Home > Community > Forums > Digital Implementation > Different OpCond for different libraries

Email

* Required Fields

Recipients email * (separate multiple addresses with commas)

Your name *

Your email *

Message *

Contact Us

* Required Fields
First Name *

Last Name *

Email *

Company / Institution *

Comments: *

 Different OpCond for different libraries 

Last post Fri, Jun 1 2007 7:52 AM by archive. 3 replies.
Started by archive 01 Jun 2007 07:52 AM. Topic has 3 replies and 1559 views
Page 1 of 1 (4 items)
Sort Posts:
  • Fri, Jun 1 2007 7:52 AM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,950
    Different OpCond for different libraries Reply

    Hello,

    I am using two different libraries for core and IO cells and the process variation range is different for these libraries. I think the way to avoid being pessimistic is to set different operating conditions for these libraries. Is it right?
    Anyone can tell me how to set different operating conditions for different libraries in Encounter (6.2)?

    Thanks,
    Cristiano.

    PS.: The command reference document (fetxtcmdref.pdf) is a little bit confuse explaining the -forceLibrary option for setOpCond command. It says "Do not use this parameter with the -min, -max, or -powerDomain parameters." but uses these parameters in its example. (?)


    Originally posted in cdnusers.org by clsantos
    • Post Points: 0
  • Fri, Jun 1 2007 9:59 AM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,950
    RE: Different OpCond for different libraries Reply

    Dear clsantos:

    I think your point is how to define Operating Corners when facing multi 'dot lib', for the bc-wc mode, you have to set 1 OC for the setup check, it means that you have to choose a OC in certain lib, although different libs may have different OC ( at the same corner), typically, all the libs (IO, Std Cell........) will have the same OC at 1 corner, like P 1.0 V 1.2 T 0. Hold analysis is the same. But at the OCV mode, things are different, you need to define derating factors, 0.9 (for example) for setup analysis, and 1.1 for hold analysis. These derating factors are offered by YOUR FOUNDRY, like TSMC, UMC or .........

    Hope this can help you! For Furthere reference, I think you'd better try seeing Prime Time user guide.


    Originally posted in cdnusers.org by eminemshow
    • Post Points: 0
  • Fri, Jun 1 2007 10:15 AM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,950
    RE: Different OpCond for different libraries Reply

    The case is that I am 2 dotlib files, one for I/O library that uses a process (P) range equal:
    0.84 (best), 1.0(typ) and 1.33 (worst)
    and other for core library that uses a process range equal:
    0.65 (best), 1.0(typ) and 1.6 (worst).

    Therefore, I want to set one PVT point to each library but I don't know how to do it using Encounter. The temperature and voltage points are the same for both libraries.

    Can anyone help me?


    Originally posted in cdnusers.org by clsantos
    • Post Points: 0
  • Sun, Jun 3 2007 8:06 PM

    • archive
    • Top 75 Contributor
    • Joined on Fri, Jul 4 2008
    • Posts 88
    • Points 4,950
    RE: Different OpCond for different libraries Reply

    Hello Friend ,

    How about using option "operating_conditions " and give an unique_names for IO and Core in .lib ,
    and load by "setOpCond"

    Ex. :
    operating_conditions ("CDNUSER_BES_CORE") {
    process : 0.543000 ;
    temperature : -55.000000 ;
    voltage : 1.650000 ;
    tree_type : "best_case_tree" ;
    }

    # on Soce procedure

    print "************** SET OPERATING CONDITION"
    setOpCond -max CDNUSER_WOS_CORE -min CDNUSER_BES_CORE

    print "******** SET DERATING FACTOR FOR SETUP"
    setTimingDerate -late 1.12 -max
    setTimingDerate -early 1.12 -max
    setTimingDerate -late 1.12 -min
    setTimingDerate -early 1.12 -min

    #
    setAnalysisMode -setup -async -bcWc -crpr


    Originally posted in cdnusers.org by mohanch007
    • Post Points: 0
Page 1 of 1 (4 items)
Sort Posts:
Started by archive at 01 Jun 2007 07:52 AM. Topic has 3 replies.