Re: model view matrix and r

New Message Reply Date view Thread view Subject view Author view

Jerry Edsall (jedsall++at++fasainteractive.com)
16 Jul 1997 10:55:11 -0500

RE>>model view matrix and r
Actualy I'm not using gl's coordinates. I'm using a multi-platform simulation
architecture that is y up. We have a very specific and defined layer between
the renderer and simulation. The idea is that any diffrences between the
simulation and performer is the video layers problem. I was doing what you just
suggested already, which is :

y=-z
z=y
roll=-roll

But this required me decompose my matrix from the simulation to figure out what
the xyz ypr was. So what you said about reording the multiplication order helped
and I figured it out. What I needed to do was a simularity transform on my
simulation matrix. Since that was orthagonal ( 90 degrees ) I just had to
reorder the matrix and change some signes.

Thanks for the help.

Jerry

------------------------------
Date: 7/16/97 5:36 AM
To: Edsall, Jerry
From: Angus Dorbie

On Jul 15, 12:06pm, Jerry Edsall wrote:
> Subject: Re: model view matrix and r
> RE>>model view matrix and rotatio
> So how do you convert from a gl matrix to a performer matrix?
>
> Jerry

There's just a view orientation difference. So there's no conversion,
there will be a pre viewing 90 degree rotate around x on the mviewing stack
so conceptually y & z mean different things.
As I've said this is for the z up coordinate system imposed by performer.

So

pf_x == gl_x
pf_y == -gl_z
pf_z == gl_y

gl_x == pf_x
gl_y == pf_z
gl_z == -pf_y

If you really insist on using gl coordinates you can add a pfSCS above
the scene graph with a 90 degree x rotate and pre & post multiply the
viewing matrix by +ve & -ve 90 degree x rotations. If you do this
then conceptually you end up with OpenGL style coordinates in all
Channel, DCS & SCS matrices.

This is what I did with Divisions Performer implementation of dPL you
may be familiar with, and why the implementation is able to easily
handle the Y = up coordinates of dPL.

There are simpler ways but performer coordinate stuff like the
isector partitioning & earth sky would suffer as a result (if you
decided you needed them).

BTW I think there is also me a row vs column difference for OpenGL, I'm
not sure about this. But it's easy to figure this part.

Cheers,Angus.
=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
            Submissions: info-performer++at++sgi.com
        Admin. requests: info-performer-request++at++sgi.com

------------------ RFC822 Header Follows ------------------
>From guest++at++holodeck.csd.sgi.com Wed Jul 16 05:19:20 1997
Received: from SGI.COM by uu10.psi.com (5.65b/4.0.940727-PSI/PSINet) via SMTP;
        id AA20715 for jedsall; Wed, 16 Jul 97 05:19:20 -0400
Received: from holodeck.csd.sgi.com ([150.166.145.108]) by sgi.sgi.com
(950413.SGI.8.6.12/970507) via ESMTP id CAA17157; Wed, 16 Jul 1997 02:19:06
-0700 env-from (guest++at++holodeck.csd.sgi.com)
Received: by holodeck.csd.sgi.com (950413.SGI.8.6.12/911001.SGI)
        for info-performer-dist++at++holodeck.csd.sgi.com id TAA11581; Tue, 15 Jul 1997
19:28:21 -0700
Received: from rock.csd.sgi.com by holodeck.csd.sgi.com via ESMTP (950413.SGI.8.6.12/911001.SGI)
        for <info-performer++at++holodeck.csd.sgi.com> id TAA11565; Tue, 15 Jul 1997
19:28:20 -0700
Received: from sgi.sgi.com by rock.csd.sgi.com via ESMTP (950413.SGI.8.6.12/910805.SGI)
        for <info-performer++at++relay.csd.sgi.com> id TAA10918; Tue, 15 Jul 1997 19:28:19
-0700
Received: from giraffe.asd.sgi.com (giraffe.asd.sgi.com [192.26.72.158]) by
sgi.sgi.com (950413.SGI.8.6.12/970507) via ESMTP id TAA11609
        for <info-performer++at++sgi.com>; Tue, 15 Jul 1997 19:28:19 -0700
        env-from (dorbie++at++multipass.asd.sgi.com)
Received: from multipass.asd.sgi.com by giraffe.asd.sgi.com via ESMTP
(951211.SGI.8.6.12.PATCH1502/951211.SGI)
         id TAA02024; Tue, 15 Jul 1997 19:28:17 -0700
Received: by multipass.asd.sgi.com (950413.SGI.8.6.12/940406.SGI.AUTO)
         id TAA27489; Tue, 15 Jul 1997 19:28:11 -0700
Date: Tue, 15 Jul 1997 19:28:11 -0700
From: dorbie++at++multipass.asd.sgi.com (Angus Dorbie)
Message-Id: <9707151928.ZM27487++at++multipass.asd.sgi.com>
In-Reply-To: "Jerry Edsall" <jedsall++at++fasainteractive.com>
        "Re: model view matrix and r" (Jul 15, 12:06pm)
References: <9707151916.AA15565++at++uu10.psi.com>
X-Mailer: Z-Mail-SGI (3.2S.3 08feb96 MediaMail)
To: "Jerry Edsall" <jerry_edsall.chic#u#pc++at++fasainteractive.com>,
info-performer++at++sgi.com
Subject: Re: model view matrix and r
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii

=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
            Submissions: info-performer++at++sgi.com
        Admin. requests: info-performer-request++at++sgi.com

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Aug 10 1998 - 17:55:37 PDT

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