Tag: GameMaker

YoYoGames charging monthly subscription for Game Maker Studio

It looks like GameMaker Studio is finally going subscription-based.

Which makes total sense, and I predicted it about 5 years ago. They denied going to a subscription model at the time, because of course they did. Users at the time weren’t in favor of it, preferring perpetual licensing. But I guess everything changes when you get bought.

Of course, for development tools, everything gets stale if you don’t keep it updated, so a subscription based licensing model probably does make the most sense.

For professionals, anyway. For amateurs and hobbyists, though, it’s a lot of money to pay for a tool if you only use it occasionally. GameMaker’s bread and butter used to be the student/hobbyist/indie developer who could readily afford a one-time price of $40, then $100. Now it’s $100/year, or $9.99/mo. Which is still not that much if you use it a lot, but if you’re just not ready for that type of commitment, there’s really nothing like what GameMaker 8 once was, over a decade ago.

YoYoGames still offers a free edition, intended for student/hobbyist users, but it lacks exports, not even for Desktop OSes, so it’s of extremely limited usefulness for anyone who wants to actually build their projects in order to get feedback from peers and friends.

PlayTech sells YoYoGames, GameMaker to Opera

Five years ago, YoYoGames was acquired by PlayTech. Today, I see the news that they have sold the company off for a loss of about $6.4M. Update: It’s come to light that the new owner is Opera.

It was a $16.4M purchase in 2015, and during this time they put a ton of development into GMS 2, released it, and so it’s a bit concerning that this hasn’t resulted in YoYoGames becoming a more valuable property.

On the other hand, I can well understand it. In the past several years, I have never felt comfortable with the new GMS 2.x UI. I find it awkward, unintuitive, ugly, and frustrating to use compared to the GMS 1.4 and earlier versions that I learned first. And at the same time, competing technologies like Unity 3D, Unreal Engine, and Godot been strong competition.

From a coding standpoint, there’s no question that the GML programming language has gotten better and better as YoYoGames continued to develop it.

From a UX standpoint, it’s been a crapshoot. The UI has some nominal improvements, but overall I feel like they changed too much too fast, and I could never get used to it. I spend way more time looking for the feature I want to use, and then wondering why it doesn’t work the way I think it should, and it completely kills my productivity and along with it my desire to work with the tool.

In fact, it’s a big part of why I haven’t done much game development in the last year, and have mostly dropped out of the pursuit.

I hope the new owner does better and continues to make improvements with GameMaker. It was very good at its original intended purpose of making it easy for game developers who are not primarily programmers to create simple 2-D games.

Many amazing games have been built with GameMaker over the last 22 years, which shows clearly the merit of putting simple, usable tools into the hands of creators who wouldn’t know where to begin with tools intended for professional programmers. Hopefully even more will be made in the years go come.

It will be interesting to see what the new owner does with the property. I want to see a product like GameMaker continue to serve the market it has traditionally done well with, while offering features that make it viable for professional game developers as a first rate tool.

GameMaker Tutorial: Configuration system

Many games have options or settings that are configurable.  The specifics can vary widely, from graphics options to music and sound effects volume to to input configuration to in-game options such as difficulty level and which rules are enabled/disabled.

This article will cover how to create a basic configuration system, using external file i/o, ds_map, and json.

Define your configuration requirements

First, before you develop it, design the system. Things to consider:

  • What variables does your game have?
  • What variables would you like the user to be able to modify?
  • What are the upper and lower bounds that make sense, or are reasonable for an enjoyable game? What values will break the game?

A little planning, and you can quickly list out all the variables you’ll want to manage with the configuration system.

Not every variable in your project needs to be subject to customization. Decide what configuration options you want, and define the variables that will be needed to control them, and decide on your default values. 

That said, having an editable, external configuration file can be very valuable for testing and tweaking the game, as well as convenient for the player. By having an external config file, you can modify game constants and variables without having to re-build the project each time. Since building the project takes time, being able to avoid building the game each time you want to test a variable can potentially save you time — for larger projects, potentially minutes per build.

Storing the config externally will also enable a tester to play around with the game variables without the need for the tester to be a programmer, or have GameMaker Studio installed. They can simply edit the config file, save it, and run the game again.

Nobody said that you must create a UI in the game program that can access every variable that you store in the config file. So, your testers can play around with configuration in a text editor, while your will eventually include a configuration UI for the game, to be used by players, and accessed through a title screen menu or a pause menu. This UI only needs to expose those configurable options that you think it should, or can have a secret mode that exposes more (or all) of them.

The default configuration should be safe to run on all devices the game is intended to run on, and should have the “standard” or “recommended” settings for the best or recommended game experience. 

Having defaults is important in case the external config file is missing or corrupted, and needs to be regenerated.  Create a ds_map called defaults, and use it to store all the default values.

Coding the config system

Coding a simple config system is actually very easy.

First, define variables to store the default configuration and the actual configuration, and create ds_maps for them.  The best time to create these ds_maps is immediately on launch, as soon as the program starts.  You may want to make these global variables, so that they will be accessible anywhere and will persist for the entire runtime.

global.defaults = ds_map_create();
global.config  = ds_map_create();

Of course, when we are about to exit the game, we will need to remember to destroy these data structures, to properly free up the RAM:

ds_map_destroy(global.defaults);
ds_map_destroy(global.config);

Next, initialize the defaults ds_map with all the variables that are part of the configuration, and their default values. It’s easiest to learn the shortcode for accessing ds_map values:

defaults[? "key"] = value;

In the code above, “key” is the name of your variable, in quotes, as a string. The value is the literal value of the variable.

So you can do this:

global.defaults[? "starting_lives"] = 3;

or

global.defaults[? "starting_lives"] = default_starting_lives;

As you can see, the ds_map’s key indexing allows you to choose meaningful names for the keys, which makes it easy to recall them later.

When you apply the configuration to the variables in your project, assign the value from the ds_map, like this:

variable = config[? "key"];
starting_lives = config[? "starting_lives"];

Once you have your ds_map populated with your configuration data, it’s time to write it to disk.  This is will save the file so that the configuration state can be read and applied to the game state the next time the program runs.

The gml function json_encode() makes it very easy to write the ds_map to a file, by converting the ds_map into json, or JavaScript Object Notation, which is just a commonly used data format, and can be thought of as a specially formatted string. You don’t need to know anything about the syntax of json in order to use the encoding/decoding functions — GML handles this for you.

Create a json-encoded string version of the config ds_map:

config_string = json_encode(global.config);

Check to see if an external config file exists already, and if not, create it and fill it with the default configuration:

if !file_exists(working_directory + "config.json")
{
config_file = file_text_open_write(working_directory + "config.json");
file_text_write(config_file, defaults_string);
file_text_close(config_file);
}

If the config file already exists:

  1. read the content of the file into a string variable,
  2. decode the json string to convert it back into a ds_map,
  3. validate the data in the ds_map, and,
  4. if valid, apply the configuration data to the game variables.
//First read the json data out of the config file
config_file = file_text_open_read(working_directory + "config.json");
config_json = file_text_read_string(config_file);
file_text_close(config_file);

//Next, decode the json, converting it to a ds_map
global.config = json_decode(config_json);

//Now, validate the configuration data to verify the settings are good
if !config_validate(global.config) //config_validate is a script that you wrote to validate your custom configuration data.
{
   //if we failed validation, destroy the config and create a good copy using defaults. 
   ds_map_destroy(global.config);
   config = ds_map_copy(global.default);

   //(Or, alternately, just correct the specific variables that were invalid, leaving any valid non-default values intact. This is more work for the programmer, but more convenient for the user.)
}

[...]

//apply data stored in global.config to the variables in your project as needed, where and when appropriate.
configurable_game_variable = config[? "key"];

The exact details will vary depending on your project, but the above is a good example to follow to get  you started.

