Path: chuka.playstation.co.uk!scea!peter_alau@playstation.sony.com
From: Gerrit Goossen <gerrit@funnytown.com>
Newsgroups: scee.yaroze.programming.3d_graphics
Subject: Re: gteMIMefunc & me: One of us is wrong - Solution!
Date: Fri, 13 Nov 1998 20:18:19 -0500
Organization: SCEA News Server
Lines: 35
Message-ID: <364CDA56.3465@funnytown.com>
NNTP-Posting-Host: 38.165.241.13
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.01-C-MACOS8 (Macintosh; I; PPC)

James wrote:

>Gerrit Goossen wrote:
>> 
>> Is there any chance that making repeated requests to the GTE to perform
>> operations on the same data could be the problem? (Does the GTE que
>> requests like the GPU?)
>
>Nope.

It was worth a shot!

>Try experimenting with known values, for a single vector and see if you >get the results you expect.

I tried this, and determined that not matter what value was being
translated (0, 1, a vertex coord, etc..) the result was getting thrown
off... When using the MIMe array my tool generated.

I thought about this and I think Craig had the right idea when he
suggested scaling the differential vectors up, but unfortunately that
wasn't helping this case because the actual difference between vertices
in frames was too small! (The models are on the small side, and some of
the motions are subtle.)

So I made the models bigger :)

I scaled all the models up 4 times, added a scale-down factor to the
local-to-screen matrix just before calling GsSortObject4() and the
problem has disappeared. (I believe it's actually still happening, but
the effects aren't noticeable.)

Thanks to James and Craig for the help and giving me plenty to think
about! Hope this information is useful to someone else out there...

- Gerrit