I have been thinking about this video when the first wave of Chat-GPT hype dried out a bit and everyone was saying that programmers soon will not be necessary cause AI will be able to do anything they do but better. When you add to that a lack of job opportunities for juniors, problems with knowledge transfer between generations (it is amazing how quotes like "premature optimization is the root of all evil" changed their meaning in the last couple of decades) and more and more layers of abstractions for your common IT person, I can't shake the feeling that we are already in the declining stage, but we either ignore the fact or just did not realize it yet.
@@divinefavour1289 The current unity events is only a problem because everyone relies on it to make games, that's exactly the point he was making. If everyone relies on Unity less and less people know how to build engines from scratch, the last events only made his point more relevant.
@@TheSeriousDog you are right but there are other replacements available like Godot or unreal. Unity no longer being a viable option is not a great loss for people who don't want to make their own engine
I'm young, I write my own game engines from scratch. I love writing and understanding every part of the code I give to the CPU/GPU. As a result, my games are well optimized. On the other hand, I spend a ridiculous amount of time programming things that don't matter from a gameplay perspective. It's a choice and that doesn't make me a "better" game developer in any way. I play games on Unity that are not very well optimized but super fun to play. And I don't consider myself part of any "elite". Gameplay programmers and Engine programmers have different challenges. Grow an ego about being able to write your own engines then start assuming that building engines from scratch is the only way to write games properly and finally enjoy building your own games alone because nobody will want to work with you.
His claim isn't that it's impossible to make fun games with Unity, his claim is that if everybody just used Unreal/Unity and never make their own game engine, the number of people like you who are able to understand what a game engine is and how it works will go down over time, and making a unity/unreal competitor (or even just finding people to maintain unity/unreal) would become harder and harder.
Yeah but he didn't say that making engines yourself is the only way , he said that if everybody uses engines to make games then nobody knows how to make an engine themselves, resulting in inexperienced programmers in game companies.
1- making game engines from scratch doesnt take that much time. If you want to make an engine like unreal or unity yes, but an engine for a basic 2d game can be made in a couple of weeks, 3d game depends on the complexity so on and so forth. In fact it could take less time to make a custom engine, if your game needs it, like factorio has a really complex system, and needed a custom engine for that. 2- hes saying there should be some people willing to make an engine, because if everyone just used an engine, the number of people who know how to make an engine go down, the competent devs unity and epic games(or other engine devs) hire go down, and eventualy we wont have competitors.
Anyone that has worked with Unity or Unreal and taken it to the limit, already has read all the source code for the engines and knows how they work to a T, and has needed to customized the source code to do hyper optimization. You do not need to reinvent the wheel all the time, it is enough to know how it works. Just because you wasted thousands of hours of your life making an engine from scratch, doesn't mean everyone else has to as well.
@TheIncredibleAverage You don't need to fully understand a game engine - you need to understand what is necessary for your purpose (just as you don't need to make a complete unity-clone, but rather a purpose-built engine). No point in making uneven comparisons here (nor in lying), it does take longer to make your own engine.
I love the phrase "Do not reinvent the wheel" in the context of game engines. Imagine people behind Unreal or Godot looking at DOOM or Build Engine and saying "yeah, we already have all the tools that we need".
Such a shaky assumption. It's like, "What happens when everyone wants to drive a car, but no one wants to understand how it all works, then we will have no cars anymore!" But obviously, there are still people around building cars, even when most want to just drive them. Same story with engines.
