vs ADS only schematic simulation I got exactly results like we produce the board!
** This can of course be for many reasons -- it is not clear to me what you are comparing.
2) I am prefer ADS because in Genesys I am hate to deal vs port/grids problems during EM simulation.
** I know of few simulators which are completely immune to port problems
But for the last year, Genesys has had Momentum available to it (and sonnet before that as well). EMPOWER, the original EM simulation tool within the Genesys suite, was a limitation. I have seen very good results with it *IF* I kept things nicely on a grid. Still, it was significantly slower than sonnet or momentum..
Momentum makes this "problem" go away.
3) Mixer (from system lib) can't work properly with HB analysis.
this is just a few problems... I work vs Genesys a few years and yes I believe his is nice, but not more!!!
** I'm not sure what you mean by this. Are you saying that you want a general purpose mixer model that will work in a HB simulation? I'm not sure this is still a limitation. You could always write a verilog A model for one and use that in Genesys. I've designed a number of mixers in genesys with good results. mostly I find the models (from the vendors) being the limitation since they are typically characterized for linear/amplifier/low noise applications (for FETs). Again, I'm not sure what you're specifically referring to here. Probably you're talking about a spectrasys model?
In other hand again Genesys have nice Sysntesis to start with! Some times help to save time!
About AWR last time I work vs this simulation was during my study, truly I don't remember much, but from my friends i know that they are simulated vs AWR with very high freq.
** I'm sure AWR is a perfectly good tool; again, I think at the core most of these simulators are very capable of most everything that anyone wants to do. I do like to address any misunderstandings that I hear about though because I think one needs to present some real data on the relative agreement of circuits. I presented a webcast in January (sponsored by agilent (no, I didn't get paid a dime from agilent to do this)) -- and presented a few cases of measured vs. modeled circuits for up to 6 GHz (this was mainly due to the limitation in my network analyzer at the time; which now is no longer a limit). In any case, you can see some good and some bad cases of agreeement (bad cases = transistor model limitation in this case) --- but since I have so much good success with Genesys I just want to provide some balance to the discussion.
In the future I'll hopefully have time to publish more positive results....
All this only me opinion and maybe I am wrong, who nows?
*** I wouldn't say you're wrong, but again the issue is really about presenting some data to support the point. the agreement probably was poor in the case you explain. There was a competition a few years back in microwave engineering europe, on the topic of a mmWave mixer. the Eagleware result actually won the competition. Really could have been any of the entrants, but they won. These are just a couple of single data points -- so maybe one of these days there should be another "throwdown" (for those fans of the food network) where we try a few circuits in different simulators...
thanks for your answer,
Lance
Regards,
David