DEV Community: Fabian Fabro

Pure Data - Visual Programming Language for Audio

Fabian Fabro — Wed, 07 Aug 2024 18:59:42 +0000

What is Pure Data?

It is an open-source programming language that is used to process and generate audio, video, 2D/3D graphics, interface sensors, input devices, and MIDI.

This was developed by Miller Puckette, who is also the author behind Max/MSP, the more well-known visual programming language for audio and multimedia.

The way the language is set up is using boxes of objects and placed in a window on a "patch" file.

Here is an example of how it looks.
Note: This is using "PlugData" program, which is more modern looking version from the original PureData program.

You can download Plugdata here: Plugdata
To check out and download the original PureData: PureData

There are 3 types of objects that are being used here for this "Hello World" app.

How to input objects:
In PlugData, at the top left menu, there is the "Add Object" button that opens a list of objects to drag and drop.

For keyboard shortcuts, I used for the following objects:
Bang: Control + Shift + B
Message: Control + 2
Print: Control + 1

At the top, is a GUI object known as a "Bang" operator.

The middle is a "Message" object. Notice at the right side with the darken edges.

And the bottom is a general object using [print].

When the user clicks the "Bang" button, it sends an immediate trigger action to its outlet (the cable connected from the bottom to the next object).

Following the strings of lines, the "Message" object can contain any type of data: a string, int, float, symbol, etc. As it is object that represents data. In this case, it represents as a string "Hello World."

And to the last box, "Print" which in code terms, is a keyword that's part of PureData, which basically prints to the console on the right pane.

The 2nd Hello World also allows to set the string as a 'parameter' to the right of "print," though the console does print it slightly different with the message being first, and printing the "Bang" operator.

Alright enough with words, Pure Data is meant to sounds, so I'll demonstrate a simple sound synthesis.

Note: Make sure to check your audio settings from PlugData/PureData that way you know you'll get audio feedback. Make sure your Audio Output "Output Device" is set to your designated speaker you're using. Sample rate should be fine with either 44100 or 44800. Buffer size is fine at 480, usually I'd go higher to at least 1024 just to get more audio buffers.

IMPORTANT NOTE: I'm glad PlugData included this, but adjusting your volume here is important since anything could happen when testing sound. So keeping at lower volumes is highly recommended. I've had moments where audio blasts and pops up while I'm experimenting around, which could damage your speakers, AND YOUR EARS!

LET THERE BE SOUND!!!!

Congrats, you made your first oscillator sine wave! OMG MAKE IT STOP!! Just disconnect the last two cables under [dac~].

Now what's going on here?

You noticed something about the colors with these objects. The sockets above the box are called "Inlets" while the bottom ones are called "Outlets." Basically they are the inputs and outputs that are connected through each box. But notice the colors too. These have orange sockets on the left side, and blue on the right side. If we look back on the Hello World, they were all blue.

The way the objects interact is through the "Audio Rate" and the "Control Rate."

In a nutshell, Audio Rate is connected through the orange inlets and outlets, while the Control Rate is connected by the blue inlets and outlets. Also if you notice the squiggly line on the audio rate objects (Tilde), is also mainly for audio rate objects.

Audio Rate objects handle anything that affects the audio signal directly, while Control Rate objects handles any informational or calculations to set values to audio rate objects.

Alright, let's at least go over what is going on with this oscillator.

[osc~ 440]

'osc' is built-in PureData that generates a cosine wave. Adding 440 gives the frequency input of 440hz, which is A4 (musical note A above middle C).

[*~ 0.5]

This multiplies the signal by 0.5. In this case, this multiples the oscillator to lower its volume.

[dac~]

This stands for "Digital to Analog Converter." The opposite of this is [adc~] for "Analog to Digital Converter."

For digital to analog, this means that we are taking a digital signal, in this case the oscillator we are feeding it, and outputting it to a speakers on the Analog side.

If it was analog to digital, it would be at the top of the patch file, and it could take in microphone input, because the analog signal is coming from what's being heard from the microphone, then translating it to digital data to be able to process the audio signals in the rest of Pure Data objects connects, and finally end with a dac~ to output it.

Notice that [dac~] is connected by 2 inlets from above. This is to output on the left and right speaker. Try to remove one of the cables, and you will hear the audio from that speaker only.

What's at least nice on this interface, and also on Pure Data's program, is that when you right click on objects, and click "help," it opens a new tab with descriptions and examples of that object.

There are keyboard shortcuts you get used to when adding objects around. PlugData is nice that it gives you the toolbox at the top bar to add all the objects.

Here are samples of my Pure Data sound design project that I worked on where I designed sound effects for a small RPG game I developed called "The Mysterious Gate."

Excuse the mess of all the objects scattered everywhere. I wish Plug Data had more features to organize sections across the patch. Or maybe there is, I just haven't figured it out yet.

Few examples:

Charge up Magic

Ice Damage

Check out the game here.
Free to play on browser at itch io: Play The Mysterious Gate

From what I learned so far, visual programming is a nice gateway for non-programmers to get into programming if code is overwhelming. I do come from a coding background and at first I was intimidated or perplexed seeing visual programming because of not seeing them as lines of code. However, I decided to keep an open mind and now I appreciate both worlds of traditional code and visual programming.

And especially for my journey in learning more audio programming, I think learning Pure Data has helped me grasp a little more understanding in this topic because going into the traditional code route for audio programming was extremely difficult for me at least.

How did I pick up Pure Data? I took a course from School of Video Game Audio.

Why I learned Pure Data? Since being interested in Audio Programming but not having a Computer Science degree feels like it's made my journey tougher to learn, but Pure Data feels like a nice stepping stone to understand a little bit of what's happening in a visual perspective with audio signal processing.

What other visual programming languages can this transfer to after learning Pure Data?

Max/MSP is a common one that I've seen musicians and media artists use, although it is not free.

MetaSounds in Unreal Engine is a pretty recent tool within the engine.

I just wanted to write this quick short introduction to Pure Data since I've been working with this recently, especially with my sound design work with sound synthesis. I hope this helped with taking a little jump start into it. I'd say YouTube tutorials really help to follow, especially since this is also for getting listening feedback that you can't always get through following a 'words only' tutorial.

