Last week I talked about using polish to turn a good game into a great one, but I deliberately avoided giving examples. That post was already 3,000 words long and there were too many disparate topics to cover that there wasn’t room to enumerate, explain, and quantify what I was talking about. So let’s do that now.
The point of the exercise is to come up with games that felt like they could be vastly improved by just a modest investment of additional work at the end of the project. Let’s imagine we’re aiming for stuff that can be fixed in ~6 weeks or less. Six weeks might sound like a long time, but in terms of AAA game development it’s not all that much. 18 months (78 weeks) is a pretty common development cycle, which means six more weeks is less than a 10% increaseWhich is probably a lot less than a 10% increase in budget, if we assume the team is reduced in size once you enter the polish stage.. Obviously this isn’t always possible. Sometimes the money just isn’t there, and sometimes you need to hit that ship date for Christmas. But if we find ourselves in a position where we can make a huge improvement to the game for only a 5% or 10% increase in budget, then that’s a move we want to make.
This short window limits what we can and can’t do. Obviously you can’t do major story re-writes or add detailed voiced characters. We can’t call famous voice actors back and have them re-do all their lines. Even if scheduling isn’t a problem, that’s not the sort of thing you can do cheaply. You probably can’t make sweeping changes to cutscenes, although maybe you can tweak things like props, timing, camera angles, and musical cues. Bonus points if we can improve the game by simply removing stuff that doesn’t work and still ship a complete experience.
We also can’t do major re-designs to the gameworld. No, we can’t completely change the layout of the Doom Fortress at the end of the game. But maybe we can remove the stupid hedge maze everyone hates. We can’t add all new guns to the game, but we can tweak what we’ve got. We can’t add a whole new village, but we could change an existing village so the player doesn’t constantly get caught on little bits of scenery as they walk around. We can’t redo all the sound effects, but maybe we can add or change a few.
So those are the ground rules. Here’s my list of games that could have been far better with just a little more effort.
You can’t fix a lot of the big problems with Rage this late in development. John Goodman’s character is our big introduction to the world and he’s monumentally flat and boring. The story gives us no motivation to oppose the “bad guys”, there’s nothing to entice the player to care about the “good guys”, and all of the characters are dull exposition dispensers.
But there’s still a lot we can do. The dialog is so verbose you can cut a bunch of it without hurting anything. If we can’t make the story good, we can at least make it concise.
We can probably tweak the weapons a bit to give them more punch. A few of them feel so lifeless that they aren’t worth using. Even if all we do is make the sounds boom a little harder and the foes ragdoll a little more, we might be able to give combat a little more kick.
But most importantly – and we should probably start with this one – we need to fix this ending. No, I’m not talking about the story ending. That’s a lost cause this close to release. I mean the ending fight, which involves gunning down more of the same freaks the player has been killing since the introduction. There was a great big zombie kaiju fight right in the middle of the game, and the finale is a mook shooting gallery? In an iD Software game? How did this happen? This is madness!
We need a new monster of similar stature for the finale. If we don’t have time to build a whole new monster, then we need to copy this one and give the second one a twist. If we can’t do that, then we need to swap the mook fight and zombie kaiju around so the big monster is at the end.
We can’t fix the most obvious problem with this game, which is that the tone is all over the place. Marketing made it look like this was Last of Us. Whoever made the opening cutscene thought the game was Saints Row. The in-game cutscenes and quests vacillate wildly between stilted David Cage style melodrama and joke-y nonsense. The gameplay is going for something over-the-top like Borderlands, and the visuals are aiming for photorealism and failing badly. Did we ever know what kind of game we were making?
While we can’t give the game a proper coherent identity, we can fix a lot of the really egregious failings.
We can get a massive gain in visual quality by disabling that stupid specular shader that makes it look like the entire world is glazed in sunscreen.
Right now the leveling system is self-defeating. Zombies always auto-level to match the player. Which means if you’ve lucked out and acquired a rare weapon, the worst possible thing you can do is level up, since it will make your awesome weapon less useful. We can just make various zones in the game locked to stay within a certain range. That’s a lame brute-force solution, but it’s easy and it’s much better than shipping the game with the existing system.
