Subject:
|
Re: Torus primitive discussion. was( Updated Primitive - 1-8t0102 1/8 torus)
|
Newsgroups:
|
lugnet.cad.dat.parts.primitives
|
Date:
|
Fri, 16 Feb 2001 17:17:36 GMT
|
Viewed:
|
4237 times
|
| |
|
|
In lugnet.cad.dat.parts.primitives, Lars C. Hassing wrote:
> Paul Easter wrote...
> > Do you think that this can be easily implimented into L3P? Lars?
> Sure, I'm just awaiting the final filename convention, so that I can make
> a programmatic (versus hardcoded) substitution covering all future
> primitives released. Something like "NtiFFFF.dat" for
> "1/N Torus Tube Inner Major Radius 1 Tube Radius 0.FFFF"
How about using:
TNNXFFFF.dat
Where T is T, NN is the fraction of the major circle, X is I/O for inner
or outer, and FFFF is the fractional ratio of tube radius to major radius.
I'm suggesting this specific format for the following reasons:
1. Putting the T on the front of the filename will provide better sorting
of the Torus p-files amongst all the other p-files.
2. Dedicating the one free filename digit/character to one of the numeric
parameters will also provide more filename consistency and sorting. And
giving this the major-circle denominator is more important than having a
fifth position in the tube:major-radius fraction.
3. Putting the I/O character between the two numeric parameters will
improve readability (because it has as a delimiter between the strings of
digits), at the expense of having somewhat poorer filename sorting.
> Which values of N do you expect? 16? - since you only got 4 F's?
I'd expect probably 04, 08, and 16. If we start getting hi-res versions of
torii, we might see higher values (smaller sweeps). With two decimal
digits, we can get down to a 3.6-degree sweep (that'd be 1/99).
If we need more range/precision in the numeric parameters, we could use
hexadecimal notation, or even base-36, for those files. But then we'd need
a way to clearly indicate which files used which notation.
Hmm. Three digits of base-36 can match 4.6 digits of base-10. So, if we
use some screwy base-36 notation from the start, we can have better
resolution in fewer characters, leaving another filename-character (or two)
for human-readable tags.
Running with this idea a little more, Paul suggested the following torii as
a start:
> I
> have a list of suggested files to be generated. Most of them are fractions
> such as 1/10(.1000), 1/9(.1111), 1/8(.1250), 1/7(.1429), 1/6(.1667),
> 1/5(.2000), 2/9(.2222), 1/4(.2500), 2/7(.2857), 3/10(.3000), 1/3(.3333),
> 3/8(.3750), 2/5(.4000), 3/7(.4286), 4/9(.4444), 1/2(.5000), plus a few others.
Using base-36 for the t-m fraction parameter, we get filenames like the
following. (these examples use base-10 for the major-circle denominator.)
tr04i3ll.dat 1/4 Inner Torus 1:10 ratio
tr04i400.dat 1/4 Inner Torus 1:9 ratio
tr04i4i0.dat 1/4 Inner Torus 1:8 ratio
tr04i555.dat 1/4 Inner Torus 1:7 ratio
tr04i777.dat 1/4 Inner Torus 1:5 ratio
tr04i800.dat 1/4 Inner Torus 2:9 ratio
tr04ifff.dat 1/4 Inner Torus 3:7 ratio
tr04o3ll.dat 1/4 Outer Torus 1:10 ratio
tr08o3ll.dat 1/8 Outer Torus 1:10 ratio
tr16o3ll.dat 1/16 Outer Torus 1:10 ratio
Notice that 1 <> l. Hmm, maybe we should disallow a few letter-digits
(such as o, i, and l) to prevent digit-confusion.
What do you think? I'm not completely thrilled about using a high-number
base to compress the parameters, but it would work fine. It would just be
a bit hard to read.
Steve
|
|
Message has 1 Reply:
Message is in Reply To:
16 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
