Re: [SI-LIST] : An Interesting Presentation

Shayle Hirschman (shayle@mho.net)
Wed, 21 Apr 1999 12:57:37 -0500

At 12:26 PM 4/21/99 -0500, you wrote:
>
>
>Interesting discussion.
>
>"Trapezoidal" waveforms are a way of getting signals around on a board
between
>pieces of IP are getting recommended at times. Once inside a chip you can
afford
>to "square things up." Complex topologies with non-zero stub lengths are more
>apt to show up the problems created by fast, hard turn-on drivers.
>
>I guess most SI'ers know this. So, what are the drawbacks with trapezoidal,
>i.e., nearly sinusoidal clocks?
>
>
>Regards,
>
>Roy

Perhaps too much uncertainty or unwanted placement in the edge crossover
detection points, eating away timing margin in the worst-case timing
analysis and increasing skew relative to other signals.

If all receivers behaved identically and there is no noise in the system,
then the effect may just be a change of duty cycle. But this isn't likely
and so, I would think, signals approaching sine waves should be avoided.

Shayle

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****