We do our view changes and animations per frame, using elapsed time to
calculate rotations in degrees/sec and movements in meters/sec. That way
our scene stays consistant even if frame rates drop. But I had long felt
things were not as smooth as they could be and never looked into it until
now.
Dave,
Switch to Linux? {grin}
If I'm not mistaken the timer for DOS (hence Win9x) is once every 18.2
seconds. Thus the timer has a resolution of 55 ms. So unless you
switch operating systems (or write your own timer in native code) you
won't get better resolution.
If someone else knows an alt
Java3D is in desperate need of a high precision micro timer. I have
mentioned this to a couple of the Java3D team but they don't seem to think
it is important as I haven't noticed it on the list of features for the next
release. This isn't that difficult a feature to implement since all the
native
John, Dave,
If you are running on Windows NT (I hope you are!) you can access the
underlying "high performance" timer, which has an accuracy of several
nanoseconds. Unfortunately Java does not do this by default on NT so you are
stuck with the default timer (GetTickCount). On Windows NT I believe
> Dave,
>
> Switch to Linux? {grin}
Yes, 1ms resolution there. Still very bad I think. I remember working on a
flightsim where 1ms resolution in time (win MM timers) was far too coarse
and caused weird effects.
I think a new time interface is needed desperately. WinXX allows to measure
time far
Hi!
Well, having the same problem I can tell you that currentTimeMillis() really
sucks and that there seems to be no way around that (except thinking of JNI
hacks).
I attached a TimerAccuracy class that shows timer resolution. My experienced
values are
50-60ms on Win9x
10-12ms on WinNT
1-2ms on
What about the javax.media stuff?
http://java.sun.com/products/java-media/jmf/2.1/apidocs/javax/media/SystemTi
meBase.html
They say the time is to the nanosecond precision. I don't know anything
about the jmf package, isn't quite large?
Transformation of observed objects seems to be ok at curr
Hello dear list-members,
I get strange results with bounds, I'm afraid.
I'd aglad if anybody could help me...
(I'm using Java 1.3 and Java 3D 1.2)
Problem 1:
Can anybody explain this to me? or is it a bug?
I tried to transform a BoundingSphere centrum with a translation, thinking
it would transfo
I've never heard anything bad about using the NVidia generic drivers
(Detonator), but I think I should relay a personal experience.
Currently I've two PCs configured with Elsa Erazor X (GeForce 256
chipset) video cards (both running Win98SE). One seems to be running
fine with NVidia's Detonator
On Sun, 4 Mar 2001, David wrote:
> looks like at around 50 fps, I am getting 3 dups for every 1 real frame
> recalc. Course everything still looks right, but I am losing 3 steps of
> smoothness.
> At 50 fps, I am calculating a frame every 20 ms. I thought the
> System.currentTimeMillis() had a
Hi!
I have a panel that contains buttons for rotating along the + and - z,x axes.
I have used the Wakeuponelapsed frame behavior listener set to 0 frames.
A 0 frame means that it will listen as a new frame is drawn?
And as part of it, I have another behavior that listens to a behavior with an I
Your loader does the trick. However I am wanting to use Java3D in applets.
I will be loading
models and textures across net connections. Currently, your importer does
not support this. To
get around this, I digitally certified my applet. I load the model over a
net connection, then save it
on
From: "David" <[EMAIL PROTECTED]>
Sent: Sunday, March 04, 2001 12:25 PM
> We do our view changes and animations per frame, using elapsed time to
> calculate rotations in degrees/sec and movements in meters/sec. That way
> our scene stays consistant even if frame rates drop. But I had long felt
13 matches
Mail list logo
