Article: Transistor Count: A Flawed Metric
By: David Kanter (dkanter.delete@this.realworldtech.com), June 2, 2020 7:53 am
Room: Moderated Discussions
Paul A. Clayton (paaronclayton.delete@this.gmail.com) on June 2, 2020 7:07 am wrote:
> David Kanter (dkanter.delete@this.realworldtech.com) on May 18, 2020 7:04 am wrote:
> > Dear Friends and Colleagues,
> >
> > Transistor count and transistor density are often portrayed as technical achievements and milestones.
> > Many vendors brag about the complexity of their design, as measured by transistor count.
> >
> > In reality, transistor count and density varies considerably based on the type of chip and especially the
> > type of circuitry within the chip, and there is no standard
> > way of counting. The net result is that transistor
> > count and density are only approximate metrics and focusing on those particular numbers risks losing sight
> > of the bigger picture. It's not about how many transistors you have, but how you use them.
> >
> > https://www.realworldtech.com/transistor-count-flawed-metric/
>
> "Effect of Design on Transistor Density" at SemiWiki provides some additional information relative
> to using standard cells, particularly the greater density achievable by using larger cells (the
> cells are effectively custom logic and so internally more highly optimized, so internal density
> will be higher than average density and larger cells will have more internal transistors).
That's a good way of framing it.
> In addition to drive strength density effects, at least before FinFET, low leakage transistors were larger (so
> a design biased toward low inactive but not power-gated power would tend to have lower transistor density).
Low-leakage devices are still much bigger. If you look at Intel's 22FFL uses a 144nm contacted gate pitch for the lowest leakage device (which has a thick gate oxide), and reduces leakage by about 100X.
David
> David Kanter (dkanter.delete@this.realworldtech.com) on May 18, 2020 7:04 am wrote:
> > Dear Friends and Colleagues,
> >
> > Transistor count and transistor density are often portrayed as technical achievements and milestones.
> > Many vendors brag about the complexity of their design, as measured by transistor count.
> >
> > In reality, transistor count and density varies considerably based on the type of chip and especially the
> > type of circuitry within the chip, and there is no standard
> > way of counting. The net result is that transistor
> > count and density are only approximate metrics and focusing on those particular numbers risks losing sight
> > of the bigger picture. It's not about how many transistors you have, but how you use them.
> >
> > https://www.realworldtech.com/transistor-count-flawed-metric/
>
> "Effect of Design on Transistor Density" at SemiWiki provides some additional information relative
> to using standard cells, particularly the greater density achievable by using larger cells (the
> cells are effectively custom logic and so internally more highly optimized, so internal density
> will be higher than average density and larger cells will have more internal transistors).
That's a good way of framing it.
> In addition to drive strength density effects, at least before FinFET, low leakage transistors were larger (so
> a design biased toward low inactive but not power-gated power would tend to have lower transistor density).
Low-leakage devices are still much bigger. If you look at Intel's 22FFL uses a 144nm contacted gate pitch for the lowest leakage device (which has a thick gate oxide), and reduces leakage by about 100X.
David
| Topic | Posted By | Date |
|---|---|---|
| New article: Transistor count: A Flawed Metric | David Kanter | 2020/05/18 07:04 AM |
| Non active transistors | Doug S | 2020/05/18 02:26 PM |
| Non active transistors | Ricardo B | 2020/05/18 09:12 PM |
| Minor quibble about fixed-performance ASIC | Paul A. Clayton | 2020/05/19 03:59 PM |
| Minor quibble about fixed-performance ASIC | David Kanter | 2020/05/21 06:58 AM |
| A complementary article about xtor density | Paul A. Clayton | 2020/06/02 07:07 AM |
| Low leakage transistors | David Kanter | 2020/06/02 07:53 AM |
| Transistor count: Metric is often GE | Chris L | 2021/01/03 09:39 PM |
| Transistor count: Metric is often GE | David Kanter | 2021/01/04 09:48 AM |
| Transistor count: Metric is often GE | Chris L | 2021/01/08 12:38 AM |
