Why are you guys so worried about these engines? I mean...Tbh the hardest part is actually getting/making the art assets o.o
Obv just art assets because its friggin fun to make sound effects...*whacks a watermelon* @_@
''/\/\'' Posted using Iphone bunni ( o.o) (")(") **This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
Originally posted by JC-Smith Having worked with Hero Engine for about a year and a half now, there is absolutely nothing wrong with the thing. It's renderer isn't on par with the likes of Unreal Tech, but the editors, collaberative editing, networking is all just fine.
you have to release your game to the public before you make this claim. no one expects a developer to criticise their own work publicly.
in terms of released games, hero engine doesnt really have a great history. no doubt it is easy to use, allows multiple people to work on the same thing at the same time in real time and has plenty of built in assets and networking but personally i would go about making an mmorpg using it. heck i wouldnt go about making an mmorpg as an indie unless it was something very simple (eg. 2d, turn based, text based or just plain simple)
So you think CCP were in the wrong to go about making EVE: Online and now there newest MMO in development? You think the creators of Dark Fall were in the wrong? You think it's wrong for Us to be making Greed Monger? You think it's wrong for JC-Smith and his team to be making The Repopulation (I've personally worked with them for a week or 2, GREAT group of guys!)? There are plenty of Indies making MMOs... Yes just as many if not more fail but oh well They learned Lessions which they can go on and use on future games!
Modern Graphics are not an issue for most gamers, anything on Par with WoW graphics, or Elderscrolls Online graphics or SWTOR graphics is more than sufficient for an MMO.
Moderns, tesselations and the such, are good for Single Player games. I do not want to be upgrading Computer every time a new MMO comes out.
And you would be making the same mistake as all of the MMO's who advertised "next generation" MMO did, limitiong your player base, creating frustration for people, and bad word of mouth from the onset of launch.
MMO's do not need latest graphics to be great. If you bargaining card is your latst Graphics then I know intuitivelly as a player based on previous experiences that you have no substance or good gameplay in your game. Or at least I know that i will not be having fun because I will be fighting with the performance issues instead of enjoying the gameplay, which comes down to not having a good gameplay anyways from a player's perspective.
What I want in MMO is a good representation of things not a realistic representation of things as to strike a balance between Gameplay and Performance without me having to upgrade to the latest OS (I hear Microsoft wants to offer DirectX 11.1 only on Windows 8) or Gear...forget it, too much hassle for most players. And I am not about to turn my PC in ot a Smartphone either with Windows 8...no way.
Just jumping in to give a perspective.
It's not about producing AAA quality graphics. That's unnecessary for a low-budget indie game. It's probably also impractical, as a low budget doesn't let you produce the large quantities of high quality artwork that an AAA game demands. Rather:
1) It's about making developers' lives easier by letting them use the right tool for the job. If a building project calls for the use of both a hammer and a screwdriver, you may be able to get by with just one of the two. In a pinch, a sturdy screwdriver can be used as a hammer, or the "claw" part of a claw hammer can be used as a screwdriver for larger screws. But it will sure make your life easier if you have both the hammer and the screwdriver available.
2) It's about optimizing performance by offloading the work that video cards are good at onto the video card rather than having to do it on the processor instead. Any image that you can compute on a video card, you could just as well compute on a processor (up to minor rounding errors) without using any sort of 3D API. The drawback of this is an enormous performance hit. Rather than frames per second, you'd likely be looking at seconds per frame. Depending on what you want to do, in some cases, it could even be minutes per frame. For a game, that's a non-starter.
The entire point of having a video card is that it's optimized to be very, very good at some things that x86 (and ARM) processors are terrible at. x86 processors are optimized for doing a large number of operations sequentially with low latency, and to still perform well with a lot of branching. Video cards completely choke on that sort of workload. Rather, video cards are very good at situations that are extremely parallel and ask for you to perform exactly the same instructions in exactly the same order on many different pieces of data at the same time.
For the sorts of data that video cards are good at, you may realistically see improvement of two orders of magnitude (i.e., speed up by a factor of 100) by doing the computations on the GPU rather than the CPU. Even if you're pairing an ultra-high-end $1000+ processor with a $60 video card, the video card may still commonly be an order of magnitude faster.
