Level design is too hard

Alright you wee lads, gather ’round the fireplace and I’ll tell you a story from days long past: when I started in games, I started as a level designer. End of story.

I don’t really like level design, because it’s so HARD. It’s like an actual job where you need skill to do actual things. You need creativity and sense of fun & pacing and you need patience to wait for game features to catch up with your vision for a level, before you can know if your idea works. Level design makes my brain melt. I think the syndrome is called toolazyforcreativitynitis.

Operation Balalaika has 50 levels, since it was a nice round number and it seemed like the absolute maximum I could might get done. Considering the years that have whizzed past, that number was too damn high, but in general, most numbers are.

A couple of weeks ago I noticed that if the game needs to be released, the levels need to be done. In light of this revelation, I chunked out the list of un-designed levels into an action plan: “design 2 levels per day until they are all done or you are dead“. I guess that death part can be taken as an option or a threat, but nevertheless, it worked, because all 50 levels now have a semi-coherent design!

I took screenshots of the levels in the game, then filled the gaps with new hand-drawn designs.

A recommendation for any budding level designers out there: if you have the self-knowledge to know you like to go to the dentist, and that you enjoy getting repeatedly punched in the face, it’s possible you have the mentality needed to become a level designer! Just kidding. You don’t actually need to like to go to the dentist.

Anatomy of a level

Thus far I’ve been just winging it with level design in Operation Balalaika and designo-implemented levels simultaneo-concurrently, but today I decided to add another layer of winging it-ness, and started using pen and paper *) to do quick design sketches. I hope this added bit of process will make development faster, as all process does. That’s why Nokia is still the greatest mobile phone company in the world, after all.

Feast your eyes on this fresh, exclusive making of material that just squeezed out of the far end of my new Process Pipeline of Agileness ™:

Michael Bay would be pleased.


That’s it for this post, thank you everyone, it was great.

*) Actually iPad and Apple Pencil, but pen and paper sounds more auteur-y, and I need to keep up appearances.

Migration complete

Hi there, you might remember me from this blog, although because the blog doesn’t have any readers (I just checked from WordPress stats, so it’s a true fact), you might not.

Let’s pretend you do, so I can tell you one thing:

Operation Balalaika has been officially migrated from ye olde Unity 5.3.5 to brand-spanking-almost-from-this-year Unity 2017.4!

The completion level of Op-Bala took maybe a tiny step backward as some funky new bugs were introduced, like the pixely art being a bit squished, and some of the collisions preventing progress – because Unity has changed some sprite and collider stuff – but this nevertheless feels like a classic, presidential #WinningSoMuchYoullGetTiredOfWinning moment!

As a reward for myself, I drew myself.

Migrating old crap into new crap is more of an art than science, especially when there’s many new versions to the platform, and on their own separate tracks add-ons have updated too. (And, if like me, you don’t really know what you’re doing.) Naturally, not all the platforms versions and add-ons sync up, and there’s no comprehensive documentation on the add-ons, so it’s magical trial and error kinda deal to get things working.

Took me many days, and eventually I decided to just randomly pick some major versions that were not too far off from each other and prayed to the almight migration gods that I would get lucky. As a sacrifice to those bastards, I did a clean install of Playmaker, which worked out in the end.

What’s the one thing you could learn from this? Backups! Backup all the things all the time! That way you have the peace of mind to just randomly push buttons and install and remove things as you go. The other thing is to make sure you’re born very lucky.

Okay, the next step is to choose one: 1) develop the game so it would get released someday, OR 2) continue tinkering on irrelevant bugs and issues. Hmm.

Back to the future!

…whoa! That break just blew past! I gotta say, I’d like to take another691 days, but mayhap it’s time to get back to Operation Balalaika. My fanbase seems to be getting restless. Or maybe I just had too much coffee?

I see that Unity has updated 45 times since version 5.3.5, so the first thing is to go through the update process 45 times. While making sure my plugins and addons don’t break while updating. Y u gotta update so often, Unity? One update per year would be plenty.

And, they took away free cloud building?! I’m not sure I want to live in the future anymore! I’m glad we at least have Teslas in space now.

Pause time

Ooh, I feel I’ve been working so, so, SO hard, I think I I need to take a break. I think maybe… 691 days would suffice. See you again in 2018!

What’s this, a milestone for ants?

Operation Balalaika is a weird little game, because it contains multiple small game mechanics, and I’m happy to announce that they’re now all… abandoned!

(‘Cuz you can’t really say “finished” about art. At least three separate guys have came up with this quote while someone was listening ready to slap a meme into Internets and print some motivational posters: Leo, Mr. Forster and Paul Valery.)

Anyways, the fact that the game mechanics are DONE, BIATCH made me smile for like more than fifty seconds, since the completion of the mechanics was one of the mini milestones I’ve set to myself. Keeping the inspiration alive with any (legal – or even illegal if you don’t get caught) means possible is imperative for a solo indie project. Most projects are slowly abandoned because if inspiration gauge hits zero, it never recovers, and with all due respect to The Abandonment Trio presented above, I’d rather finish than abandon things.

