Just occurred to me that I never updated this devlog about the release...

Hey, my game's out on Steam! https://store.steampowered.com/app/869010/Katto/

I plan to make a video talking about what I've learned from the experience, so I won't go too in depth with that stuff here. Basically I learned a lot, and while I wish more people were playing it, I'm glad I was able to push through and finish it. I'm looking forward to having more free time this winter break, where I can start experimenting with new things in life. For now I know I want to read some books, exercise harder, meditate regularly, practice cooking more, and prep myself for job interviews relating to computer science.

I'm not sure if I'll continue with gamedev. I enjoyed it, but there are still so many things I have yet to try. I think if I have an idea for a game I really like I'll pursue it, but I won't try to force anything.

Thanks to anyone who read any of this devlog! If you want to follow me on future endeavors, hit me up on Twitter: https://twitter.com/Biijii_

Progression of my art style over nearly 2 years

Dystopian stuff involving tyrannical hierarchies never fails to catch my attention. Sounds cool, good luck!

Looks great! Cool effects and character designs.

I just finished the final trailer for the game!

It comes out in 2 days

[Background Colors]

I've been making the game prettier.

Background Colors
They've been so bad for so long.

I finally got rid of the "Toddler Clip Art" aesthetic, and made the colors blend together a bit better. This mostly meant making the colors all a bit warmer.

Now I have this problem where every time I look at it, I think the colors need to be tweaked more, but then after an hour of tweaking things look exactly the same. I feel like things are a bit too dark overall, but when I try to lighten one part it throws everything out of whack. If I try to up the overall exposure the colors don't blend as well.

Guess I'll just embrace the warm comfy aesthetic.

Chopping Particle Effects
I started by taking a screencap of a katana chopping through some bamboo, and sketching ideas over top of it in Photoshop. The only thing that looked good was this pointed star.

I messed around with the scaling, opacity, and rotation until I found an animation for the star that worked. It was okay, but kind of boring, so I added a Unity particle system and found that having little trails shoot out of the particle looked good too. Using a similar process I also made some wood particle effects for my wood obstacle.

It doesn't feel perfect, but it gets the job done. Not too overstated, but still feels more satisfying than without particle effects.

Switching to Unlit Shader
Finally I changed all my obstacle materials from the standard shader to an unlit shader. I've been meaning to do this for awhile. The standard shaders didn't mesh with the background very well.

This meant figuring out how to make an unlit shader that has the ability to fade (luckily found a solution online), and going back through all of the textures to tweak the colors. I've done one pass through all of the textures, but some of it feels slightly off, so I'll need to revisit a few.

Final Thoughts
With release inching closer, I want my game to look good as soon as possible. The earlier it looks better, the longer I can market my game effectively. In future games it'll make sense to spend a little extra time on art towards the beginning of development. This time I'll give myself a pass though - I had zero experience making any kind of art prior to this. I mean come on:

I've made changes to the Interval Bamboo obstacle (again).

I've explained the Interval Bamboo in pieces throughout this devlog, but here's how the most recent version worked:

When it reached your katana, you chopped it by hitting any of the katana buttons. It would then play one of three music intervals (a whole step, a perfect fifth, or an octave), and bounce back away from the screen. Once it reached the screen the second time, you'd have to decipher which interval was played, and hit the corresponding katana orientation (whole step = horizontal orientation, perfect fifth = diagonal orientation, etc.)

This had two problems
1) Initially hitting the obstacle by pressing any katana button felt kind of awkward. The main mechanic of the game is to hit obstacles with the correct orientation, so dropping it felt out of place to me.
2) Deciphering between the three music intervals was pretty hard, even after I practiced with it for about a month. I thought this might be a problem from the beginning, but was hoping the skill would be easier to train than it ended up being.

My solution was to separate the first and second stage of the obstacle.