Resources:

School of Video Game Audio
Plugdata
PureData

QCGInteractiveMusic YouTube Playlist

Sound Simulator Playlist

Experiencing different types of Technical Interviews

Fabian Fabro — Mon, 14 Mar 2022 02:31:53 +0000

200+ applications, 100+ rejections, 100+ ghosting, maybe about 10 interviews. I've felt a part of my journey has finally been reached as I have finally landed my first software engineer role! After 3 years since graduating from a coding bootcamp, managing through a global pandemic, and trying to take on side projects, the experience has been such stress-inducing and painful at times.

Landing that interview has always been one of the most stressful parts to applying for a job. You are always greeted with the "Thank you for applying for . Although your skills are indeed impressive, we decided to follow through with another more qualified candidate," or "We regret to inform you that you are not selected at this time being." What makes it even worse is when you see the bottom of the email No reply. So, we can't even question why we are not even selected for position, left wondering in the dust like what is it that we need to improve in our application, resume, cover letter, etc.

I do just want to briefly discuss at least trying to land an interview. If you have landed that job interview, you can skip this section and scroll down to the "Types of Technical Interview Section."

Landing an Interview

Have your resume/cover letter reviewed by trusted friends & family, getting multiple perspectives if you can, either tech or non-tech perspectives
Get your resume reviewed by a professional resume writer, there is a cost for this.
Have at least 2-3 projects included, can be roughly made (for instance, having a git repo and video/gif footage demo of the project), but polished would definitely be a plus.
When adding your projects, add 2-3 brief bullet points describing the work that YOU implemented.
Add skills that you feel comfortable in, possibly intermediately where you can at least work with OOP, if you can at least create simple CRUD applications, either console app, full-stack, static sites, etc. (can differ depending on the role you are applying for)
Side note for point above: Depending on the position you are applying for, research the types of projects built with those tools that you want to learn.
Education, for someone like me who does not come from a traditional computer science college background, add your degrees you have, don't lie even if it's unrelated. But needn't include GPA, most of the time it is not necessary.
Finding a mentor or career coach. These can help out with guidance if you sometimes feel lost in the process. Heck, I would still look for mentors just to gain more insights even now.

Types of Technical Interviews

Coding Problem Interview (With Interviewer)

This is the most common one that I have experienced. These can range to maybe from 30 minutes to 1 hour. You probably are familiar with Leetcode & HackerRank. So pretty much, it's you in a video call with the interviewer, they will give you a coding problem prompt, giving some form of input, and you need to return an output.

When you get stuck or possibly drift off in the wrong direction for the solution, the interviewer may sometimes give hints. They can occasionally ask what you are trying to do at certain moments.

Coding Problem Interview (Without Interview/Automated)

This is another common one too. These are mainly timed, as if you were taking an online test and seeing that clock ticking on the corner. These can range to maybe from 1 hour to 2 hours. You would be given about 2 to 4 coding problems to solve. They can possibly have a difficulty ratio - 1 easy 1 medium : 1 easy 1 hard : 1 easy 2 medium 1 hard, etc.

There are certain rules when these are conducted:

usually not allowed to use another electronic device
google is allowed but limited to search, for instance, syntax lookup is fine, but searching how to implement something won't be allowed
only the interview window is allowed to be open, no other browser windows, tabs are fine in the interview window for the searching though
your screen can be video captured to avoid any cheating involved
Webcam video recording may be captured during the interview for cheating purposes that can occur

Projects Interview

From my experience, I've gotten this type mainly from startup or smaller companies. I have also had friends share to me their experiences with these types of interviews.

These can possibly range from 45 minutes - 2 hours. While with the interviewer, they make ask for you to build a mini functioning project, simple web app, static site, etc.

For example, a prompt could be "build a home page for a restaurant, with a navigation bar, title with a logo, description on the page, have the links go to other pages (can possibly go to blank 'under construction' pages but still based on the restaurant's site), sticky header/footer, with button interactions, etc."

The interviewer can be suggesting certain syntax or functions to include, in a way, feeling like they are testing your knowledge with the language. Of course they would allow you to search up these syntax since they would be understanding not knowing everything.

Take Home Project

This is similar to the above but with a 1 to multiple day span. I have had take-home projects that were due a few hours on the day of, literally after the interview. I had one due in 3 days. I have also had a week take-home project. The scale of these projects could be a bit bigger, from building a full-stack web app, front-end only, etc. This can vary depending on the role.

Portfolio Discussion

I was surprised to have gotten this one since I did not see this in majority of my interviews. The interview would ask about the projects you have included in your resume. They would ask for you to share any video demo footage of project and also to open the git repo code. The questions ask could range from "what was one feature you felt was a struggle to implement and how did you overcome it?" or "If you were to add ______ into this project, what would you use or how would you implement it?" This was such an interesting but also kind of tough because I had to really think outside the box for new ideas and situations the interviewer would ask about my projects.

3 years since I started my tech journey, and I can only say it has just begun with my new job. It's a painful struggle, I understand. I've had my fair share of working a side non-related job even after finishing my coding bootcamp studies since it was that difficult for me, plus with the pandemic hitting too.

I hope this blog will be helpful and beneficial for anyone in the endeavors towards their tech journey!

Unreal Engine + Wwise - Managing Music State throughout Multiple Levels

Fabian Fabro — Sun, 09 Jan 2022 16:06:01 +0000

So, for the past few months since my last blog post, I hit a brick wall in my progress because I was stuck with this problem when I finally got the chance to work on the music implementation for my game "Maliksi." This led me to take a rather long break out of near frustration and stress, especially because I wanted to move on to other projects, plus I got busy with other obligations to worry about. Now entering 2022, I'm wanting to finally try to finish this game once and for all!

Side-note: I'm still in current development, but I really felt the urge to blog this as soon as I found a working solution.
So it might be a little messy with my current way, but I wanted to preserve this information for future references.
There are probably better ways to do this, but I was at least able to get this to work for me.
For this purpose, I'm using Unreal Blueprinting.
Also, I'm no expert in Unreal Engine, as I'm still learning this engine myself too.

Problem to solve:

To play a Wwise music event and control its state when transferring to different levels in Unreal Engine.

Coming from a Unity background, I was used to the "DontDestroyOnLoad" feature attached to a game object, where I could let that run even when transferring to different scenes, all until the game closes. From my research high and low, Unreal Engine did not have a component like that.