But that's only if you you're able to have the video card process the data. Never mind tessellation for a moment; that requires understanding of some advanced mathematics, and even most AAA developers don't have the necessary expertise in-house.
Instead, let's talk about geometry shaders. There are two very important things that geometry shaders do: they let you see an entire primitive (typically a triangle) all at once, rather than only one vertex at a time. And they can emit new primitives, rather than requiring all vertices, edges, and triangles to be created by the processor and fed to the video card.
The latter lets you have one triangle come in, and then pass it through. Or one triangle comes in, and then two triangles go out. Or zero. Or 17. Or any other number up to the cap (which I think is 64 in OpenGL 4.2). And the shader can be aware of all of the triangles that you're going to output at once, rather than having to process them one at a time.
So what would you use that for? For one, it lets you mostly handle particle effects on the GPU, with the CPU only handing the video card some token data to get started. Or if you ever want to draw the same scene from more than one perspective (e.g., for stereoscopic 3D, Eyefinity without assuming that the monitor screens all lie in the same plane in real life, or a split-screen multiplayer game), that lets you save a ton of work.
More generally, geometry shaders are like mesmers in Guild Wars: they seem useless at first, but they end up providing a very elegant solution to all sorts of weird problems. And yet, if they had never existed, people wouldn't realize that they were missing and ought to be there. Developers who are still using DirectX 9.0c likely don't realize that they're making things much harder on themselves than necessary.
And then, yes, the newer APIs can produce better quality graphics. But that's largely because if you can do the same thing as before while putting a lighter load on your hardware, then that frees up resources to do more than you could before.
-----
So let's talk about hardware compatibility. Tessellation requires DirectX 11 or later, or OpenGL 4.0 or later. That means video cards from roughly the last three years. But even if your game is designed around tessellation, it's not that hard to port it back to DirectX 10 or OpenGL 3.2. (Well, I only know that for OpenGL, but I'd assume it's the same for DirectX.) You can tessellate your data once on the processor before uploading it to the video card. Doing tessellation properly means that you need so little vertex data that even if tessellating it before uploading it increases the size of the vertex data buffers by a factor of 100, that's still basically a rounding error. Now maybe you need hundreds of KB instead of only several KB of video memory.
If you're using geometry shaders much, then porting that back to DirectX 9.0c is probably impractical, or at least more trouble than it's worth. But that means your game requires a video card from roughly sometime in the last 6 years. If you're starting a project today and think you'll take 3 years to finish it, then by the time it launches, it will require a video card from sometime in the last 9 years. Is that really such an onerous hardware requirement?
Even if you could port it back further to make it playable on older hardware, would the older hardware have the level of performance necessary to run your game? People who would have bought a GeForce 7900 GTX or Radeon HD 1900 XT tend not to be the sort of people who would still be using the same video card a decade later. Rather, the older hardware that might still be in use would be low end stuff like a GeForce 6150 SE that can't run most modern DirectX 9.0c games at playable frame rates, either.
As for operating systems, DirectX 10 and 11 are both compatible with Windows Vista, 7, and 8, and presumably future versions, too. You mention DirectX 11.1, but using 11.1 is just stupid for anything beyond a tech demo. I'm not advocating that for real games. DirectX 10 does mean you lose compatibility with Windows XP, but if that bothers you, you can use OpenGL 3.2 instead and get compatibility with Windows XP, Vista, 7, Linux, Mac OS X, and likely soon Google Android.
Maybe being incompatible with Windows XP is a problem today. Think it will still be a problem three years from now, when you hope to finish your game if you start today?
So where is your robust 3D MMO engine? I'd like to take a look at it since you fancy yourself an expert.
Would my game engine be better than the Hero Engine for making Greed Monger, The Repopulation, or any other game besides my own? Nope. Not even close. They probably wouldn't be able to do much with my game engine. Artists would be completely stuck, as mine produces artwork by typing out source code without any sort of graphical interface. Even with full access to the source code, comments that assume that the reader understands, say, homology groups make perfect sense to me but would be completely baffling to anyone who doesn't. Furthermore, some capabilities that most games need are completely missing from my game engine, and not going to be added. For example, loading textures from a hard drive.
