partial torus prims


RE: partial torus prims
#22
Claude summarizes as following:

Summary of the discussion so far

The problem (René): We only have full quarter-tube tori (inside/outside). Some parts need a tube cut partway, e.g. for chamfers or partial prints. An example is already on the tracker: t04i8000-075.dat.

Where we got to:
  • Percentages were rejected (Travis). They break down at 48-res, where a quarter has 12 segments and 1/12 is a repeating fraction. Travis suggested numerator/denominator instead and was fine with +/-.
  • "+" can't be used. Willy found that files with "+" can't be downloaded from the PT, and Orion/Hageta confirmed that "+" has a special meaning in URLs.
  • Readable names. Willy asked for less cryptic names (we have 255 characters) and suggested plus/minus or CW/CCW. Peter pointed out that CW/CCW depends on how you picture the torus.
  • Alternative direction labels: p/q (Peter), anatomical terms (René), orbital terms and proximal/distal (Peter). Peter also proposed a complete overhaul of naming for all primitives.
  • N. W. Perry's synthesis: keep René's structure, use h (count from the horizontal plane of the major radius) or v (count from the vertical top of the tube), and a segment count: 1 digit for 16-res (1-3), 2 digits for 48-res (01-11). Up/down is already set by the y-factor and in/out by i/o, so nothing else is needed.

Still open: odd major sweeps (Magnus), the description format, conditional lines on the cut ends, and renders (Willy).

Proposal

I'd adopt Perry's scheme with one change: separate the suffix from the existing name with a hyphen, as the file already on the tracker does.

Filename: t + major fraction + i/o + ratio + - + h/v + segment count
  • Standard (16): -h1 to -h3, -v1 to -v3, e.g. t04i8000-h3.dat, t08o8000-v2.dat
  • Hi-res (48): -h01 to -h11, -v01 to -v11, e.g. 48\t04i8000-h05.dat

Definitions:
  • h = N segments starting at the tube's equator (the major plane): the innermost edge for i, the outermost edge for o.
  • v = N segments starting at the top of the tube.
  • A full quarter (4 or 12 segments) never gets a suffix.
  • -hN and -v(4-N) are complementary halves of one quarter-tube.
  • Lowercase only, no "+", so names are URL-safe and work on the PT.

Why this one:
  • Backward compatible. Existing tori don't change, and primitive substitution only needs a small regex extension.
  • Segment counts are exact at any resolution, with no rounding.
  • h/v (horizontal/vertical) are common words, and the description spells them out.
  • The hyphen keeps the base name searchable and is already accepted by the PT.

Description, spelled out in full so nobody needs a decoder:
Code:
Torus Inside 1 x 0.8 x 0.25 Tube 3 of 4 Segments from Horizontal Torus Outside 1 x 0.8 x 0.125 Tube 5 of 12 Segments from Vertical
(with the ratio formatted the same way as in the existing torus descriptions)

Conditional lines: same as partial cylinders. Keep condlines on the cut edges, with control points that continue the curve. This is what René's "delete faces from a full quarter" method already produces, so the pieces join smoothly with each other and with the full quarters, and generators can make them by trimming a generated quarter.

Scope:
  • Odd major sweeps (5/48 etc.) get their own thread. If they're added, they should use the existing N-D fraction style.
  • A complete overhaul of primitive naming is a separate long-term topic and shouldn't block this.
  • Before we vote, we should post one render showing h1/h2/h3 and v1/v2/v3 on an inside and an outside torus.
LEGO ergo sum
Reply
« Next Oldest | Next Newest »



Messages In This Thread
partial torus prims - by Rene Rechthaler - 2024-11-30, 15:41
RE: partial torus prims - by Willy Tschager - 2025-01-31, 13:15
RE: partial torus prims - by Rene Rechthaler - 2025-01-31, 15:54
RE: partial torus prims - by Travis Cobbs - 2025-04-26, 21:17
RE: partial torus prims - by Willy Tschager - 2025-04-27, 5:53
RE: partial torus prims - by N. W. Perry - 2025-05-07, 15:42
RE: partial torus prims - by Magnus Forsberg - 2025-05-07, 16:33
RE: partial torus prims - by N. W. Perry - 2025-05-07, 21:52
RE: partial torus prims - by Willy Tschager - Yesterday, 11:47
RE: partial torus prims - by Orion Pobursky - 2025-04-27, 6:59
RE: partial torus prims - by Hageta - 2025-04-27, 7:05
RE: partial torus prims - by Peter Blomberg - 2025-04-27, 10:02
RE: partial torus prims - by Rene Rechthaler - 2025-04-29, 19:54
RE: partial torus prims - by Travis Cobbs - 2025-04-29, 20:27
RE: partial torus prims - by Peter Blomberg - 2025-04-29, 21:36
RE: partial torus prims - by Willy Tschager - 2025-04-30, 10:13
RE: partial torus prims - by Peter Blomberg - 2025-04-30, 11:09
RE: partial torus prims - by Rene Rechthaler - 2025-04-30, 17:19
RE: partial torus prims - by Peter Blomberg - 2025-04-30, 20:22
RE: partial torus prims - by Peter Blomberg - 2025-05-06, 22:03
RE: partial torus prims - by Willy Tschager - 2025-05-07, 8:13
RE: partial torus prims - by Rene Rechthaler - 2025-05-07, 16:49

Forum Jump:


Users browsing this thread: 1 Guest(s)