How To Create your own game Engine Like Unreal,Cryengine ,and Unity Engines

Since Unity doesn't have a revenue share, this argument is irrelavant. There are plenty of games making huge dollars built with Unity, especially in mobile.

Unity games represent something in the order of 60% of the revenue made on the App Store. (Not a hard stat, just something that was mentioned at Unite.)

 

Ah my bad! You're both completely correct about that. Should've checked that before posting. I'm well aware that Unity is huge in mobile and with good reason. I love unity and I've worked at a studio where we were using it but it was for an online pc game and I'm using it for a few personal projects. I also know quite a few people working at mobile studios using unity. I also know a few large mobile studios that have their own in house solutions.

Anyways, now Im back at being confused why we're using our own engine!

 

In any case, i think my point about the OP learning about engine programming still stands. I think he will be in demand if he pursues that path.

 

There are valid reasons to use an in house engine. The two most common are

• You want to do things with the engine that can't be done with an off the shelf engine
• You already have an in house engine

The first reason is getting rarer. The capacities of the off the shelf engines are growing daily. There are still niche cases where a proprietary engine makes sense. But even some of the big traditional AAAs are starting to use off the shelf engines as a way to cut costs and development time.

The second reason is also getting rarer.

Yes, but this:

If you massively cut down your scope, a small team probably could keep up with Unity. You could develop an engine to make one type of game, on one type of platform, with far less resources.

However in many cases putting that same engineering team onto extending Unity will give better results.

 

This. Unity has a huge team, but you get huge overheads from that and you also get huge overheads from the sheer number of platforms they need to support. Just imagine creating an engine feature and having to test it on 27 platforms. Then consider that most of those platforms have many models of hardware, sometimes spread over many models / generations / architectures.

They also have to cover and support a whole lot of stuff that may not be applicable to any one particular project. Between all of those things, I'm pretty sure that an experienced team who knows what they need could still be similarly effective when creating their own tools, especially if their project is in a niche that general purpose tools aren't ideal for.

I remember hearing about the Dead Space team building a purpose-specific PC/console engine for a AAA game in a relatively short period. That's the kind of game that I'm sure you could build really effectively with today's off-the-shelf engines, but they weren't available at the time, and I think it's a good example of what can be done with a good team if you can keep tight control over your scope.

 

Unity isn't an engine "they made from scratch" either (with Umbra, Physx, Enlighten, Mono etc. etc.) and neither is there a point to doing so in most cases. What, do you want me to write my own graphics API before you'd consider it "from scratch"?

Fact of the matter is every modern engine worthy of any note uses frameworks / additions / pre-built API's.

It's simple as this, if you're at the point where every pre-made engine would need a re-write to get where you need to go then make your own. If not, carry on..

 

A year later...

 

Yes please. In assembly.

 

Yeah, give or take ~363 days.

 

The first thing you need is to get some basic knowledge on Unity and C#, which you will get by going through all the tutorials here:
https://unity3d.com/learn
It will take some time, but in the long run it will greatly speed things up for you.
If you are stuck or you have questions, please make a thread here:

https://forum.unity3d.com/forums/getting-started.82/

Now that we've gone over where to start, could you please explain how exactly you arrived at the decision to post your question in this specific thread? Thanks a lot in advance and welcome aboard!

 

Start with Godot 3 when it will be available as a base , it is open source and you won't have to re invent the wheel.

 

You'll receive far more help if you create a new thread and not revive an old one. Reviving a new thread for a completely different topic will almost never attract you the right sort of attention. Many people will simply see an old thread and close it immediately.

I highly recommend reposting this request in Getting Started as a brand new thread.

 

Dude this is nonsense and there is nothing honest about it. By all means, @PhoenixElement77, build your own engine.

 

I'm always amazed by the people who show up literally out of nowhere solely for the purpose of challenging opinions.

 

NO YOU AREN'T!!

 

You probably not the best idea to start out with a goal of making a game engine. Write even a relatively simple game. When your done, use that code to build a more complex game. You’ll find better ways of doing things to handle things inside the game. Add features that you wish you had in the prior game. Keep doing this with each game getting more complex. Before you know it, you will have a fairly generic game engine that you know well.

 

Ummm, am I the only one who notices that after 2 years he probably already made up his mind by now?

 

in 15 min even if its not what you want its a must see

 

All the time you spend coding, you would had make a game using ready to go frameworks and animation or AI tools.

