arai (arai++at++rdd.kepco.co.jp)
Thu, 27 Nov 1997 14:59:16 +0900
We are making a VR system.
And we have some questions about Performer.
Our VR system runs on IRIX5.2 on Onyx RE2 (desk-side)/128MB memory.
Our Onyx RE2 has a RM5 board and multi-channel option.
Our VR system outputs 2 displays(640x480) by MCO for right eye and left
eye.
And our VR system is inputted from 2 sensors(one globe and one magnetic
sensor).
We measured the processing time/Flame and Flame/sec by Statistics Displays.
We really need the information about real delay on Onyx from input of
sensor to output.
"Statistics Displays" is on one output(for right eye).
And only "Stage Timing Statistics Displays" is on another output(for left
eye).
For example,
"Statistics Displays" on one output(display) shows following.
12.0/30.0Hz, FREE Draw=34.6MS
MSecs : frame=234.0 app=15.3 cull=20.7 draw=34.6
Another output(display) shows following.
12.0/30.0Hz, FREE Draw=37.4MS
MSecs : frame=236.1 app=15.2 cull=15.9 draw=37.4
The picture that we got on our application is accompanied.
Questions
1. Are the video retrace intervals of both outputs at the same time or
separate?
2. Is "Application process" common on both outputs?
3. Do these "Culling process" process separately?
4. Do these "Draw process" process separately?
5. Is this "frame=234.0" the total of app, cull(right eye), cull(left eye),
draw(right eye) and draw(left eye)?
6. What does each "frame???" show?
7. Do we have to regard the other "frame=???" as the processing time/frame?
Which "frame=???" do we have to accept?
8. Do we have to regard the total of "frame=???"(right eye) and
"frame=???"(left eye) as the processing time/frame?
9. When "frame=???" doesn't equal the processing time/frame, what do we
require instead of "frame=???"?
10. How do we get the processing time/Frame?
we expect your advice.
Koichi Arai
The Kansai Electric Power Co.,Inc.
3-11-20 nakoji, amagasaki, hyogo, Japan
Tel 81 6494 9753
Fax 81 6494 9728
E-mail arai++at++rdd.kepco.co.jp
This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:56:15 PDT