Can you please make a tutorial series on the Heaps IDE, because it's very hard for a beginner to understand what's going on, yet it has a lot of potential for game development and I'd love to learn how to use it.
They should really bundle this into something like "Shiro Game Engine" like you mentioned. I find having everything spread out like this makes it a little tedious.
@@DayInDaLife Heaps doesn't have an IDE included, you can use Hide with it though. They should include a "Bundle" where you can download everything into one neat installer. So you can get the castle DB included with it as well. Would be a good marketing strategy to get more people to pick this stuff up.
That database tool is indeed something in many game companies long for! In fact, I am going to pitch it to the company I work for now. Thank you for this video. What an amazing set of tools. I look forward to diving in each of them!
This all seems overly complicated, although game design is always complicated I guess. It doesn't seem that intuitive for a new user to try and get this different parts working together well.
I agree that haxe and heaps is seriously awesome, but when I looked at it I found it hard to get into as far as documentation and tutorials go. I know there is good documentation but I still found it difficult to follow
Haxe and Heaps+HL are great. But amount of troubles with their config is intimidating... There are always some versions incompatibility of one or another. Some libs work with HL, others don't... And Heaps learning style is like "here are examples, source code, good luck"
I built a small game in Unity like 7 years ago. Ever since, any time I want to test out a new game engine I rebuild this game in it. For what it is worth, with no haxe experience, I built the game in heaps in less than a day. It was very nice to work with really. For comparison the closest to that was when I rebuilt it with Phaser a few years ago in about 2-3 days. I still haven't gotten myself to finish it in Godot, but have about 2 days of time sunk in there so far. The original Unity version took about a week. Cocos w/ Python took about 3 days. I started it with raylib as well, but haven't spent much time there. I get distracted easily... haha. Haxe overall has peaked my interest though. I really wish the ability to include native libraries in projects (i.e. a/lib/dll/dylib/etc.. c or c++) would be worked out a little more though. Ammer is a cool project attempting this, but I think if they could get that in a good position more and more people would start building things with it. There are just so many nice c/c++ based libraries that I think haxe developers would benefit greatly from being able to include easily. (I know there are some already with bindings and its possible to make bindings, but its not the greatest experience at the moment.) Cool video as always!
@@SavageArms357 I don't know if I'd call them roadblocks so much, but the extra time so far has mostly been spent learning the UI and gdscript capabilities. Even though engines with GUI editors are all the rage and definitely help non-programmers get started, I've always found it easier to just throw a library in a project and start coding. Since learning the tools involved in others requires more time. Though it is probably just me and an up front impression. Having those tools down the line and with a more complicated game is probably great once you learn it all. With all that said, I'm working with Armory3d a little now. And of all the game engines I've tested out, even though I've only finished the game with one, the top 3 as of now are Armory3d, Godot, or Heaps.
Haxe does look nice and I have tried it a couple of years ago, but for me, the issue that I have is that prefer to work with a game engine that has an editor. And I didn't know that haxe had hide. Hopefully if they do what you mentioned at 12:15 with packaging all of the components together would definitely give haxe another try.
Heaps is a nice engine but is poorly documented which kinda discouraged me. Haxeflixel has great documentation, the Vs code plugin for Haxe gives rich autocompletion and hover info. It felt very much welcoming for someone who is a noobie in programming. Highly recommended 👍
This is sort of a "Catch-22" reason for not using it though. The reason it is not well documented is because not a ton of people use it, and the reason not a ton of people use it is because it is not well documented. I think it is better to give things a try a lot of the times and maybe figure something out and share it with the community or something even if it is something easy. And, if you don't like it after that, move on
@@HonsHon I'm a noobie at game programming and also programming in general. And good documentation will give a boost to interest me. It's not like I don't like it, it's in my list.
I’ve tried lots of game engines and Haxe w/ heaps seems to be where it’s at for me. It’s free no strings attached, runs on Linux and it has a butt load of features. Haxe and heaps seems to be lacking community because several reasons such as flash exiting the arena, it does not have a heavy editor attached and when you google game engines you get results mostly for unity or unreal.
@@rytif But your comment wasn't in a vacuum was it? There was context, and there are more things to which "that" can refer, such as in this case, your reasoning. You are very stupid, aren't you?
Been using Haxe forever, it's easily my favorite language/toolkit out there, for all kinds of uses. I'm still using the HXCPP target for games projects mostly because of documentation issues around the HEAPS tech stack. There's obviously amazing stuff in here but the docs-ghost keeps haunting Haxe unfortunately. Really, really unfortunately, given the quality of the actual thing.
Hi andreas, im another long time user of Haxe from back in the day but abandoned it a number of years ago because of issues with docs. It really is a thoron in the side fo haxe
Man when I first saw Shiro I thought... wait... that Shiro? And I soon found out that the madlads involved in Evoland2 also developed an entire new language along with a game engine. Absolute units..
The biggest limitation is there is no built-in physics component. There are some dubiously supported implementations of Nape, Box2D and Bullet Physics.