If you are not sure how to play a Wwise music event into a scene, here is a youtube video for playing an SFX object, but it's essentially the same way for music.
How To Integrate Wwise With Unreal Engine + Play An Ambiance SFX by Scott Game Sounds
Time Stamps:
10:20 - Wwise Setup
15:20 - Unreal Engine Setup

For a brief summary, the life cycle of a game scene or level (this applies to both Unity & Unreal Engine),

Game objects & actors are created as soon the level is opened.
As soon as the game objects/actors are spawned, they fire their starting function:
Unity -> Start() or Awake()
Unreal Engine -> BeginPlay()

What this problem lies is, I want my Wwise music object to be playing. When it transfers to the next level, I want the Wwise music object to seamlessly change to the next track. However, that Wwise music object will get destroyed as soon as I transfer to the next level.

I could just create a new Wwise music object every level and play the music for that level, but I also had a change for vertical orchestration layering. Here is my setup for how I want my music to play.

*Rondalla is Filipino traditional music involving plucked instruments similarly compared to the mandolin.

So going back to an early point I brought up, from Unity "DontDestroyOnLoad" was something I needed to find something that's similar on Unreal Engine. Here were two options I found:

Game Instance
Persistent Level / Level Streaming

I experimented with both, but which one worked in my favor at least? The Persistent Level choice. You can skip to the Persistent Level section to read about using that one, I wanted to share my experience with the problems I occurred trying out with game instances.

Game Instance

So what are game instances? These are objects that are created as soon as the game loads, mainly like in it's own storage. You can possibly think of them as static since they can only exist once but its values can be altered.

For example, I use game instances in my game as unlocks, when Malik defeats a boss, she gains a new skill unlocked, which retrieves the game instance to tick that THAT certain skill is now unlock because the particular boss is now defeated. With this in mind, I thought I could use this to save the state of the game object that contains the Wwise music event. However, I realized that you can't save objects or actors onto game instances, only variables with data type values such as int, boolean, string, etc.

Although, I could possibly still use the game instance to save and alter the state value, but I ended up not doing it when I tried implementing the 2nd option.

Persistent Level / Level Streaming

What is a persistent level? It's a level that will be continuously running along with your main levels and maintains any objects contained in that level. For example, this can be applied to maybe having the character UI run throughout the game or certain environment assets.

Steps

1 - Create a new empty level, name it anything you want, I named mines "MusicManager".
2 - From your first level opened, mine specifically is "Title", open the levels window.

3 - At the top should say "Persistent Level," drop the "MusicManager" level under.

4 - Right click the "MusicManager" Level in the Windows Level, go through to the drop menu to "Change Streaming Method" and make sure the "Always Loaded" option is checked.

Note: Streaming Mode

blueprint: when you want to load your level manually, using "load stream level/unload stream level" in the blueprints
always loaded: as soon as the level loads, it will continuously run until the game ends

4.5 - On the bottom right, you can switch between managing the Persistent Level or the Music Manager. Make sure to set to Persistent Level when you are ready to play test.

5 - Create a new actor new blueprint, I called mines "musicObj." Attach an AkGameObject component (or Ak component, I believe either should work).

Get the current level name, use that as the condition for a switch statement for each of your levels (you can probably create an array too to manage the levels as well).

6 - The starting condition will immediately set state to the music state event I have already set up in Wwise and then post event. Make sure that the post event references itself.

7 - As the following conditions, it just has to set state.

7.5 - Note, if your music event ends, such as playing an ending stem but still want to play the next music event, just add another post event after. I did test this and it worked.

8 - For each level, I attached the "MusicManager" level on to every level, repeating steps 3 and 4.

Although, I still have not tested yet on retries when my character dies or back tracking in my levels (although for my current design I'm not allowing back tracking). I presume at least with this set up it should allow it to set state and change the music state event when you back track levels too.

Here is also my current set up for the Wwise side as I have the music states set up accordingly.

Failed Attempts I tried & problems that occurred:

I tried to utilize Switch Containers within one of the levels because I wanted to Vertical Orchestration seamless transition loop to work, however, it just did not cooperate with me just resorted to making it as state values.
Game Instances could not keep the musicObj alive as I tried to retrieve it through each level, but it only can hold variable values and not blueprint actors.
I tried to manage changing the state on the level blueprints for each level but it meant I had to keep retrieving the musicObj every level that way, and it was not able to fully access and change the state from there.

With this hurdle finally overcome, I feel that I can finally finish this game project once and for all! Thank you for reading and I hope this helps for the game audio folks because it was really difficult for me to find resources on Unreal Engine and Wwise for music, especially when working with multiple levels.

Here is a preview and link to my upcoming solo game "Maliksi."

Maliksi Itch.io game link

Maliksi Main Theme by Fabian Fabro (FirahFabe) Music Preview
Youtube Video:

Participating in Devtober 2021

Fabian Fabro — Sat, 09 Oct 2021 09:31:43 +0000

You can skip this portion as this will be just me venting here.

Its been a long time since I've posted a blog here. A lot has been going on in my life right now. Still in on the purpose of trying to land that first job, but of course I've gone through a bit of a burnout of applying for jobs again. I have also taken a break from programming the past couple months, so I feel rusty in trying to get that workflow back. It just doesn't help either when I have family who just doesn't believe in me for these goals to have that moral support, so it's tough to really maintain my mind set for this.

Alright, so I have decided to start Devtober, which started the beginning of this month. I will be continuing a game project I actually started for a different game jam back in July but didn't finish. My goal is actually wanting to finish this game by the end of this month, especially since I feel I am almost finished with it.

Maliksi

A 2D Action Platformer Boss Rush game where you play as Malik, an Eskrimador, who must retrieve the stolen light from Filipino mythological-inspired creatures to save the village.

I have been working on the art, programming, sound design, and music. I am learning Unreal Engine with unreal blueprinting (not C++ because I don't feel ready or comfortable in the language yet). Which it took me awhile to get used to visual programming.

I am using Wwise to implement the sound effects and music.

I'm keeping my scope as tight as I can in order to finish this game. I still need to polish one of the boss fights, but the final boss still needs the functionality and mechanics implemented.

