Kill your Darlings.

[DanMarshall]

Inspired by This Article (http://gibbage.blogspot.com/2006/06/scenes-from-cutting-room-floor.html) I posted this morning, I was just wondering what features other people have had to cut out of their games. And why?

[Anthony Flack]

Oh man, so much.

Some level goals just didn't work. Whole styles of play that were meant to make up a certain percentage of the levels just didn't sit right. Not to mention over half the levels I've actually made so far which just felt ugly. And the way the levels are structured, too - in fact the whole meta-structure of the game has changed completely.

Bonuses and multipliers; various ways of racking up points had to go. If something was worth more points but I didn't want to do it because I couldn't be bothered, then it's time to rip it out. I'm still ripping stuff out all the time. Although a lot of the time, if I've added a feature I won't rip it out completely, but instead I'll make it a very minor feature instead of a prominent one - it's worth that much just for variety's sake I think.

In the end I think I'll have a much simpler game than I thought I would. It's actually getting much closer to how I originally imagined it in the very early days.Trying to be too clever can get you into trouble. I guess I had to try these things to find out they didn't work. It's teaching me a lot about player motivations and what is fun and what isn't.

[princec]

3D. Hahaha!

More seriously:
- Slot management. Turns out no-one cares anyway.
- Local hiscores & batching. Didn't have time to have both in there, can't be bothered to go back and fix it now.
- Some kind of crap DirectX fallback to gain me a tiny % of new customers.
- Action replay of later levels for use in nag screens
- Ultratron and Titan needed better "level progress indicators" (maps).
- Never did implement VBO for my sprite engine. Turns out VBO is still buggy and not generally any faster though, so I'm happy I didn't bother with that either
- Never got joypad support working. Too stupid to figure it out at the moment. My key binding code is crap anyway.
- The other 30 games I've thought of.

Cas :)

[Savant]

- Slot management. Turns out no-one cares anyway.
I've often wondered just how many customers actually value that feature.

- Never did implement VBO for my sprite engine. Turns out VBO is still buggy and not generally any faster though, so I'm happy I didn't bother with that either
I suppose it depends on how many polys you're pushing. I've found them to be a huge speed up in my test cases.

[Raptisoft]

Chuzzle had LAN and TCP/IP multiplayer. And it wasn't just watch the other person play, it was competitive where you could earn locks to throw at your opponent to lock him up.

I... I... I can't talk about it. *sobs*

[James C. Smith]

Ricochet Xtreme was supposed to include a level editor and a global internet high scores list. Both were cut because development was dragging on too long and we had to draw the line somewhere and ship the thing eventually. The high scores stuff was never started. Obviously, we had a level editor that we used to make the game, but I wanted to “cleat it up” a little bit before releasing it to the public. The bigger problem was that the game itself had no concept of level packs. All the levels in the game were one long linear progression of 170 levels. If the player made new levels, they would be level 171, 172 and so on and couldn’t be played unless you played the other 170 level first.

The level editor made it into the sequel (Ricochet Lost Worlds) along with the level pack support it needed. But the high scores thing never made it into any game I worked on. In Big Kahuna Words (BKW) I had the internet high scores stuff 90% working, but you know the second 90% always takes at least as long as the first 90%. It was never finished.

BKW originally had a "mad libs" type story that was automatically filled in with the words you made. I was thinking that “fish tales” would have been a fun name for it. I got it working but I could never tune it well enough to make the story make any sense. It would have been okay if the story was just silly. But instead, it had way too many grammatical errors since my "pat of speech" information in my dictionary was very simplistic. It knew noun, verb, adjective, adverb and a few others. But it didn’t know the tense of the verb and things like that. It would be easy to code support for all those, but very difficult to get that data for every word in the dictionary. We had other ideas to make the player pick where the words go rather than the computer doing it. That way that the human could pick the right grammar (or wrong if they preferred). It could probably have turned into something interesting if we kept working on it but we decided to focus our efforts on other things.

BKW was also planned to have a way to show the definitions for the words. This one we never really started development on. It was on the todo list forever and then one day it got cut when the game seems nearly finished and the todo list was just filled with extra bells and whistles that weren't needed.

Big Kahuna Reef 2 was going to have features to name fish, mark some as favorites, set some to never show up, and other ways to customize the screen saver.