Also, players collect cash by looting containers, and then they go to a workbench and repair their weapons using handfuls of money. Who thought of this, and why didn’t we fire them on the spot? We could make this 100% less stupid by simply replacing the “cash” icon with spare parts, duct tape, screws, or whatever. It’s a five-minute fix. Stop this madness.
There are also a bunch of dumb interface problems we should hammer out. We need to give the player a way to disable the nagging multiplayer join invitations that appear during gameplay. Let’s get rid of the giant distracting message warning that you’re leaving the game world when you’re near the edge of the map, even if there are fences around that make leaving impossible. We can make our character models look less terrible by not ZOOMING IN ON THEIR DEAD-EYED FACES when the player begins a conversation. Finally, let’s re-write these cringe-inducing character bios and replace them with something that just describes the character from a gameplay standpoint.
The cars in this game are buggy as hell and their driving physics are terrible. It feels like a driving game from the 90’s. If we can’t fix this, let’s just remove the driving gameplay. The world isn’t that big and driving isn’t that useful anyway.
Oh, and then we should fix all these bugs. Those are pretty bad.
Deus Ex: Human Revolution
I have no idea why we outsourced our boss battles to another developer. I guess hindsight is 20/20, but this other firm doesn’t seem to understand anything about our game. It’s too late to give these characters the re-write they deserve, but as a band-aid we can add some sneaky / hacking type thing to allow players to skip these fights. If we can’t fix thist before release, then maybe someday we can fix it in a Director’s Cut or something.
Even if the player decides to fight naked muscle man Namir, this fight would be ten times better if Namir would shut his stupid mouth during the fight. These combat taunts make no sense, repeat endlessly, and add nothing to the character.
Can we get a few extra dialog animations for the secondary NPCs? That one where they make chopping motions in the air in front of them is really janky, and it gets used a lot.
It’s too late to really fix the colors in this game, but can we at least dial back on the MAXIMUM ORANGE filter?
And finally, is there anything we can do about this ending in just a couple if weeks? I know it’s too late for a re-write and new dialog, but can we have the player choose an ending by using an object (blowing up a transmitter, turning on a generator, opening up a flood valve, etc) rather than choosing between three buttons positioned side-by-side in the same closet? I realize this means making a little more game space, but it would really make the ending choice less blunt and videogame-y.
First we should fix the combat. Er, maybe we should fix the fussy base-building interface instead. No, the Preston Garvey dialog. Or the broken quests. The wonky weapon balance. The nonsensical skill restrictions. The strange difficulty spikes. The pointlessly immortal NPCs who aren’t even relevant to the main plot. The ponderous intro. The glitchy scenery. The terrible AI. The level scaling that prevents you from ever encountering late-game foes and gear until you’ve leveled up enough, no matter where you go. The bad lighting. The dialog glitches. The bad hitboxes. The complete inability to make meaningfully mix-maxed specialist characters. The lack of an option to spare Kellogg. The fact that the entire story falls apart if you blast Father as soon as you meet him, even though that’s a really interesting and understandable choice. The…
Oh wow. You know what? Never mind. We’ll never put a dent in this list. Just ship it. Let the modders fix everything. We’ll sell ten million copies no matter what.
What Did I Miss?
This was a fun exercise. Try for yourself. What game do you think could be greatly improved in a short time for not too much additional cost? What games could have been far better if they’d been given just a little more time to gel?
 Which is probably a lot less than a 10% increase in budget, if we assume the team is reduced in size once you enter the polish stage.
The true story of three strange days in 1989, when the last months of my adolescence ran out and the first few sparks of adulthood appeared.
The Biggest Game Ever
Just how big IS No Man's Sky? What if you made a map of all of its landmass? How big would it be?
Quakecon 2012 Annotated
An interesting but technically dense talk about gaming technology. I translate it for the non-coders.
The Best of 2013
My picks for what was important, awesome, or worth talking about in 2013.
C++ is a wonderful language for making horrible code.