Game Feel - Part 2

Why some games feel better than others – part 2

part 1 can be found here

Jumping

I chose jumping for this article as I think it’s one of the most basic functions that has been in videogames since the beginning that is still being used today. Games as old as Donkey Kong and as new as GTA all feature jumping. It has gone from the main gameplay focus in games to one of many options that your character can perform.

Games have changed from 2d to 3d and jumping has remained a constant. Even as games shifted view from 3rd person to 1st person, jumping has hung around no matter what the current trend in videogames. Something so innate to videogames and yet read any videogame magazine and I’m sure you’ll find a review of a game where people are complaining about the jumping. A basic functionality of videogames and developers are still getting it wrong.

I don’t claim to have the perfect answer but when studying jumps in as many games as I could find, I noticed one common thing. How long the character was in the air regardless of 2d or 3d, this determines if a jump feels good or not. How long the player is out of control as their character floats around is very important. Oddly enough good jumps were around the same time and bad jumps were around the same time length.

I fully admit that my tests are not the most accurate in the world but it’s a starting point for a conversation. A conversation that hopefully developers can have with one another as they have their character jump and try to nail that elusive feeling of ‘good’.

My Testing Methods

I took a stopwatch and pressed start as soon as I pressed the jump button. I then pressed stop right when the characters feet touched the ground again. I did not account for any settling that the character may do in the animation. Most games allow for the player to navigate out of this settling animation so I discounted it completely. Also since I have only 2 hands I only tested the character jumping straight up while standing still. Some games have different animations for jumping straight up or jumping toward an object and I did not test those. Since it is a very imprecise way of measuring I repeated the same jump around 8 or 10 times and what you see is my average of those attempts.

Note that a lot of games which I consider to have great jumps have controllable jumps depending on how long you hold the jump button down, such as Super Mario Bros. When testing those I tried to take an average of how long the average jump is performed. Yeah, not the most precise thing in the world but like I said I want this to be the beginning of the conversation, not the end.

My Testing Results

Time is measured in seconds.

First lets take a look at some NES games

​Super Mario Bros

• Super Mario Bros 1 – 0.75 seconds
• Super Mario Bros 2 – 0.65 (I only tested Mario)

• Super Mario Bros 3 – 0.75 seconds

• Castlevania 1 – 0.72 seconds

• Castlevania 2 – 0.68 (looks like they removed the slight hesitation at the apex of the jump)

• Castlevania 3 – 0.72 seconds

• Contra – 1.03 seconds

• Ghosts and Goblins – 0.7 seconds
• Ninja Gaiden – 0.9 seconds
• Rush N’ Attack – 0.9

Now with just a quick look at these results I think you can see a pattern emerging. Good jumps fall within the 0.7 category, anything longer and you start having a jump that no longer feels good.

Now for some Playstation 2 games

• Devil May Cry 3 – 0.75
• God of War – 0.69
• GTA3 – 1.10
• Jak & Daxter 1 – 0.72
• Psi Ops – 1.10
• Ratchet & Clank 2 – 0.75
• Shadow Of The Colossus – 1.10
• Shinobi – 0.85
• Sly Cooper 1 – 0.8

I won’t lie, I was kinda shocked to see that going from 2d to 3d a good feeling jump remains the same as far as time in the air is concerned. Personally I think that a good feeling jump outweighs any real world animation or scenario that your game is trying to present. Who cares if it feels or looks “arcadey” – players expect and need immediate response when it comes to jumping.

My numbers aren’t the most accurate thing in the world but I think it’s safe to say that somewhere between 0.70 and 0.80 is what most people consider a good feeling jump. Anything longer than this and the player starts to really notice how out of control they are during a jump. Anything less and either the jump serves no purpose or feels even worse than being out of control.

Game Feel - Part 1

Frames

Before we dive into this lets just get this out of the way – this is a very long and complicated subject to tackle. When I talk about feel – I’m trying to describe how a game plays. When you first pick up the controller and you move the character around or jump – does it feel good? Does it feel like you expect it to?

This isn’t even relegated to jumping or moving around but those are very basic functions that I think most games get wrong. One of the worst things about making videogames is that you have to re-invent the wheel with almost every new project you work on. So even though Mario jumps like a champ, when you go to make your game it’s very hard to reverse engineer Mario’s jump and port it into your game. Many games spend months on just getting their characters navigation down – going from idle to walk to run, etc. Seems like a simple thing that you should just be able to plug in some middleware entitled, ‘good navigation’ and throw your character model on top but it doesn’t work that way. I heard a story about a certain coin-op developer who did a lot of racing games in the 90’s, they had a completely separate engine for all of their racing games (which I think was around 4 different series) and none of the teams spoke to one another.

Because you have to reinvent the wheel each time out, you get people who either don’t care or simply don’t have the knowledge of what it takes to have their characters feel good when a player first picks up the controller. I completely admit this is a huge task but I think it’s one that most developers gloss over. One of my favorite Shiggy Miyamoto stories is him coming to America to play an early build of Excitebike 64. He asked the developer for an empty track with no hills, jumps, curves, etc – basically a blank road on which to drive the bike around. He picked up the controller and within minutes threw it on the desk and probably mumbled something in Japanese about how all American game developers suck. His point was that if you can't have fun in an empty room with no toys, then nothing you layer on top of it will salvage the game. I liken it to construction – if the base isn’t solid then the rest of the building will soon collapse.

So when it comes to building the foundation one of the easiest ways to measure things is the amount of frames that they take up. How many frames total is the animation or how many frames are there in the startup, hit window or in recovery. Any professional fighting game player will tell you the importance of frames, a move with 3 frames of recovery is vastly better than one with 6 frames of recovery. Things get more interesting when the move with 6 frames of recovery has 1 frame of startup.

