( DVcon 05 Item 4 ) --------------------------------------------- [ 10/25/05 ]

Subject: System Verilog

SYNOPSYS MESSES UP -- Last year users were waiting for Cadence and Mentor to
support System Verilog in their tools.  Oops!  Everyone *assumed* Synopsys
would have System Verilog fully implemented in the SNPS flow -- wrong!  This
year Synopsys *should* deliver on System Verilog, but until then it's still
only 1 in 5 engineers using/thinking System Verilog.  Oops.

 2004 - "Do you see your project using System Verilog in the next 6 months?"

          not in the
         next 6 months :  ####################################### 79%

           yes we will :  ########### 21%

 2005 - "Do you see your project using System Verilog in the next 6 months?"

          not in the
         next 6 months :  ######################################## 81%

           yes we will :  ########## 19%

Asking only those people who are or will be using System Verilog:

        2004 - "Do you plan on using the System Verilog
            design or the verification extentions, or both?"

          verification :  ######################### 63%
                design :  ## 6%
                  both :  ############# 32%

        2005 - "Do you plan on using the System Verilog
            design or the verification extentions, or both?"

          verification :  ############################ 70%
                design :  ## 5%
                  both :  ########## 26%

And asking those people in 2005 who are or will be using System Verilog:

           2005 - "Whose System Verilog tools are you using?"

          Synopsys VCS :  ################################ 79%
       Mentor ModelSim :  ###### 15%
               Cadence :  ## 6%

That 79% says it all.  Officially System Verilog is an open standard, but
with such dominance in that space, it's obvious why a lot of people use
the phrase "Synopsys System Verilog".

(Interestingly, despite its slow current adoption, if you look in the
"the Future of Specman 'e' and Vera" section of this census, you'll find
the vast majority of users think that System Verilog will replace Specman
and Vera within 5 years.  Check it out.)


  Yes, mainly for design purposes, but you do still have to be careful
  because there are some tools that have issues dealing with System
  Verilog.  Synopsys tools support it just about everywhere for the
  design extensions, and are expected to support all the verification
  extensions soon.

      - K.C. Buckenmaier of Hifn, Inc.


  Yes to some extend.  We will definitely start with VCS NTB soon within
  6 months if System Verilog is matured, we will add System Verilog test
  bench constructs since they are interoperable with Vera in VCS.

      - Azeez Chollampat of PLX Technology


  Yes.  We will use System Verilog from Synopsys.

      - Henry So of Mobilygen


  No.

      - Dave Ferris of Tundra Semiconductor


  No, we don't have plan to use System Verilog.

      - Pine Yan of ESS Technology, Inc.


  System Verilog - no.

      - Elchanan Rappaport of Lynx Photonic Networks


  No.

      - Jiye Zhao, Chinese Academy of Sciences


  We are using System Verilog for our current project.  Use is limited
  to design extensions except for assertions.  We use Synopsys VCS for
  our System Verilog designs.

      - [ An Anon Engineer ]


  No.

      - Frank Vorstenbosch of Telecom Modus Ltd.


  No.  We are far too risk averse to move to System Verilog in the
  near future.

      - [ An Anon Engineer ]


  Yes.  Mainly for verification, BFM design etc, not design.  Hey, it's
  included for free in VCS.  Can't hurt to use it.

      - Rajen Ramchandani of Mindspeed Technologies


  Yes.  We'll be using System Verilog for both design and verification.
  Currently we're using VHDL for design and verification.  We've been
  using Mentor ModelSim and will continue to use it.  The ModelSim 6.1
  beta is just out, with better SV 3.0 support, but it's not completely
  there just yet.  ModelSim comes with System Verilog -- no additional
  cost there.

      - Tom Moxon of Moxon Design


  Not within 6 months.  Probably within 12 months.  Want to use for
  verification.  Need object oriented features for maintainability.
  Probably would stick with Cadence System Verilog for business reasons.

      - [ An Anon Engineer ]


  No

      - Peng Hong of Datang Microelectronics China


  No.

      - Pascal Gouedo of STmicroelectronics


  We don't use System Verilog currently.  I'm intending on having a play
  with it when I get time.  We're mostly a VHDL house, so System Verilog
  isn't such a natural progression.

      - [ An Anon Engineer ]


  Yes to System Verilog.  We're a little concerned about full tool
  support (code coverage tools appear to be a bit behind the curve) but
  if that is resolved we'll certainly be using the design constructs in
  our next project.

  I very much doubt that the verification extensions will be fully
  functional (RELIABLE) within the next 6 months.  We're planning on
  using VCS, and Novas Debussy (if it works with System Verilog) for
  verification & debug.  Synopsys DC for synthesis.

      - Jonathan Craft of McData Corp.


  Yes.  We're already moving to System Verilog incrementally on both
  design and verification fronts (eg. use of the new "bit" type and also
  SV assertions).  That's for real use, not experimental.  We're
  currently using Synopsys VCS, but will switch to Mentor ModelSim
  shortly.

      - [ An Anon Engineer ]


  Yes System Verilog, for verification.

      - Stefano Traferro of STmicroelectronics


  No.