Implementation: Data and File I/O

It might help to explain the above, in plain English, without the code examples getting in the way.

When the game runs, the very first thing the program should do is check to see if a configuration file exists.   If it exists, we read the data out of the file using file_text_string_read(fileid) and convert the string to Javascript Object Notation (JSON) with json_decode(string).  This returns a ds_map of key:value pairs that contain all the configuration data. If the config file does not exist, then we create it, populating it with our default data values.

If the config file exists, but contains invalid data, then we restore the invalid data using defaults.

Once we’ve successfully read data out of our config file, we need to validate the configuration data that we read from the file in order to ensure that the game is being run with sensible values that won’t result in the game crashing or glitching. If the configuration data is invalid, depending on the setting and the value, the game not work properly, and may crash, or may perform unexpectedly, or break. 

How the invalid data got into the configuration, you’ll want to investigate and fix, if the program was at fault. Your program may have written the data incorrectly, or it might have a bug that results in a corrupted file, or you may have read the wrong value out of the file, and those are all errors that you would need to debug and fix.

But it’s also possible that the user may find the file and manually edit it, make a mistake, and thereby introduce errors.  So check each value in the configuration ds_map, and verify that it is valid by checking to see if the key was found in the configuration file, that its value is of the correct data type (string, number, or boolean), and that it is within the range of acceptable values for that variable. However the bad data got there, your program will need to handle it and correct it.  Write a script that does this, according to the particular needs of your game.

If validation fails, whether because the file is missing, or because one of the values found within it is incorrect, we can handle this in several ways.  One way is to reset all values back to their default values. Another way is to correct just the bad value, on a per-setting basis, restoring the invalid value back to the default value.  Or we can simply display an error message to the player and exit the program. Decide what is best for your project, and implement it that way.

Usually, I recommend checking whether the file exists, and if not, regenerate it using defaults, and if it exists, restore individual settings to default one by one if there is an invalid value, rather than resetting the entire configuration back to the defaults if any of the values is invalid.  This way, a small error in the config file doesn’t blow out all the settings, and program will leave any valid customized settings alone, so that the user will only need to correct the values that were reset to default.

If we have a valid configuration, the next step is to apply those values to the game variables. We do this by assigning values to variables in the objects that use them, and the values we assign are copied out of the config ds_map, so that they are local to the owning object, and can be modified during the lifetime of the object without altering the configuration data.

Once the game variables have been set, the game is now ready to run.

Where did that file get written to?

While you’re developing the game, you might want to know where GameMaker is writing the files, in case you want to open them with a text editor and double-check that they’re being written correctly, or if you want to edit them. If you’re running the game through the IDE, the location may be trickier to find than if you built an executable and installed it. For help in finding the location of the writeable directory for your project, see this article.

Editing the configuration

One great thing about storing the configuration data in an external file is, we no longer need to re-compile the game every time we wish to tweak the settings.  This can greatly speed up testing, as compilation can take a minute or more each time, and this rapidly adds up when  you’re quitting, coding, re-launching again and again as you develop. So use the configuration system as  you test your game design.

We can edit the configuration in various ways.

The simplest to develop is to develop nothing; just rely on the user to find the configuration file on disk, open it up with a text editor, and change values, save the file, and run the game again.  You’ll find the writeable working directory for your game somewhere inside of %appdata% in a Windows build, but if you’re building the project for other platforms, you’ll need to find it.

While this is the easiest approach, this isn’t the most user friendly solution. A better user experience would be to put a GUI in the game to change the settings, and let the game program edit the config file.  This will be safer for the user, as you can constrain the input and check it for valid values before saving the config file.

The downside is that this can take a lot of extra work to build the user interface. Sadly, GameMaker does not provide a library of user input controls, such as buttons, checkboxes, text boxes, sliders, and so on. So to build the UI, we first have to build these controls.

This is a great opportunity for beginner programmers to get some experience designing and programming their own controls.  But it’s also a time-consuming task, and can be a frustrating challenge for a newbie programmer to figure out how to make a collection of robust, bug-free UI controls that are easy to use, work together, and behave in a way that the user expects based on their experience using other software. And doing this takes time away from developing the game part of the project.

It’s a lot of work, so having an easy way out is a good thing. There are assets available through GameMaker marketplace, which can be purchased for a few dollars, and provide the needed functionality. Either way, it’s good to have a set of reusable controls that you can put into any project, so whether you buy some, or you decide to make your own, you can get a lot of value out of them.

Advanced concerns

Storing arrays, ds_structures inside json

Up until now, the examples I’ve given have all been simple data types.  If you want to store something else, like an array, or a data structure, or graphics data, it gets more complicated.  This article is just an intro level tutorial, so it won’t cover these advanced topics.  But I expect I may cover them in the future, at some point.  Many games will not need the advanced techniques to store the basic configuration data.

Applying config to application state

Certain configuration changes, such as display size settings, will require the room or game to be restarted before they take effect.  In such case, you may need to use game_restart() or room_restart(). If you are confident that the data is being applied correctly to update the game variables, but you’re not seeing the change, try a room restart and see if the changes take effect.

But any room or game restart will restart the game in progress, and that may not be what you want to happen at all!  If you have a game where you can change the configuration from the Pause screen, for example, you will not want to disrupt the running game.  In that case, you’ll need to go further and handle this in some way, such as:

  1. Display a message to the user saying that the changes will take effect after they quit and relaunch the game.
  2. Give the play the option save the present game-state and then restore it with the new configuration in effect.

Details on exactly how to do this will vary considerably depending on your game’s design, but if you need to do this, you’ll essentially be building a save state feature for your game, and then calling it to save and then restore the game after the restart.

The basic logical flow of this process is as follows:

If <config change will require restart> and <game is in progress>{ save game-state to external file; restart;}

On re-start, check for existence of a game-state file, and if it exists, clear the room of default starting instances; restore room from game-state; then delete the game-state file.

This sounds fairly straightforward and simple, and, in concept at least, it is.  The difficulty is with storing all of the instances and their state data. Depending on the game, you may have hundreds of instances, and each instance may have a lot of data, and all of it needs to be written to file, then read back out of file, in order to reconstruct the game in the state where it left off.

Making it tamper resistant

You may welcome the user tinkering with the config file, or you may want to protect the config file against unwanted tampering. Usually this isn’t critical for a config file, but for a save file or a high scores file, it might be important to prevent cheating. If your game uses password-protected user accounts, or stores any kind of financial data or purchase records, you should be storing that data securely.

This should be all you need. I won’t get into technical detail here, but will outline a few tools you can make use of.

  1. Use ds_map_secure_save() and ds_map_secure_load() to securely store the configuration data in an encrypted file. The encrypted file will not be editable by a curious user. The manual doesn’t give detail about what encryption is used.  Depending on how serious you are about protecting the data, you will want to research encryption and use an algorithm that hasn’t been broken. Don’t attempt to invent your own encryption.
  2. Create a cryptographic hash of the configuration data string, and store the hash with the file.  When you read the file, re-hash the data you read out of the file, and verify that the hashes match.  If they don’t, something has changed, and you know the file has been tampered with or corrupted.  In this case, you should re-generate the entire config file from defaults. 

    Look at GML’s md5 functions to get started.  GML also provides sha1 hashing for this purpose. MD5 hashes are no longer considered secure, and sha1 is also no longer considered secure, but may be “good enough” for non-critical needs. While not “hacker proof” they will prevent casual tinkerers from being able to modify the data.

Saving external data for other purposes

Now that we know how to store data in an external file, retrieve it, validate it, and use it at runtime, there are other applications we can use these techniques for. 

The most obvious one that comes to mind is save states.  I’ve already touched on this above, in brief.  Other ideas include:

  1. High score, leaderboard, and achievement data.
  2. Telemetry data for debugging or analytics.
  3. User profiles (to allow multiple users of the same machine to each have their own configuration and save file preference).
  4. Level Editors.
  5. Mod packs, to allow a community of players to make custom modifications to your game, such as level data, or other external resources like sprites and sound effects, etc.