The importance of frames isn’t relegated to only fighting games though. I think this is where most game developers let things fall apart. Either by not knowing the importance of frames or letting the animators dictate what looks better over what feels better. Animators will always opt for the animation that looks best.

With everything being said I am going to use 2 different fighting games examples to convey what I'm talking about. The first one is the same character in 2 different modes and how those modes play off of one another and the second is 2 different characters but with the same special move command and why one is easier than the other.

Capcom Vs. SNK2

It is not important to be well versed in this fighting game but either playing it once or twice or downloading a match video would not kill you. The important thing to know is that in this game when the player picks their character they are presented with another screen to select their ‘groove’. Groove dictates how the character will play – can they run, dodge, or how will they best use their special moves, etc. For the purpose of this post I want to talk about one important aspect that changes from some grooves to others: Low Jumping. Low Jumping is the ability to do a small jump as opposed to the normal jump that happens when the player presses up on the controller.

In a normal jumping groove, Ryu will leave the ground in 2 frames. In a Low Jumping groove, Ryu takes 6 frames to leave the ground. The game adds 4 extra frames to the same character to allow for controller input (Low Jumping command is Up then quickly tapping Down on the controller). Although at high level play these 4 frames don’t really make or break a game but when you first play the same character in multiple grooves it stands out and makes the character feel slower.

Here is a quick video of the Ryu in both grooves jumping up and down. Notice how they become off time from one another.

Zangief Vs. Thawk

Again with a Street Fighter reference but this time with a much older game, Super Street Fighter 2 Turbo. There are 2 characters that both have a complex controller motion known as 360 – you must spin the joystick 360 degrees then press the punch button to perform. Even though both characters have the same command many players have a much harder time performing Thawk’s. The reason again boils down to frames and in a very similar reason:

Thawk jumps faster than Zangief meaning that he leaves the ground faster. So when the player presses Up on the controller as part of the 360 command Thawk jumps faster than Zangief. Most players who have a hard time doing the same move with both characters will attest to the fact that Thawk ends up jumping a lot when they try to do the move and that is why. Let's look at the frames behind the problem:

Thawk leaves the ground in 3 frames after pressing Up Zangief leaves the ground in 5 frames after pressing Up

2 frames is all that is needed for players to have a hard time doing the special moves for their characters. Here is a quick and dirty video showing Thawk and Zangief both jumping. Although this isn’t as accurate as Ryu Vs. Ryu in the previous example (since I don’t know how many frames they stay in the air) I think you can quickly see them getting off time showing the importance of 2 frames.

For the next post I want to discuss more mainstream games - such as why Mario's jump feels better than Lara Croft's and some other gameplay mechanics that are common.

Thanks for reading.

Double Damage in Videogames

---

Double Damage is not your friend

So I meet a lot of designers these days – from seasoned designers who have shipped more games than me to designers whose first game is the one I’m working on them with. One weird thing that they all have in common, they all suggest giving Double Damage as a mode, attribute, attack, etc for the player or an enemy, boss, etc.

What is Double Damage

Double Damage – the concept of an attack doing twice the damage compared to normal. For Example – if an attack does 5 points, then with this ‘AWESOME!’ attribute – the attack does 10 points!! Note: Double Damage always comes with 2 exclamation points, that’s how double it is.

So why bother writing up something about this? Well I for one think that it’s a terrible idea 90% of the time. I’ll go into details about that here in a minute, but the main reason why I hate it is because it’s a very ‘designery’ solution. It deals with knowing how the details work and playing around with the numbers. It’s not something that is easily understood nor appreciated by the player.

Double Damage is almost always the first thing thought of when a power up is activated, a mode is turned on, etc. Let’s break down a little bit as to why I hate it so much.

Why Double Damage Is Bad

In most games these days – enemies don’t have health bars. Because enemies don’t have health bars, the player really doesn’t know how much damage they are doing with their base attacks. Because they don’t know how much damage their base attacks are doing, how in the world are they ever going to know they are doing Double Damage?

Let’s break it down by math a little. Let’s say an enemy has 35 points of health. Each attack from the player does 5 points, so it takes 7 attacks to kill one enemy. We now double that to 10 – so now it’s 4 hits to kill the enemy. Is that really all that noticeable to the player?

Using the same math example from above – what usually happens is that the player gets surrounded by a group of guys. No one is remembering how many hits they have dished out so far. Let’s be honest – most people are just mashing on buttons and don’t notice that 1 guy died faster than normal out of a big group.

The other down side to Double Damage is when meeting new enemies for the first time. The player has no idea how much health they have or how much damage the attack’s from this new enemy does.

When Double Damage Is Good

Like I said before, I hate it 90% of the time. So what is the 10% of the time I like it?

I think it makes a shitty power-up on its own. I do like it when it gets coupled with something else. Such as adding effects to the weapon and now the enemies explode when killed. Something that is more obvious to the player that something ‘cool’ is happening.

Obviously in situations where the player does know information – such as how much damage their attacks do or how much health the enemy has. This usually works out well in RPG’s since this information gets conveyed more often than in modern day action / adventure games.

Double Damage can be a good thing but it needs to be used properly. Don’t treat it as something ‘cool’ that the player can have and that’s it. If you are going to use it – then really use it. We used it a lot in God of War – but never as a sole attribute. We bundled it along with 900 other things that the player could appreciate: Fire Particles, Super Armor, whole new set of animations, blah blah.