      - Menno Lindwer of Philips Semiconductors


  There's efforts to get started on System Verilog.  Since it is bundled
  with the VCS package we have.  But the bottleneck is getting engineers
  to write the new System Verilog testbenches.

      - [ An Anon Engineer ]


  No, No.

      - Nuno Franca of Chipidea


  No, we can do most of what we need to do with Verilog-2001.  We're
  reluctant to add System Verilog constructs because they would likely
  break some of our legacy tools.

      - Tom Mannos of Sandia National Labs


  No

      - [ An Anon Engineer ]


  No

      - William Mills of Northrop Grumman Corp.


  Yes.  We will move to System Verilog as soon as it supports all the
  Vera constructs we use - for performance gains.  We are seeing pretty
  good performance from Vera now though.

      - Dan Joyce of Hewlett-Packard


  Yes.  System Verilog verif extension.

      - [ An Anon Engineer ]


  No

      - Ross Smith of NuCore Technology, Inc.


  Yes, System Verilog on both design and verification.  (The only thing
  stopping us right now is EDA support for the verification aspects.)

      - [ An Anon Engineer ]


  No

      - Samuel Irlapati of Unisys


  No System Verilog.

  We are evaluating the System Verilog design Constructs which are very
  promising, probably will be used within a year dependent on tooling
  support from EDA, FPGA & ASIC vendors.  We are less interested in the
  System Verilog Assertions and have limited interest in verification
  aspects of the  language.

      - [ An Anon Engineer ]


  No.

      - Jean-Paul van Itegem of Philips Semiconductors


  No, with the exception of System Verilog Assertions sprinkled in
  the code.

      - Christian Mautner of Integrated Device Technologies


  No System Verilog in the next 6 months.

      - [ An Anon Engineer ]


  No.  Just experimenting.

      - Kea Hunt of Nazomi Communications


  No.

      - Lyle Kraft of Agilent


  Not in the next 6 months, but soon after that.

      - Dan Steinberg of Integrated Device Technology


  No, not within the next 6 months.  We plan to use System Verilog in the
  longer term, for both verification and for design (one language etc).

      - [ An Anon Engineer ]


  No

      - Jan Johnson of Rockwell Collins, Inc.


  No.  We plan to evaluate System Verilog within 6 months.

      - [ An Anon Engineer ]


  Yes, we use System Verilog on a very limited basis now.   We use
  Synopsys.

      - Tom Ebzery of Hifn, Inc.


  Yes maybe System Verilog.  Not using it yet, but would like to use it for
  the verification extensions.

      - George Gorman of LSI Logic


  No

      - Michael Roeder of National Semiconductor GmbH


  Yes - verification extentions.  VCS-MX (if Synopsys implements in time)

      - [ An Anon Engineer ]


  No plans of using System Verilog.

      - Juan Carlos Diaz of Agere Systems


  No

      - Greg Tumbush of Starkey Labs.


  Not in the next six months, but we're headed that way.

      - [ An Anon Engineer ]


  No.

      - Olivier Haller of STMicroelectronics


  Not using System Verilog currently for design primarily because of
  poor Formality support.  However, any new projects will probably
  use System Verilog for design (structures, interfaces, etc.)

      - Matt Weber of Silicon Logic Engineering, LLP


  Yes, Cadence System Verilog.

      - [ An Anon Engineer ]


  No.  Designs are too small to justify System Verilog for validation.

      - Mike Bly of World Wide Packets


  Yes, we are very likely to use System Verilog in the next 6 months,
  particularly the verification extensions.  We'd be using it now except
  it's not quite stable yet.

      - Kevin Jones of Rambus


  YES.  We do some System Verilog 3.1a now, and expect extensive use
  in next 3 months.

      - Eric Holmberg of Mahi Networks, Inc.


  No.  Half the people on our verification team have very strong OO SW
  backgrounds, and retch at the thought of using System Verilog for their
  primary development language, plus having no ancillary SW development
  tools or libraries to leverage.

      - [ An Anon Engineer ]


  I seriously doubt we will ever use System Verilog.  We use mostly
  VHDL for all of our source code and testbench environments, with
  good success.

      - Tony Lanier of Harris Corp.


  No.

      - Mark Lancaster of Freescale Semiconductor


  No

      - John Gray of Atmel


  No - not using, no - not going to use

      - [ An Anon Engineer ]


  I love the System Verilog additions and am looking forward to using
  them on a future product.  Given all the troubles that Verilog 1995
  has suffered through the years, a compliance suite for System
  Verilog would certainly accelerate adoption.  Until I can depend on
  System Verilog, I'll be sticking with the old reliable stuff.

      - Niels Reimer of Agilent


  No

      - Marshall Johnson of Movaz Networks

  No

      - Ba Nguyen of SonoSite, Inc.


  No. we are evaluating TB features and benchmarking against Specman "e".

      - [ An Anon Engineer ]


  No.

      - Michiel Vandenbroek of China Core Technology Ltd.


  No, I see no future in System Verilog.

  The C/C++ basis for SystemC seems the obvious way to go for a richer
  modelling language, especially when early architectural development
  has not yet made the distinction between a hardware implementation
  or a software implementation (or even the selection of a CPU/DSP).

      - Bob Warren of STmicroelectronics


  No

      - Karthik Kandasamy of Wipro

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)