Why You Shouldn't Make A Game Engine

Many low level game programmers love to work on their game engines. It's a comforting way to avoid facing the hard questions, and ultimately your fear of failing in public. In this podcast, I propose an alternate route for folks who love to make things from scratch. I show you how to focus your efforts and actually ship something. Instead of making an engine, you need to make a game.
Wishlist Mooselutions:
store.steampowered.com/app/22...
My First Software Product, Skate Dice:
apps.apple.com/us/app/skate-d...
Pro Skier Jossi Wells Playing Ski Dice:
• 'Ski Dice' with Jossi ...
Tim Ferriss and Jerry Seinfeld Discuss Creativity, Routines
• Jerry Seinfeld - A Com...
Pico 8 Slipways:
krajzeg.itch.io/slipways
Pico 8 Celeste:
mattmakesgames.itch.io/celest...
How to Market a Game:
howtomarketagame.com/

[@Aklloll]

I’m 16 years old and I’ve barely spent 6 or so hours learning programming from a free course on python and I can already tell that this video and the one on programming the hard way will stick with me for the rest of my life. That line at the end hit very hard for me as recently I’ve tried learning yo-yoing and art and more or less gave up both before convincing myself I need to focus on or “specialize” in programming. Of course specialization is for insects and I’m more motivated than ever to pick back up art/yo-yoing alongside programming. You could read this as some kid tricking himself into believing he can do the impossible but until I try I’ll never know if that’s the case or not. I’m excited to finally be forcing myself to learn a few skills after years of doing almost nothing besides school and gaming. Overall, your videos are genuinely some of the bests I’ve seen on this platform and feel like a real human talking rather than someone spitting out lines I’ve heard a thousand times before, thank you. (Also I might update/edit this comment as time goes on to track my progress in learning these skills)

[@tedbendixson]

The only thing I would suggest is to make sure your art serves your game design. I've worked with game artists who do a great deal of decorative work before they create the game assets, or art designed to communicate gameplay. Usually, when I make art, it's to solve a specific gameplay problem. For example, the moose need to look angry. How do I draw them in such a way as to communicate their anger to the player? How do I make this block look both pushable and look like the kind of thing that won't break when the moose crashes into it? Combining art + game design + programming skills all in a single human being is very powerful. You can move faster than entire teams.

[@arminunruh452]

very good! i also took a programming course at your age. I didn’t understand anything, fast forward when i was 22 i tought myself c++. If you‘re interested and disciplined and have enough time you can learn coding at any age!

[@IBelieveInCode]

Good luck aklol7133.

[@scottishsnake9382]

Emphasis on the document and sharing your progress. Pick a platform and keep going from there

[@jeffrey5602]

You kid will go places. Learning programming is really tough, i failed a few times in the beginning. Just remember that it means you are actually challenging yourself on your learning journey. I wish i started programming that early, I unfortunately kept playing video games throughout my bachelors and only found my way to computer science at 26 :)

[@kakiii2i2i2i2i4]

20:30 you basically predicted the runtime fee😂

[@tedbendixson]

History repeats. It happened to Flash.

[@agoatheadbaby3512]

​@@tedbendixson I miss Flash so much....

[@anonimowelwiatko4455]

I understand. I think that making very generic, very broad and ambitious game engine when you don't even have product in mind is in fact stupid. I do believe that building one for specific game or type of games you want to make is in fact great idea as being familiar with tool you are going to be using for long time and having freedom to change this tool to serve your purpose and don't be bloated with hard to navigate interface, features you are not going to use etc. etc. is great thing to do.

[@namgnoi]

really like the reframe to "leftovers". has a nice effect of feeling unstuck. it releases complexity.

[@simongardner3766]

There are a lot of advantages in writing your own game engine. Firstly you don't have to spend a lot of time learning a third party interface, then you can fix low level bugs immediately not when another programming team finally gets round to it, and finally how many times have you wanted to do something only to find the third party didn't think it important to include it. Then of course there are the skills you gain from doing some of the low level stuff. It does take more time but you can be modular about it and just write the game engine modules as needed by your game. That way for the next game, you have them available for use without any major extra work.

[@tedbendixson]

I'm half trolling with the video's title. If you're coming from my other video, you already know I make my own games from scratch, which means making the engine. I like to think I approach it differently, putting the gameplay at the front of the whole thing.

