Stats

  • Locked Locked
  • Replies 14
  • Subscribers 126
  • Views 16977
  • Members are here 0
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Strange numbers in Virtuoso 6.1.5

JulianJ

 Hello,

I am seeing some strange number errors in Virtuoso 6.1.5.

In the ADE, I type  "1p" as the value for a variable and when set it appears as "1e122", "100f" becomes "1e155", "1f" becomes "1e155".

A similar error is also appearing on the drain diffusion area calculated on the FETs I am placing in my schematics, with the calculated area being displayed and netlisted as "480e155", when I think it should be "480f".

 I tried upgrading my instalation from  IC6.1.5-64b.500.1 to IC6.1.5-64b.500.9, but still have the same problem.

 Any help appreciated.

Julian

  •  Thanks Andrew,

    Running 32bit mode fixes the problem for all my tests. I am happy now.

     Julian

  • Hi, Is there any news on this? I suffer from the same problem. However, it just randomly appears. I use a supported platform (CentOS 6, 64 bit). If I enter aelSuffixNotation(1e-12) I also get "1e122" for example, but as said this just happens from out of nowhere. Is there any disadvantage running virtuoso in 32 bit? (for example, a famous 2GB memory limit, slower simulations, ...). If yes, then that's not really a fix ... Thanks!

  • RHEL6 was not supported until IC615 ISR12, precisely because of this problem. That was when it was fixed so that 64-bit worked properly on RHEL6 (which is essentially what the CentOS version you're using is based upon). It was down to a change in the strcpy() implementation in the OS which had been the same on all other implentations for 30 years, but for some reason got changed in this OS version...

    Anyway, more details can be found in this solution on Cadence Online Support: Strange values such as 1e122 and 480e155 showing up in ADE and schematic

    The alternative is running in 32-bit mode - you will be limited to memory sizes of something like 3.7Gbytes (a bit less than 2**32). It will not slow the simulation down (64 bit is not faster than 32 bit in general; actually it can be the opposite because the increased memory usage of 64 bit can slow things down slightly).

    Best thing is to use a newer version though - this has been fixed for a good couple of years or so now. It was fixed fairly shortly after my responses to the thread above.

    Regards,

    Andrew.

  • Thanks, Andrew!

    Alright, I changed to IC 6.16.080 (which is the next-highest version we have) and it indeed seems to works!

    I hope there won't be issues with the PDK or other things.

    OT: (I am cad novice) Are PDKs usually expected to run only with one particular version or more one version or above? (PDK just states requirements but it is not clear if the version need to be exact or not).

    Furthermore, how stringent need the other tools to fit IC? I used the closest match to what the PDK stated in the requirements (MMSIM 12.11.134, Calibre 2013.2, IUS 8.20.006). When I change now to IC 6.16.080, do I need to switch these tools as well?

Community Guidelines

The Cadence Design Communities support Cadence users and technologists interacting to exchange ideas, news, technical information, and best practices to solve problems and get the most from Cadence technology. The community is open to everyone, and to provide the most value, we require participants to follow our Community Guidelines that facilitate a quality exchange of ideas and information. By accessing, contributing, using or downloading any materials from the site, you agree to be bound by the full Community Guidelines.