Path: chuka.playstation.co.uk!news From: James Rutherford Newsgroups: scea.yaroze.programming.2d_graphics Subject: Re: GPU Packets... Date: Tue, 24 Feb 1998 15:01:40 +0000 Organization: The Hex Heroes Lines: 17 Message-ID: <34F2E0D4.222D@sms.ed.ac.uk> References: <6ctab6$9866@scea> <6cu4g2$as5@chuka.playstation.co.uk> <6cul1b$aq01@scea> NNTP-Posting-Host: atl4-008.publab.ed.ac.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 3.03 (Win95; I) Jeff Lawton wrote: > So I figure if they are needed for sprites there must be some logical > calculation we could use to figure out our work space. If they have > n*24 per poly there must be something similar if we are using sprites. I think my investigations led me to n*26 for sprites. I assume this corresponds to 'approximately n*24' - it would kind of make sense if all the packet entries were the same size, and 24 seems like a nice candidate... If you want to carry out your own tests, set up loads of sprites and adjust the packet area size until things start to freeze up. As to what the packet entry is actually /for/ - I've no idea. James (~mrfrosty / SCEE)