I think his main points involve mostly the people who make the game instead of the population who play them. Not everyone who uses or needs to use a car needs to be able to manufacture one first. They are just customers of the product. I indeed find some of his points valid in the sense that what happens if all car manufacturing companies ditched their own research and development and started relying solely on their party companies and their technologies? The first thing that tends to happen is that they will be able to make a car at a lower cost, which is obvious. That premise I think holds for pretty much everyone in the industry who would throw away their own R&D in favour of the switch. But over time they will no longer be able to manufacture their own vehicle without relying on their technology, or at least they will not be able to compete since their own research and development is so far behind. Moreover, since these companies make it so that producing a game is measurably easier and also at a lower cost, there will be more "manufacturers" in the market. The overall profit for companies might start to shrink. There is only so much tolerance in the market. The amount of the population that plays games and the amount of time available for people to play these games are fixed. Another problem, which I think he said somewhere, is that all these tools (Unreal, Unity, etc.) do is provide solutions for people to apply to the problems that they are trying to solve. It often gives people an illusion that they were solving the problem but in reality, they were not. If you've experience with game programming and development, you will know when the time comes, in some cases, you do need an engine that is customized for your own need. With these tools, it is more likely and feasible to make the game that these engines excel at. And there are downsides to producing games using these tools and solutions. When I watched games that were announced recently that will be made in the Unreal engine, they all gave me the same feel. It perhaps is because I had worked so many hours in Unreal, which is not uncommon these days, to notice the similarities between them since their rendering methodology and how basically the classes are set up are the same. I will take a guess that they will end up having somewhat decent graphics but are overall undercooked by the time they are finally released to the public because they never had the chance to hone their skills while using tools developed by somebody else. And all these companies that have never developed and produced a game are popping up with these trailers made in Unreal Engine and will very likely fail in redeeming what they have promised.
@@ashz2913 I think it is an illusion that there are fewer original looking games nowadays. We have this impression simply because much more games are being made than before thanks to the tools such as Unreal/Unity. So if you simply remove all those "samey" games from the list you will probably still wind up with more "original looking" games than decades ago. Also, looking at UE 5 you have to admit that the engine tech is advancing and not stagnating. There will always be someone to innovate, simply because it takes only few smart people to advance everyone.
@@SnakeEngine That is certainly right. We do end up getting more games that look unique and have innovative mechanics. Better tools contribute to the growth of this industry as a whole and not just those types of games but also innovations. It's just that probabilistically, they are less compared to how many games that had less "craftsmanship" involved (considering everything that goes into making a successful game and not just creating an engine or technology which seems to be the topic the original video was talking about). I don't know if you are involved somehow in the game industry so forgive me if you already know this. When I sat down years ago (before I made my transition) with a producer from an AAA game company, who is a close family friend of mine, I explicitly expressed my desire to create my own game which will focus on more innovative stuff. One of the more important things I learned that day is that although the number of games being created is going up, what we saw from indie games that had achieved a successful status are but a small fraction of games in that domain. 95% of indie games had failed in different ways and most don't make any profit at all. The rest 5% had either extremely talented developers with great skills and the backing of bigger publishers, or just pure luck that their game was released at the right time and people don't have other things on their minds that they need to pick up. Our problem is not with either type of game but has everything to do with the fact that most people won't see or hear the failures of these smaller companies at all unless they are in this industry and had a connection with these companies. They are empirical blindspots and they are the vast majority. As for the UE5 tech not stagnating, I partially agree but its problems are more technical. Some of the fundamental stuff in that engine was created a long time back that never changed even after a better approach has been proposed. The "new techs" of theirs are well-researched topics in the computer graphics field that has multiple implementations codes published over decades. This is only new to a commercial game engine in which game developers are merely the users of the engines and not doing R&D themselves and will find themselves constantly behind the progress. Also, we have to appreciate the advancement of our hardware because Crytek actually had the technology before Epic when they led the research and development in the industry. They failed because our hardware was simply not good enough. There are so many engines in the market right now that are still CryEngine influenced. My game engine education started with CryEngine as well. They had decent higher-level stuff and a lot more low-level stuff which developers must be required to handle on their own compared to Unreal and Unity. This is another reason people tend to find a hard time working with it. All the "tech" that we are seeing now is because Epic as a company has the advertising budget which I think is more of an "illusion" than what we pretend it to be. Though I do wish that we don't see what JB said in the video happen before our eyes in our time. I myself don't have any intention to just follow this trend either but I guess I will see how the reality unfolds.
