Path: chuka.playstation.co.uk!news
From: "Rikki Prince" <netyaroze@rprince.dircon.co.uk>
Newsgroups: scee.yaroze.beginners
Subject: Re: Sprites, OTs and dodgy wrappers...
Date: Thu, 30 Sep 1999 23:38:50 +0100
Organization: PlayStation Net Yaroze (SCEE)
Lines: 44
Message-ID: <7t0oqe$2d02@chuka.playstation.co.uk>
References: <7su34m$7f82@chuka.playstation.co.uk> <86n1u5jol3.fsf@foo.mel.cybec.com.au>
NNTP-Posting-Host: th-gt142-194.pool.dircon.co.uk
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 5.00.2014.211
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211

Well, I've been persisting with this, and it seems that the main loop goes
through once, and dies when it gets to the GsSortSprite on the second loop.
This further proves that it must be a problem with the address of the OT
that I'm passing, but I reckon - due to this evidence - that it must be the
double buffer effect that is causing problems. I don't think it is the value
in the active buffer integer, as using an int (all global btw. Toby ;)
separate from the structure still doesn't work. This must mean it's
something wrong with the OT itself. There's a number of things this could
be:
1) Not passing the address of the current (which in the double buffer) OT
correctly. At the moment, I using: &thisGame.ot[thisGame.ab] which from what
I gather should be the address of one - depending on the .ab value - of the
array of GsOT structs. I can't see anything wrong here, please correct me if
this is where I'm mucking up.
2) Something with the OT Tags. At the moment, in the struct, I'm setting a
pointer to GsOT_TAG, then in my struct initialisation function, I use
realloc, to allocate memory to the size of:
2*(1<<otLength)*sizeof(GsOT_TAG), which is equivalent (I hope) to the amount
allocated if I used GsOT_TAG otTag[2][1<<otLength]. Again this is way off,
please tell me. I really need to allocate this memory dynamically, as the
size of the array depends on the OT length passed to the init. function.
3)Similarly, I may have a problem with the PACKET area, as this is being
dynamically allocated, as well. Is it a good idea to be dynamically
allocating this memory, or should I lessen the flexibility of these
functions etc? Just to check, if I've got a pointer inside my struct, and I
need to pass the address (which has been set to that pointer by realloc() )
do I just pass: struct.pointer or do I need &struct.pointer. (for this
example, say the struct had been set up as:
        typedef struct
        {
                int *pointer;
        } ASTRUCTURE;)

Again, if anyone could aid me with this, I'd be extremely grateful. If
anyone thinks some source code would be of much help in solving this, I
could try to sort some out, though, it's a bit messy, with not enough
commenting.

Thanks
Rikki