I tried heaps recently and here is my two cents on this "engine" stack: This stack is not cross compatible with any console/PC out there. What you get is generated C/C++ code that you have to then deploy on each system alone. This is not an easy job and you have to have a Dev Console for each one of them in order to even begin porting process, so its quite expensive too. Unity/UE4/Godot/LibGDX do this for you so that you don't need dev consoles. ** heaps.io: Heaps is not a game engine, it is a graphics engine, it says so on the website. This means that many capabilities of a game engine are not available: 1. NO VIEWPORTS 2. No colliders 3. No events 4. No sound engine 5. No physics engine 6. No helper classes that allow you to calculate things like path finding, noise generators etc. 7. Many more things that make a game engine What you can do: 1. Display graphics 2. Particle system 3. Font display 4. Displaying/"create" animations 5. Compatibility with some obscure systems (CastleDB, HScript etc.) ** HashLink HashLink is a virtual machine that works just like any other virtual machine. It runs only on Windows/Linux and has the ability to generate C/C++ code (many VMs can do this). There is no HashLink port for Android/IOS/PS4/XBONE/Switch which means you have to take responsibility and port C/C++ code manually to those platforms. ** Domkit is just a UI library ** HScript is an interpreter for Haxe language ** MpMan is not available as it is an internal library ** HxBit is a binary protocol (there are many of them out there) ** Haxe language is nothing special, it uses C style syntax and reminds me a lot of Java/C#. It can cross compile to other languages but this feature is not available to the heaps.io Worst part about all of this is lack of documentation that makes it extremely hard to do event the simplest of tasks. IntelliSense plugins are available for Idea and VSCode but they are bare bones and barely work. This is a inhouse engine made by Motion Twin / Shiro Games and it makes sense for them to use it as they made it and know how every gear turns but with lack of documentation and a very elaborate deployment setup it is better to just stick to an established engine. You will waste a lot of time trying to understand how the engine works instead of making the game.
Many parts of this are not correct: - you can output a .hl file instead of C which cross platform bytecode (Window/OSX/Linux). For consoles yes you need C+SDK - heaps.io has Colliders (in 3D), Events (see Interactive in both 2D/3D), Sound engine (hxd.snd) - HashLink runtime runs on Android/IOS/PS4/XBONE/Switch, so you don't to part any C/C++ code manually, you still have to compile it though - Haxe cross compilation is used for Heaps.io for HashLink / Javascript cross compilation. You can have your games as native (desktop) and HTML5/WebGL - Plugins for VSCode is working very well. Please raise any issue you might have. This one at leas is not "bare bone and barely work", it's used for daily dev by many Haxe developers I appreciate you took the time to look into it, but please correct the errors as your report is misleading.
@@NicolasCannasse mad respect to you for creating Haxe. I don't doubt heap's capabilities at all but I've found solace in Haxeflixel. As a noobie programmer, I found haxe quite comfortable to pick up.
@@NicolasCannasse I mostly use IDEA and the plugin was very buggy Every engine out there can export to PC/MAC/Mobile, Heaps.io promises PS4/XBONE/Switch out of the box, which isn't true.
if you want a reason to use this engine, evolant 2, just play it from start to end, and you will know why, in my opinion, yes, i wish there was a just one click download
TLDR; A tech stack isn't really a game engine. About adoption, I think the concept of a tech stack is confusing to some people in and of itself. I prefer the Shiro approach myself because it lets me cherrypick what I care about. For instance I'm curious about the workflow improvements around HScript (which I already use), CastleDB and the performance of HL + its C output. However, HIDE is on first encounter an opaque hell-machine, DOM and CSS is the last place I want to do any UI ever, and I've already got my own stack that solves those problems on top of the existing CPP target. Shiro isn't presenting this as "a game engine". It's just their stack of tools. The makeup of each project built with this stack is going to have a ton of variance. There's huge value here for pretty much any game developer willing to configure and build their own stack, but if you want a spin it up and go kind of work environment for prototyping etc, this is really not going to do that for you.
I feel like it's probably not as popular as Godot cos it's a stack, which probably scares off indies who are used to the self-contained IDE/engine combo apps like Unity
Can you please make another tutorial, focussing on this technology stack and on using the many build targets (Windows, Ios, Android, Switch, Ps4, Xbox, etc)? I tried for about one month to do it myself, and even got templates and tutorials, but nothing works, or at least not for my usecase. I was trying to make an small endless runner for my first game with Haxe, but there is just no documentation for it.
Hm, it IS weird not many people talk about it, if great games like Dead Cells used it. I'd imagine people would look up the engine used. I wonder if it's a marketing problem. Unreal and Unity are well known because, well. They proved themselves, and has a great marketing along with them.
I get if you wanna write your own game engine for your own game, but I just don’t understand why there are so many “general purpose” game engines trying to get a slice of the industry.
Because there are many ways to make an omelet. If you compare Godot, Unity, UE, Defold, GDevelop, Ogre, Torque, CopperCube, etc, you'll find fundamental differences in the workflow, in the programming languages used, in the performance, the memory footprint, etc, and that will satisfy different people. The market is still evolving to better and easier ways to make games, so the more the merrier. They end up pushing each other to improve more and more. The only thing I wish existed more is specialized engines, specific to certain game genres. A problem with generalized engines is the wasteful overhead required to deal with a multitude of different game requirements. Specialized engines are more limited, but less wasteful. RPGMaker is one example of a specialized engine. But there aren't that many in the indie market. For FPS games, for example, you only have engines like Doom, Source, etc, which are actually really good, but are also tough beasts to tame because they weren't really made for indie devs.