As these features become more and more advanced and complicated, they’re really best left to professionals working on commercial projects.  But by following the approach described in this article to do a simple configuration system, you’ll have taken the first steps toward getting your skills up to that level.

GameMaker Tutorial: Audio speedup with sync

In so many games, music speedup is a great way to get the message to the player that they need to hurry up and get things done.

It’d be great if you could simply set a new tempo with a simple GML function, and have the current background music adjust on the fly. Something like audio_sound_set_speed(sound, speed) would be lovely. But it’s not as simple as that in GameMaker, as I found out recently.

Here’s how I implemented a speedup for my GMLTetris project:

First, I created two music tracks, one at normal speed, and one at double speed, and added them to the game project as sound assets.

Everything else is just programming:

if <condition> {stop slow_music; start fast_music;}

This is easy enough, the trickiest part is probably getting the condition right to switch tracks, but depending on the game, that condition could be very simple, too.  The only real complication is that if you’re checking the condition repeatedly, as you normally would every Step, you only want to trigger the changeover once.  So to do that, set up a variable to track whether the switch has happened already, and check it, too, even if the condition that triggers the changeover continues to remain true on successive steps.

if <condition> && !music_switched {stop slow_music; start fast_music; music_switched = true;}

The music speedup that happens in a game like Super Mario Bros., where the music speedup occurs when the level timer hits 100, is a typical example of such a technique. If you only need to do a single, one-way switch, this is all you need.

If your game needs to switch back and forth between slow and fast music, your conditional needs to be more sophisticated.

if <condition>
{
   if !<fast_music_already_playing>
   {stop slow_music; start fast_music;}
}
else
{
   if !<slow_music_already_playing>
{stop fast_music; start slow_music;}
}

Here, because the game can switch multiple times, when the condition check happens, we can’t get away with a music_switched variable that changes one time. What we need to do is check to see if the music we need to switch to is already playing, and if not, stop the current music and switch to the other music.

One thing to keep in mind, this basic technique will start the fast music from the beginning of the track. This might be what you want, but it would also be good if you could start the fast music at the position where the slow music was, for a seamless transition.

GML has the functions to do this: audio_sound_get_track_position() and audio_sound_set_track_position(). But in order to make use of them, we need to do a bit more work.

First, since the gml functions return the absolute time position of the track, and since two tracks play at different speeds, we need to adjust the position proportionately when we switch tracks, so that the position is at the same position percentage-wise. This is actually easy, as long as we know the tempo change, which we do. Since the fast track is double speed, we can easily calculate the equivalent position in the other track by multiplying or dividing by 2.

Slow to fast: position *= 0.5;

Fast to slow: position *= 2;

Where I ran into trouble was, I needed to be able to switch both ways. It seemed like it should be simple — just check whether the desired track is already playing, and if not, get the position of the current track, adjust it proportionately, start the desired track, set the position. Easy, right?

Let’s look at it in pseudocode first:

if <condition to switch to fast music>
{
   if audio_is_playing(slow_music)
   {get position; stop slow music; start fast music; set position;}
}
else
{
   if audio_is_playing(fast_music)
   {get position; stop fast music; start slow music; set position;}
}

This was when I discovered that audio_play_sound() returns a handle for identifying the specific instance of the sound that is playing. This is necessary to use to set the track position of the playing sound. You can’t just set the position for the sound_index; you have to set it for the specific handle of the currently playing instance of the sound. If you set the track position for the sound_index, any time that sound resource is played in the future, it will start from that position.

///Create Event:
bgm = audio_play_sound(slow_music, 10, true);
///Step Event:
if <condition>
{
  if audio_is_playing(slow_music)
  {
  var pos = audio_sound_get_track_position(bgm);
  audio_stop_sound(bgm);
  bgm = audio_play_sound(fast_music, 10, true);
  audio_sound_set_track_position(bgm, pos * 0.5);
  }
}
else
{
  if audio_is_playing(fast_music)
  {
  var pos = audio_sound_get_track_position(bgm);
  audio_stop_sound(bgm);
  bgm = audio_play_sound(slow_music, 10, true);
  audio_sound_set_track_position(bgm, pos * 2);
  }
}

I also discovered that detecting which track is playing with audio_is_playing() does not work for this purpose. I still don’t have a clear understanding of what was happening in my code, but some debugging showed that my track position calculations were being distorted by being called multiple times. This doesn’t make sense to me because the song should no longer be playing after the first step when the switch condition is met. But my theory is that since the audio is played in another process from the main program, there’s some messaging going between the two processes asynchronously, and as a result audio_sound_is_playing() can still return true a step later, even after the message is sent to stop the track playing.

By trying to set the track point multiple times in quick succession, weird and unexpected results happened, and the tracks switched but did not set to the correct position.

So I had to come up with a surer method of knowing which music is playing.

Debugging was tricky, since I couldn’t tell from listening where the audio position was. To aid debugging, I drew the position of the playing track to the screen, and then I was able to see that the position was not being set correctly as expected. Somehow, switching from slow to fast would drop the track back from about 10 seconds to 2 seconds, and then switching from fast to slow would jump from 2 seconds to 38 seconds. It wasn’t consistently the same timing, but depending on when the switch was triggered, it would jump around, seemingly at random, but within a range of a second or so around these positions.

I couldn’t figure out why that was happening, so I tried using show_debug_message() and watched the output console, and saw that the track position would update 2 or 3 times when the tracks switched; I was expecting it to only update once.

This is what clued me in to what I believe was happening due to the multiple processes communicating asynchronously.

The solution I used in the end was easy and simple: instead of checking which track was currently playing using the audio functions to directly check whether it was playing, and then switching from slow to fast or vice versa based on the currently-playing audio asset, I just added a new variable, condition_previous, and compared condition to condition_previous, and made the switch happen only when the current condition didn’t match condition_previous. This only happens in one step, when the condition changes from false to true, or vice versa, and so the track position is set once, when the bgm switches tracks and syncs up the new track to where the old track left off.

switch_previous = switch;

switch = <condition to check for switching to the fast music>;

if switch && !switch_previous
{
var pos = audio_sound_get_track_position(bgm);
audio_stop_sound(bgm);
bgm = audio_play_sound(fast_music, 10, true);
audio_sound_set_track_position(bgm, pos * 0.5)
}
else
{
if !switch && switch_previous
{
var pos = audio_sound_get_track_position(bgm);
audio_stop_sound(bgm);
bgm = audio_play_sound(slow_music, 10, true);
audio_sound_set_track_position(bgm, pos * 2);
}
}

This works, because it guarantees that the condition checks will be accurate, as they do not depend on checking the status of the audio playing in another thread.

GameMaker Tutorial: Delta Time

What is “delta”?

Delta is a term often used in math and science to indicate change. For example, in physics calculations, the term ΔV (pronounced “delta-vee”) is used to quantify a change in velocity as measured at two different moments in time. If you didn’t have an introduction to this concept in your education, it can seem mysterious why we talk about “delta” like this, but it is a common convention.

Background

In videogames, the game program incorporates a routine called an update loop, which repeatedly applies the rules of the game to the game’s present state in order to create the next game state. Normally, it is desirable that the program control its frame rate, the number of times this update loop executes per second, so that a consistent frame rate is achieved, in order to have the game run smoothly and consistently. Ideally, this frame rate would be constant and unvarying.

Achieving this in practice can be quite complicated. Back in the old days, when games were programmed in Assembly for specific hardware platforms, and the game program ran on the bare metal, the programmer relied on the CPU clock and the timing it took to perform various operations to ensure that everything worked in sync. This required great effort and intimate understanding of the hardware and program code.

For the most part, GameMaker makes things simple enough for basic game development, and you don’t have to worry about regulating the frame rate yourself.

