actually the model does not seem to have much affect on the speed...as a
matter of fact I used a 2kb model and same issues with frame rate. These
were the results of frame rates I posted before:

Ok, here is the test results on a DIFFERENT machine:
  Dell 2.4 Pentium 4, 256 Mb ram, Nvidia Geforce2, Windows 2000


                            ISS model (955 KB)         Test Model (2 KB)
(1) Normal Run                     31 fps                    35 fps
w/ animating sliders

(2) no animating sliders           38 fps                    46 fps

(3) earth texture reduced
    by 50 % ... anim. sliders      31 fps                    34 fps
   Mario

(4) reduced earth texture          39 fps                    46 fps
    no animating sliders





yes I can get to 46 fps if I get rid of the animated sliders but that's
about it....this translates to about 10 fps on the slower machines I
mentioned before. SO I'd still like to raise the overall fps so even the
slow machine can be a bit more bareable.

Oh and the sphere (earth) is absolutely crappy. If you get close to
it you can see the angles of the sphere...I think I have no more than
20 divisions in that sphere. And the sun sphere is simply a VRML object
that is the default VRML sphere and has very few polygons as well.

The only think that I can figure after all the suggestions is that my code
may be poorly written.

When I run my program it uses over 60,000 K on windows 2000...this must mean
I have way to many variables and things defined. I'm going to have to go
back and try to minimize things a lot more.

thanks for the help, everyone!

Mario

NASA Johnson Space Center


-----Original Message-----
From: Michael P. McCutcheon
To: [EMAIL PROTECTED]
Sent: 1/10/2003 9:12 PM
Subject: Re: [JAVA3D] ? Frame Rates ?

Just to let you know...here is my system:

Pentium 2 450 Mhz.
512 MB memory
TNT 1 Graphics card
Windows 2000 (and Max OS X :)
JDK 1.4.1_01
Java3D 1.3.1 beta1

For scenes of similar complexity I get roughly 30-60 FPS. (nothing
scientific,
but suffice it to say that it can be 'smooth', even on old computers).

I think that you must be using some terribly large textures or
super-super
acurate spheres or something.  Also I look at the small model...I
suspect it
was exported from a CAD tool or something with LOTS of polys.  Exporters
can be
pure evil :)

With a computer as fast as yours...you should be able to get 40-60 FPS
out of
something like that, easy.  Scale your models accuracy and texture size
down.
I bet you'll be surprised how fast you can make it.

Mike

ZACZEK, MARIUSZ P. (JSC-DM) (NASA) wrote:
> Hi,
>  I have a program (picture attached) which has a canvas3d and couple
of
> canvas2d's and a swing interface.
> When I run this program (with the canvas3d animation running) on a
Pentium
> 3, 900 Mhz, 128Mb Ram, Nvidia TNT2,
> I get pretty crappy frame rate and graphical update. I'm talking about
5
> frames/second. This includes the
> animation of the mercator projection and the plot which are at the
bottom of
> the display.
>
> What I want to know is do any of you have any suggestions for how to
improve
> the frame rate. Could my
> code be so badly written? I try to minimize the number of Transforms
and
> Groups as possible. I did notice
> that having my sliders be updated by my animation slows me down some
so I'm
> going to have flag to not
> have them get updated if the computer is too slow. But I'd still
appreciate
> any other advice...and
> also, is there any code out there that one could run and have it
output the
> framerate....so that I could
> use that code and test the machines out to see what framerates are
possible?
>
> I know there is a Java3D FAQ regarding speed up and I've read it. The
thing
> that worries me about my code
> being slow is that I know people are making Java3D games and I imagine
they
> must be fast enough to play
> so they must be doing something right.
>
>    Mario
>
> Mariusz Zaczek
> NASA - Johnson Space Center
> Automated Vehicles and Orbit Analysis / DM35
> Flight Design and Dynamics Division
> Mission Operations Directorate
> Bldg: 30A     Room: 3040A
> Phone(W): 281-244-6650
> Phone(H): 832-385-3860
>
> Disclaimer: "The opinions, observations and comments expressed in my
email
>              are strictly my own and do not necessarily reflect those
of
> NASA."
>
>
>
>
>
------------------------------------------------------------------------
>

========================================================================
===
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".

===========================================================================
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".

Reply via email to