For the first stage, I created a little music note model that can be in any orientation. Problem 1 solved. You hit it just like regular bamboo, but once hit it plays one of two intervals (a whole step of an octave) and bounces back to the main Interval obstacle. Having two intervals to distinguish between instead of three is a huge difference. Problem 2 solved. Now I only chop it incorrectly if I lose focus, which is the way things should be.

Also, I changed the obstacle's accent color. The pink music notes looked too similar to two of my other pink obstacles, and melted into the background trees a bit, so I had to switch it to black. Black music notes make sense anyways.

an easy idea with a great goal!

Thank you! It was supposed to be an easy idea at least - maybe how long it's taking to implement proves that wrong lol.


The music is FINALLY COMPLETE! Let's analyze what happened.

Look at all those little files. I made those.

I made 12 songs. I planned for it to take 2 weeks. It took me 4 weeks.

I guess I never stopped to consider how long it would take to make each song. On my most productive days, I was able to make about 1 and a half songs. This was when the entire day was dedicated to making music though, which was rare because of school. It really looked more like this: make 1/4th of a song one day, do 1 and 1/4th the next day, and then take a 1-2 day break to study for an exam or catch up on homework. Combine this with the two songs I worked on and ended up throwing away, and whaddya know, that's about 4 weeks for 12 songs.

To be honest, motivating myself to work on this music has been more challenging than motivating myself to work on anything else in the game. I'm still not sure entirely why though. Surely the biggest thing was constantly failing to meet the milestones I set for myself, because I really underestimated how long things would take. Also, all of my songs had to sound pretty similar to fit an overall theme (they all used the same chords), which could get a little stale. Coming up with the main motif for each song was a fun little challenge though.

The big positive was that it made me more confident with music composition. I played piano for 4 years in high school, but always found writing my own music a bit daunting. Maybe I'll finally learn jazz piano after I'm done with this game.

For now, I'm excited to get back to work on the actual game. I sort of ghosted the internet this past month, because I didn't want to share all of my music and spoil it, but I plan to be a bit more active again. So, hey whats up.

A little update on the music.

I've got 8/12 of the songs done. I had 9.5/12 done, but I canned one midway through, and got rid of another one entirely. They were both bad because they had too much of a melody, and didn't line up enough with the beat.

Part of the completed song I threw away

These tracks are supposed to be background music, and then the player dynamically creates the melody over top of it. Since it was hard to come up with 12 unique tracks that use the same chord progression, I ended up adding some more melodic stuff to certain tracks to give them more variety. I thought I could get away with it if the melodies were simple and repetitive.

It wasn't disastrous or anything, but it did clash a little with the players melody at times. I decided it would be better to have some tracks sound similar but be 100% functional, rather than tracks that sound different but are 80% functional. I was able to salvage one song that had a complex melody and heavily simplify it, but not the other 1.5 songs I mentioned before.

Also, in the songs I got rid of, their melodies didn't fully match up with the beat. They were still in 4/4 time, but they were using weird 32nd notes and slightly missing the on-beat for style. Turns out that's a bad idea for a rhythm game. The player can create melodies that don't fully line up with the beat, so if the background music also doesn't fully line up with the beat, it's not gonna sound so good.

Just a few more songs to go.

Good idea doing a small scale game first, I probably should have taken that advice a bit more to heart heh. I like games like this that have a simple idea you can build upon by adding new obstacles (in this case blocks). Fun to brainstorm the different ideas, and fun to play.

Also, more of a meta comment, but you don't have to worry about keeping the first post on this thread super updated. People who have been here awhile know to check the newest posts for the newest stuff, and I was slightly confused looking for the newest devlog vid in the first post. Not really a big deal though lol.

Good luck with everything!

Played the alpha version. Cool idea being able to pick up the enemies you defeat and grow your team. The gameplay itself is a bit too easy right now, but I'm sure your working on that. I also like the variety with those new enemy types. Good luck with everything!

I've been working on background music.