In older versions of GameMaker, there’s a room property, room_speed, which sets a target number of steps the update loop will attempt to execute per second.  Typically, this is set to a value that aligns nicely to the refresh rate of the display — often 30 or 60.

In GMS2, room_speed is replaced by a global option in the project’s properties.  (Cog symbol) -> Main options -> General and set “Game frames per second”.

This simple approach, simply setting a frame rate through the Game FPS or room_speed properties, works well enough for many projects.  Just set the desired frame rate and let GameMaker manage itself to try to achieve this target frame rate.  As long as the program is able to do so, everything will run as desired.

We can run into problems, though, because of the limitations of the computer hardware to perform all the calculations needed during the update loop to generate the next game state.  As long as the computer has plenty of resources, it should have plenty of time to generate the next game state.  But if those resources are not available, or if the game demands too much work be done to allow the available system resources to create the next game state in time, the game will not run smoothly.  The program will not crash, but it may slow down or stutter, or perform in a jerky manner.

It’s not really avoidable; to avoid the problem completely, you would need to write the game in such a way that it was guaranteed that the program always requires the same amount of calculations to be performed per step, and always has the same system resources available to perform those calculations.  The first is very unlikely, and would impose constraints that would make the program’s design very rigid. The second is well beyond your control, unless you’re writing software that has direct control over the hardware it’s running on, rather than writing an application that runs on top of an operating system that also supports numerous other multi-tasking programs and services.

The solution is to write efficient code that has an extremely good chance of running all the calculations for the next game state before that game state is needed for drawing the game to the display, so that there will never be a delay that results in a dropped frame.  Doing this requires that you mostly write small, simple games, and run them on hardware robust enough that it will easily be able to guarantee that sufficient resources are available in order to complete the computation required, on time, every step. Those are both significant downsides, but even so you can create a great many games within those constraints these days, without too much of a problem.

But there’s still things that are out of your control, even if you write a small program with very efficient code. You can’t control what hardware the user will run the game on.  You can’t control what other processes might also be running on that system, in the background, while the game program is running. These uncontrolled factors can defeat all your efficient programming and give a poor experience at runtime on a system with insufficient resources to perform all the instructions that it needs to, on time.

Enter delta time

When the simple approach isn’t good enough, it’s time to turn to a concept involving delta time.  This is simply measuring, each step, the amount of time that has passed since the previous step.  This amount will vary from step to step, depending on how much work the game had to do to generate that step, and how much hardware resources were available to perform that work, in addition to whatever else the operating system may have been doing.

Once you know how much time has passed, you can use this to make adjustments in various things that depend on time, such as movement speed, timers, sprite animation speed, and so on, to keep everything looking smooth.

In a simple game, you would normally just set an instances speed to some value, V, and every step the game would update by moving the instance V pixels in direction theta.

In a delta timed game, you would first measure the amount of time that has passed since the previous step, calculate a ratio between this time and the number of game state updates desired per second (frames per second), and then use that ratio to modify the speed of everything in the game, so that even if the game runs a little behind in calculating everything, the end result is that the instance looks like it moves V pixels in direction theta each step.

Make sense?  It can be a lot to grasp at first, but it’s actually fairly simple.

Let’s look at it now in detail and walk through an implementation.

Implementing delta time

GameMaker provides a built-in global variable, delta_time, which is the time in microseconds that has passed since the previous step.  Since the engine performs this measurement for you, it’s very reliable, and all you need to do is use it.

A microsecond (μs) is one millionth of a second, or 1/1000000, or 0.000001. Typical GameMaker projects target a framerate of 30 or 60 steps per second, so each step is 1/30th or 1/60th of a second, or about 33333.333…μs per step for a 30fps game, and 16666.666…μs for a 60fps game.

So, if your game were running ideally, updating the game state every step exactly on time, every step’s delta_time should be exactly 33333.3… or 16666.6…  In the real world, you won’t hit those numbers exactly, but you’d expect that they should be very close, most of the time. (And in GameMaker, the value of delta_time drops the decimal, which introduces a problem down the road that we’ll have to deal with somehow… [[Hmm… maybe it’d be better to use a room speed that yields a whole number ideal delta_time?]] We’ll get to that in a bit.)  To the extent that delta_time differs from this ideal value, you can adjust all of the time-based variables that influence the game state, and if you do it all correctly, thereby smooth everything out.

It would be nice if you could just have a “use delta time” checkbox in your project, and all the built-in time-dependent variables — speed, hspeed, vspeed, image_speed, alarm[], etc. would just receive these adjustments by virtue of the engine, automagically.  Wouldn’t that be nice?

But for whatever reason, they don’t give us that feature, but at least they give us the built-in delta_time variable to work with.  So we’ll have to do that.  And of course, every other user-defined variable that we add to our project that depends on time will also need to be adjusted with delta_time.

So, if the game is running too slow, delta_time will be higher than the ideal value, because the amount of time that passed was longer than the ideal time.  In the game, instead of moving N pixels per step for 60 steps in a second, the instance would move N pixels per step, but only hit some number less than 60 steps per second, resulting in slowdown and jerky movement and animation.

To fix this, we should be able to see how:

ideal_delta_time = 1/room_speed * 1000000; //typically 33333 or 16666

This simplifies to 1000000 / room_speed. Next, we need to compare the measured delta between the previous step and this step (delta_time) to the ideal delta time:

dt_ratio = delta_time / ideal_delta_time

Now that we have the ratio for this step, we must use this ratio for all time-dependent calculations 

speed = ideal_speed * dt_ratio;

or, since computers multiply with fewer cpu instructions than they can divide:

speed = ideal_speed * delta_time * room_speed * 0.000001;

Or in other words, to move at an average speed of N pixels per step, you adjust the actual speed every step by multiplying it by delta_time/ideal_time:

speed = ideal_speed * (delta_time/ideal_time)

(Note, many games that use delta_time eschew the built-in speed system, instead using variables that are not managed by the engine, to give the programmer complete control over the motion. This is a very common approach taken with GameMaker projects, because the way the built-in variables are managed by the runner don’t offer the flexibility and control needed.)

Once we have the dt_ratio defined for this step, we don’t want to have to re-calculate this delta/ideal ratio again and again for every calculation performed in this step, so it’s best to do it once per step as a global variable, and have every instance that needs to use it reference this global value:

global.step_delta = delta_time/ideal_time;

That’s the basics of it.

Difficulties

The main problem that people have is that they have to adjust every variable in their game that is time-dependent.  Not just speed, but alarms, image animation, and possibly even such things as the pitch and duration of sound effects.  This can get tricky and drive you nuts.  Especially when you have things going on in your game like non-linear acceleration.

Another tricky problem with delta_time is the accumulation of rounding errors.

Let’s look at what happens when we want to use delta_time to achieve an average speed of 6 pixels/step:

speed = ideal_speed * delta_time * room_speed * 0.000001;

speed = 6 * delta_time * 30 * 0.000001

Let’s say that delta time is always exactly the ideal delta_time, 33333 microseconds.  That would give us:

speed = 6 * 33333 * 30 * 0.000001

speed = 5.99994 //shit. um...

Close enough, right?  Well, it adds up over time.  And keep in mind, if delta_time varies from the ideal_delta_time, the error can be larger.

Can’t we just round it?

Well… no.  We can’t.  Because it’s entirely possible, and even likely, that you can have a legitimate fractional speed as a result of delta_time adjustments, especially in the event that your game is lagging frames!  And rounding the value in that case would actually result in even more error.

Well, so what? Can’t we just live with it?

Maybe? I’m not sure what else you can do, honestly.  But this does leave us with problems. Even in the simple demo project that I’ve created, after just a minute or so, I can observe my control object deviate out of sync with the various delta-based objects.  Even the different delta-based objects will fall out of sync with each other, due to tiny differences in the way they calculate their deltas.