[@hijarian]

16:25 is the core of EVERYTHING. "What problems do you need to solve to make this work" - this is the best question EVER you can ask yourself when learning ANYTHING.

[@tedbendixson]

Entire fields exist because there was a problem and many people offered their solutions to it.

[@FlippinFunFlips]

I can feel the level of care and effort that you pour into the things you pursue, including this video. Setting small goals and celebrating success is something I'm valuing more and more as I learn to see y projects and hobbies through. Thanks for your wisdom and energy.

[@michaelgraflmusic]

Your holistic talks are great. Especially compared to most of the fluff that KZbin is crowded with.

[@alaanvv]

You inspired me and now I'm learning Assembly! I'm very thankful, and think you got what you expected from that video. You really said good and unexpected things

[@olabodealamu5297]

I found you through the primeagen KZbin channel. I find your content to be truly inspiring.

[@JoshChristiane]

Awesome video, Theo. That was amazing to see you do a double backflip. As somebody who falls constantly while snowboarding I know exactly how hard that is.

[@soninhodev7851]

you are right, when i desing a game, i dream of success, but expect failure, i am going to adopt your strategy of always having a small goal every day, i remember my most productive days were the i did this (without even knowing), but now its going to be intentional.

[@alkuhn3939]

For years now, I've been toying with this notion of developing my dream game, an oldschool RPG for DOS. Lately, instead of trying to dive into the programming, I've committed to basically writing the whole game out in the form of a massive book of outlines and diagrams. That way I can know everything that I'm going to need before I write a single line of code.

[@tedbendixson]

That can help greatly, but don't let it stop you from just sitting down and prototyping the game. You're going to learn so much about your game's problems and hooks through the prototyping process. It'll give you a different understanding of what you are working with.

[@alkuhn3939]

@@tedbendixson I have prototyped some things like the rough graphical rendering system at least, but you're right, I need to do a lot more of it before the whole thing becomes practical

[@Headtalk]

Your advice echoes some really good writing advice I heard recently: don’t start with universals when trying to get at universals. For example, everyone understands the concept of a mother’s love, but if you want to write a strong piece about it you should stay away from generic or universal experiences. Don’t say, “I was her entire world.” Say, “She worked 12 hours and then came home and made me my favorite meal, even though it was something she didn’t particularly like to eat, just in case I’d had a bad day at school again.” Building a strong foundation for any creative work starts at the specifics and builds up and out into generalizations.

[@tedbendixson]

Interesting how this applies outside of programming. There's something more natural, more human about this approach. You're more like a carpenter than an academic

[@Headtalk]

@@tedbendixson the reasoning I heard behind this is that people by nature want to work for their meal. It’s more compelling to be given reasons to feel the same way as a character than to be told how that character feels directly because we want to feel like we’re solving a problem or putting together the puzzle pieces ourselves. To tie this back into your video, it’s more compelling as a creator to solve specific problems that, beyond just making your game possible on a practical level, instruct as well as illustrate the specific game you’re trying to make. And just like writing, if you are compelled by what you create, chances are you’ll find an audience who feels the same way. Thanks for the video by the way. I often struggle with defining why I feel the need to create and with motivating myself. What you said and the confidence you project really help.

[@tedbendixson]

It's just like creating puzzles. You want to give people the maximum amount of freedom inside of a puzzle so they can work for their meal. If they can try out a variety of things that don't work, it's more satisfying to find the principle or set of moves that does work. Constrain the player to handful of choices and they quickly discover what works without having to work for it, breeze past the puzzle, and the whole exercise becomes rather pointless. Games and fiction have more in common than I would have expected.

[@russmack11]

Thank you for this! I watched your other video yesterday. I am not a programmer by trade, but also no stranger to programming and games. Video games are my passion and I have wanted to make my own game forever! Your words are so encouraging for someone who has always thought of programming as being so daunting. I have always been afraid to just START something. Its nice to hear about your successes and failures. Makes me feel like its ok to just go for it and not worry about whether or not I succeed first try. edit: also, love the Heinlein quote. Starship Troopers is a cultural treasure!

[@tedbendixson]

Go check out Chris Zukowski's blog. He has a great post that discusses why your first game probably won't be a hit and why that's okay. You're rolling a big snowball. It's gonna take a while. The key is to survive long enough to see the snowball at its biggest.