@@ashz2913 "We do end up getting more games that look unique and have innovative mechanics." And I think this is the main metric to measure the benefit. As long as this trend continues things certainly don't get worse. I potentially see the danger of an effective monopoly of UE in AAA space though. If the vast majority of devs relies on UE and the engine fails to meet the expectations at one point (can also be nasty deep rooted bugs no one dares to fix), everyone (including the gamers) will have to deal with the flaws.
Idiotic take. Unity has thousands working for them. Unreal has hundreds. How on earth will any indie game team comprised of 10-20 come up with something that can match Unity or Unreal plus tackle 3A’s? Removing a couple libraries? Right. Good luck with that. Even the companies who develop their own engines use Nvidia and Intel frameworks. It isn’t the 80’s anymore.
If an indie developer is trying to make a game, it's probably not going to be on the level of AAA game studio, no matter the engine they use. Unity and Unreal are designed to make big, huge, and powerful games. Games that will impress people because the realistic graphics and huge worlds. But, on the other hand, if there is an indie game company that is trying to make a simple 2D platformer, they're probably not going to make a Unity or Unreal. Why would they? They only need the 2D functionality and all of the fancy 3D stuff. In my opinion, it falls on the developers themselves. If they want to use a game engine, that's great. If they don't want to and decide to make their own small engine that will serve their purpose, then that's fine as well. If both parties are having fun, then nothing else matters.
Because they aren't trying to come up with something that "can match Unity or Unreal". They're building a game engine suitable for the constrained and specific use case of driving their specific game. That being said, a handful of developers, or even solo devs, can and HAVE come up with engines that compete well with Unity in large sub-segments of the "general purpose" game engine market. Imagine how profoundly ignorant you must be to think the quality or suitability of a tool is proportional to the number of people employed working on it.
@@fennecbesixdouze1794 He starts by saying "Video games at least used to be about maximizing what the machine could do". Huge slide over his head says the same thing. He doesn't say "you can build your own game engine for the constrained and specific use case of driving your 'specific' little game". Which is what he did a decade ago. And he is a TH-camr now and didn't come up with anything for years. Why? Because it is god damn hard to come up with something worth playing, even with those "small snippets plugged into Unity". I am profoundly ignorant. I think the quality of a tool is proportional to the number of people working on them. Those people are employed by those companies because there are millions of people using their tools for more than a decade which creates an innumerable amount of data and knowledge revolving around them. Imagine how profoundly dumb you must be to underestimate the power of such a collective. Geniuses like you keep talking about some highly exceptional specimen("a handful of developers, or even solo devs, can and HAVE come up with engines that compete well with Unity") but never list a couple things that can't be done with Unity and Unreal and decide to degrade using those tools instead.
@@fennecbesixdouze1794 The quality or suitability of a tool may not necessarily be proportional to the number of people employed working on it, but it is damn close. I would pay good money to see something like Zbrush or Photoshop be made by only a handful of people, and those aren't even as complex as Unity or Unreal.
Unity/UE are generalized kitchen sink engine. If you're making your own engine it should just what your game absolutely needs. I saw a car analogy in another comment, but I think of it like go-karts vs cars. Unity, UE can be used to build both a F350 supercab, or a Ferrari. If all you're trying to build is a go-kart, then it's something that is achievable without the unity and unreal, and will get you a lot more benefit in terms of knowledge and experience. I don't think it's an idiotic take, I think it's a nuanced perspective trying to reconcile the convenience of using a prebuilt mega-engine versus building a from scratch efficient engine that is efficient for your purposes and can run on lower spec hardware than the big boy engines. Where is the line being drawn between convenience and the toil of building something from scratch ? I think that was the point he was making, or that is what resonated with me.
I have been thinking about this video when the first wave of Chat-GPT hype dried out a bit and everyone was saying that programmers soon will not be necessary cause AI will be able to do anything they do but better. When you add to that a lack of job opportunities for juniors, problems with knowledge transfer between generations (it is amazing how quotes like "premature optimization is the root of all evil" changed their meaning in the last couple of decades) and more and more layers of abstractions for your common IT person, I can't shake the feeling that we are already in the declining stage, but we either ignore the fact or just did not realize it yet.
