Previous Page: *** NONE *** This page: readme.1 (also called diss.txt) Next Page: intro/dissinfo.txt --------------------------------- BSc (Hons) in Computer Science Dissertation Archive '3D Stereo Vision and Realism in Computer Graphics' Ian C. R. Mapleson BSc. Last Change: 15th January 1997 Heriot Watt University, Riccarton, Edinburgh, Scotland. What is this archive? ^^^^^^^^^^^^^^^^^^^^^ This readme file describes an 'ASCII' (ie. plain text) version of my BSc Honours Dissertation in Computer Science, which I completed towards the end of June 1994. Also included are some extra GIF and JPG images, plus a few additional articles of interest. The text files in this archive have been designed for DOS systems. Many people have expressed interest my dissertation because the second half of it deals with the "side effects" of playing the wildly popular computer game 'Doom'. The first half of the dissertation deals with 3D stereo vision in computer graphics and, as a result, some will not be so interested in part one; however, the two areas are closely connected. Please read ALL of this file before deciding what to do next. How to use this archive. ^^^^^^^^^^^^^^^^^^^^^^^^ The rest of this page describes how the archive is structured, where the files are, how to use the archive, where to begin reading, what extra information has been included, points of interest, etc. Sorry it's a bit long, but I feel it's wise to take the time to make things as clear as possible. Unlike a book, it's rather easy for one to dive in at any point when confronted with a collections of files - I don't want my readers getting lost along the way. :) The dissertation archive as a whole consists of three separate zip files: diss1.zip (1376890 bytes) diss2.zip (1196420 bytes) diss3.zip (3649841 bytes) _All_ of the main dissertation files are in diss1.zip. The other zip files contain _additional_ image files which are NOT necessary to understand the basic dissertation. I have created diss2.zip and diss3.zip for those who wish to see further examples of certain subject areas highlighted in the dissertation. Hence, don't feel you _have_ to download diss2.zip and diss3.zip if you don't want to; they aren't required in order to read and understand the main dissertation contained in diss1.zip. I strongly recommend you do download them if you can though; the images contained in diss1.zip have been kept small to limit the size of the archive, but it definitely helps to see much larger images in order to fully appreciate a point being made. diss2.zip contains example real-time computer graphics screenshots, from various high-end SGI systems, mostly obtained from www.sgi.com. The single Reality Engine II image in the main dissertation (Appendix G, which is the file apendi/apendixg.jpg in the diss1.zip archive) is very old and in no way indicative of what today's SGI systems can do (SGI's latest system, the Onyx2, is between 10 and 100 times faster than the old Onyx: ten times faster for textured polygon rendering and 100 times faster for operations such as image processing). So, if you want a more up-to-date impression of what is possible with real-time graphics today, then download diss2.zip as well. diss3.zip contains further example autostereogramme images. Appendix N in the dissertation (the file apendi/apendixn.jpg in diss1.zip) is an autostereogramme of the Starship Enterprise. However, this image is not the clearest example I have seen; thus, should you so desire, you can download ten additional and much clearer images in the form of diss3.zip. No doubt some of you maye have seen the 'Magic Eyes' books which are large collections of autostereogrammes, or even the videos of moving autostereogramme sequences. As a result of the above, the rest of this document only discusses the structure of the diss1.zip archive, since both of the other zip files are just collections of images and need no such explanation. In case the archives are unzipped in the wrong order, they each have differently named 'readme' files, ie. readme.1, readme.2 and readme.3. NOTE! If you are unzipping these archives on a PC, then you MUST use the -d option with PKUNZIP since the archives will want to create directories. For example, you would enter: pkunzip -d diss1.zip Version 2.04g of pkunzip is available by ftp from: oak.oakland.edu in the directory: /pub/simtelnet/msdos/arcers and the file name is: pkz204g.exe This is a self-extracting archive. Run the program and the various PK utility programs will be extracted for you automatically. File Index ^^^^^^^^^^ This archive (diss1.zip) should contain the following directories and files (indentation used to show directory heirarchy and contents): README intro/ dissinfo.txt title.txt contents.txt abstract.txt part1/ intro.txt backgrnd.txt analysis.txt design.txt implemnt.txt testeval.txt concdisc.txt part2/ pt2sec1.txt pt2sec2.txt pt2sec3.txt pt2sec4.txt pt2sec5.txt pt2sec6.txt pt2sec7.txt pt2sec8.txt pt2sec9.txt pt2sec10.txt apendi/ apendixa.txt apendixb.jpg apendixc.gif apendixd.gif apendixe.txt apendixf.txt apendixh.gif apendixi.txt apendixj.txt apendixk.txt apendixl.txt apendixm.jpg apendixn.jpg refrnces.txt prgguide/ obstrcts.txt cube.txt adder.txt cobra.txt sorscode.txt proggide.txt extras/ alien.gif dragon.jpg extras.txt futrtech.txt futurevr.txt lockpnt.txt office.jpg stpeters.jpg vrnews.jpg If any of the above are missing, then you have received an incomplete version of my dissertation, in which case please email me with details and I'll tell you where to get an original version. Please do _not_ modify _any_ of the file names contained in this archive or move files around from one directory to another, especially not if you are considering giving a copy of the archive to someone else. This is because the file names and their locations are the means by which some parts of the text refer to other parts. If you change the names, you could lose your place when reading. You are free to distribute this archive as widely as you wish, as long as: a) the archive is distributed complete and in it's original form. b) no modifications are made to any of the text or images