But you know what my game engine is better than Hero Engine for? My game. Because that's exactly what it's designed for. I want to do a bunch of stuff that Hero Engine won't allow, but my game engine does. This isn't peculiar to me, either; an awful lot of game engines that are designed for some particular game are the best engine for that particular game, or at least, far better than any off-the-shelf game engine that doesn't give you full access to modify the source code.
Why are you guys so worried about these engines? I mean...Tbh the hardest part is actually getting/making the art assets o.o
Obv just art assets because its friggin fun to make sound effects...*whacks a watermelon* @_@
There's a lot of truth to that, though it depends on what you're doing. If you're trying to create a purely text-based game, then no, creating the art assets isn't that hard. But that's not what you meant.
Originally posted by JC-Smith Having worked with Hero Engine for about a year and a half now, there is absolutely nothing wrong with the thing. It's renderer isn't on par with the likes of Unreal Tech, but the editors, collaberative editing, networking is all just fine.
you have to release your game to the public before you make this claim. no one expects a developer to criticise their own work publicly.
in terms of released games, hero engine doesnt really have a great history. no doubt it is easy to use, allows multiple people to work on the same thing at the same time in real time and has plenty of built in assets and networking but personally i would go about making an mmorpg using it. heck i wouldnt go about making an mmorpg as an indie unless it was something very simple (eg. 2d, turn based, text based or just plain simple)
So you think CCP were in the wrong to go about making EVE: Online and now there newest MMO in development? You think the creators of Dark Fall were in the wrong? You think it's wrong for Us to be making Greed Monger? You think it's wrong for JC-Smith and his team to be making The Repopulation (I've personally worked with them for a week or 2, GREAT group of guys!)? There are plenty of Indies making MMOs... Yes just as many if not more fail but oh well They learned Lessions which they can go on and use on future games!
It's kind of like being asked to evaluate NFL draft picks on draft day, and saying that you'd rather wait about a decade to see how good the players end up being in the NFL before you give teams their draft grades.
Have there ever been any good games that used the Hero Engine without having full access to the source code? As far as I'm aware, no. Will there ever be? Find me in a decade and I can give you a better answer than I can today.
Originally posted by JC-Smith Having worked with Hero Engine for about a year and a half now, there is absolutely nothing wrong with the thing. It's renderer isn't on par with the likes of Unreal Tech, but the editors, collaberative editing, networking is all just fine.
you have to release your game to the public before you make this claim. no one expects a developer to criticise their own work publicly.
in terms of released games, hero engine doesnt really have a great history. no doubt it is easy to use, allows multiple people to work on the same thing at the same time in real time and has plenty of built in assets and networking but personally i would go about making an mmorpg using it. heck i wouldnt go about making an mmorpg as an indie unless it was something very simple (eg. 2d, turn based, text based or just plain simple)
So you think CCP were in the wrong to go about making EVE: Online and now there newest MMO in development? You think the creators of Dark Fall were in the wrong? You think it's wrong for Us to be making Greed Monger? You think it's wrong for JC-Smith and his team to be making The Repopulation (I've personally worked with them for a week or 2, GREAT group of guys!)? There are plenty of Indies making MMOs... Yes just as many if not more fail but oh well They learned Lessions which they can go on and use on future games!
i really dont know how you would define indie or non indie but the people who were making eve and darkfall seem to have a fairly strong engineering base. eve and darkfall use their own game engines and server technnologies. eve was an extremely well thought out practial and feasible mmorpg -->
- very few assets ( art, models, textures) due to space theme
- no story cutscenes and cgi stuff
- simple, slow paced combat
- no/minimal animation requirements
i cant say is wrong for you to make greed monger or above and beyond studios to make repopulation but i dont recommend tackling huge projects such as mmorpgs unless you have the knowledge and experience and business mindset required. i have no idea what you guys are capable of but choosing an unproven engine, surrendering server control to a third party and the forfeiting 35% of your profits makes it hard for me to take you guys seriously. dont get me wrong, i would like for you guys to succeed, i really hope everything works out fine for you guys and i harbor no ill wishes i just dont find the whole idea of herocloud and small groups of inexperienced dudes making mmorpgs very promising. having fun and learning lessons is great but it means nothing to an end user.
just a few questions:
- what are you doing about hackers?
- herocloud servers have provide enough bandwidth, how are you ensuring they will?