[@PeidosFTW]

9:49 im having these thoughts right now, but more because of a lack of interest in what I'm currently doing and bad pay. Feels good to know im not alone in that

[@tedbendixson]

Most of the work in the software industry isn't interesting, nor should it be interesting. What bothers me is the particularly American "toxically positive" charade where we're supposed to pretend it is. I once worked a job writing the code that makes credit card ads appear in a mobile app. I don't use credit cards, and the programming itself wasn't interesting (just downloading some JSON from an API and gluing stuff together with frameworks), but we're meant to put on this show like it is. Like, can we just say this is digital janitorial work, get the job done, and go home so we can do the interesting stuff? Do we really need to dress it up with fancy Model View View Models and call ourselves intellectuals?

[@PeidosFTW]

@@tedbendixson I'm starting to think that's true lol. I only have (almost) a year of experience so that's why I have those feelings. But I never thought about that expectation you mentioned, although I'm not American, it's definitely something I think I need to have, and not having it is slightly worrying. Thanks for the comment, opened a new way of thinking about this stuff

[@tedbendixson]

Just be aware that most employers won't help you advance your career. That's been my experience for nearly a decade working software jobs. You have to take the initiative and build your own things and leave when you're bored. Longterm boredom is usually a sign that you should be challenging yourself more in a job that will probably pay more, so start seeking those out. The longest I've ever stayed at a job was two years. I've been a freelance contractor for most of my employment in software, and it's better this way because I get plenty of time to pursue my own projects between gigs. People who leave their jobs every 2-3 years make more money on average.

[@PeidosFTW]

@@tedbendixson thanks for the advice!

[@captainfordo1]

How different will the steam release be from the itch version?

[@tedbendixson]

I'm basically "finishing" the game, which means putting in more animations to "juice" up the game, adding a world map so you can play the game nonlinearly, making it so certain puzzles don't block your progress. I've also got Steam achievements. I might do another pass on gameplay and include a few more puzzles that I excluded earlier due to not knowing quite where they fit in with the rest of the game.

[@captainfordo1]

@@tedbendixsonsounds exciting!

[@khoavo5758]

I love everything that was said in this video.

[@Roxaro]

wise advices thanks

[@fu_games]

Great talk

[@ZarHakkar]

7:40 Which is funny, because as of last night, after years of trying to figure out enough conceptually about my game to convince myself I had something workable, I finally did. Now I'm ready to start making the engine for it, which is why I'm looking into videos like these. I always planned to make my own engine for this game (that's just the way I am), but I did take so long being stumped concept-wise (and stricken with brain fog and zero motivation) that I kinda have to relearn programming. Wish me luck! As for the question of "What if nobody cares about this game?" Well, I know one person who cares very much: it's me. As long as I end up making something I'm happy with, I don't really care what reception it recieves. 22:22 I hear this a lot, and I can see how it's useful advice, but I don't agree that it's applicable in 100% of the case. Game feel isn't just about programming, it's also about presentation. We see it all the time in the modern industry where satisfying animations and sound effects consistently pull past their weight to significantly enhance what would be otherwise dull and uninteresting mechanically.

[@tedbendixson]

Good luck! The reason I offer advice at 7:40 and 22:22 is because I see so many "game engine" developer types who never make a simple prototype of their game, which focuses on the gameplay. Instead, I see lots and lots of technical demos that have no gameplay in them or some small amount of gameplay that isn't all that deep. It's good to get a small demo out early to see if your game can hold anyone's attention at all, even with low-fidelity artwork. I know this because my first game prototype (Cove Kid) couldn't hold peoples' attention for more than about seven minutes. Mooselutions, on average, holds attention longer (53 minutes median playtime with some outliers at over ten hours). You should know these numbers, and you should care about them. Median playtime is one of the biggest predictors of a game's success.

