"Kenneth" <kdw### [at] gmailcom> wrote:

> I like this; I might print it out and hang it over my own machines.

Well, what I would do is use the 1/64th scale with the metric equivalents.

And now that I'm telling you to do that, I should look into going in the reverse
direction.

And Metric.

I had a conversation way back with clipka about print resolution data in the
file output.

If we can work out a way to print to scale, then you could also draw tori with
specific diameters, boxes to show the thickness of sheet metal, etc.

And like I did with my (unfinished) slide rule, you could define arrays of
special sizes - number and letter drill bits, and plot those on the scale too.

(I just scored an L. S.  Starrett No. 245 Engineer's Gauge at he flea market)

> An improvement would be to align/stretch each conversion line so that the
> '1.0-unit'-ends on the right will all line up vertically...so that it's clear as
> to where each conversion's values match the others. When I first looked at your
> graph, I assumed that this was already the case.

No, these are just arbitrarily scaled lines that get divided up.
Just take some x-value and set that as the length, and you can calculate
conversion factors for all of the Multipliers in the calls to DrawGraduatedLine
().

> I'm also curious about the divide-by-11 line; the practical use of that one is a
> mystery to me. What's a real-world example?

But you were ok with 18ths and 5ths?

To test if the algorithm works.
What if you wanted to fold a line up into an hendecagon/undecagon?
You'd need to divide it equally into 11 parts.
What if you needed to measure things for The Six-Fingered Man?
Or your amp went up to ELEVEN?

Why tau?  or pi?  or 18?  16?  32?  64?  24?  4?

It needed to be value agnostic.


- The Great Northern Walker

Post a reply to this message