- do you know what server hardware is being used and can you prove that it is good enough?
- have you tested your servers with the targetted number of concurrent users?
- how are you dealing with latency (do u have client side prediction or something in place)
Originally posted by JC-Smith Having worked with Hero Engine for about a year and a half now, there is absolutely nothing wrong with the thing. It's renderer isn't on par with the likes of Unreal Tech, but the editors, collaberative editing, networking is all just fine.
you have to release your game to the public before you make this claim. no one expects a developer to criticise their own work publicly.
in terms of released games, hero engine doesnt really have a great history. no doubt it is easy to use, allows multiple people to work on the same thing at the same time in real time and has plenty of built in assets and networking but personally i would go about making an mmorpg using it. heck i wouldnt go about making an mmorpg as an indie unless it was something very simple (eg. 2d, turn based, text based or just plain simple)
So you think CCP were in the wrong to go about making EVE: Online and now there newest MMO in development? You think the creators of Dark Fall were in the wrong? You think it's wrong for Us to be making Greed Monger? You think it's wrong for JC-Smith and his team to be making The Repopulation (I've personally worked with them for a week or 2, GREAT group of guys!)? There are plenty of Indies making MMOs... Yes just as many if not more fail but oh well They learned Lessions which they can go on and use on future games!
i really dont know how you would define indie or non indie but the people who were making eve and darkfall seem to have a fairly strong engineering base. eve and darkfall use their own game engines and server technnologies. eve was an extremely well thought out practial and feasible mmorpg -->
- very few assets ( art, models, textures) due to space theme
- no story cutscenes and cgi stuff
- simple, slow paced combat
- no/minimal animation requirements
i cant say is wrong for you to make greed monger or above and beyond studios to make repopulation but i dont recommend tackling huge projects such as mmorpgs unless you have the knowledge and experience and business mindset required. i have no idea what you guys are capable of but choosing an unproven engine, surrendering server control to a third party and the forfeiting 35% of your profits makes it hard for me to take you guys seriously. dont get me wrong, i would like for you guys to succeed, i really hope everything works out fine for you guys and i harbor no ill wishes i just dont find the whole idea of herocloud and small groups of inexperienced dudes making mmorpgs very promising. having fun and learning lessons is great but it means nothing to an end user.
just a few questions:
- what are you doing about hackers?
- herocloud servers have provide enough bandwidth, how are you ensuring they will?
- do you know what server hardware is being used and can you prove that it is good enough?
- have you tested your servers with the targetted number of concurrent users?
- how are you dealing with latency (do u have client side prediction or something in place)
You have to have an awful lot of stuff in place before you can meaningfully test server capabilities. If you're launching a game next week, then yeah, you need to have answers to all of those questions by now. But if the first beta is a year or two away, the game could easily be in a state where the most you can do is have plans that you're fully aware are unlikely to survive contact with reality.
Hmm... while some of the more experienced people around here who are against the use of premade engines, are offering some valuable insight and information.
I still think that the spirit in which the OP is presentingthem is still valid and good. The premade engnes are a good start of the new people to get in to making a game. Not everyone can start coding their own engine just like that. And not everyone can make a game and grasp many of the notions that are involved. But everyone can learn and premade engines are a good way to learn in my opinion.
So unless some here have somethng to fear, i do not think that discouraging people from getting their feet wet with these engines which at 99$/year are a very affordable endeavor, is actually a good thing.
Many players have much better ideas than existing Devs (mainly due to constrtraints imposed on them by the industry), which these engines could help some to express without those constraints. It can only be good for the industry and I do not hink that there is anything to fear of.
I say anyone who ever wanted to put some of their Ideas to the test, go for it, pick one of these engines and let your creative juices guide you!
Don't read too much into criticism with pre-made engines. They are industry standard for a very good reason. Namely, building a game engine up from scratch is very time intensive, which means expensive to do. I offered criticism for Hero Cloud not because I think building it yourself is a better choice, just that a lot of people go into this not realize just how monumental a task creating a game really is, regardless of the engine used.