[@prawtism]

Wow, double backflip, respect

[@PharmacyBrain]

Do you have any advice for someone that can't seem to decide on a specific game to make? Like many, I've been stuck tinkering with code more than making an actual game. I start a lot of projects (enough to get the core "engine" and mechanics working) and don't finish any of them once it comes to creating levels and filling out the content. When I think about what I want to make, my "simple" ideas seem to always somehow slowly morph into a full blown exploration RPG, dungeon crawler, survival, or other highly complex game. Making one of those is, of course, extremely ambitious and probably unrealistic. I want to make a simpler game (I think?), but I have a hard time connecting with simple games. I don't typically play simple games, not in recent time anyway. It's funny, because this is exactly why I quit game programming as a teenager and didn't pursue it as a career. I wanted to make shooters like Quake. Once I learned enough about doing that I felt like it was an insurmountable task as a solo developer; I'd never be able to do it on my own, so I lost interest and motivation. I also have a great deal of conflict based on a game's potential for income. I have a nice work-from-home IT job, so I'm going to be fine no matter what. However, I want to either transition to indie game development, or at least make it a retirement/early retirement activity that can produce meaningful income. As such, I am always weighing my ideas against what will actually sell well. This, unfortunately, further pushes me down the path of making more highly complex RPG games. Finally, as someone who also just hit 40 with two kids I'm daydreaming about my legacy. I'd love to create something with a real story and adventure; something artistic that gives insight into me and who I am that I can share with my kids long after I'm gone. All of these things together are causing me to think about saying "screw it" and just making the RPG exploration game even if it's difficult or takes a long time.

[@tedbendixson]

I'm also 40. I don't have kids, but I have two adorable nephews and a niece. I also have two crazy German Shepherd dogs. I think you are right to go for something that is likely to sell. I'm not one of those types who says just make what you want and to hell with others expectations. I've seen people who have retired and are making games full-time who took that route and become unhappy because nobody seems interested in their projects. They make it sound like they don't care, but actually they really do care so I take it that's kind of a bullshit copout. Make what people want. This will sound simple, but study the market, pick a genre that is popular, and then execute something well inside of that genre. Roguelike Deckbuilders, City Builders, Automation games, Visual Novels, Horror games. At the moment, I'm making a roguelike deckbuilder because I enjoy playing games in that genre, and I have a game concept / hook that fits the genre. I don't know what to tell you about making simple games vs. complex ones. The truth is, Steam players prefer deep complicated games with a lot of systems that interact. If you don't deliver on that, it won't sell, and you probably won't enjoy your retirement that much. You can start with a smaller more stripped down version of the concept, and if people like the core thing, expand on it until it's competitive on Steam. Slipways is a good example of that strategy.

[@iki_kaf]

Great video! Sometimes it's too hard to focus on listening to you, because of the gameplay at the background. It's too interesting to watch!

[@tedbendixson]

This has been my secret plan all along! Pretend to be some kind of game development expert but actually use it as a front to sell people video games! Mua hahaha!!!

[@anon_y_mousse]

Aha, and now I know why the version on Itch is restricted, because you plan to release it on Steam. You could change the payment model on Itch instead, because I think Steam takes too big of a cut at 30%. I have a few game related projects I'm currently working on, but I won't be selling through Steam because of how much they take, and if I could encourage others to use alternatives I will. Even though I think Epic also takes too much, I'd encourage going with them over Steam. However, as I know that you can set the percentage that Itch takes, I'd recommend them over everything else. Just my attempt at a little nudge.

[@tedbendixson]

But Steam has such a big audience that it makes it worth the cut they take. For what it's worth, I will put the game back on itch.io once it is completed. I just didn't want to keep the older version up because it's so unpolished compared to what I plan to release as a final product.

[@anon_y_mousse]

@@tedbendixson Unpolished is perfectly fine. I played the PICO-8 version of Celeste, and it's still a good game. What you've shown of your game looked pretty polished. For what it's worth, I'd rather support developers more by buying their games on Itch than ever using Steam.

[@johnterpack3940]

Is the time spent learning the ins-n-outs of an existing engine, and dealing with its shortcomings or baggage, really any less detrimental than the time spent working on your own low-level stuff?

[@tedbendixson]

Exactly. You're gonna wrestle with the computer anyway. Might as well wrestle in a way that gains you longterm skills and control over your business. Get comfy.

[@johnterpack3940]

@@tedbendixson I probably need to learn to program first. Keep getting bored and frustrated when I set out to learn a language. Any thoughts on simple projects that teach useful skills? Seems everything is "Hello World"... which is practically useless and massively uninspiring.

