Re: @above in analyses other than tran and dc sweep

Hi,

I haven't contributed to this mailing list before, but I have been 
quietly watching it. So, I'm ready to be shot down!!

Two observations:

1. The LRM section that Paul quotes, says:

"The cross() function will not generate events for
non-transient analyses, such as ac,
dc, or noise analyses of SPICE (see 4.6.1), but the above() function
can."

IMHO this is very badly written (ducks under desk!). "can" in colloquial 
English could mean "may" or "shall". I think the meaning here is "may", 
which really shouldn't be in a standard. If the authors meant "shall", 
they should have said so.



2. There is no reason why causality should be assumed in AC analysis. By 
which I mean that the frequency points could be evaluated in any order. 
The same is true of a DC sweep. It might be convenient (from an 
algorithmic point of view) to perform the sweep in an ascending or 
descending order, but I don't see why a sequence should be assumed. 
There is also a question about what might be meant by a crossing in an 
AC analysis. AC analysis is small signal. The bias point is fixed. What 
would we expect to cross zero?

Mark


Geoffrey.Coram wrote:
> Verilog-AMS doesn't provide a way to access ac quantities:
> frequency, the ac current/voltage, etc.  So, I'm not sure
> what expression you imagine could cross zero.
> 
> Some simulators do allow you to do a single-frequency ac
> analysis with a swept operating point; I would expect that
> to be treated like a dc sweep for the purposes of @above.
> 
> -Geoffrey
> 
> 
> Paul Floyd wrote:
>> Jonathan David wrote:
>>
>>> @above should be 1 or zero in any dc, op, ic, (static) simulation. 1 
>>> if above is true , 0 if not. The point of @above is that it must work 
>>> in these situations where @cross is undefined.
>>>  
>>>
>> I appreciate that. My question mainly concerns ac rather than ic. Should 
>> that be considered like dc sweep or not? ac has some similarities with 
>> dc sweep (a fixed number of static calculations), and I can imagine the 
>> expression crossing zero from below at some point.
>>
>> Regards
>> Paul Floyd
> 

-- 
===================================================================
Professor Mark Zwolinski
Electronic Systems & Devices Group       Tel. (+44) (0)23 8059 3528
Electronics & Computer Science           Fax. (+44) (0)23 8059 2901
University of Southampton                Email.  mz@ecs.soton.ac.uk
Southampton SO17 1BJ, UK             http://www.ecs.soton.ac.uk/~mz

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.