Things that I need to work on for this last push for this month:

polish one of the boss fight's mechanics
final boss still needs functionality & mechanics implemented
Waiting for musicians to record their parts for the music so I can start mixing
Implement the music into the game
Few more sound effects to design
Slowly implementing sounds in each level
Ending still needs to be created

1st Video Footage

2nd Video Footage

I want to publicize this to hold myself accountable to finish this game. I did participate in Devtober 2020, but I never finished that game, but this one, this one for sure I want to complete, since I've been working on this one for a long time already. Plus, it is also Filipino American History Month, and this game showcases my Filipino culture that I have incorporated into this.

If you want to participate in this game jam, here is the link to the itch.io:

https://itch.io/jam/devtober-2021

Setting up an FMOD Project + Unity for Collaboration (Git)

Fabian Fabro — Mon, 01 Feb 2021 21:09:53 +0000

This blog post is based on FMOD ver. 2.1

So I just participated in this year's annual Global Game Jam. This marks my 4th year in this specific jam (besides doing other in between jams over the year).

My first 2 years, I would take on the role as a composer/sound designer. These past 2 years, since getting into software development, I've been aiming towards wanting to do my technical sound design and audio implementation.

Okay, enough backstory about myself, let's get to the topic!

Topics

Gitignore
Source Control
Build Files
Other Resources

Setting up the Gitignore

For many collaborations I've worked in, most programmers have complained how much Git behaves with Unity. Git is notorious with working with Unity. This is also based on my personal endeavors with Git and Unity as well. I'm super glad to finally have found the template to handle FMOD's files into Unity for Git.

Here is the Gitignore template:
Reference from Paalo's post in the FMOD Forums

### FMOD Unity Integration ###
# Never ignore DLLs in the FMOD subfolder, so make sure that the 
# FMOD-folder is in the correct path.
!/[Aa]ssets/Plugins/FMOD/**/*.dll

# Don't ignore images and gizmos used by FMOD in the Unity Editor
!/[Aa}ssets/Gizmos/FMOD/*
!/[Aa}ssets/Editor Default Resources/FMOD/*

# Ignore the Cache-file since it is updated locally either way
/[Aa]ssets/**/FMODStudioCache.asset
/[Aa]ssets/**/FMODStudioCache.asset.meta
/[Aa]ssets/Plugins/FMOD/Cache.meta
/[Aa]ssets/Plugins/FMOD/Cache/Editor.meta

# Log-files
fmod.log
fmod_editor.log

# FMOD 1.10.x (Legacy)
/Assets/FMODAssets/*
/Assets/FMODStudioCache.meta

With web development, Git is nice with because of the size of files. When it comes to Unity, Unreal Engine, or other game engine, it's a tricky story because now we are dealing with an LFS (Large File Storage) territory. That's for a whole other topic, I've had some failures from a past game jam collaboration trying to incorporate LFS. You can give it a try here.

Audio Files are no joke because we mostly work with .wav & .ogg format. And .wav files can get pretty big, especially in music when dealing with instrumentation, time length, sound libraries, etc.

Personally, I use .wav when I compose music or sound design in my DAW. Except if I'm using Wwise, I'll still work with .wav, but compress the files into .ogg in Wwise, for compressing file size, but that's going a bit off tangent.

This may differ for other people's setup, when our team was setting up the game project for a Git repo, we had 2 gitignore files. One gitignore was on the outside folder from the Unity Folder, and the other in the actual Unity project folder. I would just copy this FMOD template in both gitignores just in case.

Source Control

When using the Git Terminal, the most common commands to utilize is:

This will add all files you want to be pushed to the repo

git add .

You can customize your message, that will be seen on the repo, best to add a message that relates to your latest progress you made

git commit -m "add commit message here"

You are grabbing the files from the repo in case other collaborators pushed their files and changes.

git pull

You push up your files and changes.

git push

When collaborating with others, it's safer to git pull before git push in order to avoid less merge conflicts (which will happen, it WILL HAPPEN).

=============
3/15/21 Edited

git reset --hard <commit code number> //

The commit code number usually looks like a string of random letters and numbers like this (without the < > surrounding):
64f31ej5q1684028501w1389ui67nv2z745h7234

Which can be found in the git repo, That's what those git commit messages are for!

Git reset is used to revert back to a previous commit, or previous version if you find yourself messing up on your current progress.

=============

Just based on experience, the other collaborators will somehow get some meta or cache FMOD files from your pushes. Your FMOD project is the main one that you will be merging anyway to resolve since you are in charge of making changes to those FMOD files.

Build Files

Creating the export files differs on which route you take on how you import them into your Unity project. It's listed in the FMOD Documentation here: Accessing Your FMOD Studio Content

Which they do a wonderful job explaining it, but I can try to add on my experience as well.

So in Unity (after installing FMOD into your project), on the top menu, go to:
FMOD -> Edit Settings

And inspector should open FMODStudioSettings

There are three ways on how you want to share your FMOD files into the collaboration space.

Project
Single Platform Build
Multiple Platform Build

Project: This way, you can drag your whole FMOD Project Folder and add it into your Assets Folder in Unity (to be in the folder level next to scripts folder, prefabs folder, scenes folder, etc.)

Looking for the .fspro project file.

This would probably be used if you are collaborating with multiple game audio members who need access to FMOD project, because you are essentially adding the whole project folder into the Unity folder. This also means that there might be complications with the other non-audio members that don't want to deal with merge conflicts of accidentally altering any files with FMOD, whether it could be from hidden meta or cache files.

Single Platform Build: This way is instead of bringing your whole FMOD Project folder into the Unity Assets Folder, you just need to bring in the bank files you generated from FMOD. In case you're not sure how to generate the bank files:

File -> Build

It should have generated these files.

When you have installed FMOD to your Unity Project already, you should have two folders called "Gizmos" & "Plugins" generated in the Unity Main Assets folder:

Gizmos carries the an FMODEmitter file, while the Plugins carries the cache data.

You can create a new folder in the Unity Assets Folder, calling it "FMOD Banks" (feel free to name it the way you want, as long as you negotiate to your collaborators for file/folder name conventions) for organizational purpose.

For me, I dragged the FMOD Master.bank & Master.strings.bank files into the FMOD Banks folder. I created the FMOD Banks folder in Gizmos (you don't have to only place it there). I would have probably just place FMOD Banks Folder in the main Assets Folder.

You can also set up an automation location of where you want your FMOD builds to generate.