[@tedbendixson]

Yes. If you are interested in making games, just try to follow along with the first 50 days of Handmade Hero. See where that gets you. From there, try making a really simple game like Pong or Asteroids. If you can do that from scratch, you've probably got the equivalent of a C.S. degree (arguably more than that since they don't give you much time to actually practice at university). Handmade Hero is such a great resource because Casey builds the whole thing from scratch without looking anything up. It gets you in the mindset of someone who can build from first principles.

[@johnterpack3940]

@@tedbendixson Thanks! I'll check that out.

[@thewhitefalcon8539]

​​@@johnterpack3940Very basic programming concepts are almost the same in any platform or language or system. Any type of programming anywhere will teach you basic programming, whether it's something like RPG Maker, Garry's Mod, Minecraft, Roblox, or something from scratch. I remember as a little kid there were two kinds of games I tried a lot. One was 2D tile based Windows games that were the style at the time - e.g. similar to Drainstorm. The other was MUDs. I never finished anything as a kid, but I had an idea of how to get from zero to something that was kind of like one of those types of games. But you can make anything you want to make. Just actually do it. Don't sit around thinking about what to do. You can always abandon it if you hate it.

[@alaanvv]

mf just got the snowboard pill

[@KANJICODER]

Never make a game engine if you haven't already used a game engine.

[@KANJICODER]

If I am up at 12:00 PM tomorrow I will definitely stop by and watch some of this.

[@tedbendixson]

Wait isn't that noon? I'm on East Coast time. I want to be sure it premieres at noon my time.

[@KANJICODER]

@@tedbendixson I always get confused with noon-vs-midnight. I am in Michigan, so that would also be noon for me. 11:45 AM is 15 minutes before 12PM . ( I just double checked ) Anyways, the original comment was to say "my sleep schedule is weird" not "your premeir time is late".

[@thewhitefalcon8539]

Never make any reusable software library without at least 3 clients. A non reusable library only needs 1. If you're really good it could be a hypothetical 1 but that's still not very good.

[@KANJICODER]

@@thewhitefalcon8539 Why 3 ( I assume 2 others than yourself ) ? I see why you'd want at least one other person besides yourself . It is impossible to imagine what "other people" would want unless other people are actually using your code and telling you what they want .

[@fu_games]

what is your time horizon where you have to have start making money? How many years of financial freedom do you have? ( assuming you're doing this full time)

[@tedbendixson]

Without revealing too much, I can confidently say I made some smart financial investments over the past decade which have put me in a position that allows me to reallocate some of that capital from shares in businesses I don't run to shares in a business I do run. I'm never going to reveal my net worth publicly, but I have made a small amount of money since starting, and my audience is growing. Although I don't expect my first game to be a blockbuster, it will help to grow my audience. I would be doing this even if I was working a job to keep the lights on, just more slowly. Also, financial freedom starts to mean a lot less to you after you go through a health problem. At some point, you have to decide to live in the present, and that means taking risks.

[@fu_games]

@@tedbendixson thank you for taking the time to reply. I hope my question was not rude. I come from a very different production experience. Mobile, Hyper-Casual. Where publishers expected 3 prototypes a month from us. Which was hard on the developers. Your way of evolving your skills seemed very interesting to me. I wanted to know what it takes to start doing that. But i quess it really depends on where you live and what your costs are :)

[@tedbendixson]

It's not rude at all. I just don't think it's prudent to answer it publicly. I started out while working a job as a mobile app developer. All of the time I put into it was during my free time or when we didn't have much work coming in. I gradually built up my skills while buying up stocks. A point eventually arrived when my spouse and I decided to try going full-time with it, so now I'm part homemaker, part game maker. I'm totally self-published and have a different business model that's more oriented towards the long term. I recognize this isn't for everyone, but I enjoy it and it works for me. Ideally, I want to build a business like Wube software, just one very popular hit game (like Factorio) that evolves over the course of a decade. That's the sort of thing I'm trying to do.

[@fu_games]

​@@tedbendixson That's awesome. Just looked up Tube Software. I had a friend that was crazy about Factorio back in college. But I personally never played the game. Maybe now I will, to understand better what you are trying to achieve. It really gives me different a perspective seeing what other individuals doing in the industry. Especially the independent ones, like yourself. They are usually less corporate, more free to innovate. Btw; I'm also a part time homemaker. After closing the hyper-casual business over a year ago. My spouse and I had a similar conversation. Had a little left over money. Building up my art & animation skills over the past year.

