OCEAN Parametric analysis

Hi Stefano,

By the looks of the error message, something has ended up as an integer, but you're using %f in the fprintf statement to print it. Not really sure quite where the data is coming from, but that's my guess. If you're going to use %f, you could case the value to a float by using float(L_opt) or float(eff_opt) in the arguments to fprintf.

Regards,

Andrew.

Hi Andrew,

Thanks for the reply. Since this parametric analysis is inside a while loop, I tried to simplify the problem studying a simpler script with only one parametric analysis. And surprise: it works! I tried also to put some printf before and inside the while cycle and the printf inside the while are not shown on the screen! The simulations are run in the while, but it seems that something of strange happens.

So, I was thinking to use a script in another languange to call the OCEAN script, implementing the while cycle with this new languange.

What do you think?

Thanks a lot,

Stefano

Hi Stefano,

I don't see why you should need to use a different language for the loop. It's almost certainly an error in the code of your loop, rather than a language bug.

Perhaps  you can send the entire script (send it to me as a personal message if you don't want to post it on the forum)?

Regards,

Andrew.

Thanks a lot, Andrew!

Stefano,

No problem!

In case anyone else is interested, the problem was caused by a parenthesis mismatch in the code.

Regards,

Andrew.

 Hi Andrew,

 I am experiencing the same sort of problem although I am not using any special OCEAN or SKILL script. I was trying to print out a signal to a text file to use it as a PWL source later on. As the number of significant digits used by the waveform calculator is limited to about 4 (despite my having changed this number under "Options->Significant Digits"), I decided to use ocnPrint through the CI.

 I try

ocnPrint(?output "myOut.txt" ?precision 10 ?numberNotation 'none clip( i("I0:3" ?result "tran-tran" ?resultsDir "pathname") 2n 10n) )

 Note that the expression is simply copied from the calculator buffer. I get the following error message:

 *Error* fprintf/sprintf: format spec. incompatible with data - "Format is '%g', argument #1 is 170"

I get the same problem if I try to print a voltage, and so it is not caused by the current not being saved with the correct port mapping or so. 

I even tried to run it by directly invoking OCEAN in the terminal, and selecting the results with openResults(), selectResults() etc., but to no avail. I don't get an error message there, just an empty output file. Of course, when I run OCEAN in the terminal, I leave out the ?result and ?resultsDir parts, since they have been already taken care of with openResults() and selectResults().

 I am working with Cadence IC 6.1.4.

Thanks,

Vivek

Hi Vivek,

If you still have this issue, please contact Cadence Customer Support and file a service request.

Best Regards,

Andrew.