That's because it's not a game engine. What we're talking about here is a collection of open source libraries (and one closed source) and tools that they use to make their games. It's not a single game engine like Unity or Godot.
CastleDB crashes consistently right away just inserting an image (jpg) (i also find it strange that it tries to be a mapeditor too), the documentation is really sparse and workflows are obscure and things are still experimental as it seems. When i opened this video i was enthusiastic about finally trying it out, but it doesn't feel that much different from year/s ago when i checked it out after it was mentioned here, unfortunately
yeah... the haxe world just suffers from not feeling like a single product, but really though, isn’t all software made used other libraries ‘n code? i think they’re just too explicit about it. nme, openfl, lime, kha, is def all confusing.
Thank you for the video it's really nice because I really like to have language. Oh yes you would be really good if you could do more tutorial. So very much look forward to it.
4 года назад
Can you give new haxe resources? Why is it not used very much?
Almost three years later and the problem persists: there is very little documentation, what does exist is shit, and if you have any questions about the heaps engine, you have to ask a quirky neck beard in the haxe discord who hates answering questions.
@@akshayazariah Thanks for the info. I hope someone makes a free exporter for everyone, or maybe make a tutorial /book/video for everyone to follow. Oh, well... just dreaming.
@@JL-mw4mo That would be great, but everything about console development (apart from the Xbox ONE) is protected. I'm no expert on this subject, but I assume that you can get help in writing an exporter, once you become a licensed developer for that platform that you have in mind. Again, it would be best to ask someone who's knowledgeable about this, so why not contact Lone Wolf Technology and ask about it? Also, Ariel Manzur, one of the Godot developers, talked about porting his game to the Switch over here - bit.ly/3bV4Ojw .
@@JL-mw4mo By law, the code required to export to consoles cannot be made open-source, as the console makers want that information proprietary, and I believe (correct me if I'm wrong) require you to sign an NDA to be able to see the information needed to port your games.
So, Godot and this one are pretty prominsing, right? Mind talking about how you see each one? I'd like to settle with one or the other (or maybe both if each one has their own niche). I've tried heaps in the past but it was pretty difficult to set up honestly (that was before HashLink went on stable, though)
Hey I'm making own indie game and hope that it will be displayed on this channel cause i love this channel so much plz send back if you wanna get a free play when it's finished
They're so similar that the only difference I can think off the top of my head is that TIC has a wider screen (which I personally prefer). I don't know if pico8 imposes a token limit in your code, but TIC does. Which means games with too much code can get a bit cryptic due to abbreviations. github.com/Skaruts/Game-of-Life-in-Multiple-Languages/blob/master/Lua/TIC-80/tiny_life.lua That's because TIC limits your code to ~64k characters (though the paid version gives you 64x8). Limiting by words would be better, imo. If by any chance pico8 does that, then I'd recommend it over TIC. Pico8 has no free version, though, as far as I'm aware. I'm personally not too fond of such a small screen though. I've been meaning to try PixelVision8 because of that.
@@skaruts that's interesting that's way i asked such video idea. I'm new to.these cool stuff, i am curious cuse im just seeing really great.games made with pico 8 than others consoles like the retro super hot demake.
@@damonsalvatore792 Yea I think pico8 is the most popular. It has a lot of cool games indeed. I was impressed by this one in particular: hackernoon.com/pico-8-lighting-part-1-thin-dark-line-8ea15d21fed7#.q578i21cc Meanwhile I think a good place to get information about any fantasy consoles is the Fantasy Consoles discord: discord.gg/aD8Z52 . There you can get an overview of most (maybe all) existing consoles and how they work, or even ask for info.
"Why the hell aren't more people using it?" is wrong question. "Why would anyone use this engine over UE/Unity?" is the right one. And are there any reasons outside of empty things like "free and open source". I don't see any.
That's not really a great question at all. If there weren't any reasons then 100% of the world would be using UE and Unity. There's plenty of reasons. Preferring a different workflow, a different licence, a different business model, wanting actual 2D rendering, hating C# or C++, having a language preference, not needing a bloated engine, etc. Two major advantages in FOSS is that you have the liberty to tweak the engine itself, and you can also look into how the engine does things under the hood, such that you can better optimize your games. It's not a matter of vanity or elitism. It's actually a pretty big deal. I'm developing text-based roguelikes and I'll tell you, Unity and UE are pretty shitty for this kind of thing. And so is Godot, for that matter, at least for now.
@@foxto100 That doesn't mean anything. 90 times out of 100 the only difference is that you must install additional tools for use with frameworks and jump through additional hoops to make it work. Other than that it's almost always same process of developing subengine inside the framework/engine.
@@skaruts A bit idealistic, but understandable. For all intents and purposes we can say that 100% of the world are using ue/unity now, other engines are getting tinier and tinier, their development is getting less and less financially viable. Reality is simple, for commercial dev, you don't really have much choice, using anything other than most developed and supported engine for your need is like shooting yourself in the foot and risking everything. Your example is a bit weird, yes ascii roguelikes are a tiny bit special when it comes to render dev, but i don't see why unity can't be used for that. After initial setup it will not be harder than whatever engine/framework you are using. Only objective reason for not using unity/ue is if you have incredibly advanced programming skills, big team of professionals, enough time and budget. Or if it's just a hobby.