Most distressing of all, I’ve seen the objects glitch at their reversal point, getting stuck in the edge of the screen.  This isn’t consistent, and seems to depend on fluctuations in delta_time at just the right time to adjust the instance far enough outside of the room that the Intersect Boundary event will be triggers on two successive steps, resulting in the instance oscillating, becoming stuck on the boundary.

I’m still looking for a solution for how to fix this issue.

Manual errata

The manual gives us the following incorrect example:

speed = spd * (ot - delta_time);

Where ot is the value of delta_time from one step ago, e.g. a “delta_time_previous”.

This is wrong. The manual is in error, this is not how delta_time works.  It seems whoever wrote the manual was under a mistaken impression that delta_time is basically a point-in-time value of System.DateTime(Now) that is updated every Step, and that you would then need to calculate the actual delta by subtracting the current value for Now by the value of Now from the previous step.

The way delta_time really works, it is the number of microseconds since the previous step, essentially it is already the result of the expression “ot - delta_time” used in the example.  The correct implementation for the example should be:

speed = spd * delta_time/ideal_delta_time;

Alarms and animations

If you want to use delta_time to smooth out your movement, you should also consider whether you need to also consider delta_time with other time-based variables, such as image_speed.

Sprite animation is controlled by image_speed, and is normally set to a value of 1 by default, resulting in the sprite cycling through its sub-images at a speed of 1 per step.

An instance’s image_speed can be adjusted to speed up or slow down the animation speed.  At speeds slower than 1, the same sub-image will be drawn for more than 1 step in a row.  At speeds higher than 1, the animation will skip over sub-images.  At speeds less than 0, the animation will cycle in reverse.

If you’re using delta_time, and want your animation speed to also be delta_time-based, then the best thing to do is set image_speed to 0, and control animation manually.  Each step, increment image_index by 1 * ideal_time/delta_time:

image_index += ideal_time/delta_time;

Sprites can be set up and used in too many different ways to achieve customized ends, so if you’re storing sub-images in a sprite resource and selecting them in some other way than cycling through them as an animation, this is not necessarily going to apply.

For Alarms, you’ll want to stop using the Alarm Events that are built into GameMaker, and implement your own timer system. This is simple to do.

Step Event:

if timer >= -1
{
    timer -= ideal_time/delta_time;
    if timer <= 0
    {
          //do stuff
    }
}

You can have as many timer variables as  you need, and name them however you want, store them in an array if you want, or another data structure if you want.

Once you get used to it, you’ll probably like this better than the built-in Alarm system. YellowAfterlife’s article on Custom GameMaker Alarms is a great read for more information.

Capping the delta

Delta_time is intended to smooth out minor variations in fps, but if you have a longer interruption, delta_time will break your game. It’s useful to understand why and how, so you can prevent problems.

You’ll need to cap the amount of delta that you want the game to react to.  For example, if your game loses application focus, it will stop processing, essentially pausing the action.  The timer that delta_time is based on still keeps track of time however. This means that when you return application focus to the game, so much time will have passed that the ratio between ideal_time and delta_time will be enormous.  If your game uses this enormous ratio to adjust your game speed, it will almost certainly break the game.

Say you have switched application focus for a second to respond to an instant message you receive in another app. The instant you return to the game, the delta_time between the next step and the step your game paused at is going to be several seconds, instead of a value close to 1/30th or 1/60th of a second.  This means that in that step, everything in the game is going to move the equivalent of several seconds worth of distance, missing collisions that would have happened during those “missing” seconds”, and skip several seconds worth of animation frames, and count down several seconds worth of alarms, in just one step.  Things will go haywire — unless you cap the ideal:delta ratio.

You’ll want to experiment with your game and see where this limit should be set. It should be a value that is well short of any game-breaking bugs that a too-large ratio would cause. The exact value will depend on the specifics of your game.

To implement the cap is simple, though. Revising our earlier dt_ratio expression, we just use clamp() to cap it:

dt_ratio = clamp(delta_time / ideal_delta_time, min_cap, max_cap); 

The value of max_cap is your maximum allowed delta_time/ideal_time; min_cap is your minimum allowed delta_time/ideal_time.

Controlling the tempo

Delta_time can be used for more than just smoothing out your frame rate.  You can use these same ideas in slightly different ways to “control time” — that is to create a slow motion effect, or to speed things up like a “fast forward” effect.  To do this, just add in an extra factor, which I’ll call tempo:

speed_this_step = ideal_speed * (delta_time / ideal time) * tempo;

  • If tempo is equal to 1, the game will run at normal speed.
  • If tempo is >1, the game will speed up, moving faster than normal.  You can do this within some limits, but if you speed up the game too much, you’ll start having to deal with “high speed collisions” — situations where an instance moves so far that it can skip over objects that it should have collided with — which will require more work to handle correctly.
  • If tempo <1, the game will slow down.
  • if tempo == 0, the game will freeze.
  • If tempo < 0, the game will run backwards! (Not perfectly, though; any use of randomization will cause a loss of determinism which means that things won’t run backwards exactly as they ran forwards, only in reverse — and negative tempo won’t cause events like Collision, Create, and Destroy to “undo” or run in reverse.  But for certain uses, especially if used briefly and sparingly, this may not matter.)

How cool is that!  You can do a lot of things with this.

You can have a global.tempo, for controlling the overall speed of your game, and an instance.tempo factor that is unique to each instance.  Change the value of global.tempo, and the entire game will speed up or slow down; change the value of instance.tempo for selected instances, and only that instance will be affected.

speed_this_step = ideal_speed * (delta_time / ideal time) * global.tempo * instance_tempo;

A good range for tempo, for most games, is probably anywhere from 0.10 to 2.0 or 3.0 or so. But it will depend a lot on the specifics of your project, so you’ll need to experiment and test.

You use tempo to create a pause system easily, by applying delta_time to objects that are used in the game, setting global.tempo to 0, and controlling the global.tempo using objects that are not bound by delta_time so that they will remain responsive when the rest of the game is paused.

You could also create a script that performs this function:

///delta(ideal_value, global.tempo, instance_tempo) 
///@description returns value_this_step computed from ideal_value based on delta time, global and instance tempo values.

return argument[0] * (delta_time / ideal_time) * argument[1] * argument[2]; // We assume here that ideal_time is defined elsewhere in the project, probably as a #macro, or is represented inline with the direct calculation here.

Then when you use this script, you avoid code duplication, and it will make the code more readable:

speed = delta(10, 1, 1);

Beware bogus approaches

There are a few common approaches to implementing delta_time-like solutions that you may run into if you look at other people’s code.  I’ll touch on them briefly here, in order to explain why they’re not ideal.

Changing room_speed.  In older versions of GameMaker, before Studio, the program could change the room_speed and the GameMaker engine would adjust its speed on the fly.  This was very easy to do, but wrong.  For example, if you set the room_speed to 0, or a negative number, that would break the game.  A room_speed of 0 would freeze the game, and make everything stop, even the runner, so that the game would never update again.  And negative room_speed values are not allowed.

Later on, GameMaker was updated so that the room_speed variable could be changed during runtime (so it wasn’t a read-only variable) but that changes did not take effect at runtime for the current room, meaning you could not change room_speed in the current room and see immediate results; only by leaving and returning to the room, or restarting the room would the new room_speed be in effect.  This effectively ended adjusting room_speed as a viable approach.

Programmers who have the idea of how to do a technique like delta_time, but haven’t heard of it before, will sometimes “re-invent the wheel”, usually by using fps_real and room_speed to create a ratio similar to delta_time/ideal time.  This approach can seem at first to work, but it is always off because fps_real is the actual, achieved fps for the previous frame.  For the current frame, however, fps_real could actually be a very different value from what it was one step ago, meaning that if your game just did a huge amount of work in the previous frame, say to initialize new instances that were just created, and it doesn’t need to do that work in the current frame, the calculations could be way off, giving a poor result.  As long as fps_real is not changing dramatically from one step to the next, and is reasonably close to the target fps for the game, it will appear to work well, but when fps is very inconsistent, which is when you really need it, it will not work well.