 

I know I'm late on this, but don't let yourself be put down by people saying it's impossible. Notch made Minecraft with nothing but Java and OpenGL, and Minecraft's Physics and Graphics are not bad at all. Now, it will be very difficult of course, but it won't take 50 years to create like what some people say. However, it still can take several years to build an engine like what some have advised. Start simple, decompile and read minecraft's source renderer and manager first, then go on to read sources of bigger game engines like Unreal. Or if you want to be a legend, build a game purely on code like what Minecraft's developers did

 

...Minecraft's physics and graphics are the worst. The physics engine is a total hack job and for the longest time didn't even have a terminal velocity system for players. The only reason you can stand on a fence without getting booted for flying is likely still because of a F***ing flag. Minecraft is nowhere near as complex, well programmed, or even as good as you're making it out to be.

 

With Xenko, Godot and some others in Beta i don't understand the need to spend years re inventing the wheel, and finally not make any game

 

Actually, Minecraft was based on LWJGL framework. Does that still considered nothing but?

On the topic:
In my time I did some "research", tried to implement a game on LWJGL, before that on Slick.
It was a real mess. Something that took me month to implement could be done in Unity in two clicks.

XNA Framework was a decent "success" for me, at least something was coming along right with it.

Switched to Unity after that, never looked back at creating own engine. It's just not worth it.
Even all that pseudo-"low-level" stuff, can be learned and experienced through Unity.

Solid coding practices and knowledge of the existing popular engines are way more potent nowadays than any in-house engine.

One suggestion though, if anyone wants to get "low-level", try writing shaders.
They're fun and enjoyable to learn.

 

Yes. The LWJGL framework is just a lightweight wrapper around the Khronos APIs. It's comparable to the Managed DirectX framework Microsoft was developing before they replaced it with XNA. If it provided any functionality to assist with game development my answer would have been different but it's just an easy way to access the Khronos APIs. Nothing more.

 

Minecraft's graphics are not bad.....

Thanks for the laugh

 

Game development forums in a nutshell:
Don't build a game engine, it's too hard.
Don't build MMOs, they're too hard.
Don't even bother with multiplayer, it's too time-consuming.

These are often accompanied by snide remarks and thinly-veiled insults.

There are still some people (some of whom have contributed to this discussion) who believe that knowledge gained > income potential. I think the notion that only projects that can be sold on Steam are worth anyone's time is exactly why the industry is bloated. For me, I'd rather spend my time building multiplayer frameworks than build a game engine, but I certainly wouldn't fault someone whose personal interests favored the latter. Not every engine needs physics or complex rendering or even lighting, and I'd be willing to bet that there are plenty of cases in which new game developers, who feel like Unity is overkill for their particular project, would find a simpler engine fits the bill.

Worst-case scenario is you come out of it knowing more than you went in. I'm just saying, it seems like a lot of responses here focus on what you will or will not end up with as a product when you're done, and that is the only thing of worth to be gained.

 

That's not exactly true. Building a game engine isn't a good idea because the cost/benefit ratio is rather poor. It will take years to build and is unlikely to be a better engine than something freely available like Unity, and then if you were building the engine for a specific game then that game is a couple years behind in development than where it would have been. It is very difficult to create your own game engine today where your time could not have been better spent doing something else.

The problem with building an MMO is often it is the new game developer just getting started that wants to take on an MMO project without understanding the amount of work and enormity of the project they are committing to. The same thing goes for any networked multiplayer project, as to the new game developer it seems like just adding a minor feature to their game without understanding that they are committing to multiplying the development effort.

The worst case scenario is they put months or even years into their project, finally understand the scope of what they are trying to tackle, get discouraged, and walk away from game dev.

 

Again, not everything is about income or is subject to cost/benefit analysis.

Understanding the scope of what they are trying to tackle = knowledge gained, and that doesn't even take into account the actual game development knowledge they would learn along the way, especially if they spend years doing so.

Personally, I don't buy into the whole multiplayer is too hard or MMO's are too hard. Game development in itself is hard work, and while those two types of games are harder, I don't see difficulty as any reason to give up before starting. If we all took the easiest route then we'd all be making Flappy Bird clones, and maybe that's the route you've decided to take, but I will never understand telling someone else that their ambitions aren't within their grasp simply because you think the ends don't justify the means.

Making a game engine is one of many, many doors that any developer, beginner or veteran, can choose and the possible fruits of that labor - be they a working engine, knowledge gained, or even a job offer - aren't something someone else can quantify for you, which seems to be what a fair number of developers try to do on behalf of beginners.

 

That's because multiplayer and MMO building is NOT too hard, but creating them takes a very long amount of time, and requires experience not yet gained by the novice. A gaming dev novice shouldn't take on an enormous project in the beginning for the same reason a new writer should start with a short story instead of a novel trilogy, and a budding director should start with some YouTube length videos rather than a feature length film as a first project. The reasons are plainly obvious.