I decided that each of the 12 obstacles would get it's own music. Mostly because there was no other music schema that made sense to me. One song for all levels? Boring. A small pool of songs to randomly select from each time you play a level? Could work but kind of lame.

I already have a problem where all of my levels feel similar because they have the same environment in the background. Giving each level set their own song would help differentiate them a little, and could also be something to look forward to when you encounter a new obstacle.

I've already come up with 12 rough drafts for each song. I basically sat at a piano for a couple days punching out random rhythms to see what stuck. The stipulation was each song had to use the same FMajor7 CMajor7 chord progression. 1: because I knew it worked with the pentatonic scale the melody uses, and 2: I still wanted each song to have something in common so they felt like they were part of the same game.

So far I've made 5/12 of those rough drafts into full songs using FL Studio. Well I'm sure I'll tweak them a bit later. Here's an example of one of them:





Each song is broken up into 8 sections, ranging from 60 bpm to 200 bpm. Every 32 beats, when the level in the game speeds up, the track with the next highest bpm is played until you reach the max speed for that level.

One problem I've already come across was making the music too complex. It can't compete with the melody that the player will be creating in-game. This is mostly a good problem to have, just don't put as much effort into the songs lul. The challenge is to make the songs simpler, without making them too similar.

Anyways ya, it's taking awhile to make these songs. Even if I could manage 1 song a day (which I haven't so far), that's 12 straight days of making songs. Guess we'll see where I am a week from now

I made a video!





I talk about a few game design problems I've overcome. I'm still not content, but I do feel like my videos are steadily improving over time. This was the first video I've fully story boarded, and that worked out pretty well. Separating the step where I decide what footage I want from the step of collecting footage made both go by much faster.

I've made some updates to the UI, and added an exciting new "life" system!

For the UI, I added some spinning obstacle models to the level select screen. This shows what obstacles a level will contain.

The transparent UI really messes with my gif software

I've been planning to do this for awhile, but wasn't sure what approach to take. My original idea was to spawn a separate camera for each obstacle I wanted to have spinning, and then I could project each camera onto a RenderTexture that I then put on top of the UI. Turns out having that many cameras would have been horrible for performance, so I only use one extra camera to capture all of the spinning objects. Also, chroma keying a RenderTexture wasn't easy to figure out, so instead I make the background transparent by using camera culling.

Now the life system: you no longer get a game over for the first mistake you make.

Currently you have to mess up 3 times to lose, which seems fair. Then for every obstacle you chop correctly, you get a little life back.

This makes the game more fun. Keeps you in the action longer, and allows you to lose focus for a second (which tends to happen after awhile) and still have the chance to get back on track.

And ya, I made another schedule. I feel weird uploading it because then you'd know exactly when my college classes are, but it's similar to that last schedule I uploaded. I've been lazy about the meditation sessions lately, and I'm realizing they were actually pretty important. It's easier to get myself to meditate than to do work, and after I meditate I'm always ready to work. Without it, I've been late to start a few of my work sessions. Gotta start meditating again. I love schedules   ﹏

I've added a daily challenge mode!

You can't play the daily more than once outside of dev mode

After making 12-9, my final level, I had a function that would randomly generate a level with all of the obstacles in the game. This does most of the heavy lifting for the daily challenge logic, but there were a few changes I had to make.

To make sure everyone got the same challenge, I used Steamworks' SteamUtils.GetServerRealTime() to generate a unique daily seed, and fed those seeds into C#'s Random class.

Also, level 12-9 spawns every object in the game, but I wanted the daily challenges to spawn objects from a smaller randomly selected pool. I made it more likely for daily challenges to have a small pool (2 to 6 objects) than a larger pool, because obstacle combinations are more interesting that way. Like, if you have 15 different obstacles, adding a 16th doesn't make things that much different.

It's hard to know if it's working 100%, because I can't test every seed. There are certain obstacle combinations that would break the game, so it'll take a bit more testing before I'm really confident.


