[info-performer] some problems linux->onyx2

New Message Reply Date view Thread view Subject view Author view

From: Ive Akkermans (Ive.Akkermans++at++esat.kuleuven.ac.be)
Date: 10/17/2002 02:47:55


Hi Performers,

I ported my application from pf2.5.2 linux to our Onyx2 system.
However, some very weird problems are occuring, to which I don't have a
solution.

For instance, at a certain moment I want to follow a path that moves
away from the camera position, makes a slight curl to the left and then
a big one to the right. On Linux, this works fine. However, when I run
my app on the Onyx2, instead of moving forward, the object makes a 180
degree turn and starts moving backwards (I use Hermite interpolation, so
the object follows a curve).

Now, my first impression is that on the Onyx2 the direction of the axis
'straight forward' is inverted in respect to the Linux's.

Secondly, I am having some trouble with .iv/.pfb format. On Linux, our
models are created in maya and by some means converted to .iv. I use
pfconv to convert these models to .pfb and they look perfect under Linux
Performer.

However, the same models used on onyx2 give an error message saying that
the version [23] is too recent. When I try to use onyx2's 'pfconv'
function on my iv models, I get the same message, but [21] instead of
[23]. I can solve this by writing a very basic performer app which
reads in the .iv file with pfdLoadFile and immediately stores the file
again with pfdStoreFile.

However, when I do this, my models are not good anymore. One, which is
the representation of a human head, has it's right eye shifted to the
right at about half a face length out of is head. The left eye is also
slightly shifted to the left. I figure this has to do with some
inconsistency in Performer's load/convert/save functions on our Onyx2.

This brought me to checking the Performer versions on the Onyx2. This
is the strange output I get from 'versions | grep performer' on the
onyx2:

I performer_demo 06/26/2002 OpenGL Performer 2.5.1 Demos and
Demo Data
I performer_demo.man 06/26/2002 Performer Demos and Data
Documentation
I performer_demo.man.demo 06/26/2002 Performer Demo Man Pages
I performer_demo.sw 06/26/2002 Performer Demo Programs
I performer_demo.sw.demo 06/26/2002 Performer Demos & Utilities
I performer_dev 10/19/2001 OpenGL Performer 2.4 Development
Environment
I performer_dev.books 10/19/2001 Performer On-Line Books
I performer_dev.books.Guide 10/19/2001 Performer Programming Guide
I performer_dev.books.Start 10/19/2001 Performer Getting Started
Guide
I performer_dev.man 10/19/2001 Performer Developer Documentation
I performer_dev.man.pages 10/19/2001 Performer Man Pages
I performer_dev.man.relnotes 10/19/2001 Performer Developer Release
Notes
I performer_dev.sw 10/19/2001 Performer Development Environment
I performer_dev.sw.debug 10/19/2001 Performer Debug Libraries
I performer_dev.sw.hdr 10/19/2001 Performer Header Files
I performer_dev.sw.tools 10/19/2001 Performer Development Tools
I performer_eoe 06/26/2002 OpenGL Performer 2.5.1 Run-Time
Environment
I performer_eoe.man 06/26/2002 Performer User Documentation
I performer_eoe.man.relnotes 06/26/2002 Performer User Release Notes
I performer_eoe.sw 06/26/2002 Performer Run-time Libraries
I performer_eoe.sw.2_2 06/26/2002 Performer 2.2.15 Run-time
I performer_eoe.sw.2_4 06/26/2002 Performer 2.4.4 Run-time
I performer_eoe.sw.performer 06/26/2002 Performer Run-time Required
Files

Look at this:
I performer_dev.sw 10/19/2001 Performer Development Environment
I performer_eoe.sw 06/26/2002 Performer Run-time Libraries

I'm no expert at sgi's performer, but it seems to me that there are some
very nasty mix-ups in these versions ? Or is this a pretty consistent
Performer installation ?

Many thanks in advance for all advice,
ive.

-- 
Ive Akkermans
K.U.Leuven - ESAT-PSI     tel: +32-16-32.17.06  fax: +32-16-32.17.23
Kasteelpark Arenberg 10   e-mail:ive.akkermans++at++esat.kuleuven.ac.be
B-3001 Leuven-Heverlee    http://www.esat.kuleuven.ac.be/psi/visics/


New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Oct 17 2002 - 02:48:04 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.