[@tedbendixson]

lol we're in the same boat. I can't say I will come up with anything that's as good as Factorio, but I figure I can use my experiments to provide some entertainment value through this channel. I can share what I am working on while tackling topics related to software and game development. The next game I want to pursue has to do with blood circulation and hearts. I have a valve problem I'm getting fixed soon, and the whole experience has become an inspiration for a Factorio-like game involving hearts, blood, logistics, bottlenecks, clogged arteries, statin-like drugs, clotting / platelets, and so on. I wonder if it's possible to turn blood circulation into some kind of logistics management game. I'll be starting that project after I launch Mooselutions at the end of January. My surgery is next month, and they're saying it'll take me about a month to recover, assuming all goes well (it's 99.8% likely to). :-)

[@KHodow]

I feel like the same argument can be said about NOT making a game engine. Sure, center your entire workflow writing C# snippets and not understanding the backend of rendering, object pooling, etc. You would be staying in your comfort zone and continue to fear those concepts. Writing a game engine is an excellent way to improve your skills in game development and is not necessarily a waste of time. Also, most people don't "make and engine" without a game or games in mind, most of them actually start making a game and build some tooling to help them make the game (This is what I do). This is my opinion btw much love! 😄 TLDR: Let people do what works best for them, making their own engine or using a commercial engine. If they fail at least they'll learn.

[@tedbendixson]

Yeah it's amazing how many people fail by doing neither :-)

[@KHodow]

@@tedbendixson Yea for sure! This was the first video I watched and I checked out some others! I really think your moose game is really cool! Very inspiring!

[@Ortagonation]

Lol, actuallyI came from "Why You Should Learn To Program The Hard Way" and want to see what project you are doing and see this title with contradict with previous video.

[@storm14k]

Wish I'd thought about it that way many years ago. 🤣

[@tedbendixson]

But you do now! :-)

[@storm14k]

@@tedbendixson definitely. It was a good exercise but I abandoned the engine up game idea. If I'd only done that before kids and life lol.

[@caedicoes]

where's the patreonnnnn

[@tedbendixson]

www.patreon.com/tedbendixson Many thanks!

[@0x0404]

Thinking of this puzzle game makes me think of Helltaker and how I would rather play that again. Yours is technically more complex but helltaker has a better hook.

[@tedbendixson]

I'll have to give helltaker a play. And yes, I agree about hooks. My next game definitely needs a better one. Working on that

[@thewhitefalcon8539]

Software engineers talk about the rule of 3 for libraries. To make a truly reusable software library, you need at least 3 clients. If your engine doesn't have to be reusable and is just for one game, it's ok to have 1 client, but you only end up with an engine that is usable by that 1 game and not a lot else. Which is fine. But when you have 0 games, you make a game engine that is useful for 0 games and that is just a waste of time.

[@tedbendixson]

I like the spirit of this, but it contradicts my experience. I can usually reuse quite a bit of old code from project to project. I did that with my last game, keeping most of the 2D rendering, sound processing, and platform layers for both Mac OS and Windows. You can reuse most of the code.

[@thewhitefalcon8539]

@@tedbendixson Copy/pasting code *and changing it* isn't the same as making a library where the SAME library is useful in multiple projects.

[@tedbendixson]

You're right. It isn't. I think I am saying I don't find the concept of libraries all that useful for most of the code reuse I do. Most of the time, I just copy the code from a previous game project and then scrape out the specific game content until I'm left with the basic stuff like rendering, sound, etc. I don't do this that often, since I tend to work on years long game projects. It usually only takes a couple of hours to start a new thing with my existing "engine" code anyway, so no big loss.

[@sumofat4994]

So you have some good messages here but there is no reason to be forced to do one or the other. People poking around learning how game engines work is totally fine. Not everyone is relying on their private time output as a means to put food on the table. If you already have a job spend your free time doing what you enjoy simple as that. Making games are not for weary and if you expect to make money off of one you need to be hardcore. Making games for most people should be fun and if bikesheeding a game engine is fun for them than its fine. So I never am a fine of these this is how you do it videos.

[@tedbendixson]

Thanks. I get the sentiment. The algorithm loves this kind of genre, but I do want to explore other content genres like, "Here are the programming concepts I ran into while making a game from scratch."