That aged like wine
this video is in no way relevant to the current events surrounding unity
@@divinefavour1289 The current unity events is only a problem because everyone relies on it to make games, that's exactly the point he was making. If everyone relies on Unity less and less people know how to build engines from scratch, the last events only made his point more relevant.
@@TheSeriousDog you are right but there are other replacements available like Godot or unreal. Unity no longer being a viable option is not a great loss for people who don't want to make their own engine
@@divinefavour1289 They are, but game engine development is something people should learn, so the knowledge doesn't go away with generations.
@@Maraus92 I agree with this but if people don't want to learn, I am doubtful we could force them
When nobody knows how to make cars, your country just gets conquered by a country of those who do.
I'm young, I write my own game engines from scratch. I love writing and understanding every part of the code I give to the CPU/GPU. As a result, my games are well optimized. On the other hand, I spend a ridiculous amount of time programming things that don't matter from a gameplay perspective.
It's a choice and that doesn't make me a "better" game developer in any way. I play games on Unity that are not very well optimized but super fun to play.
And I don't consider myself part of any "elite". Gameplay programmers and Engine programmers have different challenges.
Grow an ego about being able to write your own engines then start assuming that building engines from scratch is the only way to write games properly and finally enjoy building your own games alone because nobody will want to work with you.
His claim isn't that it's impossible to make fun games with Unity, his claim is that if everybody just used Unreal/Unity and never make their own game engine, the number of people like you who are able to understand what a game engine is and how it works will go down over time, and making a unity/unreal competitor (or even just finding people to maintain unity/unreal) would become harder and harder.
Yeah but he didn't say that making engines yourself is the only way , he said that if everybody uses engines to make games then nobody knows how to make an engine themselves, resulting in inexperienced programmers in game companies.
"Unity sucks because its so useful everyone will use it and no one will know how to do low level programming anymore in order to keep Unity updated."
Great advice for people who want to make 2 games in their entire life
You misunderstood what he's trying to say.
1- making game engines from scratch doesnt take that much time. If you want to make an engine like unreal or unity yes, but an engine for a basic 2d game can be made in a couple of weeks, 3d game depends on the complexity so on and so forth. In fact it could take less time to make a custom engine, if your game needs it, like factorio has a really complex system, and needed a custom engine for that.
2- hes saying there should be some people willing to make an engine, because if everyone just used an engine, the number of people who know how to make an engine go down, the competent devs unity and epic games(or other engine devs) hire go down, and eventualy we wont have competitors.
I love how idiots filter themselves
I just want to make one good game in my life
Anyone that has worked with Unity or Unreal and taken it to the limit, already has read all the source code for the engines and knows how they work to a T, and has needed to customized the source code to do hyper optimization.
You do not need to reinvent the wheel all the time, it is enough to know how it works.
Just because you wasted thousands of hours of your life making an engine from scratch, doesn't mean everyone else has to as well.
Way to miss the point. Also, the fraction that reads and uses source code like that is insanely small.
@TheIncredibleAverage You don't need to fully understand a game engine - you need to understand what is necessary for your purpose (just as you don't need to make a complete unity-clone, but rather a purpose-built engine).
No point in making uneven comparisons here (nor in lying), it does take longer to make your own engine.
I love the phrase "Do not reinvent the wheel" in the context of game engines. Imagine people behind Unreal or Godot looking at DOOM or Build Engine and saying "yeah, we already have all the tools that we need".
Such a shaky assumption. It's like, "What happens when everyone wants to drive a car, but no one wants to understand how it all works, then we will have no cars anymore!" But obviously, there are still people around building cars, even when most want to just drive them. Same story with engines.
I think his main points involve mostly the people who make the game instead of the population who play them. Not everyone who uses or needs to use a car needs to be able to manufacture one first. They are just customers of the product.