@@DarkDao Idealistic? Which reason I enumerated is idealistic? *_"For all intents and purposes we can say that 100% of the world are using ue/unity now, other engines are getting tinier and tinier"_* Actually, Godot is getting bigger, not tinier. It's not even tiny at this point. Some other engines have also been growing, not shrinking. That inherently means Unity and UE are having some market share taken away. *_"Reality is simple, for commercial dev, you don't really have much choice"_* Yes you do. There are a ton of games that aren't made in Unity or UE. *_"Your example is a bit weird, yes ascii roguelikes are a tiny bit special when it comes to render dev, but i don't see why unity can't be used for that. After initial setup it will not be harder than whatever engine/framework you are using."_* 99% of text-based roguelikes aren't made in Unity or UE. If any at all. I know one made in Godot (Door In The Woods), but it's a rather simplistic game. You actually can make such a RL in Unity, that's not in question. The problem is 1) you don't have proper 2D rendering to emulate a terminal, and 2) unless you're making something rather simplistic -- with simple AI, with no lighting, no thousands of entities to update, etc, then you're gonna have trouble with performance. RLs are performance hogs. Most text-based RLs are made in their own engines or with frameworks tailored for RLs with zero overhead (like libtcod), for that very reason. Unity is waaaaay too bloated for this, and so is UE. You wouldn't need 90% of it's tools, and the wasteful engine overhead would take a toll. *_"Only objective reason for not using unity/ue"_* Stop right there. For the most part people don't choose engines for objective reasons. The tool of choice is a very subjective thing, and that's what you're failing to realize. Still, there are plenty of objective reasons. There's not one universal tool. There's only the right tool for a given job. The very reason you have so many engines on the market is because many people needed their own right tool for the job and developed them, and then made them available for others. The very reason why you have frameworks like SDL, SFML, RayLib, etc, on the market is because many people need to go down to the metal in order to make certain games. Unity and UE are neither the right tools for every job nor the anatomically perfect tools for every hand.
@@SavageArms357 The combination of those... When you make a game engine and use a language that nobody really knows or uses, People would NOT be so interested in this game engine. When someone makes a game engine, this someone should make it in a popular languge that everybody knows or uses. That person can include his or her own language too. Blender 3d is available both in C++ and Python.
I have to say, it's not as easy to make, say, a Flappy Bird clone, but I love it! It works great, and plus, it's always fun making your own level editors.
Setup sounds too convoluted. While it sounds interesting, it differently doesn't seem like an "E Rated" engine. Stack? Maybe if there was a one stop shop download/install. Biggest reason I push Godot is its size and simple setup. Which for those that don't know. it's < 300mb and there is no setup. You just run an .exe. you can run it on a thumb drive from 2004.
@@5minutemovies977those games are not even close to a successful project. I have checked them all, the most successful is mobile game with 10k downloads, which costs 1 euro.
"For the most part it is entirely open source!" uh "It is 100% absolutely kind of definitely maybe sort of for sure guaranteed to be almost completely true."
![Megan Thee Stallion - Mamushi (Remix) [feat. TWICE] [Official Audio]](http://i.ytimg.com/vi/5c0_u9PsrsQ/mqdefault.jpg)
Link:
www.gamefromscratch.com/post/2020/04/07/Shiros-Game-Technology-Stack.aspx
I've never heard of this one before. I feel I should've known about this one.
I want your opinion on the relatively old Reality Factory.
Can you please make a tutorial series on the Heaps IDE, because it's very hard for a beginner to understand what's going on, yet it has a lot of potential for game development and I'd love to learn how to use it.
They should really bundle this into something like "Shiro Game Engine" like you mentioned. I find having everything spread out like this makes it a little tedious.
@@DayInDaLife Heaps doesn't have an IDE included, you can use Hide with it though. They should include a "Bundle" where you can download everything into one neat installer. So you can get the castle DB included with it as well. Would be a good marketing strategy to get more people to pick this stuff up.
@@DayInDaLife Heaps only does the game bits. The db, the virtual machine, the editor,haxe are not included
That database tool is indeed something in many game companies long for! In fact, I am going to pitch it to the company I work for now. Thank you for this video. What an amazing set of tools. I look forward to diving in each of them!
i don't understand. Why is this need??
Yes, I would love a new tutorial series on the different tools and how to use the together!
This all seems overly complicated, although game design is always complicated I guess. It doesn't seem that intuitive for a new user to try and get this different parts working together well.
@Chipmunk Hm, didn't think of that. In that case, maybe I'll like it once I'm more experienced.
Honestly, this has me kind of interested. I'd like to see a full tutorial series for this engine.
The games using parts of this are seriously impressive. I'm interested if only for that, dead cell is fantastic.
*Nice to see you again.* _Now I am regular on this platform._
Welcome
@@daveann3426 Thank you
I agree that haxe and heaps is seriously awesome, but when I looked at it I found it hard to get into as far as documentation and tutorials go. I know there is good documentation but I still found it difficult to follow
Haxe and Heaps+HL are great. But amount of troubles with their config is intimidating... There are always some versions incompatibility of one or another. Some libs work with HL, others don't... And Heaps learning style is like "here are examples, source code, good luck"
@Francisco Teixeira really? Would you please, please please write a guide on how to build for Android? Pretty please :)
I built a small game in Unity like 7 years ago. Ever since, any time I want to test out a new game engine I rebuild this game in it. For what it is worth, with no haxe experience, I built the game in heaps in less than a day. It was very nice to work with really. For comparison the closest to that was when I rebuilt it with Phaser a few years ago in about 2-3 days. I still haven't gotten myself to finish it in Godot, but have about 2 days of time sunk in there so far. The original Unity version took about a week. Cocos w/ Python took about 3 days. I started it with raylib as well, but haven't spent much time there. I get distracted easily... haha.