If you see these techniques in use in a project, it can be a pain to convert from the bogus delta time technique to the good technique.  You’ll need to meticulously review the code for the entire project and find every place that uses the bad technique, and refactor it to use the good technique. Depending on how well documented the code is, and how expressive the code style is, this can easily get out of hand.  The best thing to do is to use version control and create a branch, and work on the conversion in the branch, and just focus on converting the delta time system before you go about making any other changes.  Take notes and put comments in the code to keep track of your progress, remind you of things that you need to do, need to test, or need to figure out.

Demo

In writing this article, I created a demo project to help me understand the concept of delta_time, and test various approaches. You can download it for free and play with it yourself.

Summary

  • Delta_time is simply the number of microseconds that have elapsed since the previous step. This value can be used to smooth out unevenness in frame rate if the program drops below the ideal frame rate.
  • It’s easier to implement delta_time if you decide to use it from the very beginning.
  • The basic delta time expression is: 
    N * delta_time * room_speed * 0.000001
  • You can adjust the tempo of the game even more by using global and instance tempo variables, and multiplying by those as well:
    N * delta_time * room_speed * 0.000001 * global.tempo * tempo
  • You can create a simple pause system (in part) by setting the global.tempo (or a custom pause factor varable) to 0.
  • Remember to consider things like animation speed and timers as well as movement.
  • Avoid using the built-in Alarms Events, movement functions and speed variables; it’s easier to manage everything if you take complete control over the position, speed, etc. by using your own system.
  • Cap the amount that delta_time can adjust your game, to keep things from going out of control after a long pause between frames.
  • Delta_time will result in values that are slightly off the ideal target value due to decimal imprecision/rounding errors.
  • Test thoroughly!

GMLive by YellowAfterlife: code and run gml live in your browser

I just discovered GameMaker Community Forums moderator YellowAfterlife’s wonderful in-browser utility, GMLive.

I’d long been wanting YoYoGames to produce a GML console that I could type in a simple snippet of code that I could run to see what it does without the need of having to set up a project, build, and run it, similar to the REPL in Python, so I’m thrilled to learn that this exists. It’s very well done, and while not 100% compatible with GML, it’s very close, and very well documented.

Other People’s Code

A GameMaker user recently posted an interesting question in one of the communities that I watch:

So, I’ve known for a long time now how helpful it is to go through source code you haven’t written, just to pick up good habits, and generally understand how things work.

I got the source code for several games including Home and Ink from a Humble Bundle deal a while ago, and am just starting to take a look at them now. I’m finding it really difficult to get a grasp of what’s going on/which objects and scripts relate to what though – mainly because I’m having to switch between several windows of code.

Does anyone have tips about how to learn to understand GM projects? I usually have no problem understanding singular sections of code, it’s more getting to grips with how everything fits together, especially in full-length games. And generally, any tips about how you work through other people’s code.

This happened to be a very timely question, as I’ve recently been going through this myself.

I find it very enjoyable to code my own projects from scratch. I know what I’m trying to do, and I know why everything I put into the project is there. Especially when the project is something I’ve recently worked on.

It’s quite a different thing for me to try to look at a project that someone else has coded. I don’t know what the author was trying to do, or why they did anything they way they did it. Oftentimes the code is not well organized, and very frequently the code is done in a style very different from how I code, or the programmer’s approach is very different from my own, or they use math that I don’t understand well without explaining what they’re doing or why or how it works, and so on. This can make it very difficult to get anywhere.

It’s of course vitally important to be able to work with code that other people have developed unless you have no aspirations to be anything more than a solo coder for your entire life. But even if you never work with other people collaboratively, it can be very valuable to look at other people’s code and learn from it.

If the code is well-documented, this shouldn’t be too hard. I like to say it’s the original author’s  responsibility to ensure that the code they write is easy to understand.  But well-documented code is unfortunately rare. Documentation (ideally) tells you what you need to know in order to understand the code. How does the author know what you need to know to understand their code? No project can afford to be a 101-level tutorial for new programmers. People will come to the project code with all levels of knowledge and from many different backgrounds, and you can’t anticipate everyone’s needs. Nor should you have to — it should be expected that someone reading your code comes prepared. Working code is rarely intended to be a textbook example for neophytes. Most often, the programmer, if they document their code at all, documents it for themselves, and doesn’t care about anyone else’s needs if they should ever read the code.

So it’s quite understandable, it’s still all too common for code to not be well documented, or just not documented in a way that is useful to you.

Getting into big project that you didn’t author can be overwhelming.

What to do:

  1. Before doing anything else, make sure the project builds and runs.

    Play it for a while, and try to catch any runtime errors that may already be in the code. Don’t make any changes to the project until you’re satisfied that it is working. But recognize that there may still be bugs that are hidden more deeply than your playtesting was able to reveal. These may not be the fault of the programmer, either; it’s always possible that some update in GameMaker caused a bug.

    It can be really frustrating to open an (unknown to you) already-broken project and make changes to it, then try to build it and find that it’s broken. The assumption will be that it’s your changes that broke the project, but if the bug was already there before hand, you’ll waste a lot of time focusing on your changes, not where the problem really is. So to avoid that, test the project first before you do anything else to it.

    While  you’re running the project, observe it and take note of any objects that seem important.  Guess at what these might be called in the project, and look for them when you start exploring the project in the IDE.

  2. Start with the first Room in the project.

    Open it, and see what’s in there. In GameMaker, the project always starts with the first room in the resource tree. Start taking notes about what you find in this room. What objects are present in the room editor, or being created in code? These should be the first objects to examine.

  3. Look at how resource groups are used.

    The project may use folders to group things together, especially if it’s a large project. If the author used folders to organize their projects, this will give you some insight as to how the author thought about the project and its parts, and how they relate to each other.

  4. Look for parent objects and see how they are organized.

    The inheritance structure of objects will reveal a lot about how the game works. The game may have a lot of objects, but if the author organized them effectively, it’s likely that many of them are related to each other through inheritance, so you may have a comparatively smaller number of object families that you’ll need to familiarize yourself with, and understanding the family broadly will convey much of what is needed to be understood to understand the project as a whole.

  5. Look for dead code and other resources that aren’t used.

    It’s very common for developers to leave stuff in a project that never gets used at runtime. This is the “cutting room floor” of the project; it represents stuff the developer worked on for a little bit, but for whatever reason never finished and didn’t end up using. Sometimes there can be a lot of it. Identifying the dead code can be helpful because then you can ignore it, and focus on the live code.  Once you’re sure the code isn’t used by the project, you may want to move it to a folder of dead code where it is out of the way and can be ignored, or just delete it from the project entirely.

    There isn’t really an easy to way to identify dead code.  But a few things to look for are code that is commented out, scripts that are never called anywhere in the project (use GMS’s “search in scripts” feature to find everywhere a script is called in the project.) Objects that are not present in any of the rooms, and are not referenced in code anywhere. Logical conditions that can never be satisfied can lead to dead code branches in if or switch statements, also, but these are a little harder to spot.

  6. Look at scripts.

    Scripts (often) support objects. Scripts are also (usually) well isolated and self-contained, so as to be reusable. Depending on the project’s size and the programming style, scripts may call other scripts.  This can become difficult to follow if you  have to jump between a bunch of different codefiles, but at least on your first pass this shouldn’t matter as much — just recognize that a script call is abstracting some detail, and you don’t necessarily need to get into the detail yet.

  7. Look at objects.

    First, scan the resource tree for objects that have sprites that you recognized from playing the game. Those are probably a good place to start.

