Doug Twilleager wrote: > The big part that is miising is explicit low level buffer > control - something like you suggested below. We could do > it in the current system, but how we allocate buffers would > be just a best guess. We would need new API to let the application > help tell us which buffers were most important. This is one of > the possible additions we are looking at for 1.4. So, no there > in no reason we can't do it today, but we need some new buffer > control API's to complete the story.
It seems to be a piece of code quite independent from rest of system. In fact buffer control api could be developed independent from java3d by 'third party'. Of course java3d-endorsed standard would be best, but for the time being... I know that this is probably too late for 1.3, but maybe 1.3.1 or something like that ? I'm not talking about buffer api, just about some property flag for java3d, which would tell it to use asynchronous loading of all NIO+BY-REF geometry. Somebody else would write open source buffer management, which you could later use or drop in 1.4. Artur =========================================================================== 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".
