This may be a slightly strange subject for LockerGnome but I found it interesting. Fellow Gnomie Kat Shaw and I have started designing test chambers — after helping Matt Ryan with his article yesterday — and we’re trying to make them as devilishly deadly and difficult as possible. The test chambers we’re designing will use momentum elements. Momentum is the way you make large jumps without the use of the aerial faith plate, light bridge, or excursion funnels.
The way to gain momentum is by shooting a portal into an opportunely placed hole — with a portal surface on the bottom — then shoot another portal onto a portal surface above you. Thus, when you jump into the first portal (the one that’s in the opportunely placed hole), you keep all that momentum as you pass into the portal that’s above you. This will then shoot you across an expanse of toxic goo, bottomless pit, or other death trap. The short answer is: speedy thing goes in; speedy thing comes out.
The problem with this is that there is no place to find a calculation or the VALVe algorithm to define how far the player will travel when falling into and out of the portals. This lack of information — as much as you can fudge with your chamber design with trial and error — isn’t good enough for us. So we came up with a simple test chamber that allowed us to make some interesting discoveries. However, first we’ll go through our chamber and the science before we get to our conclusions.
Kat and I built our own chamber to test the effects of momentum within Portal 2. This Portal 2 chamber is avaliable to download, scrutinize, and have fun with. The maximum chamber size within the Perpetual Testing Initiative is a 25 by 25 cube. This is the size Kat made and the chamber design was Kat’s as well. The design of the chamber is made up of a staircase-like structure that reaches from the top to the bottom of the chamber with a platform at the top, pictured below. You will notice that there are non-portal surfaces. These non-portal enabled surfaces are there to aid the calculation of distance travelled.
Momentum is mass times velocity, but we don’t know mass or velocity. We can guess that VALVe probably used a generic mass for an adult human. Some of you may think that Chell was chubby because GlaDoS kept refering to her weight. However, near the end of single player when Wheatley decides to play that saying “fatty, fatty, no parents” GlaDoS — the potato — says “look at her, you idiot, she’s not fat.” So generic weight seems to fit. The only problem here is that you’re not playing as Chell and neither are you playing as Atlas or P-body — the two robots for the co-operative testing initiative, aka multi-player. You are playing as a generic stick figure. The other point is that we have no idea about velocity.
Kat and I, via testing, have found out that jumping into a portal between 10 and 25 blocks off the ground will result in your character reaching terminal velocity. How can you possibly call terminal velocity? The graph below shows the average of the tests that we performed. As you can see from 10 to 25 blocks, the distance traveled via momentum is the exact same. The main difference in where you land from this momentum is when we started jumping from five blocks and below. The reason that the chamber is blocked off in fives is because, when we jumped in singles, there wasn’t any change. As such, we decided to make the job easier on ourselves and run in fives.
I then thought about proving terminal velocity, and it turns out that either VALVe fudged the numbers to get a good game, or my assumptions are wrong. Either way, the height you have to fall to reach terminal velocity is around 490 meters — or 1,600 feet. I came up with a chamber tile size of around seven or eight feet because your character would probably be around an average height of 5’8″ tall and you can’t see over a single tile when you are standing in front of it. These may be fudged numbers on my part, but they certainly seem to make sense in my head. Assuming for 8′ high tile sections, multiplying that by 25 tiles gives us 200 feet — or nearly 61 meters. I hope this isn’t moving too fast for you guys.
We have published the chamber for public consumption and our results will be published at the bottom of this page. I would like to point out that the published results will be all nice and neat and tidy in giving the average numbers. I wouldn’t honestly think you would be interested in the full set of results. However, our conclusions when it comes to the science of momentum in Portal 2 is that VALVe probably fudged the numbers for the good of the game. However, the numbers that we publish below should help anyone build momentum into their test chambers without too much trial and error to get the chamber right. We searched for the VALVe algorithm or for any kind of information regarding momentum and we found nothing. Hence this article and why we spent all this time testing the function of momentum.