    Look at the player object(s), then enemies, and other things like terrain objects, weapons, pickups, or whatever. There may be some important objects that are invisible during gameplay and don’t have sprites. Look at the name for a clue about what the object is for.

    As you get into each object in detail, probably the most important Events to understand are the Create, Step, Collision, and Draw events. Start there if the object has these Events defined, and take note of any other events it may use, and what it is doing in them.

    You may find GML code, or you may find Drag-and-Drop actions, or a little of both.

    As you start looking through the code, it’s going to be very tempting to make small changes, perhaps to conform to your ideas about code style and good practices. But try to resist the temptation to do too much of this at first. Even seemingly simple fixes can have consequences that you don’t realize. And you might find after “cleaning up” a bunch of code that it’s not needed at all, and can just be discarded.

    The only thing you should really do on your very first pass through the code is leave comments. I like to prefix my comments with my initials, and what the comment is about: Question, reminder to do something later, explanation, bug, possible bug, etc. This helps me as I go through the whole project and find answers to the questions I initially had when I first started digging into the project. And my initials help me to keep straight whether the comment is mine or the original author’s.

  8. Take notes.

    As you review the project code, add comments with your notes, explaining where needed. Comments can be questions, eg “What’s this for?” or notes for things to do later (“This looks interesting, come back later and understand this fully.”
    You can (and should) also take notes outside of the project. It depends a bit on the purpose — are you just looking at this project to learn from it, or are you working on the project to maintain or modify it? If you’re conducting a learning exercise, the notes should be the takeaway of what you learned by exploring the project files. If you’re actively working with the project, then the notes can be a guide to accumulate the understanding that you gained as you explored the project, a list of questions you still have, problems you intend to work on, bugs that need to be fixed, or other work to be done.

  9. Focus on the interesting parts.

    (The stuff you don’t yet know, but are interested to know/figure out.) Don’t worry about stuff you don’t understand, but aren’t interested in. Don’t worry about the stuff that’s simple/obvious to you.

  10. Don’t assume perfection.

    The programmer knew what they were doing. Well, maybe. Programmers make mistakes. Or do experiments. Or half-formed thoughts.  Or just don’t know a better way to do something.

    Code often has bugs. Code is usually sub-optimal.

    As you take ownership of the project, you will need to start being the decision maker who knows what they’re doing.  It’s easiest and best to carry forward with a good understanding of the previous author’s thinking, but it’s almost never available to us directly, and we have to infer it from what we do have.

    If you’re new to programming and lack experience and are looking at advanced, complex code in a big project, it’s hard to do this.  But start with small changes, test constantly, and build confidence as you learn and grow familiar.

  11. Ask questions.

    Ask questions about the stuff you can’t figure out. Contact the author directly if you can, be respectful and appreciative if they’re open to talking about it, and understand if they’re not interested. If they’re not responsive, or you still have questions, take them to the GMC Forums. Lots of people are there who will take an interest and try to help. Find a friend who is willing and able to mentor you, or even work with you on the project.

GameMaker tutorial: high speed collisions

If you have a small object, such as a bullet, collisions can be problematic for GameMaker if the bullet’s speed is faster than the length of the bullet. Any time an object moves faster than the size of its collision mask, it is possible for the object to skip over object in its way that it should have collided with.

The illustration below shows why:

high speed collision skipping

Here’s a simple, elegant solution for your high speed object to avoid this problem, regardless of its top speed.

Make the bullet sprite’s length as long as the bullet’s speed. If your bullet moves 16px per step, then its sprite should be 16px long.

If the bullet’s speed is variable, then the thing to do is vary the length of the bullet as the speed of the bullet varies. You can use this using the image_xscale property to scale the bullet.

Say the base bullet speed is 16px. If the bullet can somehow accelerate to 32px per step, then it should lengthen to 32px, by setting the image_xscale to 2. We can see from this that a general expression that solves the problem for any sprite width and speed would be:

image_xscale = speed/sprite_width

But it’s probably a good idea to use image_xscale = max(1, speed/sprite_width), to keep the bullet from shrinking if it is moving slowly.

A nice thing about this effect is that it makes high speed bullets look like elongated streaks, very similar to a blurry action photo where the subject of a photo is moving too fast for the shutter speed. In effect, that’s exactly what’s happening in GameMaker, if we consider the game speed to be like the shutter speed of a camera. So this creates a very natural looking visual effect, and solves a problem with high speed collision skipping.

This technique may be applicable to other objects besides bullets, but it works best with smaller objects because the stretching isn’t as noticeable. As the instance is moving at a high speed while being stretched, it can trick the eye into thinking it’s just a motion blur effect. With larger objects, this illusion is not as convincing, and the apparent distortion is more obvious.

Pinpointing the point of impact

Let’s say you need to know the exact point of impact from your high speed bullet. Since the bullet is stretched out, the x, y coordinates are no longer approximate enough.

high speed bullet point of impact

We an use point_direction to find the angle at which the impact occurred.

point_of_impact = point_direction(other.x, other.y, bullet.x, bullet.y);

If the target object is a circle, then calculating the point of impact is simple from whatever angle:
impact_x = other.x + lengthdir_x(other.radius, collision_direction);
impact_y = other.y + lengthdir_y(other.radius, collision_direction);

The above will work as long as the high-speed bullet isn’t positioned such that the x,y origin of the sprite is on the far side of the target object. If it is, the point of impact will be on the opposite side of the target object. If this is a problem, you can do additional calculations using the direction of the bullet’s travel as a clue to determining where the true point of impact was. You can also reduce the problem by making the bullet sprite’s x-origin 0, so the bullet’s [x,y] position will always be at the rear of the bullet.

If the target object is non-circular, or non-centered, you’ll need to do a bit more work to determine the point of impact, and exactly how to do that is outside the scope of this article. But using built-in variables image_xoffset, image_yoffset, bbox_top, bbox_bottom, bbox_left, bbox_right, you should be able to figure it out, at least approximately enough to be useful.

Achievement unlocked: $100 in revenue on GameMaker Marketplace

On July 20, 2014, I released my first asset for GameMaker Marketplace, MMap Mini Maps.

Today, nearly three years later, I have finally reached the $100 author revenue milestone, meaning I’ll finally get paid.

In celebration, I’ve put all of my paid assets on the GameMaker Marketplace on sale for just $0.99 for one week, ending 4/10/2017. Update: I’ve decided to extend the discounts by one day per asset sold during the sale. So currently the sale will run through 4/12/17. We’ll see how long we can keep this going.

You can buy them here.

Enjoy! And share! And thanks!

A tale of two GML scripts: code optimization through iterative development

Today I wanted to share two versions of a function that I wrote, in order to show how my iterative approach to software development works when I am doing code optimization to improve performance.

This example comes from my iMprOVE_WRAP asset. It’s a function that returns the shortest distance (taking into account the wrap capabilities of the calling object) between the calling instance and a target object.

The first implementation works, in that it correctly does what it’s supposed to do, but I never released it, because I wasn’t satisfied that it was good enough code to ship.