Preferences -> Build Tab -> Browse where you want to generate the build files

In this case, you would place them into your FMOD Banks folder in your Unity Assets.

Multiple Platform Build: This way is when importing from multiple sub directories. Personally, I have not used this yet because I have not built a game in this scope yet. However, in the FMOD documentation, its use is when you want to use separate .Bank files for separate builds based on each platform (console, PC, Mac, mobile, etc.).

The main goal is having the files in the Unity Project folder in order to keep it in scope of where the Git Repo is located.

I've been curious about looking at alternatives of Git with Game development. I've been suggested Perforce, which I'll probably look into the future since I've heard that is AAA studio standard. I think it's because Git is open source and easily accessible to a wide audience, which commonly most people go to for source control.

My way of working with FMOD may not fit for everyone, feel free to add your workflow with it. I'm mainly sharing my experience that I've had a rough time before with dealing with setups for collaborations but now finding something that actually worked for me.

I just hope this helps with game audio folks getting into collaborative projects.

Here are more resources for things discussed in this blog.

Gitignore

How to use gitignore file

Git LFS (Large File Storage)

Getting Started in FMOD

Perforce

2020 Update & currently recovering from burnout

Fabian Fabro — Mon, 19 Oct 2020 20:31:07 +0000

Welp. I haven't blogged since the beginning of this year. It's funny, a self reflection blog I wrote back in January, ready to make changes in my life and conquer, then COVID hits right near the end of February. What timing huh? Did I feel I was able to get through things? I wouldn't say the least. So much has happened in these couple of months.

I'll be ranting and just talking about my life update, so if this doesn't interest you, I don't mind if you stop reading and skip this article. This is just another one of my life rants.

March - April:

Slowed down in productivity progress on personal projects
Joined and collaborated in a few game jams
Constant job rejections, no interviews
Animal Crossing

May:

Even slower progress on productivity
Playing games
Stopped applying jobs

End of May - June:

George Floyd happened
Developed and deployed a web app out of anger & remorse for the Black Lives Matter movement
Submit more job applications, more rejections, no interviews

July:

Participate game jams
Work on game projects
Finally got an interview for a technical sound designer position, submitted the take home technical audio test

August:

Waiting for results for the technical sound designer position, sent follow-up emails...no response.
Work on personal projects
Submit more job applications, more rejections, no interviews

September:

Participate in another game jam
Work on personal projects
Play games

Now it's October, what am I doing now?:

I'm participating in yet another game jam, but this time it's a solo game project
Picked up pixel art drawing, for the game project, but also because I've been wanting to draw too
Starting FreeCodeCamp's Data Structures & Algorithms Course, because I'm feeling really rusty in those topics, especially since I haven't been practicing coding problems either.
Started Twitch Streaming
Submitting some job applications, rejections as usual

I wish I can share a success story about myself, but no, a year of finishing a coding bootcamp & COVID happening, it feels even harder to find a job.

My title also mentions my burnout. Well, I'm tired. I'm tired of applying for jobs and just getting rejected. I'm tired of questioning my self-worth because of these rejections, to not even be able to talk to a human being. I would rather get rejected at the interview stage, that way I know I talked to a human and got feedback, but no, just a robot email echoing to me:

Unfortunately, "Insert Company Name" has moved to the next step in their hiring process, and your application was not selected at this time.

I just want to create stuff. But I'd like to also be able to be financially be stable to create stuff for people.

Self Reflection 2019, Improve Myself for 2020

Fabian Fabro — Wed, 01 Jan 2020 05:04:11 +0000

This is going to be a small, casual post. Less tech but still tech-related topics. And probably a big rant as well.

The past two months have been a very stressful endeavor, especially this month. I won't go into full details since I feel like they are a bit too personal to share.

These past few days for the new year, I decided to take some steps back, self reflect on myself, ask myself many questions, watch videos on personality traits and the psychological self.

I think the biggest cause that made me have this self reflection was seeing constant rejections from job applications, which made me start doubting my skills and my self worth. I noticed my doubt as I started picking up C++ and the JUCE Framework, and saw that trying to debug the tutorials was stressful, especially when I was barely familiar with the language, and the videos mentioning how I should have some better background in DSP (Digital Signal Processing), which I barely have (I have written a blog on it before that was not even scratching the surface). So I decided to take a step back from the JUCE Framework and relearn C++ and dive into DSP, which did not go well. C++ is a difficult for me to grasp and most DSP sources always started off with math equations and algorithms, and my math is actually really terrible, so I couldn't grasp that side. What I'm saying is that DSP is not a beginner friendly topic according to one of my friends that studied it is currently a data scientist for Microsoft. So now, I feel like I have to down another rabbit hole and start learning the math requirements to understand DSP. This has discouraged me so much that it started affecting my programmer self. Am I good enough to even tackle on this? Why am I even tackling this when I could be working on another thing? I'm struggling with this language and this topic that I sometimes forget why I'm studying it.

My biggest stress along with this endeavor is balance. Because I'm trying to tackle in two industries for two different positions, a Software Engineer in the Tech Industry and an Technical Sound Designer in the Game Industry, but if I were to combine it, it would be an Audio Software Engineer.

I feel like I've lost focus because trying C++ has been a struggling beast for me. My focus has been all over the place this month, I've had over 100 tabs open, for every single error I'm always trying to debug, a chapter from DSP readings, youtube tutorials of trying to do certain things in C++, Linkedin Jobs, Indeed Jobs, social media, JUCE Framework, game-dev related articles, miscellaneous Youtube videos. I'd be constantly switching over tabs to a different task every 10 minutes or so. I'd be scrolling my phone every 5 minutes through social media, refreshing, refreshing. I'd be switching through different programs to my IDE for coding work then on my music work if I got a C++ bug and couldn't solve it in the next 5 minutes and wouldn't return back to my IDE for a long while. I wouldn't even shut down my computer when I go to sleep, just to keep all my tabs and programs open so that way I see it once I wake up the next day. I felt that this had ruined me. I felt broken as I felt scattered with no system. I felt when I graduated from my bootcamp, I had the drive and energy to keep learning because my target of topics was more open, but I think when I tried to dive into topics that I should be going for career wise, I've been straining to drill these topics but they were just not sinking in as I was tackling a way over out of comfort zone. I think it's because I've mostly worked with web development that I could say it is my comfort zone, and tackling new topics within web development doesn't feel too daunting even when it gets hard. However, I think since tackling C++ and trying to actually work with stand-alone application, or not web (can't think of another word for it), development, this was a huge step out of my comfort zone with programming.