Haxe overall has peaked my interest though. I really wish the ability to include native libraries in projects (i.e. a/lib/dll/dylib/etc.. c or c++) would be worked out a little more though. Ammer is a cool project attempting this, but I think if they could get that in a good position more and more people would start building things with it. There are just so many nice c/c++ based libraries that I think haxe developers would benefit greatly from being able to include easily. (I know there are some already with bindings and its possible to make bindings, but its not the greatest experience at the moment.)
Cool video as always!
@Francisco Teixeira Thanks for the tip. I'll check that out.
As someone who's still deciding on which engine to settle on, what sort of roadblocks were you experiencing with Godot?
@@SavageArms357 I don't know if I'd call them roadblocks so much, but the extra time so far has mostly been spent learning the UI and gdscript capabilities. Even though engines with GUI editors are all the rage and definitely help non-programmers get started, I've always found it easier to just throw a library in a project and start coding. Since learning the tools involved in others requires more time. Though it is probably just me and an up front impression. Having those tools down the line and with a more complicated game is probably great once you learn it all.
With all that said, I'm working with Armory3d a little now. And of all the game engines I've tested out, even though I've only finished the game with one, the top 3 as of now are Armory3d, Godot, or Heaps.
@@Pewsplosions Ah, good to know there wasn't any major problems with it. I'll take a look into Armory 3D as well, cheers for the info! :D
Haxe does look nice and I have tried it a couple of years ago, but for me, the issue that I have is that prefer to work with a game engine that has an editor. And I didn't know that haxe had hide. Hopefully if they do what you mentioned at 12:15 with packaging all of the components together would definitely give haxe another try.
Heaps is a nice engine but is poorly documented which kinda discouraged me. Haxeflixel has great documentation, the Vs code plugin for Haxe gives rich autocompletion and hover info. It felt very much welcoming for someone who is a noobie in programming. Highly recommended 👍
This is sort of a "Catch-22" reason for not using it though. The reason it is not well documented is because not a ton of people use it, and the reason not a ton of people use it is because it is not well documented. I think it is better to give things a try a lot of the times and maybe figure something out and share it with the community or something even if it is something easy. And, if you don't like it after that, move on
@@HonsHon I'm a noobie at game programming and also programming in general. And good documentation will give a boost to interest me. It's not like I don't like it, it's in my list.
I’ve tried lots of game engines and Haxe w/ heaps seems to be where it’s at for me. It’s free no strings attached, runs on Linux and it has a butt load of features. Haxe and heaps seems to be lacking community because several reasons such as flash exiting the arena, it does not have a heavy editor attached and when you google game engines you get results mostly for unity or unreal.
What's keeping you away from Godot?
@@rytif That's circular logic.
@@rytif But your comment wasn't in a vacuum was it? There was context, and there are more things to which "that" can refer, such as in this case, your reasoning.
You are very stupid, aren't you?
Been using Haxe forever, it's easily my favorite language/toolkit out there, for all kinds of uses. I'm still using the HXCPP target for games projects mostly because of documentation issues around the HEAPS tech stack. There's obviously amazing stuff in here but the docs-ghost keeps haunting Haxe unfortunately. Really, really unfortunately, given the quality of the actual thing.
Hi andreas, im another long time user of Haxe from back in the day but abandoned it a number of years ago because of issues with docs. It really is a thoron in the side fo haxe
@@MichaelCannUk You should give it a new look. Modern Haxe is an absolute beast, and the VSCode extension is great now too.
More engines need to use CSS and HTML like structure for UI. Will check this out in the future
Oh my, this engine is so cool! I'm learning Heaps and it is SO FUN. I love it :D
Haxe would be so much more popular if there was more learning material and tutorials/docs.....
Srsly tho, we need more tutorials on this
Man when I first saw Shiro I thought... wait... that Shiro?
And I soon found out that the madlads involved in Evoland2 also developed an entire new language along with a game engine.
Absolute units..
The biggest limitation is there is no built-in physics component. There are some dubiously supported implementations of Nape, Box2D and Bullet Physics.
Thank you for introducing us to this stack.
I tried heaps recently and here is my two cents on this "engine" stack:
This stack is not cross compatible with any console/PC out there. What you get is generated C/C++ code that you have to then deploy on each system alone. This is not an easy job and you have to have a Dev Console for each one of them in order to even begin porting process, so its quite expensive too. Unity/UE4/Godot/LibGDX do this for you so that you don't need dev consoles.
** heaps.io:
Heaps is not a game engine, it is a graphics engine, it says so on the website. This means that many capabilities of a game engine are not available:
1. NO VIEWPORTS
2. No colliders
3. No events
4. No sound engine
5. No physics engine
6. No helper classes that allow you to calculate things like path finding, noise generators etc.
7. Many more things that make a game engine
What you can do:
1. Display graphics
2. Particle system
3. Font display
4. Displaying/"create" animations
5. Compatibility with some obscure systems (CastleDB, HScript etc.)
** HashLink
HashLink is a virtual machine that works just like any other virtual machine. It runs only on Windows/Linux and has the ability to generate C/C++ code (many VMs can do this). There is no HashLink port for Android/IOS/PS4/XBONE/Switch which means you have to take responsibility and port C/C++ code manually to those platforms.