 

So the notion is that you should only write short stories until you prove yourself to yourself, and then and only then can you tackle a novel? I think the line of authors who would argue that point is a very lengthy one, as is the number of directors who started making full-length movies at home before and after YouTube existed. In fact it is quite common for a young director to recreate their favorite movies as a starting point. Chances are none of these are ever sold, not a single dollar is ever made, next-to-nobody sees them, a fair portion of them never complete their project, and yet it is a system that has proven to work time and time again. You seem to be suggesting they should all start out with commercials, then when they've earned it they can start to think about TV episodes and then maybe a movie after they've paid their dues? To whom are these dues due?

 

When you learn math in school, should you start with calculus? When learning about money, should you start with a class on macro-economics? When you want to learn a new language, should you just apply for a job as a translator? If you want to learn about physics, should you start with M-Theory?

There is a reason why when you want to learn something it is very important you start with the basics and not start with the most advanced topics or projects in that field. Game development is not some exception to this. Multiplayer games, MMO's, building your own game engine, these are all the most advanced topics game development has to offer and as such simply are not where a novice developer should start.

 

That's a false equivalency, but I think you're already aware of that. In your analogy, both the "math" you want to learn and the calc are both parts of a whole. Learning rudimentary math would be equivalent to learning the basics of creating a game engine; one leads to the other. I never suggested someone start with learning how to create a physics engine (calc in your analogy), I only suggested that saying that someone who doesn't understand basic arithmetic has every right to set out to learn calc from the onset.

I'm not having trouble with this at all, although I did find it a bit taxing to follow your logic. Yes, you would need an understanding of the written word to write a novel, but there is no prerequisite that you have written short stories beforehand, especially if you are only doing so for the sake of learning something new or accomplishing something you've never accomplished before. Google can provide you with examples of authors who skipped the whole short-story thing and went directly to writing novels.

I know how difficult multiplayer/MMO frameworks are to build, it's pretty much the only thing I get hired for these days. If I had listened to everyone then I'd have considered myself a fool for even learning how to solo an MMO, and yet here I am making a nice living off of what I learned. I never released an MMO myself and never intend to, but the knowledge I gained has been priceless in advancing my game development career. In fact, dealing with database tables-worth of data as I learned completely changed how every one of projects handles storing information for units/characters/etc. I could toss up my MMO framework today and have people playing it, asking me for tuts, and even work towards a Patreon setup to fund its progress, but it's not where I want to take my career. The OP has the added benefit of knowing what they want to make and that they will use it in the end

It seems to me someone came to the forums to ask for direction on how to build an engine and instead got a fair number of responses that were simply recycled nonsense that the person who responded got when they first asked the same question: tell new devs what they can and cannot do and someday when they wear big pants, they'll be able to tell the new kids on the block what they can and cannot do as well. Rinse and repeat.

Seems we went from someone asking for direction on how to create a game engine to them being told they shouldn't even try by the fourth response. I think the community as a whole deserves a round of applause.

 

I agree with the first half, but not the second. Yes, some things aren't about income. Everything can still be subject to cost-benefit analysis, the cost and benefit just aren't always going to be measured in dollars.

Paying a cost purely because you think you'll enjoy the activity or the outcome is a perfectly valid decision to make.

 

I never claimed it wasn't, and a willingness to pay a cost doesn't mean it needs to produce benefits; this current thread is a perfect example. I do, however, always enjoy when someone takes a single line from one of my responses to argue a point that seems to run parallel with the overall point I was making anyway.

 

I'm struggling to come up with anything that someone would pay for, in any way, that doesn't provide benefits of some kind. I think we may be using the word "benefit" differently?

All I meant is that I don't do much of anything without first thinking what I'll get out of it. Sometimes the thing I get out of it is for someone else, or is for pleasure rather than practical value, but there's still something there.

 

I think so, because you're using the term to cover benefits to others whereas I was not, although considering the context of this thread, I'm not sure how benefits to others plays into this discussion. Everyone who said building a game engine is a bad idea pointed out the costs to the developer, and I'm not sure how creating a game engine for the sake of learning would provide a benefit to others unless it was open-sourced or something similar. Again, that doesn't seem to be the case in regards to this discussion.

As far as examples of things that have a cost personally and no tangible benefit to the one paying said cost, the first thing that comes to mind given recent news is hurricane relief. There is no benefit for me but it comes at a cost. I am also mentoring my sixth aspiring game developer, and again, it comes at a cost to me but there is no benefit on my end.

I can't imagine living my life this way, so I think it's safe to say that we are having two separate discussions and aren't at odds with one another whatsoever.

 

I find thinking like that all the time to just be stressful. If I find myself considering what I get out of something, I find myself being concerned with the end result, rather than the process to get there. When doing something for fun or just to relax, I find its generally going to be the process I enjoy most. In that case, thinking about what I get out of something is kind of a distraction, if that makes sense.

 

Ok, it's not the word "benefit" that was causing the misunderstanding, it's the word "everything". I am not talking specifically about making game engines. I am genuinely talking about everything.