With all the above stressing me and my mentality losing it, I kinda broke down this Mid-December. That's why these past few days I took the time to not actually plan out what to do, but question myself and question everything I'm doing at this point in my life right now.

Here are some questions I asked myself:

What do I want to do in the game industry?
What do I want to do in Audio?
Why did I go into software engineering?
Why do I do music?
How do I achieve parts of goal?
What are the steps I need to take?
How do I break down topics to work on?
How do I balance audio and tech?
What do I need to work on for myself?
How do I work on my soft skills?
How do I work on my introvertness?
What do I see myself doing as a living?
Sometimes I feel like I’ve disconnected myself from friends.
How do I reconnect with friends again?
How do I overcome my introvertness?
How can I improve my mentality to overcome imposter syndrome?
How do I try not to burn myself out?
Have I stacked upon so much on myself on what I want to do in life?
How do I recognize my self-worth after facing so much rejections at once?
I try to do things on my own and don’t reach out for help when I should because of being afraid of being rejected or ghosted to not ever get a reply back as I would probably be waiting for a long time.
How should I approach my job search if I’m trying to pursue in two industries simultaneously (Game Audio & Tech)?

I'm sure sure not all of you have the same questions about yourself with these, but just think about, and really break down what you are trying to question about yourself. Why do you exist? Why do you do what you're doing? Why did you choose programming? How are you still going even after facing so much rejection?

That is why I want to try an approach for myself. I've never been a tact with working with schedules. I usually work in whatever flow I'm in, and sometimes I felt that sometimes a disorganized approach. It's nice because that flow is a nice driving feeling. But now, I wanted to try this for 2020. I want to schedule in every detail that I know what I already do and work on them in a timely manner.

I decided to try out the Taskade Firefox Extension App to help me plan out my daily agendas and calendar events. I'm going to try out the StayOnTask Android Phone app to notify me if I'm being productive with my time. I'm trying to plan out more agendas, like breaking down the topics to study that way it's not vague like "Continue learning C++." I feel like I would need to break it down where it's like "work on the next 2-3 modules in the C++ Udemy Course," with "Practice building a small app based on those 2-3 modules." My goal is that if I can check off at least 50% of daily agenda tasks a day, then I'd feel accomplished in that sense.

I want to try out the #100DaysofCode as I feel that will fix my learning drive again.

I actually do want to start DSP blog series, especially to make it beginner friendly, but that is after I am able to comprehend the starting topics for it, or when some of it starts sinking in.

I noticed that my other last blog post was also a rant, I actually had more tech related topics saved as drafts, but I haven't gotten the chance to work on them. So, I apologize for listening to another rant, but I also thank you if you've reached this part of the blog.

To myself for 2020:

I want to be able to organize myself better
Improve my mental well-being
Get good

Sources

Taskade - www.Taskade.com

StayOnTask - StayOnTask App

Experiencing a Game Jam as an Audio Integrator & Sound Designer

Fabian Fabro — Mon, 11 Nov 2019 21:42:36 +0000

So for those unfamiliar to what a game jam is, it is an event where people of different artistic backgrounds from coding, visual art, story writing, audio, program management, game design, etc. come together to build a game revolving around a given theme within a scheduled amount of time. It is similar to a hackathon, but game-focused. One of the most well-known game jam is the "Global Game Jam" where countries all across the world spend 48 hours to build a game, from Friday early evening to Sunday late afternoon. Live streamed on Twitch to view the progress going on from France, Korea, Australia, Brazil, around many US States, and many more.

I have worked with Global Game Jam twice, but as a Composer and a Sound Designer. I just participated in a recent game jam, my third one, by Seattle Indies, a meetup for local game devs around Seattle. I'll say, music is my strong suit to my skill set to offer. However, it seemed that there were more composers that attended this game jam, so it was actually difficult for me to jump on a game project for music, so I decided to jump in as a sound designer and audio integrator. Audio integrator, I find many terms for this, Audio Implementer, Tech Sound Designer, Audio Engineering. All-in-all, they essentially do the same task for game development, program the audio assets into the game.

I have spent the last week learning Wwise, an audio middleware that makes it easier to design more control with sound assets for a game. And I had to pick up FMOD, another audio middleware like Wwise, a few days before the Seattle Indies Game Jam was coming up since I had problems installing certain Unity extension packages for Wwise in order to incorporate it into Unity projects. I've only worked with FMOD a bit back then, but only working on my original music to manipulate and control on FMOD, but not knowing the Unity side of connecting FMOD. So that was my short few days spent learning the Unity integration side of FMOD. I realized it was already rough learning the integration to FMOD, because most tutorials I found on it would only focus more on just FMOD alone, or a little bit of FMOD & Unity because FMOD is geared towards composers and sound designers, tutorials focus less on coding and more on audio manipulation, which I understand because those were helpful of my music part of me, but the coding part of me wanted to learn the nitty gritty work of connecting those sounds into the game. Working with the FMOD API and watching tutorials are always working with outdated versions, especially with the Unity update versions, so always seeing myself trying to follow along the code and finding out it poses errors that give no helpful error messages. So debugging FMOD errors is always a pain in the sense that there is not enough answers. I was basically learning Unity at the same time learning the integrating side of FMOD in such a limited amount of time before the jam.

Jam Day Kickoff!

Arriving to game jam, preparing for an intense 2 1/2 days of cramming to build a game, starting from Friday 5pm to Sunday 5pm. Ready......GO!!!

//////
Quick side note if unfamiliar of the process for game jams:
Before starting, first there is a mingle and meet, getting to know what skill backgrounds people are coming from. Then we are given a prep talk, discussing past game jammers experiences and stories, the judges, the prizes, and the THEME.
//////

The theme was "Intentionally Broken." Build a game in inspiration based on this theme.

I joined with a group, where we built a game called OSHA Coaster, Department of Laborers was a our team name for fun.

Synopsis:
OSHA Coaster lets you live out your dreams of inflicting inflicting as much pain and suffering to allow you cash in on that sweet medical bill money. Find a way to balance the rising costs of an ever-expanding theme park against the need to out compete your competitors in the industry of thrills and joys.

