Alex Bowden wrote: > Initial load call as in > > aVRML97Loader.load(aURL)
But which time? Is it the first time you call this it is slow or is it every time you call this? There is a lot of setup work being done the first time around for a lot of the static structures. As I said in the last one, the Rhino startup takes a long time and maybe that is impacting your calls. Also, what flags have you got set? If you are loading geometry only and don't want anything else, and yet leaving the flags set as the default LOAD_ALL, then don't blame the loader code for the sluggishness as we are doing exactly what you ask for. > However Alan pointing me toward untangling version 1.0 from the XJ3D CVS > archive isn't a lot of help when the archive appears to have been truncated > at version 1.1! What do you mean by truncated? Xj3D has a continuous development right from the start and the CVS logs show that. When we started with the V2 code, we built around the outside of the sun core. All the XML work used the Sun code for the rendering core. Eventually we got to a point where it was not capable of dealing with the requirements of XML and the new X3D spec, so we rewrote that portion. At that stage we issued a "cvs remove" on the code to say that it is no longer relevant to the current mainline development and that was it. There was no truncation, we'd just come to the end of the useful life of the code and so marked it as being no longer current. AFAWK, that was the most modern, highly developed version of the code. Now, the issue of whether the VRML97.jar that comes with the J3DFly app is the same as the one that was in CVS, you will have to ask Sun that. If not, why not, and why were they not interested in putting their changes back into CVS for the rest of the world to work with. (Mind you, post April last year I could see why they wouldn't put those changes in because we weren't maintaining it, but my understanding is that nothing has changed). > 1 is inappropriate to mislay or hide the existing generally pretty good > loader We have done no such thing. It is there for the world to use. We, as the maintainers have moved beyond that code. If someone wants to work with the old code, it is there. We have no obligation to maintain old code that we see as being inappropriate or generally not-useful. > 2 it is inappropriate to channel all enquiries about VRML97 loaders toward a > facility that is not intended to be one and apparently never will be a > satisfactory one. Two different points there. Firstly, Xj3D *is* intended to be a VRML97 loader, among many other tasks. That point is invalid. The second argument about appropriateness can go either way. We think the current code is more appropriate than the old code because we support more parts of the spec, and we are way more spec compliant. In addition, we have the bonus points for supporting the newest spec, which none of the other VRML loaders do. Your measure of appropriateness is different to ours and therefore accusing us of hiding or being malicious with the Sun code is unfair and inappropriate in and of itself. We have stated quite clearly where to find the source code and that we have no interested in maintaining it. Its Open Source - take the code and run with it. Do what you like, we don't care. If you do, send me the URL and I'll put it on the loader page along with the other VRML97 loaders we already have there. Above all, we have the right to market and advertise our own products to people on the list. Are you saying that we are not allowed to say "here is something that we think does what you want" and then let them decide as to whether it is good enough? If so, why is our loader not allowed, but the others like NCSA or Satoshi's are allowed? In fact, up to a week ago, I believed we were the only ones maintaining a VRML loader. Satoshi's update post from last week now says otherwise. If these maintainers are not on the list pushing their projects/products then that's not my problem. Most (all?) people know where the loader page is on j3d.org. That's always kept up to date with the latest (+/- 1 week) information. -- Justin Couch http://www.vlc.com.au/~justin/ Java Architect & Bit Twiddler http://www.yumetech.com/ Author, Java 3D FAQ Maintainer http://www.j3d.org/ ------------------------------------------------------------------- "Humanism is dead. Animals think, feel; so do machines now. Neither man nor woman is the measure of all things. Every organism processes data according to its domain, its environment; you, with all your brains, would be useless in a mouse's universe..." - Greg Bear, Slant ------------------------------------------------------------------- =========================================================================== To unsubscribe, send email to [EMAIL PROTECTED] and include in the body of the message "signoff JAVA3D-INTEREST". For general help, send email to [EMAIL PROTECTED] and include in the body of the message "help".