In other news, I've moved my release date from Summer to October . I've had a few new ideas (like this daily challenge mode) that I think are worth taking the time to put in. Well and this is my first game so I don't really have a gauge for how long things will take heh. I haven't had much time to work the last 2 weeks, but with classes starting on August 20th, I should be back on a steady (although reduced) game dev schedule again. I'll probably make another turbo geek schedule to make sure I finish by the end of October, so stay tuned for that

I like the art It's already got a polished feel to it.

With finals coming up and plans to visit family, my game dev cocoon mode is officially over. Thought I'd give a quick summary of how it went.

The same schedule pic from before

For game dev, the schedule was a success! I skipped one day to go hiking, but other than that I only missed 2 of the 45 remaining game dev sessions. It's the stuff other than game dev that I was lazy about, because I didn't find them as important I guess.

I worked at a pretty good pace during my morning work session, but there was about a 50% chance of me being slower during the "afternoon slump" sessions. The meditation break was supposed to help with that, and it was better than my old strategy of working for 3 hours straight in the afternoon, but the problem was still there.

In the future it might be better to take 20 minute breaks every hour instead of 30 minute breaks every hour and a half. I've realized the point of taking more breaks isn't to rest your brain more, but to break up your work more. That way you need to have a small goal for each work session, so your less likely to lollygag. I can't go full pomodoro technique though. Only being able to work for 25 minutes makes me angry, and 5 minutes barely feels like a break.

I didn't expect too much from the meditation, it was really just a distraction-less break from working. I got slightly better at holding my attention on my breath, but that's about it. Could be because I already knew about the concept of mindfulness, so I'd already gotten all the nooby gains. Good for calming your brain and bones tho.

The bigest takeaway is that this schedule improved my relationship with work. I think I like it. I thought this schedule might make me go crazy, but I'm more content then I've been in ages (I might go crazy without it tbh).

There is one thing that makes this schedule different from previous ones I've tried: I got to choose all of it. I didn't have to dedicate certain hours to school or a job - which are sprinkled with personal goals, but ultimately someone else is telling you what to do. Instead I knew that at any given time, I was doing exactly what I determined was most important for reaching my goals. With spending more time during the day in flow, and seeing real progress towards my goals, I felt a lot less anxious than usual (and I'm a pretty anxious guy). This gives me hope that I could work well with freelance or entrepreneurial stuff, which might be more fun than a software development job idk.

Anyways ya. I work well with a schedule, maybe u do 2.

I've added a combo system and brought my game to a semi-stable state.

The combo system keeps track of how many obstacles you hit with perfect timing in a row. You can still chop an obstacle a bit before or after the beat, but it'll give you a tenth of the points. I'll probably change it a bit so combos instead give you extra points on top of what you get for chopping an obstacle, but the desired effect is already there. Now on top of hitting obstacles correctly, there's a meta game you can play of trying to increase your combo.

I thought it might be hard to get the "perfect timing" window to be precise, but the easy approach I gambled on seems to have worked out. I used Unity's AudioSettings.dspTime to compare the time an obstacle was chopped with the time of the current music track being played.

I'm not sure if it's a psychological thing, input lag, or both, but I kept finding myself hitting the buttons slightly before the actual beat of the music. To combat this, I just made the perfect timing window start 0.055 seconds before the beat, and end 0.035 seconds afterwards. Maybe that sounds janky, but here's how I rationalized it: when you're playing piano, you begin pressing a note down slightly before it should be played because it takes time for the note to be pushed down . All that matters is that it feels right, and for that I'm about 90% of the way there.

To get my game in the semi-stable state I just had to fix a bunch of small bugs that mostly aren't worth mentioning. I did implement those leaderboard buttons I was talking about in the last post, but I'm the only person on the leaderboards, so idk if they work heh. Hopefully I can get some people to test the game now.

