Stats

  • Locked Locked
  • Replies 4
  • Subscribers 93
  • Views 13838
  • 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

Why there is a difference in Optimisation when we change clock period directly instead of changing the uncertainty?

NAADHAN

Hi ,

 In our project we have a single clock....Here we run mutiple placement runs by targetting higher frequencies. This can be done in two ways          1.By changing the uncertainty each time keeping the clock period constant.
                  2.By Changing the clock period each time keeping the uncertainty constant.

Here in the both the approach when we target for the same frequency(Eg: 1)Period:2ns & uncertainty:100ps 2)Period:1.9ns& uncertainty:0ps) ....we get entirely different results.....

 So can anyOne Pls explain me how does the tool undertands the  change in uncertainity and change in time period..????

Thanks & Regards,
Naadhan

  • The 1)Period:2ns & uncertainty:100ps 2)Period:1.9ns& uncertainty:0ps are quite different during "hold" timing check.

    When calculate "hold" timing: 

    For case (1).  the capture clock pin "see" the period 2.1ns (2.0 original period plus 0.1 uncertainty)

    For case (2). the capture clock pin "see" the period 1.9ns. (1.9 original period plus 0.0 uncertainty).

    So, your case (1) result may have much more buffer added to avoid hold violations.

     Gele

  • Thanks a lot Gele....But FYI we dont to any hold optimisation in our Design...We do only setup Optimisation ....So am still not sure why this difference is comming...thats why i doubt how the tool interprets the time period and uncertainty while optimising for timing(Setup)?!

    Regards,
    NAADHAN

  •  Based on my understanding of Cadence placement, I don't think tool generates different placement quality because of your different timing constraint setting. You should use this 2 sdc file based on 1 same placement database to report timing and it won't show you any different slack. Even if you provide the same timing constraint and run multi placement, Cadence placement will give you supurise.

    It's random for no reason. 

     You can try 10 runs with exact same floorplan same timing constraint (everything same) and do placement. 

    You will see the interesting thing. :)

  • Hi frankz....thats correct we have also observed that we cannot recreate the same placement and placement timing with similar constraints.... its both an advantage and a disadavantage too....;D

    Thanks and Regards,

    NAADHAN  

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.