Designer’s Guide Community

jimmy

Community Member

Offline

Posts: 32

error when use "Ocnprint"
Nov 25^th, 2004, 5:03pm

Hi,

I always use Ocnprint to save data in cadence. Sometime I encountered
such a problem. Let me describe it using an example.

If I want to run a transident analysis for 100us. During the long
simulation, I like to use direct plot to view the on-going results.
The problem is: if I viewed the data at 20us. Then if I want to use
direct plot to view the data again later, I can not see data after
20us. If I use Ocnprint to save data, I can only save valid data up to
20us. However, i can still use result browser to view updated data.

I have no idea what is wrong?

Thank you for your help,

jimmy

Back to top

IP Logged

Andrew Beckett

Senior Fellow

Offline

Life, don't talk to
me about Life...

Posts: 1742
Bracknell, UK

Re: error when use "Ocnprint"
Reply #1 - Nov 25^th, 2004, 9:24pm

Are you saying that you're using ocnPrint() during a running
simulation? If so, that won't work.

Some parts of ADE will work by doing this (for example, the plot outputs button), because they tell spectre to pause, flush its outputs, read the results, and then tell the simulator to continue. Other methods (like ocnPrint) will just read the data at that time - and will then probably not realise that it has changed when you do ocnPrint again.

Without more details of precisely what you're doing, it's hard to say...

Andrew.

Back to top

IP Logged

jimmy

Community Member

Offline

Posts: 32

Re: error when use "Ocnprint"
Reply #2 - Nov 25^th, 2004, 10:19pm

Andrew:

Actually this thing randomly appeared, I don't know what is the reason.

Somtimes I have no problem to use Ocnprint several times during simulation, while sometime it doesnt' work, I mean it can't continue to update the results.

Simiar thing happened to the waveform window. Sometimes I can't update the waveform anymore. For example. the transient waveform is shown only to 20us while the simulation has been runing to 100us. And I found these two things are correlated. Whne one of them happened, the other happened too.

The reason I use this command several times during simulation is to post-process the data to see if the result is what I want. If it is not good, I can stop the simulation.

thanks,
jimmy

Back to top

IP Logged