View Poll Results: Do you agree with this view?

Voters
16. You may not vote on this poll
  • I agree with this view

    7 43.75%
  • I' in the middle

    6 37.50%
  • I disagree with this view

    3 18.75%
Page 2 of 2 FirstFirst 12
Results 11 to 12 of 12

Thread: ILDA file usability

  1. #11
    Join Date
    Dec 2006
    Location
    Netherlands
    Posts
    983

    Default

    Thanks for these latest contributions, they pinpoint the problem well. Not expecting ilda files to work, the need for modern software to interpolate / reinterpret the point data to reproduce decent output, needing to know the design scan rate all point in the direction that ILD file format is being misused for graphics definition instead of raw output data.
    The creation of OLA is a partial answer to these issues. It indeed uses a subset of SVG for the definition of true scalable vector graphics. Looking at what we have today in terms of software, the following levels of 'sophistication' can be identified:

    1 - User designs a 'world' in either 2D or 3D and uses modern software tools to convert the frames of the animation to laser output. Objects that the user designs are either 2D shapes (triangles, circles, drawings) or 3D geometry. Examples of software are the flash converter, and 3D Max plugin. The software must detect outlines, remove hidden lines etc to produce the arcs and curves along which the laser beam must travel. These steps can be quite complicated. Then, the point data for the scanner needs to be generated. File formats at this level are from Flash and 3D geometry files.

    2 - User designs the arcs and curves directly him/herself, taking care of the hidden line removal etc. This is basically the approach advertised by OLA. It is also the reason random SVG art from the net won't turn into laser art by using OLA. OLA just does the point placement to draw the arcs and curves provided by the user. The user is still designing art and all hardware specific translations are done by the software. SVG is one possible file format, CDR is another - as long as they define true vector graphics.

    3 - User has to edit points, add dwell points / anchor points and needs to have a design scanrate in mind when designing. He is creating the raw data for the scanners, which is completely hardware dependent. This is the level at which the ild file format exists.

    Currently ILD file format seems to be used somewhere between levels 2 and 3. Mixing hardware dependent data and hardware independent graphics definition.


    As for defining a new format, I think that would be a poor choice now. Simply using the subset of SVG that can sensibly be used to define laser lines and colors, makes more sense to me. Maybe we can identify where SVG has shortcomings wrt to laser art we can make our own definitions. since SVG is an open format and XML, that is easily realized in a comprehensible way. The last thing we should do however, is create a new format just for the sake of having it and possibly introducing new shortcomings.

    Finally, If you take for example the 3D max plugin the software has to pass through level 2: generating the curves for laser display (can also be done by a cartoon renderer like Illustrate!). It would be great if we can come to some agreement that everyone, who is willing to work on (partially) open projects, can use a `level 2 file format’ (e.g., SVG) for intermediate output. So the art produced is hardware independent, and can be converted to raw laser points to the liking and requirements of each individual in the laser community.

  2. #12
    Join Date
    Oct 2006
    Location
    Cleveland, Ohio
    Posts
    2,342

    Default

    This is a good idea, the use of a 'true' vector format could improve the quality of shared art. One benefit so far for the ILDA standard is that it is relatively simple. Just points in succession organized into frames. A 'true' vector format is not as simple for each programmer to implement, as such it would benefit us to have a library that deals with loading and perhaps saving these files where the structure, shape math, and file binary operations are contained. We've talked about this to some extent via PMs, specifically the vector loading aspect.
    For this to be a success there should be broad adoption and an ease of going between the formats, even if some data must be lost in the translation process to achieve interoperability. If we got together on this the result could be nice.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •