The Cutting Room Floor

All ideas are not created equal. If you follow me on Twitter you will be privy to the more detailed aspects of Waves development and will have seen lots of screenshots of stuff that is in development and in some cases been cut from the game.

I thought I’d share some of them with everybody else along with why they might have ended up on the cutting room floor. This is a bit like the Deleted Scenes part of a DVD.

If you want up to the minute updates on what I’m working on then you should really follow me on Twitter.

The Sniper

Owly Images

This is going back quite a while now to before the background art existed and enemies were still just particles.

The Sniper was inspired by the lasers in Gridrunner and was supposed to recreate the effect of splitting the arena into quarters periodically with an impenetrable beam of death. The beam particle effect was pretty cool and I spent longer working on that than I did the actual enemy behaviour.

The Sniper was glued to the walls of the arena and would zoom around trying to keep the player targeted – every so often it would stop to charge and then fire it’s beam weapon. It worked and it did exactly what I wanted it to in creating moments of tension where you couldn’t get to the other side of the arena. There were 3 problems though:

  1. You could only have one Sniper at a time due to being on the arena wall minimising the angle to the player meant they all ended up in the same place.
  2. You couldn’t kill it at all.
  3. While the beam effect looked good the enemy itself looked ass. I had no idea what to make it look like. This is one of the times I wish I had an artist on board.

So for the above reasons the Sniper got canned. I don’t miss it because with the game the way it is adding it back in would require significant changes to how the game plays to compensate.

The Sniper may return as a boss but no longer tethered to the wall.

Gratuitous Lightbeams

Owly Images

The gratuitous lightbeam background effect.

This was an indulgence and something that looked way cooler in my head than in reality. Wait that’s wrong – it looked awesome in reality but it impeded gameplay massively. You couldn’t see anything that was over the lightbeams and it ruined the contrast between enemies and background.

It did however lead to the Hex based background that the game now has so it was absolutely worth it.

Owly Images

The Bouncer

Whatever happened to the Bouncers?

They were the purple baddies that were funded by the biscuits sent by the Rock Paper Shotgun grant comity (Quinns).

They would bounce around the arena merrily getting in the way but nobody ever got to see them moving except me.

They are still around!

Bouncers have been on my Todo list for ages now and I finally got around to fixing them not long after I put out the last video. They are now red not purple.

The Bouncer had one big problem – The Arena. Being a circle they tended to end up bouncing along the sides of the arena after a short while rather than as I had hoped off the walls and back to the opposite side. This made them entirely non-threatening so I took them out of circulation for a while. The fix was to have them plot a route through their origin (Normally the middle of the arena but sometimes where the player was when they spawned) and keep going in that direction until they can go no further. Then they turn around and go back the way they came. They are now much more dangerous and I often find myself getting hit in the side by one.

Scaleform

Ah the old UI. It was very ugly and got replaced by one that is much less ugly. It was also using Scaleform and I have no idea how Flash works so it was a real pain in the arse to implement. I have since stopped trying to use Scaleform and wrote my own UI system that is much simpler and easier to work with.

Outbreak

Outbreak is a game mode that I’ve actually started to dread working on. It was conceived of long ago before any of the games mechanics had stabilised into their current form and was based on the idea of the player trying to survive as long as they can against an incredibly aggressive version of the green self-replicating “Virus” enemies in the other mode. It sounded great when I had just finished the virus and had things like this happening:

Owly Images

The problem is that in the modern version with combos, bombs and some much more interesting new enemies the Virus has become more of a background thing. Yes it’s massively integral to getting a good score but it’s not exciting. On it’s own it’s very dull to play against and with bombs you can clear the arena of every last one in no time at all. It’s just not dangerous.

So the Virus is like salt. It makes whatever it’s added to taste better but on it’s own it’ll just dry you out.

Outbreaks future as a game mode is questionable. I don’t enjoy working on it and I don’t enjoy playing it but it nags at me that there must be a seed of something in there that might be worth developing. It’s not dropped yet but it’s on the list.

Multiplayer

I only ever mentioned this and never actually took any pictures but I did at one point in the past have online multiplayer working with 4 players.

However I’m going to tell you why you won’t see a game like Waves shipping with online Multiplayer anytime soon.

The internet hates it.

The problem is one of packet loss and bandwidth. With your Unreal Tournament or your Quake 3 you have a dozen players interacting with each other and that’s about it. Most of the time they aren’t even interacting or can see half of the other players on the server. This makes them excellent for the internet as you can optimise this massively. If two players can’t see each other and have no possibility of affecting each other then you don’t need to send them updates about the other person. This is called “Being Relevant” and you see it in MMOs alot too. MMOs will only show you the 50 or so nearest things (enemies/players) because it saves on bandwith.

At any given point in Waves there are around 100 or so things that are relevant to every single player (including the other player) and not including bullets. Not only are these things relevant but they are unpredictable as well. So many things change so frequently during a game of Waves that with just two players logged in on the same machine the clients couldn’t stay synchronised. If you put that on the wild Internets then packet loss is going to make the game unplayable very quickly.

Could I make Waves work online? No. I wouldn’t be able to claim that the online and single player versions are the same game because I would have to reduce the number of enemies and speed of the online game in significant ways that would in fact stop it from being anything like Waves.

2 thoughts on “The Cutting Room Floor

  1. I beleive multiplayer in such games is possible, but most likely not on Unreal engine and not soon. The game can be fully deterministic (including random events generation), so that each computer simulates everything independently (and has equal results) and just sends and receives mouse and keyboard input. The input also has to have some global timestamp with it, and after receiving input, one has to discard all simulation after its timestamp and recalculate it with the new input data. I think getting such design to work well would be a challenging engineering task.

  2. asd: The method you describe does work for some games, I’ve used it myself in a game where players aren’t usually too close to each other. It breaks down a bit when there’s much lag and the players are close together. I don’t know all the details of Waves, but if the enemy movement depends a lot on exactly where the players are, the simulations could diverge pretty quickly and you’d really perceive the lag.
    (At least Waves is fairly continuous, so you could smoothly interpolate the other players to their correct positions; my game was not.)

Leave a Reply

Your email address will not be published. Required fields are marked *