[James C. Smith]

The things mentioned above are some of the bigger features that are easy to explain. But every game has always has dozens of things that get cut. I am not talking about wild brainstorming ideas we decided not to go with. I am talking about things we decided to do, added to the todo list, and often implemented some or most of it. And then later, these features were cut to get the game out and/or keep the game focused. It’s not that there is a deadline when the game has to ship. It just that eventually you have to wrap things up and call it done or it will NEVER ship.

I always say that one of the most important and hardest jobs of a producer is deciding what to cut. On every game I have produced I have cut lots of great features that could have made the game better and I am glad I did. One of my favorite saying is “save it for the sequel.”

[James C. Smith]

I forgot to mention, Wik on the PC had a Mouse Party™ mode that was playable but not finished. It got cut. The Xbox 360 version of Wik ended up having lots of “same console multi player” modes which is the Xbox equivalent of Mouse Party™. In Mosaic, the level editor (and player created content pack selector) was almost shippable and then got cut in the last month of development. Mosaic also had a mostly playable Mouse Party™ mode that was cut.

[Jason Chong]

For a moment there I thought this topic is about dumping your girlfriends so you can focus full-time on being an indie, uninterrupted..

:eek:

[yanuart]

Big Kahuna Reef 2 was going to have features to name fish, mark some as favorites, set some to never show up, and other ways to customize the screen saver.
that'll be awesome, as a diver myself I'll always want to make a game that also educate or tease the audience (subliminally of course) about the fascinating underwater world. It's not always about sharks and clown fish I tell ya :D
My plan is always stopped at the brainstorming session when no artists or member of my team can actually understand what I'm trying to bring inside the game and giggles upon hearing that the plural form of octopus is actually octopussies (there are other known form of it : octopy is one of them that doesn't cause silly jokes to emerge)

[DanMarshall]

For a moment there I thought this topic is about dumping your girlfriends so you can focus full-time on being an indie, uninterrupted..

It is. I'm crap at metaphors.

[Anthony Flack]

I always say that one of the most important and hardest jobs of a producer is deciding what to cut. On every game I have produced I have cut lots of great features that could have made the game better and I am glad I did. One of my favorite saying is “save it for the sequel.”

My experience has been that cutting things is not only necessary for practical reasons - it generally makes the game better. Even features that are good in themselves, and added value in some ways, can make the game a bit fat and unfocused. I had to lose a lot of stuff that, while perfectly good in itself, just didn't match with the main style of the game - and so it felt like an unwanted intrusion.

Now, that's my cue to go and trim down to a very simple end result, release and come up with a new, tighter design for next time.

[Mike D Smith]

My experience has been that cutting things is not only necessary for practical reasons - it generally makes the game better.

I totally agree, but it's oh so fun to put those extra features in (especially with only self imposed deadlines)! I like to make a priority list and try to stick with doing what is more important first and then if there is time, I'll get to the rest.

It takes a lot of dicipline not to add yet another shadow rendering technique or some cool particles effect you thought up the other night.

However, some times I just go ahead and indulge. After all, if making games stops being fun, then why are we doing it?

[MrMark]

For a moment there I thought this topic is about dumping your girlfriends so you can focus full-time on being an indie, uninterrupted..

:eek:

lol I'm one worse. I thought it was a thread about killing your children (not literally) so you can focus on your work :D

[GBGames]

Well, this this thread has come back from the dead, I'll say that I recently finished a simple Pong clone after spending almost a year trying to write a component-based game engine. I realized that I was just like one of those newbie game developers who want to make the "best MMORPG ever". It was a hard thing to swallow, especially when you believe you had "been there and done that" and so you should know better.

While the component-based game engine would be really nice to have, I really needed to start with the "Hello, world!" of games before I could be a good judge of what features would be needed and how I would implement those features.

On the bright side, I feel a lot less stressed since I'm not getting down on myself for not knowing better about how to get something as unwieldy as a generalized game engine finished. And now that Pong is finished, I can move on to something that might actually be somewhat marketable. Or not, if I decide to try another basic clone like Asteroids or Tetris.

[mot]

The original link is dead. Dan, where's the article now? I am intrigued by your
ideas and want to read more of the things that you write.

I don't feel bad for cutting stuff from my game. I'm not killing it; I don't want
to spoil the little brat. Also, not all of my ideas turn out to be completely
brilliant. Only some of them do.