I would use Unity or Unreal 3 rather than the hero engine, hell I prob wouldnt even bother if I had to use it. It still doesnt support muticore/threads by default, alot of processes are kinda backward because features have been slapped in badly, rather than the engine being designed with things in mind from the ground up. It's kinda like a house that was built with no doors, windows or plumbing, then had these whacked in a year or two later by bazza the tradie for a few slabs of beer.
Originally posted by JC-Smith Having worked with Hero Engine for about a year and a half now, there is absolutely nothing wrong with the thing. It's renderer isn't on par with the likes of Unreal Tech, but the editors, collaberative editing, networking is all just fine.
you have to release your game to the public before you make this claim. no one expects a developer to criticise their own work publicly.
in terms of released games, hero engine doesnt really have a great history. no doubt it is easy to use, allows multiple people to work on the same thing at the same time in real time and has plenty of built in assets and networking but personally i would go about making an mmorpg using it. heck i wouldnt go about making an mmorpg as an indie unless it was something very simple (eg. 2d, turn based, text based or just plain simple)
So you think CCP were in the wrong to go about making EVE: Online and now there newest MMO in development? You think the creators of Dark Fall were in the wrong? You think it's wrong for Us to be making Greed Monger? You think it's wrong for JC-Smith and his team to be making The Repopulation (I've personally worked with them for a week or 2, GREAT group of guys!)? There are plenty of Indies making MMOs... Yes just as many if not more fail but oh well They learned Lessions which they can go on and use on future games!
i really dont know how you would define indie or non indie but the people who were making eve and darkfall seem to have a fairly strong engineering base. eve and darkfall use their own game engines and server technnologies. eve was an extremely well thought out practial and feasible mmorpg -->
- very few assets ( art, models, textures) due to space theme
- no story cutscenes and cgi stuff
- simple, slow paced combat
- no/minimal animation requirements
i cant say is wrong for you to make greed monger or above and beyond studios to make repopulation but i dont recommend tackling huge projects such as mmorpgs unless you have the knowledge and experience and business mindset required. i have no idea what you guys are capable of but choosing an unproven engine, surrendering server control to a third party and the forfeiting 35% of your profits makes it hard for me to take you guys seriously. dont get me wrong, i would like for you guys to succeed, i really hope everything works out fine for you guys and i harbor no ill wishes i just dont find the whole idea of herocloud and small groups of inexperienced dudes making mmorpgs very promising. having fun and learning lessons is great but it means nothing to an end user.
just a few questions:
- what are you doing about hackers?
- herocloud servers have provide enough bandwidth, how are you ensuring they will?
- do you know what server hardware is being used and can you prove that it is good enough?
- have you tested your servers with the targetted number of concurrent users?
- how are you dealing with latency (do u have client side prediction or something in place)
You have to have an awful lot of stuff in place before you can meaningfully test server capabilities. If you're launching a game next week, then yeah, you need to have answers to all of those questions by now. But if the first beta is a year or two away, the game could easily be in a state where the most you can do is have plans that you're fully aware are unlikely to survive contact with reality.
any good multiplayer is built with multiplayer in consideration right from the start. from the looks of repopulation they easily have enough stuff in the game to test servers. all you need is players in an area performing actions and seeing if the server can handle relaying/processing information between clients to each other assuming authoratative server model is being used. this should really be done asap and im sure most multiplayer games are subject to such testing asap. maybe swtor wouldnt have been so crap if they taken server/multiplayer capabilities seriosulsly right from the go.
Hmm... while some of the more experienced people around here who are against the use of premade engines, are offering some valuable insight and information.
I still think that the spirit in which the OP is presentingthem is still valid and good. The premade engnes are a good start of the new people to get in to making a game. Not everyone can start coding their own engine just like that. And not everyone can make a game and grasp many of the notions that are involved. But everyone can learn and premade engines are a good way to learn in my opinion.
So unless some here have somethng to fear, i do not think that discouraging people from getting their feet wet with these engines which at 99$/year are a very affordable endeavor, is actually a good thing.
Many players have much better ideas than existing Devs (mainly due to constrtraints imposed on them by the industry), which these engines could help some to express without those constraints. It can only be good for the industry and I do not hink that there is anything to fear of.
I say anyone who ever wanted to put some of their Ideas to the test, go for it, pick one of these engines and let your creative juices guide you!
