( ESNUG 445 Item 4 ) -------------------------------------------- [05/24/05]

Subject: ( DAC 04 #24 ) A Boatload of Detailed FishTail User Benchmarks

> Here are the data from two designs that we have run FishTail Focus on:
>
>     Design #1
>                  Size: 2M gates    Clocks: 14
>           false paths: 9840          false path runtime:  3h 38m
>     multi-cycle paths: 3896    multi-cycle path runtime: 21h 58m
>
>     Design #2
>                  Size: 1.2M gates  Clocks: 30
>           false paths: 2661          false path runtime:  2h 54m
>     multi-cycle paths: 1554    multi-cycle path runtime: 65h 13m
>
> False paths and multi-cycle paths are the number found by Focus.  The
> runtimes are the wall time for Focus to run on a 64 bit Linux machine.
>
>     - David Vinke of LSI Logic


From: [ Mighty Mouse ]

Hi John,

We used Focus to generate timing exceptions from RTL for the critical
endpoints on a 200 K instance, 4 clock design.  Focus found 6,266 false
paths in our design in 3 hours, 57 minutes.  The exceptions were verified
using Solidify and then read into Design Compiler.  The netlist and SDC
generated by Design Compiler were then fed into PhysOpt.  QoR data was
compared for the same flow with and without Focus exceptions.

                             no Focus       with Focus
      WNS                    -0.82 ns         -0.03 ns
      # Violations             682               19
      Instances            193,147          192,123
      Route length      11,841,259       10,634,302

Please handle me as anonymous.  Hope this helps.

    - [ Mighty Mouse ]

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

From: Vikram Somaiya <somaiya464=user company=oki spot gone>

Hi, John,

We did an eval of Fishtail Focus.  In it, Focus was used to generate timing
exceptions from RTL.  These exceptions were verified using a property
checking tool.  The exceptions were read into Design Compiler.  QoR (WNS,
total number of violations and instance count) was compared after logic
synthesis with and without Focus exceptions.

Design A: 10 K instances, 2 clocks

   Exception Generation:

          5 false paths generated in 2 hours
          114 multi-cycle paths generated in 6 hours 26 minutes

   Synthesis Impact:
                        Without Focus        With Focus
   WNS                     -0.57 ns            -0.26 ns
   # of Violations           688                 272
   Instances              10,796              10,613


Design B: 20 K instances, 2 clocks

   Exception Generation:

          39 false paths generated in 3 minutes
          29 multi-cycle paths generated in 7 hours 22 minutes

   Exception Verification:

          False paths verified: 39/39
          Multi-cycle paths verified: 29/29

   Synthesis Impact:
                        Without Focus        With Focus
   WNS                     -2.75 ns            +0.02 ns
   # of Violations           729                   0
   Instances              19,771              17,265


Design C: 50 K instances, 1 clock

   Exception Generation:

          1,059 false paths generated in 2 hours 21 minutes
          93 multi-cycle paths generated in 2 hours 48 minutes

   Exception Verification:

          False paths verified: 1059/1059
          Multi-cycle paths verified: 93/93

   Synthesis Impact:
                        Without Focus        With Focus
   WNS                     -0.22 ns                0 ns
   # of Violations            64                   0
   Instances              49,966              50,532


Design D: 120 K instances, 4 clocks

   Exception Generation:

          0 false paths generated in 5 minutes
          527 multi-cycle paths generated in 12 hours

   Exception Verification:

          Multi-cycle paths (MCPs) verified: 95/527
          (the remaining 432 timed out)

   Synthesis Impact (taking only the 95 verified MCPs):

                        Without Focus        With Focus
   WNS                     -0.91 ns            -0.78 ns
   # of Violations           469                 447
   Instances             118,142             118,247


All run time data for these 4 benchmarks are from a Linux platform.

    - Vikram Somaiya
      Oki Semiconductor                          Sunnyvale, CA

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

From: Danny Traynor <dtray=user company=jennic spot gone>

Hi John,

Enclosed is the 15 page eval we did of the Fishtail Focus tool when we first
used it last year.  Your readers might be interested in it.  The eval has
all of our benchmark data plus the scripts we used.  Cheers.

    - Danny Traynor
      Jennic Ltd.                                Sheffield, UK


  Editor's Note: This eval is #58 in the DeepChip.com Downloads.   - John

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)