(This article is meant to be a collaborative effort, rather than written by a single person. Please improve it if you are able to do so!)
What demos are “really about” is as contested as always and taste differs. No matter what standpoint we have in this issue, everyone would agree that one of the important aspects (historically) of the art of demos has been to push the capabilities of the machine. That is, many demo effects have been aiming to do things previously considered impossible, or things previously not even thought of, or to do something better than someone else. This article is an attempt to collect such information into one single place. Curiously, this has (to my knowledge) not been summed up in a systematic way before, apart from being stored in the freaky brains of various demo coders. :) The aim of this article is to teleport this info from the brains of demo coders (and others), into an article right here on this wiki!
As one could expect, it is not always straightforward to determine who did what and when. For example, someone may have discovered something by accident, and was perhaps not able to use the discovery in a completely bug free way, while someone else managed to do that at a later point. Another example; it may not be clear whether two effects should be counted as being “the same” or not, even if they are both using some quirk of the VIC chip in a similar way (i.e., the quirk may be used for achieving completely different things). This means that some of the information here may not be accurate (in that case you should improve this article), and it also means that some of these issues are subject to dispute (for example who was really the first to open the borders?). In cases of dispute, please do not just list one release as being THE first, but list several and add a note about the nature of the dispute. In fact, it is a good idea to add several early occurring instances of something, since this will avoid the question if another often mentioned demo was first.
So, please, go ahead and fill this article with info about these matters. For example, correct misstakes, give a brief definition of some effect, add links to the relevant productions or to source code for achieving the effects.
This section should list the first time, or times, a certain trick appeared or was invented. Examples would be who were the first to open the border(s), who invented sprite stretching, and so on.
To start this list off, I (FTC) took a little list posted by someone called Rost at the CSDb forum and pasted it below. I take no responsibility for it. As said, if you find incorrect information below, please improve this article. (And some people have done so already, so the list below is not 100% Rost's list.)
Is this the one? Release: 1985-10-11
TSI/1001: Border Letter I
Thomas Larsen: “The Vikings” - game. (“10 points” full screen FLD)
Release: before 1987-03-xx (according to Rost; probably february according to Rough).
Thomas Larsen: “Tiger Misson” - game.
Release: before 1987-03-xx (according to Rough; probably february)
Mule/The Cream Crackers: Mega Jive
Release: 1987-03-xx - “around march 1987” (according to graham).
The Omega Man/Teeside Cracking Service: Game Over - cracktro
First use of FLI as moving-color effect
ASP/Blackmail: So-phisticated III
First FLI picture (not 100% sure this was the first one, anyone knows better?)
Bones, Triangle 3532, Upfront: Random
Pernod/Fairlight: Rutig Banan
SounDemon made some “new waveforms” coming out of the sid by using an undocumented quirk of the SID chip to be able to reset the noise waveform. The “new waveforms” where created by doing this periodically, and thus getting non random waveform(s), based on the noise waveform data sequence.
As stated above, people tend to compete with each other and try to show that they can do an effect better than someone else. This section belongs to these kinds of deeds. Examples would be highest number of sprites on screen simultaneously, longest scrolltext in a onefiler demo, and so on.
Crest: 2 Years Crest
112 normal sprites, was this really the first 112-sprites?
Release: October 1990
Crest: Ice Cream Castle
120 normal sprites, but is this really a valid record? Crossbow writes long texts about why other attempts to do more than 112 sprites are not valid, based on the theory that a sprite has to hold one single object (not be split into 2-3 objects). Considering the height of the screen and the size of (normal) sprites, it's easy to calculate the maximum number of *whole* sprites to 112. Another 8 sprites can of course be displayed partly, but does that count? Crossbow says ”..but with a little trick” you can have up to 120 sprites, but he still has not explained how. So let's say INVALID RECORD until we have a better explanation.
Release: April 1991
144 sprites, using sprite crunching.
Megastyle: The Great Kloakkman
$1dcbb characters (the scroll states “bytes”) long scrolltext in a single load
Bones: The Larch 3
600 sine-dots moving in Y direction over multiple charsets
672 sine-dots moving in Y direction over multiple charsets
600 plotted sine-dots moving in X and Y direction
Oxyron: Natural Wonders
960 plotted sine-dots moving in X direction
Noice: Teen Dreams
1123 plotted sine-dots moving in Y direction in a 16×16 char area (128×128 pixels)
Camelot: Pimp My Snail
1536 plotted sine-dots moving in X direction over a full screen