I indeed find some of his points valid in the sense that what happens if all car manufacturing companies ditched their own research and development and started relying solely on their party companies and their technologies? The first thing that tends to happen is that they will be able to make a car at a lower cost, which is obvious. That premise I think holds for pretty much everyone in the industry who would throw away their own R&D in favour of the switch. But over time they will no longer be able to manufacture their own vehicle without relying on their technology, or at least they will not be able to compete since their own research and development is so far behind. Moreover, since these companies make it so that producing a game is measurably easier and also at a lower cost, there will be more "manufacturers" in the market. The overall profit for companies might start to shrink. There is only so much tolerance in the market. The amount of the population that plays games and the amount of time available for people to play these games are fixed.
Another problem, which I think he said somewhere, is that all these tools (Unreal, Unity, etc.) do is provide solutions for people to apply to the problems that they are trying to solve. It often gives people an illusion that they were solving the problem but in reality, they were not. If you've experience with game programming and development, you will know when the time comes, in some cases, you do need an engine that is customized for your own need. With these tools, it is more likely and feasible to make the game that these engines excel at.
And there are downsides to producing games using these tools and solutions. When I watched games that were announced recently that will be made in the Unreal engine, they all gave me the same feel. It perhaps is because I had worked so many hours in Unreal, which is not uncommon these days, to notice the similarities between them since their rendering methodology and how basically the classes are set up are the same. I will take a guess that they will end up having somewhat decent graphics but are overall undercooked by the time they are finally released to the public because they never had the chance to hone their skills while using tools developed by somebody else. And all these companies that have never developed and produced a game are popping up with these trailers made in Unreal Engine and will very likely fail in redeeming what they have promised.
@@ashz2913 I think it is an illusion that there are fewer original looking games nowadays. We have this impression simply because much more games are being made than before thanks to the tools such as Unreal/Unity. So if you simply remove all those "samey" games from the list you will probably still wind up with more "original looking" games than decades ago. Also, looking at UE 5 you have to admit that the engine tech is advancing and not stagnating. There will always be someone to innovate, simply because it takes only few smart people to advance everyone.
@@SnakeEngine That is certainly right. We do end up getting more games that look unique and have innovative mechanics. Better tools contribute to the growth of this industry as a whole and not just those types of games but also innovations. It's just that probabilistically, they are less compared to how many games that had less "craftsmanship" involved (considering everything that goes into making a successful game and not just creating an engine or technology which seems to be the topic the original video was talking about).
I don't know if you are involved somehow in the game industry so forgive me if you already know this. When I sat down years ago (before I made my transition) with a producer from an AAA game company, who is a close family friend of mine, I explicitly expressed my desire to create my own game which will focus on more innovative stuff. One of the more important things I learned that day is that although the number of games being created is going up, what we saw from indie games that had achieved a successful status are but a small fraction of games in that domain. 95% of indie games had failed in different ways and most don't make any profit at all. The rest 5% had either extremely talented developers with great skills and the backing of bigger publishers, or just pure luck that their game was released at the right time and people don't have other things on their minds that they need to pick up. Our problem is not with either type of game but has everything to do with the fact that most people won't see or hear the failures of these smaller companies at all unless they are in this industry and had a connection with these companies. They are empirical blindspots and they are the vast majority.
As for the UE5 tech not stagnating, I partially agree but its problems are more technical. Some of the fundamental stuff in that engine was created a long time back that never changed even after a better approach has been proposed. The "new techs" of theirs are well-researched topics in the computer graphics field that has multiple implementations codes published over decades. This is only new to a commercial game engine in which game developers are merely the users of the engines and not doing R&D themselves and will find themselves constantly behind the progress. Also, we have to appreciate the advancement of our hardware because Crytek actually had the technology before Epic when they led the research and development in the industry. They failed because our hardware was simply not good enough. There are so many engines in the market right now that are still CryEngine influenced. My game engine education started with CryEngine as well. They had decent higher-level stuff and a lot more low-level stuff which developers must be required to handle on their own compared to Unreal and Unity. This is another reason people tend to find a hard time working with it. All the "tech" that we are seeing now is because Epic as a company has the advertising budget which I think is more of an "illusion" than what we pretend it to be.
