( ESNUG 555 Item 5 ) -------------------------------------------- [01/22/16]

Subject: Sanjiv on MENT Veloce, Ansys PowerArtist, and FinFET power issues
             DAC'15 Troublemakers Panel in San Francisco, CA

   Cooley: Ok.  Sanjiv.  You're friends with Mentor.  You're tight with
           Mentor.  Yet when it came time to buy a RTL power tool or
           team up with an RTL power tool, Mentor Veloce teamed up with
           Ansys not you in RTL power estimation.  Why?

   Sanjiv: It's called being customer driven.  One, I think, the Veloce
           Ansys integration is an open one.  There can be multiple RTL
           power analysis tools becoming part of that, including ours.
           This first tool was driven by a particular customer who wanted
           a particular integration and so we've been talking to the
           Veloce team.  We're working with the Veloce team and they
           said that's the first integration and that's exactly what
           happened.

   Cooley: So it was a customer that was driving it?

   Sanjiv: One customer who has and who's using a PowerArtist and uses
           Veloce and wanted to see the two of them to work together.

   Cooley: Do you see...  OK, I'll be honest here.  What's the difference
           between these three tools?  I mean RTL power optimization
           pretty much is a wash.  You guys are pretty much getting the
           same numbers...

   Sanjiv: No, when you look at it, most people who are looking at RTL
           level power have been using power analysis tools.  And in power
           analysis, I think there is a big change coming with FinFET
           designs.  The kind of technologies that worked in the past,
           were given adequate power numbers at RTL level...

   Cooley: Wait a minute, wait a minute.  RTL level analysis for
           FinFET is different?

   Sanjiv: Yes it is.  If you would look at what's going on with FinFET
           the dynamic power becomes a much large portion of the power.
           And until FinFET, leakage power was almost as big as dynamic
           power.  And so if you were able to estimate the dynamic power
           within a certain level of accuracy, your overall accuracy
           would be still OK.

           But now when the leakage power comes down dramatically and the
           dynamic power is now a much bigger proportion, the accuracy of
           your RTL power analysis tools has to get much, much better.
 
           And so that's one area that they're going through a big change.
           The area that Calypto has been good at has been power
           optimization at the RTL level.  That's an area that we feel
           really good about technology.  We don't lose benchmarks doing
           the RTL level power optimization.

           But the true challenge for dynamic power, for FinFET design, is
           that because the dynamic power is becoming such a large portion
           of the overall power, you need to give designers (because today
           right now power experts are doing a lot of the optimization) a
           lot of the analysis.

           What you have to do is to give RTL designers the tools to be
           able to iterate and try different micro-architectures and
           different RTL configurations -- and be able to get their RTL
           power down.  And that's where we the market is going right now.

Join    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)