The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> Current Access by Cadence Ocean
https://designers-guide.org/forum/YaBB.pl?num=1451046840

Message started by cheap_salary on Dec 25th, 2015, 4:33am

Title: Current Access by Cadence Ocean
Post by cheap_salary on Dec 25th, 2015, 4:33am

(1) Cadence Spectre 11.1.0.312.isr5
Ocean function of IC5.10.41.500.6.151

(2) Cadence Spectre 14.1.0.826.isr13
Ocean function of IC5.10.41.500.6.151

I want to access currents for transient analysis with data format of sst2.
Assume save data is vdd:p.

I can access current by using i("vdd:p") or getData("vdd:p") in (1).

But I can not access current in (2).
On the other hand, I can access voltage in both (1) and (2).

Strangely, I can access current for result of "pss-fd" even in (2).

If I use Ocean in IC6.X, I can access current for result of "tran" using Cadence Spectre 14.1.0.826.isr13.

Is this know issue or bug for Ocean of IC5.X ?

Title: Re: Current Access by Cadence Ocean
Post by miles2018 on Nov 18th, 2018, 7:40pm

Finally, I find someone with the similar issue! I don't have the answer, but a clue for this.
Reading tran-tran by ADE Result Brower, i found the current saved differently between your (1) and (2).
For example, current of a vdc V1, saved as /V1/PLUS(a file PLUS in directory V1) in (1) and \v1:p(only one file) in (2). This difference causes caculator can not work with the data.
I'm use the IC5.x and MMSIM15, it's strange that my college using the same software have result like /V1/PLUS in (2), caculator works for them!
Did you solve the problem?

Title: Re: Current Access by Cadence Ocean
Post by Andrew Beckett on Dec 4th, 2018, 3:24am

I thought this might have been a difference caused by the use (or not) of the -sst2usecolon command line argument, but I think it appears to have been caused by a change in behaviour of the underlying SST2 libraries. I seem to find that if -sstusecolon (which corresponds to a setting on the ADE Setup->Environment form) then I can't read the currents in IC5141 OCEAN with either version, whereas without this option only the older spectre works.

The SST2 result format has been deemphasised now, as PSF XL is a better format for analog data. However, PSF XL isn't supported in IC5141. I think the challenge here is using an IC version where the last hotfix was in 2011, and you're using a simulator from 2014 - so you're in the land of combinations that are not being tested together and these kind of compatibility problems undoubtedly creep in. Unfortunately SST2 always had some challenges because they normally used . as the delimiter for both currents and hierarchy, and I think this is where this arises from.

Note that it doesn't occur in other analyses because even if you pick SST2 as the result format, it's only used for transient; analyses such as ac, noise, dc etc continue to use PSF format (SST2 can only store transient data, as it originates from the digital simulation tools). SST2 was introduced as an attempt to have better capacity for large signals, but that's really been superseded as I mentioned earlier by PSF XL.

I think the only workaround is to use psf (psfbin) if you're going to read the results in IC5141, or use OCEAN from (say) IC617 - in which case you should switch to PSF XL anyway...

Regards,

Andrew.

The Designer's Guide Community Forum » Powered by YaBB 2.2.2!
YaBB © 2000-2008. All Rights Reserved.