Don't read too much into criticism with pre-made engines. They are industry standard for a very good reason. Namely, building a game engine up from scratch is very time intensive, which means expensive to do. I offered criticism for Hero Cloud not because I think building it yourself is a better choice, just that a lot of people go into this not realize just how monumental a task creating a game really is, regardless of the engine used.
Licensing game engines may be common, but launching an MMORPG on what is basically a trial version of a game engine that doesn't let you touch the source code? Have any games done that and succeeded, ever?
-----
As for my own engine, it's not AAA graphics quality by any means. Here's a few screenshots I just took:
Looks decent enough to my eyes, especially for a game with exactly zero artists working on it. The weird cylinder in all of the screenshots is my placeholder character for use in testing collision detection, and won't be a part of the final game.
Also, at around 300 KB, each of those screenshots is larger than the entire game itself is at the moment.
Hmm... while some of the more experienced people around here who are against the use of premade engines, are offering some valuable insight and information.
I still think that the spirit in which the OP is presentingthem is still valid and good. The premade engnes are a good start of the new people to get in to making a game. Not everyone can start coding their own engine just like that. And not everyone can make a game and grasp many of the notions that are involved. But everyone can learn and premade engines are a good way to learn in my opinion.
So unless some here have somethng to fear, i do not think that discouraging people from getting their feet wet with these engines which at 99$/year are a very affordable endeavor, is actually a good thing.
Many players have much better ideas than existing Devs (mainly due to constrtraints imposed on them by the industry), which these engines could help some to express without those constraints. It can only be good for the industry and I do not hink that there is anything to fear of.
I say anyone who ever wanted to put some of their Ideas to the test, go for it, pick one of these engines and let your creative juices guide you!
Don't read too much into criticism with pre-made engines. They are industry standard for a very good reason. Namely, building a game engine up from scratch is very time intensive, which means expensive to do. I offered criticism for Hero Cloud not because I think building it yourself is a better choice, just that a lot of people go into this not realize just how monumental a task creating a game really is, regardless of the engine used.
Licensing game engines may be common, but launching an MMORPG on what is basically a trial version of a game engine that doesn't let you touch the source code? Have any games done that and succeeded, ever?
-----
As for my own engine, it's not AAA graphics quality by any means. Here's a few screenshots I just took:
Looks decent enough to my eyes. The weird cylinder in all of the screenshots is my placeholder character for use in testing collision detection, and won't be a part of the final game.
Also, at around 300 KB, each of those screenshots is larger than the entire game itself is at the moment.
Hmm... while some of the more experienced people around here who are against the use of premade engines, are offering some valuable insight and information.
I still think that the spirit in which the OP is presentingthem is still valid and good. The premade engnes are a good start of the new people to get in to making a game. Not everyone can start coding their own engine just like that. And not everyone can make a game and grasp many of the notions that are involved. But everyone can learn and premade engines are a good way to learn in my opinion.
So unless some here have somethng to fear, i do not think that discouraging people from getting their feet wet with these engines which at 99$/year are a very affordable endeavor, is actually a good thing.
Many players have much better ideas than existing Devs (mainly due to constrtraints imposed on them by the industry), which these engines could help some to express without those constraints. It can only be good for the industry and I do not hink that there is anything to fear of.
I say anyone who ever wanted to put some of their Ideas to the test, go for it, pick one of these engines and let your creative juices guide you!
Don't read too much into criticism with pre-made engines. They are industry standard for a very good reason. Namely, building a game engine up from scratch is very time intensive, which means expensive to do. I offered criticism for Hero Cloud not because I think building it yourself is a better choice, just that a lot of people go into this not realize just how monumental a task creating a game really is, regardless of the engine used.
Licensing game engines may be common, but launching an MMORPG on what is basically a trial version of a game engine that doesn't let you touch the source code? Have any games done that and succeeded, ever?
-----
As for my own engine, it's not AAA graphics quality by any means. Here's a few screenshots I just took:
Looks decent enough to my eyes. The weird cylinder in all of the screenshots is my placeholder character for use in testing collision detection, and won't be a part of the final game.
Also, at around 300 KB, each of those screenshots is larger than the entire game itself is at the moment.
opengl or directx?
also is directx free to use?
OpenGL 4.2 currently. If I ever finish it, I'll eventually port it back to OpenGL 3.2, but no further.