contained in the archive or to the structure of the archive. c) no money is charged for the distribution of the archive. You can print out the files if you want, but you might have some trouble with the colour images, as finding a good printer that can do them justice isn't easy. You can get good printing results by printing images as Greyscale, though. I recommend using XV v3.xx for this purpose. At some point I may construct a postcript version of this archive if the demand for such a thing becomes sufficient. Anyway, that's the legal nonsense done with. Dissertation Archive Structure ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ This archive actually contains a lot more than the original dissertation. As is natural when writing on such a subject matter as technology and computer graphics, there have since been relevant developments and so I have included extra articles and images which I'm sure you'll find interesting. These include articles about the Nintendo64 games console (N64), a list of World Wide Web (WWW) links to places with further information on the computer hardware mentioned in the dissertation (eg. the machines made by Silicon Graphics Inc.), and a number of other things. The dissertation is organised as a set of files that are loosely meant to represent different 'pages'. All the '.txt' files are in plain ASCII text format. I decided not to use any kind of word processor format as that would have prevented many people from reading them. The top and bottom of each text file contains information showing which file is the 'previous' page and which file is the 'next' page. In this way, you should always be able to keep track of 'where' you are in the archive. Look at the top or bottom of this file for an example. IMPORTANT: in order to prevent this archive being just a mess of seventy odd files, there are separate directories for different parts of the dissertation as well as for the supplementary articles. Note that all file names conform to the name length limit imposed by DOS (8 characters plus 3 character suffix). The various directories are: intro - contains introductory file, titles, abstract and contents page, part1 - contains all the files for Part 1 of the dissertation, part2 - contains all the files for Part 2 of the dissertation, apendi - contains all the Appendix files, prgguide - contains all the files which constitute the Programmers' Guide, extras - contains all supplementary articles, images, etc. Page headers and footers always show complete file paths. A complete index, listing where all the files are in this archive, is given above, so refer back to this page if you're not sure where to look for a file (the file intro/contents.txt also has a useful file-equivalences list for the various sections of the main dissertation text). For those cases where a page is actually an image file (ie. a GIF or JPG file), then the nearest previous text page will list all such image pages up to and including the next available text page. The dissertation in its original form actually begins with the file intro/title.txt; however, I recommend you read intro/dissinfo.txt first as it contains some explanatory information which I was unable to include in the original dissertation due to writing style, space considerations and time constraints. Also note that the files which are described as 'pages' in this archive often do not correspond to the actual physical printed pages in the original dissertation. This is because one page of original text takes up very little space as ASCII. Having a separate ASCII file for each page would result in over a hundred files! Clearly not a good idea. Hence, the various page fields ('Previous Page', 'This Page' and 'Next Page') at the top and bottom of each file have listed after them what range of physical page numbers they cover from the original text. The format is: [x-y] ie. from page x to page y, or just: [x] ie. just page x. where x and y are numbers or Roman Numerals. If a page field has no such information after it (eg. the 'This Page' field of the this file) then that page was not a part of the original dissertation text. This information should enable you to layout the files properly if you ever decide to print them out. It also allows you to see what files constitue the original dissertation as opposed to those which I have added later. Archive Construction Information ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Information for the curious... :) All the text files in this archive were created using the 'Jot' and 'xedit' editors on an SGI Indy R4600 (134MHz, 64MB RAM, 8bit XL, 20" monitor), with later work (everything after 1995) done on an Indy R4400 (200MHz, 64MB RAM, 24bit XL, 17" monitor). All scanned images were done using an HP ScanJet II, connected to a 486DX50 PC. All subsequent image processing work was done using XV v3.00, imgview and imgworks on the aforementioned Indy. The original dissertation was written on a 486DX2/66 using WinWord. My thanks to Heriot Watt University for permitting me access to the R4600 Indy, and to UCLAN for getting me the R4400 Indy for my office machine :) Oh, in case you're wondering why I only mention SGI when discussing real-time graphics, it's because their top systems are more than an order of magnitude more powerful than anything else available from any other vendor, period. This performance gap is likely to widen as time goes on, especially since SGI's purchase of Cray Research in June 1996. Please now go to intro/dissinfo.txt, which contains an introduction plus details of how to contact me should you wish to do so. Have fun! :) Ian. The Doom Help Service (DHS): http://doomgate.gamers.org/dhs/ FutureTech Reference List: http://doomgate.gamers.org/dhs/tmp/s/futrtech.txt SGI Network Admin, University of Central Lancashire, Preston, England, PR1 2HE. mapleson@gamers.org | Tel: (+44 -0) 1772 893297, Fax: (+44 -0) 1772 892913 "There is no magic, only stuff." - Nakor, "The King's Buccaneer" (R.E. Feist) --------------------------------- Previous Page: *** NONE *** This Page: readme.1 (also called diss.txt) Next Page: intro/dissinfo.txt