Path: chuka.playstation.co.uk!news From: "Alan Marshall" Newsgroups: scea.yaroze.programming.3d_graphics,scee.yaroze.programming.3d_graphics Subject: Re: Bounding box 3d collision detection / (delurk!) Date: Thu, 11 Mar 1999 19:06:56 -0000 Organization: PlayStation Net Yaroze (SCEE) Lines: 40 Message-ID: <7cdji8$cij22@chuka.playstation.co.uk> References: <7bc8t3$abn7@chuka.playstation.co.uk> <01be67ab$4e1c5220$0a3ddec2@sonia> NNTP-Posting-Host: ppp-1-12-19.access.telinco.net X-Newsreader: Microsoft Outlook Express 4.72.3110.5 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 Xref: chuka.playstation.co.uk scea.yaroze.programming.3d_graphics:356 scee.yaroze.programming.3d_graphics:1264 Ok cool, cheers, I'll play with that idea over the weekend. the problem I'm having at the mo, is that because I'm using the centres of the objects (they were created in SoftImage 3D) some objects have parts outside the grid. while other behave themselves :) At the moment I get funny blobbly effects when blocks merge, 'cos I'm trying a 3d array full of 0s and 1s approach, which is being more of a pain than it should be! Chears Al Sean Kennedy wrote in message ... >Use wall detection by cheating. Use the Ordering table and the resulting >Z-Sort perspective of the camera, to find the "Closest" to walls, then >just do the good old slide. > >Remember Q-bert? Or a retro reflective Surface? the closest walls are the >ones closest to the walls you cannot see --> the object you are putting >into place. >Since your camera is already doing the Z-Sort, you just want to monitor that. > >:) > >Then it is a vertice check to make sure you do not lose your 3-d object IN >the completed Tetris tiles. > >2-3 walls and a vertice check. Faster than a 4 point vertice check and an >overlap jitter increment decrement. > >Sorry no source. It's a bit hairy with 2 matrices and the vertice in >"Freemove" areas. But think about it. Then try it. > >-sean