I have no idea about DirectX. I know that it's proprietary to Microsoft, but I think their plan on making money off of DirectX is by making it Windows-only so everyone has to buy Windows to play DirectX games. I don't think they're trying to charge developers for it directly, though you will have to pay for Visual Studio if you're using that, and I'm not sure if you can use DirectX without using Visual Studio.
OpenGL is definitely free to use, though. So is Eclipse, which is what I'm making it in. JOGL (OpenGL bindings for Java) is also free, and open source, even. And yes, I'm using Java, not C++. It was kind of a haphazard decision, as when I started, I expected the project to go basically nowhere, and was surprised to learn that it was a lot easier than I expected.
Visual Studio express is free * (corrected by bishbosh)
DirectX is free as well. And tbh youre not losing much at all by going java. At least you wont have to go through the hellish api that is the win32 api -.- god forsaken win32 api was hell the first time I tried using it.
Edit 2: dont expect to use DX11.1. You need to pay for the pro version of ual studio 2012 to be able to make anything other than metro apps. Stick to 2010 express if you were to use visual studio and want to make anything other than metro apps
''/\/\'' Posted using Iphone bunni ( o.o) (")(") **This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
Visual Studio express is free * (corrected by bishbosh)
DirectX is free as well. And tbh youre not losing much at all by going java. At least you wont have to go through the hellish api that is the win32 api -.- god forsaken win32 api was hell the first time I tried using it.
Edit 2: dont expect to use DX11.1. You need to pay for the pro version of ual studio 2012 to be able to make anything other than metro apps. Stick to 2010 express if you were to use visual studio and want to make anything other than metro apps
Internet lore seems to have it that C++ is a little faster than Java on CPU-side things, but not massively so. My understanding, which could easily be wrong, is that Java has a little bit of overhead for just-in-time compilation, and C++ doesn't have that.
Regardless, the real question is whether it's fast enough. When I can get 50 frames per second on an AMD E-350 based laptop/netbook and that's GPU-bound, not CPU-bound, I think Java is fast enough. And that's outside of cities, so the game world is basically complete except for adding mobs.
A lot of processor performance is how well you can scale to more CPU cores. Give me 20 CPU cores clocked at something stupidly low like 500 MHz (so as not to immediately be GPU-bound) and I can put them all to good use. Java has some very nice threading tools, so there's no problem there. I'd assume that C++ has some nice threading tools, too, but I don't know.
the repopulation is using the hero engine and just from watching their dev videos you could tell bioware had no idea how to code for the engine....the engine itself got a bad rap for biowares incompetence. It's not a flawless engine however it does alot of things very well. Hero is trying to be the unity for mmo devs...they may or may not succeed.
Don't read too much into criticism with pre-made engines. They are industry standard for a very good reason. Namely, building a game engine up from scratch is very time intensive, which means expensive to do. I offered criticism for Hero Cloud not because I think building it yourself is a better choice, just that a lot of people go into this not realize just how monumental a task creating a game really is, regardless of the engine used.
Licensing game engines may be common, but launching an MMORPG on what is basically a trial version of a game engine that doesn't let you touch the source code? Have any games done that and succeeded, ever?
It's all about target audiences. The Hero Cloud thing is designed to be more of a hobby engine that's cheap enough for anyone to try out, but still robust enough to actually make a game with, provided you have access to the proper art assets. More importantly, it can be a stepping stone for licensing the full Hero Engine source code and migrating your project to it if you so choose.
All in all, it's a pretty good deal. That said, of course there's going to be no AAA studios trying to launch a game on Hero Cloud. That doesn't mean a smaller "studio" couldn't put it to good use, with little risk involved.
The same can be said for any of the "free" licenses of various engines, like UDK. The big selling point for Hero Cloud, in particular is that they provide a very strong set of world building tools designed around MMO workflows, rather than something like UDK, which has great tools for FPS stuff.
For it's target audience, Hero Cloud is a much better prospect for a small dev studio with little to no real funding, looking to try and create the game they have in mind, for practically nothing.
The main difference between java and c++ is that you have more control over memory in c++.
