My week in game dev #19

I did a podcast interview thing the other day and was asked for my advise on getting into game dev. This was a good question for me as I think about this stuff quite a lot. I think I managed to answer as I would wish to, but as I kept thinking about it afterwards and so I’ll reiterate a little on here.

Now then, back in 1986, aged 17, when I left school, (was ejected from school is a more accurate description) there was no game dev course at university as there is now. The powers that be knew nothing of the fast growing games industry and wouldn’t do for a long time to come. And so, everyone going into the industry was self-taught and, most likely, a bit of a dropout. In my case I knew what I wanted to do and had been preparing for it for at least a couple of years.

Games in those days were 8 bit and the good thing about that was that there were limitations imposed by the hardware. It was hard to get screwed by ‘feature-creep’ when you only had 48k to start with. Limitations are good and require creativity to overcome them and that is what drove rapid innovation in graphic tech and gameplay.

Obviously it’s different now. On a PC you have infinite CPU, memory and storage and so developers can fall into the trap of trying to create something too big that ends up crushing their spirit before the project is finished. And this happens at all ends of the industry from the lone dev to the (so-called) AAA mega studios with their schedules and spreadsheets.

I’m not sure why people would want to to work for one of these studios, but plenty do. You can go to Uni, run up huge tuition fees, then be worked to the bone for three years before you are burnt out and ejected from said studio because some Metacritic score was slightly lower than they wanted. Sounds great!

So, before doing all of that, here’s what I think you should do…

First, go here and grab a copy of Pico8 which is a really cool virtual game console with a hard limit of 32k memory. It uses a modified version of the Lua coding language so it’s easy to learn and get started. You can do really cool things with it. Despite the spec, the only true limitation is your creativity and imagination. What you now do is write a game. It can be something random from your head, or a classic with a twist.

You can write down a design on paper if you wish. And even a schedule of how long you think it will take (it will take longer). The important thing is to start the game and then finish the game. Like, really finish it. Now you show it to your friends. If you’re afraid to do that then it’s because deep down you know something is wrong with it. But do it anyway. Maybe you go back and fix it up a bit based on feedback. Now it really is finished.

But you’re not.

Next you do some social media promotion for the game and you publish it. I’d suggest just putting it on a dedicated itch.io page and maybe wrap it up and put it on Steam too. And you’re not putting it out for free, it needs to be for real money even if it’s $1 because it must be worth at least a dollar, right? Ideally you’ve promoted it all through development by tweeting, blogging and TikToking the ups and downs of your development journey.

Now you’ve designed a game, written it, marketed it and published it. That’s the entire cycle and there’s no mystery to it anymore. You don’t need to be in awe of anybody now. There is no mystery.

Having done all that, feel free to head off to Uni, or, you know, save some time and money and just start making games.

Further steps? Learn C (you’ll love pointers!) and write a better game using some framework - like raylib perhaps. Maybe a bit of Python for tool writing purposes. After all that take a look at C++ / C# I suppose.

Do all of that and you’re ready to go. Or, perhaps you know now that it’s not for you and this was at least a useful cost free exercise. Either way, thank me later :)

What else is going on?

UrbX progresses nicely. I spent a lot of time mopping up the fallout from implementing a Z-sort on the Next engine sprites. A days predicted work turned into a week! I’m still not sure it’s 100% working. Or rather, it seems fixed but the last thing I did there doesn’t make any sense and shouldn’t have fixed anything. That stuff is scary!

Beyond that we implemented new player run and fire anims that depict the weapon the player carries. These are very cool, but really hit the Next engine hard as it has to swap out all the old player sprites for new ones as it goes. It works, but I feel further optimisations are coming for the sprite cacher stuff.

I also added support for potions. These are basically one-use weapons, but that generated a bit of work as they have to remove themselves from the inventory, which gets smaller, and so on. Nothing big, or difficult, and very good for designing gameplay strategy.

What else?

Had a lot of fun at the Revolution staff wrap party for BS-R. I went as a +1 :)

Also got myself one of the new MiSTer Pi’s. A bit of a faff to set up but quite an impressive device that will configure itself to almost any 8 or 16 bit device you can think of. I tried it with Amstrad CPC and GBA first off and it worked a treat. Recommended.

Have a good Christmas. See you in 2025 - the year of UrbX :)