Neither of those things make you feel good? I like helping other people, so that is a benefit in and of itself. But now we're getting philosophical...

That's what that cheesy but true quote "it's the journey, not the destination" is all about.

If I watch a movie I'm not thinking "what do I get out of completing this movie?" I'm thinking "will I enjoy watching this?"

 

This is what I meant when I said we have different ideas of what "benefit" represents in the context of this discussion. Knowledge gained from developing a game engine - even one that fails - is a personal benefit. Donating time and/or money benefits others. Honestly, if donations made me "feel good" I think it would take away from the whole thing, almost taint it. I do what I can and go back to business as usual. I suppose now that I think of it there is a benefit to mentoring in that teaching someone to do something always helps you understand it better, but even thinking of it in that terms muddies the waters; like pointing out to a teacher that while teaching the leaders of tomorrow they're earning a salary.

Then for the sake of the OP, we should probably get back on the topic of developing a game engine.

 

I have to point out that you and I had a recent discussion in which you took the complete opposite opinion and stated very clearly that you always take into account what you are getting out of anything before ever doing it. We had differing opinions on what benefits were. If you're playing Devil's Advocate then by all means, play away, but I felt compelled to get your take on this and where you truly stand. Again, I may be reading too much into this and maybe you didn't actually mean that you don't consider benefits anytime you undergo a task, and I am referencing another thread, but it still is on point insofar as this discussion.

 

That's not "opposite". I'm saying here that the "journey" can be the benefit. That's the whole point of the quote.

If I take a drive down the coast the benefit could well be enjoying the drive in and of itself, regardless of where I end up.

On the topic of the thread, someone could write a game from scratch, or write an engine, because they enjoy the act of doing so rather than because of some future outcome.

 

This feels like I'm getting sucked into another debate in which we will again both concede that we are hung up on a single word. We've been there and done that.

I agree someone can create an engine for fun, but I don't agree (nor am I implying you're suggesting) that if they choose to do so in a way that is not a hobby that it gives people the right to tell them they shouldn't. I don't agree that someone can make the call as to whether or not they will fail based on the OP either. I don't agree with this notion of everyone sharing their own personal trials and tribulations over and over again in the hopes of discouraging people. And I'm not saying that everyone does it with that intent, but it doesn't take a very long look around these forums to find instances of that very thing occurring with disheartening regularity.

It seems the consensus is that if you're successful in the industry you have the right to assess whether someone else will be with only so much as an forum post. Again, I am not implying you said this specifically, I'm just saying that has been and continues to be my point.. I'm just responding to ridiculous off-topic replies until someone steps in and continues that particular discussion.

 

You're doing this to yourself, and it's based on misunderstandings (I think?) which aren't even from this thread. It's been fun, but I'm out.

 

I'm inclined to both agree and disagree. I believe it entirely depends on the context of the thread but generally speaking I believe people have the right to make the recommendation just like the person asking for advice has the right to ask how to do it. Just as long as they're polite and keep it to a minimum since it won't be the topic of the thread.

That said the ideal post in my opinion wouldn't just be a recommendation against that approach but would include ways on how to get started. That way if they don't succeed and decide to choose an existing engine after all they'll be more inclined to be receptive to our advice since we at least tried to help them. If they do succeed then that's awesome too.

 

Stay on topic.

 

If you look at it seriously you'll realise it renders very well compared to other games that don't use an engine and the graphics are clear. If you think the textures are bad that's another different case

 

I'm not saying it could compare to better Engines with entire teams of developers behind them, but it's quite good when you look at it. I'm not sure whether you're referring to the block like textures when you say the graphics are horrible so i can't say anything else. Yes, sure, the physics engine is simpler to a certain extent than in an engine like Unity due to lack of joints and other essential parts, but gravity and pushes from attacks and movement in water work just fine, and aren't trash especially compared to the coded games that other people simply put together for attention. It's not as well programmed as Unity or Unreal of course, but it would be a lie to say it isn't complex. My point was that if Minecraft could be made then a Game Engine is possible, just very, very hard to make. In other words some encouragement for him in case he wants to actually try.

 

Minecraft has nothing remotely complex to render.

It's not like Markus Persson went in with no prior experience either. Everyone thinks of Minecraft when they think of him but he had extensive experience prior to developing the game. He began programming at the age of seven, he worked at King Ltd (the company known for Candy Crush), he was one of two developers that started an MMO, etc.

Unlike many people who ask questions like this he had the qualifications to do just that. If you want to encourage new developers you might want to use a different example. Ideally one that wasn't already very skilled when they made it.

https://en.wikipedia.org/wiki/Markus_Persson
https://en.wikipedia.org/wiki/Wurm_Online

 

Fair point i guess

 

"Standardise"