I got steam leaderboards working!

First I made a spot in the level select UI for a leaderboard.

↓

I was originally gonna put the highscore, max speed, and leaderboard for a level in that UI panel on the right, but it created a lot of unnecessary white space. Also, The leaderboards are the most important thing on the screen, so it made sense for it to take up the whole panel on it's own. That left panel is eventually going to have spinning 3D models of the obstacles that show up in the selected level.

The leaderboards store your score and the amount of obstacles it took to get that score. I ripped that idea straight from Spelunky - seeing someone that died at an earlier level than you but got a higher score shows how much you could improve by playing better. This'll make sense once I program obstacles to give you more points for chopping them closer to the beat.

Ignore the highscore in the top right. I reset my highscores every time I launch.

Also, I wrote the code for the leaderboard buttons, but didn't hook them up yet. I want to finalize the look of the buttons first. This is another one of those situations where I'm spending more time thinking and writing then it would take to just program it 

I've been working to get steam cloud saves to work, and fixing a bunch of tiny problems along the way.

I was originally using Unity's PlayerPrefs to score highscore data, but that writes to registry files I can't access. I needed to save game data to a regular file so it could be uploaded and downloaded with the steam cloud.

I ended up using this FileBasedPlayerPrefs asset I found. It uses the same getter and setter methods as PlayerPrefs, so it was very easy to implement. Ex:

Code:

// Original
PlayerPrefs.SetInt("highscoreLevel1", score)

// New
FileBasedPrefs.SetInt("highscoreLevel1", score)

The FileBasedPrefs outputs a single encrypted .txt file with all of the save data on it. Most games seem to use .sav files for save data, but I can't figure out what makes that any better than .txt. Maybe it's the implication that .txt files are meant to be editable by users? It's probably taking me longer to write these sentences then it would to change the output file's extension from .txt to .sav.

While setting that up, I started messing with Steamworks.NET. It provides a C# wrapper for the steamworks SDK (which is written in C++). I also started tracking how many obstacles a player has chopped (separated by obstacle type), and refactored some code.

Here's the little implementation roadmap I'm currently working on:

Save to file instead of registry -> Steam cloud implementation, leaderboards -> level select menu, stats menu

I'm gonna start planning these roadmaps a bit more now. I've been burned too many times from implementing something specific that doesn't work when I make a more general change . As my tasks become smaller this becomes less important, but better late than never!

[Level 11-1:]

I've created what should be the final obstacle, and logic for the final 2 levels in the game.

New obstacle: The "Decimate" obstacle

It's a tree trunk

You hit it at least once with any button, and then mash all of your buttons to chop it as many times as you can. Each extra chop gives you extra points.

This obstacle was made to replace the "Long Bamboo" obstacle I created about a year ago, which basically had the same logic. The problem was long bamboo took up both sides of the screen, which was needlessly limiting, and the chopping wasn't very satisfying. It's annoying because it's the obstacle I put the most work into modelling and animating, but I couldn't justify keeping it when I knew how to improve it.

pathetic

The decimate obstacle still has a temporary texture, and I need to make custom katana animations to go along with it, but I did make those sweet animations of the wood slowly getting chopped up over time.

Using the new decimate obstacle, I created level 11-1 and 12-9. None of the other levels had so many obstacles, so I had to make some new functions to make them spawn randomly but in ways that are challenging. So I finally made it quick and easy to code up logic for a level, now that all the levels are done

Level 11-1: All obstacles except control flip and dynamite. Decimate shows up for every 16th beat. Gives you a chance to experience all of the obstacles without control flip and dynamite, which tend to be the most challenging.

Level 12-9: All obstacles.

So that's the first draft of all the obstacles and levels complete! I'll of course make some changes as I test the game more, and every obstacle is gonna need at least some texture work, among other touch ups. I'm gonna neglect that for a bit though, I'm excited to start workin on some new stuff.