I’ve found these mini milestones to be a small, but handy tool in getting things done, as long as they’re concretely written down somewhere where I can tick a box after the milestone is completed. (I actually do this in two separate places: my masterplan Excel sheet and in my general To-Do app. Two boxes to tick for one milestone… such happiness … such meaning to life…)

The next milestones now are finishing the small batches of levels for each mechanic. Each batch is a milestone, and each batch is a box tick. I only try to look at the nearest mini milestones to try to avoid paralysis by analysis of all the remaining work. After taking some Valium, I nevertheless have listed also all the remaining milestones for the whole project too, I just don’t look a it. In fact, the overall plan is like a well-repressed childhood trauma: I know it happened, and it’s still somewhere there, I just can’t remember the details anymore and it only surfaces as two-week-long drinking binges and bouts of crying and wailing when someone asks why the game is not released yet.

Hmm, all this makes me wonder if you can ever finish a blog post either, or can you just

Old Games and Windows

Here, as promised by the topic:


I think you can see the monitor is off, i.e. I’m not developing Operation Balalaika, just sitting on my butt playing Day of the Tentacle, while a Hobot robot from Japan is cleaning my windows.

Never mind Op. Balalaika, since DOTT is still, after 23 years, delicious! Compared to the remastered (or should I say who-the-hell-came-up-with-that-fugly-art-style’stered) Monkey Island games, DOTT Remastered is well done and actually deserves the tag “remastered”.

The new graphics work so well that I basically constantly accidentally think the game always looked like that. The story & dialogue is still fun, the characters great, the music magical and it plays well even on a phone. Only the inventory UI sucks, but isn’t a showstopper.

I dunno who owns the IP, but a Maniac Mansion 3: Day of the Tentacle 2 would be in order right about now. It doesn’t need anything fancy, just good writing and funny animations.

PS. The window cleaning robot also works well, but if you plan to buy only one thing this year, buy DOTT, and get Hobot next year.

Orcs & Humans

Strange to think that just 21 years ago I was playing this:

… And now I’m watching this:

The text on the screen seems to be getting just harder to read. Well, I guess that’s getting old for you.

PS. The movie was better than expected. I went mainly because I didn’t want Duncan Jones to feel bad, but I actually ended up enjoying it more than Angry Birds Movie.

PPS. Now where’s my Mass Effect movie?

Debug Menu V2: Revenge of the Debug Menu

Because most of you wrote considered to write in demanding more information on the sequel of the debug menu, here’s an update:

debugmenu2New features:

  1. Shows which level you are currently in (a feature most requested by the future beta testers) painting it black
  2. If you try to load a level that doesn’t exists, it will be marked with a.. slightly deeper red

Now, feature 2. literally took me hours and hours to create, probably eliminating any time-savings the debug menu could have brought.

I wanted the menu to actually pre-emptively mark levels that can’t be loaded, which would make more sense. But, turns out this short of trickery just is not possible in Unity, because you cannot read the scene list without opening all the scenes. To find that bit of disappointing information took hours, mainly because I just could not believe it.

One could also maybe write a pre-processing script to read the file names and create a separate index of the scenes and blahhety blah, but the guiding principles of Planet Jone prohibit such actions. If you’re unsure what the principles were again, here’s a visual representation of it:

I like that picture, it’s a very good life lesson for anyone, and after the saddening hours lost it also gave me the strength to just give up and accept the fact that an underwhelming slightly dark red button appears only when you press it, because the level fails to load, and then the system knows that hey, it doesn’t exist.

The silver lining of the struggle was that I also modified the level changing routine to not crash if the next level is not found, but to try again with a “level number+1” until a level is found and can be loaded. This allows me to plop in new levels randomly at any point and they will just work.

Because the blog post seems to be ending, I feel the need to somehow inject a point too. Here it is: when making a game by yourself, any automatic system that eliminates any routine work is always (okay, let’s say 99.3% always) , since if you need to spend even a couple of minutes on a task every time you add a level, you have a couple of minutes of less time and energy for the creative part. And if you average a total of one hour per day of development, those minute are pretty valuable. (Note that I only promised to inject a point, not an astonishing revelation.)

In any case, the debug menu is now done, and I shall not touch it until I need to start on the very secret Super American Mode that adds to the needs of the menu.


It’s finally complete! Done! Yes!

Oh, not Operation Balalaika, no no. I mean my Pelit collection. I just received my 1990 edition of the Pelit book, and I now have a complete collection of all the Pelit publications from books to magazines from 1987 to 2016!

I basically would have always had the complete collection, if it wasn’t for an unknown scoundrel who loaned this 1990 book from me in the, well, 90s, never to retun it. Mr. Thief, if you’re reading this: you caused me decades worth of unfulfillment, which only now has subsided. I hope it was worth it. (Btw, you should still return it. This copy is not in a great shape.)

Here’s a picture:


The first article I’m gonna read is “Can you play on a Mac?” since I’m pretty darn sure you can’t.

PS. This is the sort of post that comes from the Unity update jitters; I’m currently waiting for Unity to update from 5.2.3 to 5.3.4, to maybe fix a thing that prevents me to getting the info if a scene exists in the build, which I need to make the Debug Menu V2: Revenge of the Debug Menu.