///iw_distance_to_object(target_obj, x1, y1, x2, y2, do_wrap_h, do_wrap_v,)

///@description Returns the distance_to_object from an improve_wrap object calling this function to another instance. 
///Compares all relevant points for the iw_object and returns the nearest distance, taking the wrap range into account.
///@param target_obj id of the target object to determine the distance to.
///@param x1 left x boundary of wrap range
///@param y1 top y boundary of wrap range
///@param x2 right x boundary of wrap range
///@param y2 bottom y boundary of wrap range
///@param do_wrap_h set whether the horizontal wrap is on (true) or off (false)
///@param do_wrap_v set whether the vertical wrap is on (true) or off (false)


//get the distance from the nine virtual positions
//return the shortest distance
var obj = argument[0];
var iw_distance, iw_distance_up, iw_distance_down, iw_distance_left, iw_distance_right, 
    iw_distance_up_left, iw_distance_up_right, iw_distance_down_left, iw_distance_down_right;
var tempx, tempy, shortest;
var x1, y1, x2, y2, range_width, range_height, do_wrap_h, do_wrap_v;

//keep track of original location of target object
tempx = x;
tempy = y;

//set up wrap range
x1 = min(argument[1], argument[3]);
y1 = min(argument[2], argument[4]);
x2 = max(argument[1], argument[3]);
y2 = max(argument[2], argument[4]);
range_width = x2 - x1;
range_height = y2 - y1;

do_wrap_h = argument[5];
do_wrap_v = argument[6];

//check distances
//check center
iw_distance = distance_to_object(obj);

if do_wrap_h && do_wrap_v //wrap vertical and horizontal
{
  //check corners
  x = tempx - range_width;
  y = tempx - range_height;
  iw_distance_up_left = distance_to_object(obj);
 
  y = tempx + range_height;
  iw_distance_down_left = distance_to_object(obj);
 
  x = tempx + range_width;
  iw_distance_down_right = distance_to_object(obj);
 
  y = tempy - range_height;
  iw_distance_up_right = distance_to_object(obj);

  //check left and right
  y = tempy;
  x = tempx - range_width;
  iw_distance_left = distance_to_object(obj);
  x = tempx + range_width;
  iw_distance_right = distance_to_object(obj);

  //check up and down
  x = tempx;
  y = tempy - range_height;
  iw_distance_up = distance_to_object(obj);
  y = tempy + range_height;
  iw_distance_down = distance_to_object(obj);
 
  shortest = min(iw_distance, iw_distance_up, iw_distance_down, iw_distance_left, iw_distance_right, 
                iw_distance_up_left, iw_distance_up_right, iw_distance_down_left, iw_distance_down_right);
}
if do_wrap_h && !do_wrap_v //do_wrap_h
{
  //check left and right
  x = tempx - range_width;
  iw_distance_left = distance_to_object(obj);
  x = tempx + range_width;
  iw_distance_right = distance_to_object(obj);

  shortest = min(iw_distance, iw_distance_left, iw_distance_right);
}

if do_wrap_v && !do_wrap_h //do_wrap_v
{
  //check up and down
  y = tempy - range_height;
  iw_distance_up = distance_to_object(obj);
  y = tempy + range_height;
  iw_distance_down = distance_to_object(obj);

  shortest = min(iw_distance, iw_distance_up, iw_distance_down);
}
if !do_wrap_h && !do_wrap_v
{
  shortest = iw_distance;
}

//return calling instance to original location
x = tempx;
y = tempy;

return shortest;

Let’s take a moment to appreciate this function as it’s written. It’s well-structured, documented, and expressive. First we declare a bunch of variables, then we do stuff with the variables, then we get our answer and return it. And this gives a correct result…

So what’s wrong with the above? It’s an inefficient approach, which checks each virtual position of the wrapping object. If the calling instance wraps vertically and horizontally, it has to temporarily move the calling instance 9 times and check the distance from each of 9 virtual positions, then return it back to its original position, only to return the shortest of those 9 virtual positions.

There’s also a lot of code duplication.

Still, it’s not horrible code. But it’s up to 9x slower than the distance_to_object() function it’s based on, if you’re wrapping in both directions, which will probably be common. I didn’t think that was good enough.

Rather than check each virtual location to see which is the shortest distance, we just need to know whether the horizontal and vertical distances are more than half of the width and height of the wrap region. If they are, then it’s shorter to go around the wrap. To know this, you simply take the x and y values of the two positions, subtract one from the other, and compare to the size of the wrap range. Once you know which virtual position is the closest one, you can temporarily place the calling instance there, and use distance_to_object() to get that distance. Put the calling instance back where it was, and then return the distance.

I realized as well that depending on whether the calling object wraps in both directions, you may not need to check for a wrap shortcut in the horizontal or vertical. So we can potentially avoid doing some or all of the checks depending on whether the do_wrap_h and do_wrap_v arguments are true or false. As well, this means we can avoid declaring certain variables if they’re not needed, which conserves both execution time as well as RAM.

I usually create local script variables in a var declaration, and assign the arguments to them so the code will be more readable, but I wanted to avoid doing that so that this function could be as lean and fast as possible. This might be an unnecessary optimization, but that’s hard to predict since I have no way of knowing ahead of time how this function might be used in a future project. In a project with many wrapping instances, it could very well be called many times per step, and every optimization could be critical. Since the script is intended to be included as a function in an extension, once I have it working properly it shouldn’t be opened for future maintenance, so making the script readable is not as important. So I opted to remove the local variable declarations as much as possible and just use the argument[] variables directly.

Also, to ensure that the wrap range is defined properly, in the non-optimized version of this function, I declare x1, y1, x2, y2 and assign their values using min() and max() so that (x1, y1) is always the top left corner, and (x2, y2) is always the bottom right corner of the wrap range. Technically for this function, we don’t care precisely where the wrap range is, only what the width and height of the wrap range are. That being the case, I can further optimize what I have here, and rather than use min and max, I can just take the absolute value of the difference of these two values.

It turns out that the process I went through to optimize this function is pretty interesting, if you care about optimizing. So I’ll go into greater detail at the end of this article about the approach I took to get there. But for now, let’s skip ahead and look at the finished, optimized function. Here it is, re-implemented, this time doing only the minimum amount of work needed:

///iw_distance_to_object(obj, x1, y1, x2, y2, do_wrap_h, do_wrap_v)

///@description iw_distance_to_object returns the shortest distance in room pixels between two objects in the wrap range, 
///taking into account the horizontal and/or vertical wrap properites of the calling object.
///@param obj the id of the target object
///@param x1 left x boundary of wrap range
///@param y1 top y boundary of wrap range
///@param x2 right x boundary of wrap range
///@param y2 bottom y boundary of wrap range
///@param do_wrap_h set whether the horizontal wrap is on (true) or off (false)
///@param do_wrap_v set whether the vertical wrap is on (true) or off (false)


if !(argument[5] || argument[6]) //not wrapping actually
{
 return distance_to_object(argument[0]);
}
else
{
 //We're going to figure out which virtual position is the nearest to measure from
 //To do that, we have to compare the h-distance and v-distance of the calling instance and the target position
 //If this distance is <half the range size, then the original position of the calling instance is closest
 //Otherwise we have to use one of the virtual positions
 //Then we're going to temporarily put the calling instance in that location, get the distance, and put it back 
 
 //arguments
 var tempx = x, tempy = y;
 
 if argument[5] //do_wrap_h
 {
   var range_width = abs(argument[3] - argument[1]);
   if abs(x - argument[0].x) > (range_width * 0.5)
   {
     x -= sign(x - argument[0].x) * range_width; 
   }
 }
 
 if argument[6] //do_wrap_v
 {
   var range_height = abs(argument[4] - argument[2]);
   if abs(y - argument[0].y) > (range_height * 0.5)
   {
     y -= sign(y - argument[0].y) * range_height;
   }
 }
 
 var d = distance_to_object(argument[0]);
 
 //return calling instance to where it was
 x = tempx;
 y = tempy;
 
 return d;
}

We don’t need to measure all nine distances to know which is the shortest; we can tell by comparing the direct distance to the size of the wrap zone — if it’s less than half as big as the wrap zone, the direct distance is the shortest. If not, then we need to wrap. We can check the x and y axes separately, and if both are called for then we can just combine them.

The second function should be much faster to execute, and uses less RAM. How much faster? Well, let’s do a test project using my Simple Performance Test and compare.

Download the iMprOVE_WRAP distance_to_object test project

It turns out that the improved code runs about 50% faster than the old code! That’s a measurable and worthwhile improvement. Although, that said, the old function ran well enough that I could have released it, and likely it would not have been a problem for many uses, particularly in Windows YYC builds.

Appendix: Optimization through Iteration

(more…)