( DAC 04 Item 23 ) --------------------------------------------- [ 02/09/05 ]

Subject: PrimeTime, PrimeTime-SI, Cadence CeltIC

THE ETERNAL STRUGGLE -- It looks like (for now) in the Cadence CeltIC vs.
Synopsys PrimeTime-SI battle, Synopsys has the upper hand:

      Dataquest FY 2003 Signal Integrity Analysis Market (in $ Millions)

     Synopsys PrimeTime-SI  ##################### $20.9 (46%)
            Cadence CeltIC  ################# $17.3 (38%)
                  Sequence  #### $4.1 (9%)
               Quantic EMC  ### $3.2 (7%)

But then again, don't put too much into this.  No one is really ahead here
until they've had at least 3 years of being in a solid lead.  There's just
an 8% difference between 46% to 38% -- this could easily reverse itself
next year given two or three big business deals.


    Yes, I am a hands-on user of CeltIC.

    Strengths:

    1. Accuracy to SPICE sims.
    2. Decent run times.
    3. No capacity issues with the 64-bit executable.
    4. Simple environment.
    5. Fits into our noise fix flow.
    6. Have not had a chip come back with any noise issues.

    Weaknesses:

    1. When CeltIC crashes it does a poor job of reporting why.
    2. Difficult for casual users to decipher the results.
    3. Difficult for casual users to figure out what the appropriate
       noise fix is.
    4. The html reporting could be improved tremendously.
    5. Need multiple licenses to use other features - I think they are
       going to change this.
    6. No local expert AE support.

    PrimeTime-SI: We ran alot of correlation experiments 3 years ago and
    CeltIC came out more accurate.  Synopys has made alot of headway since.
    But we do not have the resources to evaluate PrimeTime-SI again.

    Also, since CeltIC does a SPICE-like sim, I trust the glitch analysis
    from CeltIC much more.

    Magma Blast Noise: We do not use this feature although we use Magma.

        - Chad Narkchareon of Broadcom


    We are using Cadence CeltIC as the SI sign-off tool.  The results from
    the newer versions of the tool are more accurate with a significant
    reduction in runtime.  Also, we have seen that the combo of SI
    prevention TD routing with Nanoroute and CeltIC is a great 1-2 punch
    for SI closure.

        - Mani Gopalakrishnan of Fujitsu


    We use PrimeTime in onchip variation mode plus worst case incremental
    SDF generated by Cadence CeltIC.  This has so far seemed good enough
    to avoid SI issues.

        - [ An Anon Engineer ]


    I was a hands-on CeltIC user before, but it was almost 2 years ago.

        - Weikai Sun of Volterra


    PrimeTime is the gold standard for this. The biggest drawback of PT is
    that Synopsys analysis and implementation tools aren't completely
    integrated.  Because PT is more sophisticated, it may report problems
    that DC, PhysOpt, or Astro don't see.

        - John Busco of Nvidia


    Here are my impressions on CeltIC:

    - Noise library characterization is part CeltIC, unlike the version of
      PrimeTime-SI we'd used previously.  (May have changed now.)  So as
      long as you have SPICE and Verilog model, CeltIC generates its own
      cdb database.

    - For use with Primetime-SI, we need SPDM models (or derated K factors),
      which library vendors are still trying to refine.  Most lib vendors
      tend to prefer the 'traditonal' Liberty format.

    - The noise is given is by the amplitude of the voltage swing across
      the gate as well as its gain.   When | gain | > 1, there is tendency
      for signal noise to propagate.

    - Writes out repair file for both Cadence as well as Synopsys tools,
      which is nice.

    Generally CeltIC is not as pessimistic as PrimeTime-SI (again, may have
    changed with newer versions of PrimeTime-SI.)

        - Neel Das of Tallika Corp.


    The strength of CeltIC

    1. Accuracy is better.
    2. It is a cell-level tool but seems to have a transitor engine
       underneath, so it claims to cover some transistor behaviors we
       worry about - pass transistors, charge-sharing, etc.

    The weakness of CeltIC

    1. Not too easy to use due to its many switches.  It also needs
       well-characterized cell/pad/memory models.
    2. We could not use its noise-avoidance feature, mainly due to
       have to go in-and-out among tools.

    We currently use CeltIC for sign-off on coupling peak noise,
    sensitivity, clock slope, clock jitter, and also SDF-backannotate
    to PrimeTime.

        - Chiping Ju of Broadcom


    CeltIC is more accurate than PrimeTime-SI.

        - [ An Anon Engineer ]

Index   
Next->Item







   
 Sign up for the DeepChip newsletter.
Email
 Read what EDA tool users really think.


Feedback About Wiretaps ESNUGs SIGN UP! Downloads Trip Reports Advertise

"Relax. This is a discussion. Anything said here is just one engineer's opinion. Email in your dissenting letter and it'll be published, too."
This Web Site Is Modified Every 2-3 Days
Copyright 1991-2024 John Cooley.  All Rights Reserved.
| Contact John Cooley | Webmaster | Legal | Feedback Form |

   !!!     "It's not a BUG,
  /o o\  /  it's a FEATURE!"
 (  >  )
  \ - / 
  _] [_     (jcooley 1991)