Play OSHA Coaster here

So, Here I am, in charge of Audio Integration

Sound design was not too bad since I've done it before, experimenting with sounds to get that right punch to that sound. It's a tiring process for my ears listening to the same 1 sec or 2 of SFX because of constantly playing back and editing.

Now the Audio Integration Part!!

If you reached to this part of the blog, I want to thank you for taking the time to reading this so far. So to start, we started up a github repo as the Unity project was starting up too. We mainly stuck with the master branch as I asked one of the experienced game programmers of why we don't use other branches, coming from a web development experience of utilizing multiple branches, but it's because of many of us having to send different types of assets, we need to see immediate feedback towards the unity project straight rather than having to send assets to multiple branches. Plus with the unity project, we were having multiple game scenes being built, which would be like the multiple projects, but it's like prototype after prototype to integrate updated assets every commit coming in.

Entering FMOD project instantiated! So I installed the Unity FMOD integration package in order to attach my FMOD project into this Unity project. Little did we know, all hell was gonna break loose soon. It started off fine, I was able to attach music to the FMOD Listener along with a SFX playing with player interaction on click events placing objects. It's all good, all good. Then...taking it on the next level of trying to control parameters in FMOD to Unity FMOD Event Emit Listener, and adding another SFX for another type of player interaction using another FMOD Event Emit Listener became a problematic start. I wanted to be able to debug log incoming FMOD events in a C# script to see what information I was getting back from those calls, but the debug messages were not helpful, then I found out FMOD had their own debugging log, but it was difficult to utilize.

Then came the Git flow control. Committing my changes with FMOD and pushing them up. Ok, just a bunch of warnings but no errors, the game can still run. Other pushes occur from the other team members. I pull down.

Everything is broken. Where is the game scene? Where are all the game objects in the hierarchy?! Why isn't anything running?? Merge-conflict?!!? Merge conflict on certain assets?? Edit them on a text editor?!?! But....it's on Unity, and these types of files are not text files? How can I resolve these merge conflicts? "FMOD Editor Log" merge conflict?!? "FMOD meta files?!" What is going on?!?!?! Umm help!! I'm sorry!!!! I'm sorry for breaking things!!!!

This was everything that was happening to me on my side with Unity. The FMOD meta files, cache, and editor log would constantly be changing with every time the game would run, which caused so much merge conflicts when I updated assets into FMOD and pushed. So the other team members had to adjust the gitignore file to handle some problems with FMOD:

.gitignore
/Assets/FMODStudioCache.asset
/Assets/FMODStudioCache.asset.meta
fmod.log
fmod_editor.log

Maybe it was the location of where I saved the FMOD project? And whenever I linked the FMOD to Unity, I think it was linking from my local machine folder and not the folder within the Unity project where the git repo was living, so this caused with breaking the connection and mixing up between my FMOD local files, to the FMOD in the Unity project directory. I thought I had to make it separate because of FMOD's project files probably clashing with Unity's directory. That was not too bad to resolve since I just had to work with the FMOD project I saved in the Unity project directory since that's where the connection is.

So I thought keeping the FMOD Project Folder separate from Unity Project Folder is how it works, but when working with a team and a git repo, adding the FMOD Project Folder within the Unity Project Directory is to do it. It sounds like such an obvious beginner mistake but I made that mistake so I'll have to accept that.

However, even with the FMOD Project folder within an Assets folder in Unity, there were still merge conflicts happening to the "fmod_editor.log" file even though it was in the .gitignore, which we were getting annoyed on why that kept showing up all the time.

Some git commands I didn't know about that I learned:

git checkout --theirs FILENAME
git checkout --ours FILENAME
git checkout :/

--theirs FILENAME = taking their changes over yours
--ours FILENAME = taking my changes

These are quick auto-resolves when handling merge conflicts. I can't remember if it worked for some of the issues or not but it was something we tried.

git checkout :/ was an interesting one, where I believe it was to ignore the conflicted commits that you wanted to push, that way you were only pushing the ones that were added and not ticked for conflict.

With all this mess in hands after facing the last few tries of working with FMOD, I decided to do something I didn't want to but knew I had to quickly adjust just so we can get the audio assets implemented into the game.

Abandon ship FMOD & work with Unity's AudioSource

So, we decided to delete the FMOD files, disconnect the FMOD components connected to the game objects, to switch to just working with the AudioSource components, which was fine even though it was basic and I wanted to customize and control the sound much more through FMOD, but had to adapt quickly and make do learning parts of the AudioSource API. I switched gears, trying to attach audios to new game objects, to prefabs, coding a script for switching music (with help from one of the game programmers). And let's update:
git add .
git commit -m "here are some sweet audio stuff in now"
git push

'git push failed to push some refs'
ok ok...

git pull
git push
pray...

MERGE CONFLICT! MERGE CONFLICT!!

WHAT IS GOING ON?!?!?!
FMOD??!!?! I Thought we killed you??!?!?!

Apparently the FMOD cache and meta files would still continue living on even not in use. Looking up some other forums of people trying to get rid of the FMOD package after installing it into their project. It was more of a hassle getting rid of than just ignoring it altogether. So we had to literally purge FMOD files, ignoring more meta and just burying it until it does not come back the light of day. And then we reached to the end with minor repercussions. I was causing less conflicts with not using FMOD and just small problems here and there. But we pull through, in 2 days, we built a game.

If you have reached here, thanks for listening to my rant about unity audio integration with FMOD. To what it's worth, I don't hate FMOD, I still like it as an audio middleware. I can see the potential of its use, same with Wwise, but hearing from experienced game programmers stories about game jams and audio middleware, and it is still one of the most challenging experiences for them that most usually don't succeed in implementing it.

But this just makes me think, there is not enough tutorials in learning the middle ground side of working with FMOD in Unity, like fully. As I see it, mostly composers and sound designers solely focus on the FMOD/Wwise/audio middleware software while the game programmers focus on Unity, or Unreal Engine, seeing that the job would be for the game programmers to be in charge of integrating the audio using the audio middleware packages. The main thing the game programmer and composer/sound designers that would be collaborating about is the parameter control, game objects, cues, triggers, variables to be able to connect and talk to each other from both softwares.