Though I do wish that we don't see what JB said in the video happen before our eyes in our time. I myself don't have any intention to just follow this trend either but I guess I will see how the reality unfolds.
@@ashz2913 "We do end up getting more games that look unique and have innovative mechanics."
And I think this is the main metric to measure the benefit. As long as this trend continues things certainly don't get worse. I potentially see the danger of an effective monopoly of UE in AAA space though. If the vast majority of devs relies on UE and the engine fails to meet the expectations at one point (can also be nasty deep rooted bugs no one dares to fix), everyone (including the gamers) will have to deal with the flaws.
when using unity you're not "driving a car", you're "making a car". And you're using a flawed software to do it. Jon is right
Jonathon Blow: "Who will be our future low level engineers??"
GPT5:
Idiotic take. Unity has thousands working for them. Unreal has hundreds. How on earth will any indie game team comprised of 10-20 come up with something that can match Unity or Unreal plus tackle 3A’s? Removing a couple libraries? Right. Good luck with that. Even the companies who develop their own engines use Nvidia and Intel frameworks. It isn’t the 80’s anymore.
If an indie developer is trying to make a game, it's probably not going to be on the level of AAA game studio, no matter the engine they use. Unity and Unreal are designed to make big, huge, and powerful games. Games that will impress people because the realistic graphics and huge worlds. But, on the other hand, if there is an indie game company that is trying to make a simple 2D platformer, they're probably not going to make a Unity or Unreal. Why would they? They only need the 2D functionality and all of the fancy 3D stuff. In my opinion, it falls on the developers themselves. If they want to use a game engine, that's great. If they don't want to and decide to make their own small engine that will serve their purpose, then that's fine as well. If both parties are having fun, then nothing else matters.
Because they aren't trying to come up with something that "can match Unity or Unreal". They're building a game engine suitable for the constrained and specific use case of driving their specific game.
That being said, a handful of developers, or even solo devs, can and HAVE come up with engines that compete well with Unity in large sub-segments of the "general purpose" game engine market.
Imagine how profoundly ignorant you must be to think the quality or suitability of a tool is proportional to the number of people employed working on it.
@@fennecbesixdouze1794 He starts by saying "Video games at least used to be about maximizing what the machine could do". Huge slide over his head says the same thing. He doesn't say "you can build your own game engine for the constrained and specific use case of driving your 'specific' little game". Which is what he did a decade ago. And he is a TH-camr now and didn't come up with anything for years. Why? Because it is god damn hard to come up with something worth playing, even with those "small snippets plugged into Unity".
I am profoundly ignorant. I think the quality of a tool is proportional to the number of people working on them. Those people are employed by those companies because there are millions of people using their tools for more than a decade which creates an innumerable amount of data and knowledge revolving around them. Imagine how profoundly dumb you must be to underestimate the power of such a collective.
Geniuses like you keep talking about some highly exceptional specimen("a handful of developers, or even solo devs, can and HAVE come up with engines that compete well with Unity") but never list a couple things that can't be done with Unity and Unreal and decide to degrade using those tools instead.
@@fennecbesixdouze1794 The quality or suitability of a tool may not necessarily be proportional to the number of people employed working on it, but it is damn close.
I would pay good money to see something like Zbrush or Photoshop be made by only a handful of people, and those aren't even as complex as Unity or Unreal.
Unity/UE are generalized kitchen sink engine. If you're making your own engine it should just what your game absolutely needs. I saw a car analogy in another comment, but I think of it like go-karts vs cars. Unity, UE can be used to build both a F350 supercab, or a Ferrari. If all you're trying to build is a go-kart, then it's something that is achievable without the unity and unreal, and will get you a lot more benefit in terms of knowledge and experience. I don't think it's an idiotic take, I think it's a nuanced perspective trying to reconcile the convenience of using a prebuilt mega-engine versus building a from scratch efficient engine that is efficient for your purposes and can run on lower spec hardware than the big boy engines. Where is the line being drawn between convenience and the toil of building something from scratch ? I think that was the point he was making, or that is what resonated with me.