( SNUG 05 Item 11 ) ---------------------------------------------- [12/20/05]

Subject: Sequence Power Theatre vs. Synopsys Power Compiler

SEQUENCE WINS (SORT OF) -- If you're looking at the percentage of customers
using Synopsys vs. Sequence power optimization tools, Synopsys wins:

  "Do you use 'power tools' like Synopsys Power Compiler and PrimePower?"

            don't use power tools:  ######################## 61%

          Synopsys Power Compiler:  ################## 46%
              Synopsys PrimePower:  ########### 27%
           Sequence Power Theatre:  ##### 12%
           Cadence Get2chip Power:  ## 6%

But if you believe that "quality" beats "quantity", Sequence easily wins:

             "How do they fare vs. Sequence Power Theatre?"

            Synopsys fares better:  ############# 32%
             Synopsys fares worst:  ########################### 68%

Still, I was surprised to see that 46% of users used Power Compiler.  That's
an awful lot when you consider there are an estimated 10,000 to 16,000 DC
licenses being used worldwide.  (5K to 7K Power Compiler licenses roughly.)

         ----    ----    ----    ----    ----    ----   ----

  Sequence tools must be better.  We have not got any good results from
  the Synopsys power tools.

      - [ An Anon Engineer ]


  We occasionally use Synopsys Power Compiler.  Works well for most part
  since it is well integrated in our DC flow.

      - Sunil Malkani of Broadcom


  Power Compiler is adequate.  Power Theatre is far superior to PrimePower.

      - [ An Anon Engineer ]


  We are using Power Compiler and PrimePower.  We don't have plan to
  switch even though Sequence tool might be considered more advanced.

      - Wanhao Li of Zoran


  We used Power Compiler on our latest design.  It reduced the power by
  27% over a previous version of the same part - very effective and exceeded
  all predictions.  PrimePower took several sessions with the FAE/apps
  engineers to truly be able to understand and predict results but resulted
  in good correlation with actual silicon.

      - [ An Anon Engineer ]


  Power Compiler is a great tool.  Get about 40% reduction with clock
  gating, another 10-20% with power aware PhysOpt, sometimes up to 50%
  less leakage.  PrimePower is okay, too.

      - [ An Anon Engineer ]


  Power (uW)            DC          Get2chip     delta
  Leakage..............127 ........... 81 ....... 36%
  Dynamic..............178 ...........166 ...... 6.7%

  clk period 4.5 nsec, 130 nm, post-layout

      - [ An Anon Engineer ]


  Currently using Cadence Get2chip for low-power synthesis.

      - [ An Anon Engineer ]


  PrimePower and Power Compiler are fair tools, the results are based on
  the .lib thus garbage in->garbage out.

  Power Theatre might have improved since last time I looked at it but
  back then, too much $$$ for the capabilities it offered.

      - [ An Anon Engineer ]


  We use Power Compiler.  We never tried Power Theater.

      - Joe Xu of Huawei Technologies 


  Power Compiler and PrimePower are good.  Still like RTL analysis in
  Sequence Power Theatre better.

      - [ An Anon Engineer ]


  We are a primary user of Power Compiler and Get2chip power optimization
  features.  We haven't used Sequence Power Theatre so I can't comment.

      - [ An Anon Engineer ]


  Power Compiler is more useful than the near-useless Power Theatre.

      - [ An Anon Engineer ]


  Sequence Power Theatre is good.

      - [ An Anon Engineer ]


  We are using Power Theater for power estimation because of accuracy.
  We are using Power Compiler for low power implementation.

      - [ An Anon Engineer ]


  We use Power Compiler for gate-level optimization and analysis.
  RTL analysis in Power Compiler needs improvement.

  We use Sequence Power Theater for RTL-level analysis.

      - [ An Anon Engineer ]


  Sequence Power Theatre > Synopsys Power Compiler and PrimePower

      - Benjamin Chen of Socle Tech.


  We are presently comparing PrimePower with Power Theatre and power 
  results appear equivalent, with PrimePower being a little easier to
  use at the onset since the use model is similar to PT.  I believe
  we are leaning towards Power Theatre, however.

      - [ An Anon Engineer ]


  The Sequence tools are more accurate, but a pain in the #@! to setup
  and get running.  If you are a PT user you can be running w/ PrimePower
  very quickly.

      - [ An Anon Engineer ]


  We do use Power Compiler but mostly for reporting.

      - [ An Anon Engineer ]


  We use Power Compiler, don't know Power Theatre.

      - [ An Anon Engineer ]


  I'm eager to try out PrimePower on a tapeout; it's dynamic analysis
  capabilities in particular.

      - Neel Das of Tallika Corp.


  Interested, but can't afford.

      - [ An Anon Engineer ]


  I only use Power Compiler under the hood inside DC or PhysOpt.

      - [ An Anon Engineer ]


  We use Power Compiler for clock gating.  It mostly does the right thing,
  but could do more and also have better awareness of physical design.

      - [ An Anon Engineer ]


  Used Power Compiler once a long time ago.  Plan to use it again for our
  0.13 micron designs.

      - Tom Mannos of Sandia National Laboratories


  Yes we use Power Compiler.  Better than than Power Theatre because of
  ease to use and fit with our current DC.

      - [ An Anon Engineer ]


  Power Compiler yes.  PrimePower evaluating.  Traditional user of the
  Sequence analysis tools.  Besides performance, important aspects are
  correlation as well as price.  SNPS behind on these points.

      - [ An Anon Engineer ]


  Yes use Power Compiler.  Comparable to Sequence.

      - [ An Anon Engineer ]


  Package deal.  Didn't get a chance to evaluate Power Theatre.

      - [ An Anon Engineer ]


  We're starting to use Power Compiler and PrimePower to a lesser extent.
  Power Compiler seems to work OK, although getting Formality to agree with
  some of its optimisations (clock gating in particular) can be a real PITA.

  I don't have any experience with Sequence Power Theatre.

      - David Smith of STmicroelectronics


  Yes, I use Synopsys Power Compiler and PrimePower.
  They do their their job.  Can't compare to any other tool.

      - [ An Anon Engineer ]


  We use Power Compiler, although mostly for assessment, not to manipulate
  the clock domains. 

      - [ An Anon Engineer ]


  Power Compiler, PrimePower used a bit.  When we last looked at Power
  Theatre, I think I remember that its biggest selling point was the
  ability to assess and adjust for power at the RTL level, pre-synthesis.
  At the time, we weren't ready to pay for that capability.

      - [ An Anon Engineer ]


  Use Synopsys Power Compiler and PrimePower.  No eval was ever done.  So
  far we are happy with the results we are seeing.

      - [ An Anon Engineer ]


  Use Power Compiler, but for better results use Azuro Powercentric.

      - [ An Anon Engineer ]


  Can't compare, but PrimePower is a good tool IMO.  Runtimes is an issue
  though.

      - Are Arseth of Atmel


  Have used Power Compiler for dynamic and leakage optimization.  Have some
  good results.  I have not used Sequence tools though.

      - [ 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)