Transistor count: Metric is often GE

Article: Transistor Count: A Flawed Metric
By: David Kanter (dkanter.delete@this.realworldtech.com), January 4, 2021 8:48 am
Room: Moderated Discussions
Chris L (not.delete@this.real.address) on January 3, 2021 8:39 pm wrote:
> David Kanter (dkanter.delete@this.realworldtech.com) on May 18, 2020 6:04 am wrote:
> > Many vendors brag about the complexity of their design, as measured by transistor count.
> > ... and there is no standard way of counting.
>
> Actually, the industry tends to use Gate Equivalent for the metric, e.g. when
> purchasing some IP in RTL form, the post-synthesis will be quoted in GE.
>
> That said, your other remarks about the overall design and architecture are spot on.

Thanks for sharing!

How is gate equivalent calculated in these tools? Again, decap, dummy devices, etc. seem like they would throw off the count.

David
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
New article: Transistor count: A Flawed MetricDavid Kanter2020/05/18 06:04 AM
  Non active transistorsDoug S2020/05/18 01:26 PM
    Non active transistorsRicardo B2020/05/18 08:12 PM
  Minor quibble about fixed-performance ASICPaul A. Clayton2020/05/19 02:59 PM
    Minor quibble about fixed-performance ASICDavid Kanter2020/05/21 05:58 AM
  A complementary article about xtor densityPaul A. Clayton2020/06/02 06:07 AM
    Low leakage transistorsDavid Kanter2020/06/02 06:53 AM
  Transistor count: Metric is often GEChris L2021/01/03 08:39 PM
    Transistor count: Metric is often GEDavid Kanter2021/01/04 08:48 AM
      Transistor count: Metric is often GEChris L2021/01/07 11:38 PM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell avocado?