//Rant Note
I wanted to challenge myself to be able to be that bridge since I was familiar with audio middleware (from short time crunch cramming) and familiar a bit with Unity (with small crunch time learning as well). Especially when I hear from game programmers with game projects, and personally I don't like hearing this, "music and sound is not a priority in a game." Music & sound is just as an important asset to the rest of the team. It is part of the team as it is the front-end that people will hear when playing a game. Sure you can play a game on mute, but when sound effects play a big role into a game, like interactive cues or music cues, then it is just as essential as the art, story, and coding.
//Rant Note Done

Game Jam is DONE

So we built the game! There is a 2 week jam period starting now where we can polish this game and will be showcasing at Seattle's Pacific Science Center.

I want to personally thank the game programmers on my team for being understandable and patient with my mess ups that occurred throughout this journey.

My goal: I really want to implement FMOD back in because I've been wanting to work with audio middleware with games.

OSHA Coaster

You can try it out here at itch.io

https://terrehbyte.itch.io/osha-coaster

Reference a Pointer to Point to a Reference

Fabian Fabro — Mon, 04 Nov 2019 16:38:15 +0000

I have been intimidated by C++ for such a long time because of how long this language has been around and how it is a standard for most developers coming from a computer science background (I don't come from a traditional computer science background for context). I was not sure if I would have been able to tackle a low level language with my caliber of mostly using high level languages. Although, there is debate that C++ is sometimes being considered a high-level language, but it would then be kept as mid-level, as it could have traits of both low and high level languages. However, I knew I would want to try C++ at some point because many music & audio tools are built using this language. But that's besides the point. Point.

So, Pointers and Reference! Something I always heard about whenever I find topics about C++. Let's start with references.

Reference

What is a reference? Reference is the memory address of a variable. This uses the & operator with the variable name. Let's show how it looks:

int num = 3 //instantiate int variable

cout << &num;
//Output: 0x7ffcc1b8a35c

Why doesn't it return the integer 3? And what is that long bunch of characters? This is the location of memory address that is being stored into the computer data. Every variable instantiated gets stored into all these different memory addresses and we can 'reference' them using this reference operator.

The long address output, Programiz has a good explanation on that:

"The 0x in the beginning represents the address in hexadecimal form.

Notice that first address differs from second by 4-bytes and second address differs from third by 4-bytes.

This is because the size of integer (variable of type int) is 4 bytes in 64-bit system. "

Since working with Javascript & Ruby on Rails especially, I'd like to think of this analogy as every variable you are creating also auto-generates an ID to that. Then you can access those 'ids' like &variable_name. So how do we access these? We point to those references using Pointers.

Pointer

What is a pointer? According to the official Cplusplus website:
"...variables have been explained as locations in the computer's memory which can be accessed by their identifier (their name). This way, the program does not need to care about the physical address of the data in memory; it simply uses the identifier whenever it needs to refer to the variable."

Pointers Source

So we can say that a pointer is what is used to access that reference.

The syntax for a pointer is adding an * next to the variable name or to the data type, like so:

int *num;//Valid Pointer
int* num2;//Valid Pointer

And to use the pointer with the reference.

  int num = 3; //create variable

  int *ptr = &num; //create pointer and assign num variable's reference

  cout << endl << "Pointer: " << *ptr;  //Output: 3
  cout << endl << "Reference: " << &num;  //Output: 0x7ffcc1b8a35c

You can see that when *ptr printed out 3, while &num printed the memory address of num. If we tried to access num as itself:

  int *ptr = num;//ERROR: cannot initialize a variable of
                 //type 'int *' with an lvalue of type 'int'

  //or if we tried to access the memory address without using a pointer:
  int ptr = &num; //ERROR: cannot initialize a variable of
                  //type 'int' with an rvalue of type 'int *'

  cout << *ptr;

It's invalid because you are trying to access one another without using the other syntax. I would say that the pointer and reference go hand-in-hand together for access, but there could possibly be exceptions to this.

I just wanted to a basic introduction with pointers and reference, but I can POINT to some good REFERENCES to learn more about this topic, since you can work with Pointers with structs, with arrays, with functions, passing pointers, passing references, etc.

Pointers with Arrays
Pointers with Functions
Another Pointers with Functions
Passing Reference to a Pointer

Final Fantasy XII's Gambit System was like programming for beginners

Fabian Fabro — Tue, 15 Oct 2019 22:10:04 +0000

As someone who grew up with video games, especially with the Squaresoft, Square-Enix titles, I thought that Final Fantasy XII's approach to their battle system was fascinating compared to any other of the titles. If you have not played any of the Final Fantasy games, here is a rundown on how the gameplay mechanic works for each title.

In traditional RPGs, battles are engaged through 'Random Encounters.' This is generated by the amount of steps taken by the player before a battle is initialized.

//using c# to describe this

Random random = new Random();

int steps = 0;
int battleTrigger = random.Next(64, 255);
bool inBattle = false;

public void PlayerMove() {
   //Get player's controller input, if playerMove was a method a Player class
   playerMove = step + 1;
   battleTrigger -= 4;
}  

while(!inBattle){
   PlayerMove();
   if(battleTrigger < steps){
     inBattle = true; <-- Random Battle Engaged! *Cue Battle Music*
   }
}

This was based on looking at the algorithm in the Random Encounter Wikipedia link I provided, where it states:

Set X to a random integer between 64 and 255.
For each step in plains, decrement X by 4. For each step in forest, swamp, or desert, decrement X by 8.
When X < 0, a fight ensues. Go to step 1.

I felt this is more true to the Final Fantasy Random Encounters compared to the first algorithm given.

There are the exceptions of battle encounters in Final Fantasy I where battles are triggered when you step on a tile adjacent to a dungeon treasure box, (which was super annoying but you knew there was always a battle waiting for you even though you were low on health traversing in a dungeon with low MP!).

Final Fantasy I - III: Wait System

The player has all of the time in the world to select a command in their array of actions:

Example:

Attack
(Magic) or (Skill) <- According to their class/job
Defend (or Guard which is the same thing just worded differently sometimes)
Item
Flee

Final Fantasy IV: Slight Active System (Wait System Choice Optional)

Attack
Exclusive Skill to Character
Additional Exclusive Skill to Character
Item

with side commands Run, Defend, change Row (which affects all party members)