Re: interpolation and algorithms of multiple axes

Jeff wrote:

>I don't have it here with me but I'm pretty sure one of Knuth's books
>covers this stuff. Is that the "numerical methods" book being
>referred to, perhaps?

Nope. Knuth's books are good, and he covers a lot of territory, maybe
this also. Numerical methods is a different book.

>I'm getting the jist of this correctly, the so called "pulse
>generating black box" that we're talking about is simply another
>(much slower)hardware implementation of this algorithm.

No again :-( This refers to Mariss' posted pulsegen.doc in the CCED
files section. There is no multi-axis interpolation inherant in the
design. It's a hardware pulse generator which has an
even "granularity", unlike PC timer/counter circuits where most of
the the available values (to time with) are "crammed into" one end of
the scale. Now, if you "feed" the pulsegen box with appropriate
values, interpolated moves should result! But someone still needs to
look at the "available" angles of lines, IF we're just going to set
it and go... And if not it gets pretty tricky, pretty quickly, IMO.

>What I think I've learned is that if you don't take the time to get
>the system partitioning right before you start writing code and
>designing hardware you end up working way too hard to make something
>that only sort of works.

BIG ditto!!!

>It seems clear to me that a better partitioning choice is to have the
>software calculate the pulse rates while the generation of the step
>pulses themselves is best done in hardware. Since a standard PC does
>not include any appropriate hardware we need the "pulse generating
>black box" that Jon Elson and Mariss, it appears, are both now
>working on.
>Jeff

Yes. But let's not forget Mr. fenerty and his S/W based effort.

Lotta good people on this list!

Ballendo