[soniCron]

I find it extremely productive to use a game to drive the development of a framework. (Rather than spawning a framework from thin-air.) The originally released "beta" of Jeweltopia had no engine to speak of. (It was all hard-coded spaghetti.) This made adding or modifying features unbearable. In the last few months, I've been porting the game over to a more generalized framework that I write as I need features. I started with the base - the "core" class - and add components as I need them.

For example, input management was one of the last things to be coded because the existing code in JT was adequate for my needs at the time. However, when I decided to add keyboard support (for debugging - not playing,) I created an input manager that handled hooking keys to a function. Now if I want to print a debug message, I just use core.input.AddKeyDown(PrintCurrCoordinate, "F12"); and I'm set. It can get more involved (typematic settings, etc.,) but this is all I need right now, so this is all I coded.

There are still many features I'd like to add, but once I add a component to the framework, it's very reusable (in almost any situation I can concieve of,) and extendable. It may take a bit of time designing and developing each component (one or two days,) but the resuability makes it worthwhile. Since porting Jeweltopia to the new framework, my productivity has skyrocketted. For example, I added 3 major changes to the gameplay in a matter of 2 days, not to mention a plethora of "neat!" polish and effects. Now my bottleneck isn't the code - it's the media! :)

[ravuya]

I really wanted to release my latest with scripting, but after taking a look at the amount of work I'd have to do, and the amount of cool content I'd have to make to justify it, I just ended up releasing a "dumb" game.

Stuff like friendly NPCs were also cut out. :( Ah well, more for the sequel.

[DanMarshall]

The link is now http://www.gibbage.co.uk/2006/06/scenes-from-cutting-room-floor.html

I'm publishing to my own ftp now, rather than the old blogspot address... like a proper grown up.

[amaranth]

I ditched my action battles. It was nice to fight monsters directly on screen, but when you have four fighting characters in one party, it's a bit tricky to make sure they all get a turn to fight.

[GBGames]

I find it extremely productive to use a game to drive the development of a framework. (Rather than spawning a framework from thin-air.) The originally released "beta" of Jeweltopia had no engine to speak of. (It was all hard-coded spaghetti.) This made adding or modifying features unbearable. In the last few months, I've been porting the game over to a more generalized framework that I write as I need features. I started with the base - the "core" class - and add components as I need them.


It's good to know. I feel better about my "setback".

[Uhfgood]

My two cents (as if it matters),

I find they're not really my darlings at all. They're sort of like a cat with kittens, you can give them away because you know you won't be able to take care of them all.

On my current game I had to rewrite a little bit about how it handled because it was too boring one way, and you were keeping an eye out for this feature which basically meant the gameplay was kind of dragged down with it. Sometimes you just don't know until you've done it.

All of my (3 and 1/2) games have been like this. You think a feature will be cool and it turns out to be boring or too much or whatever. That is the process. Really that's what prototyping is all about. You think of lots of cool ideas try to implement them and if they don't work, drop them.

I mean a game is sort of like making dinner. Sometimes you want to experiment with new flavors. Let's put in this spice and see if it's good. Sometimes it's terrible so you order out, and sometimes you have nothing in your fridge but scraps and leftovers and it turns out great. Just like game development. Sometimes you have alot of cool ideas which are thoroughly undoable, and sometimes you don't have enough time, or don't think you have enough imagination and come up with something truely cool.

Game development is an art. Until it becomes a science there won't be a set way of making them.

[Davaris]

I just cut two things from my game in my last update that I put a lot of effort into.

My game is an RPG and the player can have up to 5 followers. Previously you could control all of the characters by clicking on them or their protraits. But to me the characters didn't feel alive when you could click on them and move them around. So I got rid of this capability and scripted them to follow the player and fight his enemies. I also make them moan when they are hurt. :)

My game also used to have real time and turn-based combat. I just cut real time combat out, because it was hard to use and I think having two systems is confusing.

[zoombapup]

A great deal of time on Worms, Karl would spend a night coding a particular weapon, play it the next day, decide it was crap and ditch it the day after.. rinse and repeat for the best part of a year :)

Almost all features are subject to "try and see" methods. I much prefer that. Have an overall plan, but always be prepared to cut something, because after implementation, sometimes things just dont pan out.

Right now, our main one for Air Ace, is the AI. Just dont have time to get it done for the first release, so no bombers to shoot down in v1.0. Next version will be super-special with Ai and stuka attacks and whatnot.