Skip to main content

Open source tiny 3D engine for tkinter

Project description

Tiny 3D Engine

logo

This package is a python 3D engine based on the Tkinter Canvas. It uses Numpy for math handling.

It is available on pipy, documentation is on readtthedocs, sources are mirrored on gitlab.com

Install this using

pip install tiny_3d_engine

It create simle 3D rendering for bars, tri and quad elements and store the scenes in Ensight's ASCII .geoformat. A trivial grid would look like:

trivial_example

Loading a model

In this simple example, a .geo file is loaded load_file_as_scene into a 3D scene. This scene is given to a new engine object Engine3D. We apply a rotation .rotate() before rendering .render()the scene on the screen, then leave the interaction to the user .mainloop().

scene = load_file_as_scene("myfile.geo")

test = Engine3D(scene)

test.rotate('y', 45)
test.render()
test.mainloop()

The SCENE and the ENGINE

The SCENE is an object storing the 3D model. A void scene is simply None. You can update a scene with the method .update(). Each scene handle several parts identified by a name, a string looking either as -tag- (e.g. "ceiling") or as -family-.-tag- (e.g. "house.ceiling").

The ENGINE is used to project the scene on the 2D screen. Once started, the view point can be contrlled by methods such as .translate() or .rotate(), then refreshed with .render(). The scen cane be update with .update(). If you want user interaction with the result, finish with the typical TK .mainloop().

What if I already have my vertices and polygons?

In the following example, two squares are appended to an initially void Scene3D object, using the method scene.add_or_update_part.

  • The first, in blue, is made of edges (2 vertices connectivity)
  • The second, in red, is made od squares (4 vertices connectivity)

This scene is passed to the Engine3D object, triggering a window.

from tiny_3d_engine import (Scene3D, Engine3D)

scene = Scene3D()

SIZE = 2
LENGTH= 200.
points = list()
conn = list()
dx = LENGTH/

for i in range(SIZE):
    for j in range(SIZE):
        index = len(points)
        points.append([i*dx, j*dx, 0])
        points.append([(i+1)*dx, j*dx, 0])
        points.append([i*dx, (j+1)*dx, 0])
        points.append([(i+1)*dx, (j+1)*dx, 0])
        conn.append([index, index+1])
        conn.append([index+3, index+1])
 
scene.update("square1", points, conn, color="#0000ff")

points = list()
conn = list()
for i in range(SIZE):
    for j in range(SIZE):
        index = len(points)
        points.append([i*dx, j*dx, LENGTH])
        points.append([(i+1)*dx, j*dx, LENGTH])
        points.append([i*dx, (j+1)*dx, LENGTH])
        points.append([(i+1)*dx, (j+1)*dx, LENGTH])
        conn.append([index, index+1, index+3, index+2])
scene.update("square2", points, conn, color="#ff0000")

test = Engine3D(scene)
test.rotate("x", 45)
test.rotate("y", 45)
test.render()
test.mainloop()

(It would have been easier in numpy, but I wanted to keep this readable for non-numpy programmers)

Command line

A small command line interface is available:

Usage: tiny_3d_engine [OPTIONS] COMMAND [ARGS]...

  ---------------    TINY_3D_ENGINE  --------------------

  You are now using the Command line interface of Tiny 3D engine, a Python3
  Tkinter lightweight 3D engine, created at CERFACS (https://cerfacs.fr).

  This package is likely as a dependency of other packages, to
  provide a light 3D feedback for small 3D scenes <100 000 polygons. This
  CLI is given here for developers perusal and demonstrations. Find the
  script of these small tools in the /examples folder of the package.

  This is a python package currently installed in your python environement.
  See the full documentation at :
  https://tiny-3d-engine.readthedocs.io/en/latest/.

  DISCLAIMER: Tiny 3D engine is a brute force flat renderer. As it is NOT
  using your graphical card,  do not excpect anything fancier than a 1980
  video game.

Options:
  --help  Show this message and exit.

Commands:
  bench   Run a short benchmark on your machine.
  load    Load a 3D scene from FILENAME.
  rabbit  Run a demo with the Stanford Rabbit.

Performances

Do not expect more than an early 90s videogame. During mouse interations, the frames per second is roughly 30 000 / nb. of polygons (i.e. 15 fps for 2000 polygons).

The engine is by default limited 100 000 polygons in a static view and 2 000 during mouse interactions. If the model goes beyond these limits, the engine ramdomly remove polygons at the loading time, to keep the window responsive.

Requirements

The present library require Numpy and Tkinter. The Tk aspects are limited to the screen object. In the future I might write extensions for PyQT4 Canvas or Matplotlib... or not.

Origins

This work stems from a mix between a Pure Tcl/Tk Engine of mine pure TK 3d engine and the the pyEngine3D-master of henry Haefliger pyEngine3D , because I really liked the API.

The present one allow several parts to be loaded, and uses numpy. Scenes can be dumped or read from the Ensight .case/.geo files.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tiny_3d_engine-0.3.0.tar.gz (394.5 kB view details)

Uploaded Source

Built Distribution

tiny_3d_engine-0.3.0-py3-none-any.whl (28.6 kB view details)

Uploaded Python 3

File details

Details for the file tiny_3d_engine-0.3.0.tar.gz.

File metadata

  • Download URL: tiny_3d_engine-0.3.0.tar.gz
  • Upload date:
  • Size: 394.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.11.4

File hashes

Hashes for tiny_3d_engine-0.3.0.tar.gz
Algorithm Hash digest
SHA256 960fc845c0a8d5a798322ca2542052ee7a41fbff48c5eda26280c502ba371477
MD5 4891599f0c2c55a886536e3e68a57047
BLAKE2b-256 0bab7075ea7ef0fee0a2a25e3956c0fd71a754239cbf08f0d48820049e5100b0

See more details on using hashes here.

File details

Details for the file tiny_3d_engine-0.3.0-py3-none-any.whl.

File metadata

File hashes

Hashes for tiny_3d_engine-0.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 1eaf170235f3e97197e483221c6e8f69cfc129ebbe9ddb979879de4b817f1aa0
MD5 6c6979e858dc178e890801319374708b
BLAKE2b-256 c130f9787e411aa00b2b3410bb008850d74752aa1e07e8a581e00985a506e902

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page