** Domkit is just a UI library
** HScript is an interpreter for Haxe language
** MpMan is not available as it is an internal library
** HxBit is a binary protocol (there are many of them out there)
** Haxe language is nothing special, it uses C style syntax and reminds me a lot of Java/C#. It can cross compile to other languages but this feature is not available to the heaps.io
Worst part about all of this is lack of documentation that makes it extremely hard to do event the simplest of tasks. IntelliSense plugins are available for Idea and VSCode but they are bare bones and barely work.
This is a inhouse engine made by Motion Twin / Shiro Games and it makes sense for them to use it as they made it and know how every gear turns but with lack of documentation and a very elaborate deployment setup it is better to just stick to an established engine. You will waste a lot of time trying to understand how the engine works instead of making the game.
Many parts of this are not correct:
- you can output a .hl file instead of C which cross platform bytecode (Window/OSX/Linux). For consoles yes you need C+SDK
- heaps.io has Colliders (in 3D), Events (see Interactive in both 2D/3D), Sound engine (hxd.snd)
- HashLink runtime runs on Android/IOS/PS4/XBONE/Switch, so you don't to part any C/C++ code manually, you still have to compile it though
- Haxe cross compilation is used for Heaps.io for HashLink / Javascript cross compilation. You can have your games as native (desktop) and HTML5/WebGL
- Plugins for VSCode is working very well. Please raise any issue you might have. This one at leas is not "bare bone and barely work", it's used for daily dev by many Haxe developers
I appreciate you took the time to look into it, but please correct the errors as your report is misleading.
@@NicolasCannasse mad respect to you for creating Haxe. I don't doubt heap's capabilities at all but I've found solace in Haxeflixel. As a noobie programmer, I found haxe quite comfortable to pick up.
@@NicolasCannasse I mostly use IDEA and the plugin was very buggy
Every engine out there can export to PC/MAC/Mobile, Heaps.io promises PS4/XBONE/Switch out of the box, which isn't true.
It would be amazing if there was a video of setting up the whole package to work as a unit.
this could be a whole bother series, lol... get the engine working!
It seems weird that castledb is a map editor but ok. Does this support android and iOS? It does look pretty cool though.
Interesting
Never heard of it
This is both very cool, and an inevitable maintenance nightmare at some point.
if you want a reason to use this engine, evolant 2, just play it from start to end, and you will know why, in my opinion, yes, i wish there was a just one click download
Interesting. I'll definitely keep an eye on this if I get serious on making games cause I just got into Godot.
TLDR; A tech stack isn't really a game engine.
About adoption, I think the concept of a tech stack is confusing to some people in and of itself. I prefer the Shiro approach myself because it lets me cherrypick what I care about. For instance I'm curious about the workflow improvements around HScript (which I already use), CastleDB and the performance of HL + its C output. However, HIDE is on first encounter an opaque hell-machine, DOM and CSS is the last place I want to do any UI ever, and I've already got my own stack that solves those problems on top of the existing CPP target. Shiro isn't presenting this as "a game engine". It's just their stack of tools. The makeup of each project built with this stack is going to have a ton of variance. There's huge value here for pretty much any game developer willing to configure and build their own stack, but if you want a spin it up and go kind of work environment for prototyping etc, this is really not going to do that for you.
thanks for this! this makes sense...
The problem with the whole haxe ecosystem is that, other than for haxe itself, the documentation is either poor, incomplete, or missing.
I feel like it's probably not as popular as Godot cos it's a stack, which probably scares off indies who are used to the self-contained IDE/engine combo apps like Unity
A tutorial series about Heaps would be great
Yes, make tutorial on the stack.
Can you please make another tutorial, focussing on this technology stack and on using the many build targets (Windows, Ios, Android, Switch, Ps4, Xbox, etc)? I tried for about one month to do it myself, and even got templates and tutorials, but nothing works, or at least not for my usecase. I was trying to make an small endless runner for my first game with Haxe, but there is just no documentation for it.
Do you think HashLink could be useful for gameserver hosting of various games?
Hm, it IS weird not many people talk about it, if great games like Dead Cells used it. I'd imagine people would look up the engine used.
I wonder if it's a marketing problem. Unreal and Unity are well known because, well. They proved themselves, and has a great marketing along with them.
Never heard of Hide but it sounds interesting. I'll be sure to check it out.
I would e interested in seeing an example of how all of the tools would link together to make an isometric game
yooo, that paper, rock, scissors code was sick tho. what was that lol!! a switch on a 2d array?
The Evoland bois haxe-oring their way to success
I get if you wanna write your own game engine for your own game, but I just don’t understand why there are so many “general purpose” game engines trying to get a slice of the industry.
Because there are many ways to make an omelet. If you compare Godot, Unity, UE, Defold, GDevelop, Ogre, Torque, CopperCube, etc, you'll find fundamental differences in the workflow, in the programming languages used, in the performance, the memory footprint, etc, and that will satisfy different people. The market is still evolving to better and easier ways to make games, so the more the merrier. They end up pushing each other to improve more and more.
The only thing I wish existed more is specialized engines, specific to certain game genres. A problem with generalized engines is the wasteful overhead required to deal with a multitude of different game requirements. Specialized engines are more limited, but less wasteful. RPGMaker is one example of a specialized engine. But there aren't that many in the indie market. For FPS games, for example, you only have engines like Doom, Source, etc, which are actually really good, but are also tough beasts to tame because they weren't really made for indie devs.
Yes, I also get the feeling that haxe is way over underrated, not sure why...
Hmmm i didn't know about this game engine .
That's because it's not a game engine. What we're talking about here is a collection of open source libraries (and one closed source) and tools that they use to make their games. It's not a single game engine like Unity or Godot.
I'd be interested to see more tutorials on it!
Yeah for the French dudes behind Haxe !
CastleDB crashes consistently right away just inserting an image (jpg) (i also find it strange that it tries to be a mapeditor too), the documentation is really sparse and workflows are obscure and things are still experimental as it seems. When i opened this video i was enthusiastic about finally trying it out, but it doesn't feel that much different from year/s ago when i checked it out after it was mentioned here, unfortunately
Odd, I never experienced an issue in my brief time with it. What platform are you on?
yeah... the haxe world just suffers from not feeling like a single product, but really though, isn’t all software made used other libraries ‘n code? i think they’re just too explicit about it. nme, openfl, lime, kha, is def all confusing.
Wow, how i dont know for this, this look very good!
I would to see a tutorial combining parts of this stack with Armory3D!
What are you guys using for Haxe just visual studio code?
Thank you for the video it's really nice because I really like to have language. Oh yes you would be really good if you could do more tutorial. So very much look forward to it.
Can you give new haxe resources? Why is it not used very much?
Almost three years later and the problem persists: there is very little documentation, what does exist is shit, and if you have any questions about the heaps engine, you have to ask a quirky neck beard in the haxe discord who hates answering questions.
More successful titles are shipped with tech stack than Godot! Interesting!
I wonder how this engine can port your game to gaming consoles while Godot couldn't.
Godot can; you need to either contact a developer ( www.lonewolftechnology.com/ ) or write the exporter yourself.
@@akshayazariah Thanks for the info. I hope someone makes a free exporter for everyone, or maybe make a tutorial /book/video for everyone to follow. Oh, well... just dreaming.
@@JL-mw4mo That would be great, but everything about console development (apart from the Xbox ONE) is protected. I'm no expert on this subject, but I assume that you can get help in writing an exporter, once you become a licensed developer for that platform that you have in mind. Again, it would be best to ask someone who's knowledgeable about this, so why not contact Lone Wolf Technology and ask about it? Also, Ariel Manzur, one of the Godot developers, talked about porting his game to the Switch over here - bit.ly/3bV4Ojw .
@@JL-mw4mo By law, the code required to export to consoles cannot be made open-source, as the console makers want that information proprietary, and I believe (correct me if I'm wrong) require you to sign an NDA to be able to see the information needed to port your games.
@@SavageArms357 Too bad for Godot, but his makes a great case for using Shiro's Game Tech.
Thank you.
So, Godot and this one are pretty prominsing, right? Mind talking about how you see each one? I'd like to settle with one or the other (or maybe both if each one has their own niche). I've tried heaps in the past but it was pretty difficult to set up honestly (that was before HashLink went on stable, though)
Hey I'm making own indie game and hope that it will be displayed on this channel cause i love this channel so much plz send back if you wanna get a free play when it's finished
did u say u can compile haxe to html? xd how would that work lol
Hey, sorry if I mocked you, if you're listing game engines, you're strongly a good a guy
Could you make a video review comparisons pico 8 vs tic.computer?
They're so similar that the only difference I can think off the top of my head is that TIC has a wider screen (which I personally prefer). I don't know if pico8 imposes a token limit in your code, but TIC does. Which means games with too much code can get a bit cryptic due to abbreviations.
github.com/Skaruts/Game-of-Life-in-Multiple-Languages/blob/master/Lua/TIC-80/tiny_life.lua
That's because TIC limits your code to ~64k characters (though the paid version gives you 64x8). Limiting by words would be better, imo. If by any chance pico8 does that, then I'd recommend it over TIC. Pico8 has no free version, though, as far as I'm aware.
I'm personally not too fond of such a small screen though. I've been meaning to try PixelVision8 because of that.
@@skaruts that's interesting that's way i asked such video idea.
I'm new to.these cool stuff, i am curious cuse im just seeing really great.games made with pico 8 than others consoles like the retro super hot demake.
@@damonsalvatore792 Yea I think pico8 is the most popular. It has a lot of cool games indeed. I was impressed by this one in particular: hackernoon.com/pico-8-lighting-part-1-thin-dark-line-8ea15d21fed7#.q578i21cc
Meanwhile I think a good place to get information about any fantasy consoles is the Fantasy Consoles discord: discord.gg/aD8Z52 .
There you can get an overview of most (maybe all) existing consoles and how they work, or even ask for info.
@@skaruts Pico has a cartridge size limit as well as a Lua token count limit.
Well I can't run heaps hide editor.
"Why the hell aren't more people using it?" is wrong question.
"Why would anyone use this engine over UE/Unity?" is the right one.
And are there any reasons outside of empty things like "free and open source". I don't see any.
That's not really a great question at all. If there weren't any reasons then 100% of the world would be using UE and Unity. There's plenty of reasons. Preferring a different workflow, a different licence, a different business model, wanting actual 2D rendering, hating C# or C++, having a language preference, not needing a bloated engine, etc. Two major advantages in FOSS is that you have the liberty to tweak the engine itself, and you can also look into how the engine does things under the hood, such that you can better optimize your games. It's not a matter of vanity or elitism. It's actually a pretty big deal.
I'm developing text-based roguelikes and I'll tell you, Unity and UE are pretty shitty for this kind of thing. And so is Godot, for that matter, at least for now.
"Why would anyone use this engine over UE/Unity?" because its not an engine its a framework
@@foxto100 That doesn't mean anything. 90 times out of 100 the only difference is that you must install additional tools for use with frameworks and jump through additional hoops to make it work. Other than that it's almost always same process of developing subengine inside the framework/engine.
@@skaruts A bit idealistic, but understandable. For all intents and purposes we can say that 100% of the world are using ue/unity now, other engines are getting tinier and tinier, their development is getting less and less financially viable. Reality is simple, for commercial dev, you don't really have much choice, using anything other than most developed and supported engine for your need is like shooting yourself in the foot and risking everything. Your example is a bit weird, yes ascii roguelikes are a tiny bit special when it comes to render dev, but i don't see why unity can't be used for that. After initial setup it will not be harder than whatever engine/framework you are using.
Only objective reason for not using unity/ue is if you have incredibly advanced programming skills, big team of professionals, enough time and budget. Or if it's just a hobby.
@@DarkDao Idealistic? Which reason I enumerated is idealistic?
*_"For all intents and purposes we can say that 100% of the world are using ue/unity now, other engines are getting tinier and tinier"_*
Actually, Godot is getting bigger, not tinier. It's not even tiny at this point. Some other engines have also been growing, not shrinking. That inherently means Unity and UE are having some market share taken away.
*_"Reality is simple, for commercial dev, you don't really have much choice"_*
Yes you do. There are a ton of games that aren't made in Unity or UE.
*_"Your example is a bit weird, yes ascii roguelikes are a tiny bit special when it comes to render dev, but i don't see why unity can't be used for that. After initial setup it will not be harder than whatever engine/framework you are using."_*
99% of text-based roguelikes aren't made in Unity or UE. If any at all. I know one made in Godot (Door In The Woods), but it's a rather simplistic game.
You actually can make such a RL in Unity, that's not in question. The problem is 1) you don't have proper 2D rendering to emulate a terminal, and 2) unless you're making something rather simplistic -- with simple AI, with no lighting, no thousands of entities to update, etc, then you're gonna have trouble with performance.
RLs are performance hogs. Most text-based RLs are made in their own engines or with frameworks tailored for RLs with zero overhead (like libtcod), for that very reason. Unity is waaaaay too bloated for this, and so is UE. You wouldn't need 90% of it's tools, and the wasteful engine overhead would take a toll.
*_"Only objective reason for not using unity/ue"_*
Stop right there. For the most part people don't choose engines for objective reasons. The tool of choice is a very subjective thing, and that's what you're failing to realize.
Still, there are plenty of objective reasons. There's not one universal tool. There's only the right tool for a given job. The very reason you have so many engines on the market is because many people needed their own right tool for the job and developed them, and then made them available for others. The very reason why you have frameworks like SDL, SFML, RayLib, etc, on the market is because many people need to go down to the metal in order to make certain games. Unity and UE are neither the right tools for every job nor the anatomically perfect tools for every hand.
I think I know why people don't really use it...
The language? Or the lack of a singular package?
@@SavageArms357 The combination of those...
When you make a game engine and use a language that nobody really knows or uses,
People would NOT be so interested in this game engine.
When someone makes a game engine, this someone should make it in a popular languge that everybody knows or uses. That person can include his or her own language too.
Blender 3d is available both in C++ and Python.
good case
This hasn't impressed me enough to change from unity.
I have to say, it's not as easy to make, say, a Flappy Bird clone, but I love it! It works great, and plus, it's always fun making your own level editors.
Can we, like, stop it with the meme languages?
C and family work just fine
What qualities make a language a meme language for you?
I don't understand this 1 game engines sepert game engines ane technology?
And this free to use?
Setup sounds too convoluted. While it sounds interesting, it differently doesn't seem like an "E Rated" engine. Stack? Maybe if there was a one stop shop download/install. Biggest reason I push Godot is its size and simple setup. Which for those that don't know. it's < 300mb and there is no setup. You just run an .exe. you can run it on a thumb drive from 2004.
Found instructions on how to compile HIDE:
github.com/calderarchinuk/hide
Can someone get Tom a life?
Bot being botty, just finally hid from channel. Not sure why there is such an uptick in comment bots lately.
@@gamefromscratch Yea, I've been noticing quite a lot of them. At first I thought they were just spammers...
0:20 emm... more than zero?
What are you talking about ?
ruclips.net/video/JF60CfSLjew/видео.html
@@5minutemovies977those games are not even close to a successful project. I have checked them all, the most successful is mobile game with 10k downloads, which costs 1 euro.
"For the most part it is entirely open source!"
uh
"It is 100% absolutely kind of definitely maybe sort of for sure guaranteed to be almost completely true."
looks bad