And the biggest dealbreaker is if that you can not use a data driven approach in java because arrays are not necessarily stored in contiguous memory location. Not a problem considering you probably wont be using data oriented paradigm except for when you code for consoles or you really have nowhere else to optimize with the OO paradigm and you completely bottleneck on your memory speed.
And thats not me saying Java sucks at all. Java works well for 90% of the stuff. Unless you were trying to make some state of the art ID tech/ cryengine/udk - like tech, dont bother thinking "i shouldve gone c++..."
CPU speed is hardly relevant because of JIT.
''/\/\'' Posted using Iphone bunni ( o.o) (")(") **This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
i dont know jackshit about this topic, but imo, if u want to make a game with unseen, unique and complex stuff, its best to create ur own shit to mess with it.
Comments
Why are you guys so worried about these engines? I mean...Tbh the hardest part is actually getting/making the art assets o.o
Obv just art assets because its friggin fun to make sound effects...*whacks a watermelon* @_@
''/\/\'' Posted using Iphone bunni
( o.o)
(")(")
**This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
So you think CCP were in the wrong to go about making EVE: Online and now there newest MMO in development? You think the creators of Dark Fall were in the wrong? You think it's wrong for Us to be making Greed Monger? You think it's wrong for JC-Smith and his team to be making The Repopulation (I've personally worked with them for a week or 2, GREAT group of guys!)? There are plenty of Indies making MMOs... Yes just as many if not more fail but oh well They learned Lessions which they can go on and use on future games!
Company Owner
MMO Interactive
It's not about producing AAA quality graphics. That's unnecessary for a low-budget indie game. It's probably also impractical, as a low budget doesn't let you produce the large quantities of high quality artwork that an AAA game demands. Rather:
1) It's about making developers' lives easier by letting them use the right tool for the job. If a building project calls for the use of both a hammer and a screwdriver, you may be able to get by with just one of the two. In a pinch, a sturdy screwdriver can be used as a hammer, or the "claw" part of a claw hammer can be used as a screwdriver for larger screws. But it will sure make your life easier if you have both the hammer and the screwdriver available.
2) It's about optimizing performance by offloading the work that video cards are good at onto the video card rather than having to do it on the processor instead. Any image that you can compute on a video card, you could just as well compute on a processor (up to minor rounding errors) without using any sort of 3D API. The drawback of this is an enormous performance hit. Rather than frames per second, you'd likely be looking at seconds per frame. Depending on what you want to do, in some cases, it could even be minutes per frame. For a game, that's a non-starter.
The entire point of having a video card is that it's optimized to be very, very good at some things that x86 (and ARM) processors are terrible at. x86 processors are optimized for doing a large number of operations sequentially with low latency, and to still perform well with a lot of branching. Video cards completely choke on that sort of workload. Rather, video cards are very good at situations that are extremely parallel and ask for you to perform exactly the same instructions in exactly the same order on many different pieces of data at the same time.
For the sorts of data that video cards are good at, you may realistically see improvement of two orders of magnitude (i.e., speed up by a factor of 100) by doing the computations on the GPU rather than the CPU. Even if you're pairing an ultra-high-end $1000+ processor with a $60 video card, the video card may still commonly be an order of magnitude faster.
But that's only if you you're able to have the video card process the data. Never mind tessellation for a moment; that requires understanding of some advanced mathematics, and even most AAA developers don't have the necessary expertise in-house.
Instead, let's talk about geometry shaders. There are two very important things that geometry shaders do: they let you see an entire primitive (typically a triangle) all at once, rather than only one vertex at a time. And they can emit new primitives, rather than requiring all vertices, edges, and triangles to be created by the processor and fed to the video card.
The latter lets you have one triangle come in, and then pass it through. Or one triangle comes in, and then two triangles go out. Or zero. Or 17. Or any other number up to the cap (which I think is 64 in OpenGL 4.2). And the shader can be aware of all of the triangles that you're going to output at once, rather than having to process them one at a time.
So what would you use that for? For one, it lets you mostly handle particle effects on the GPU, with the CPU only handing the video card some token data to get started. Or if you ever want to draw the same scene from more than one perspective (e.g., for stereoscopic 3D, Eyefinity without assuming that the monitor screens all lie in the same plane in real life, or a split-screen multiplayer game), that lets you save a ton of work.
More generally, geometry shaders are like mesmers in Guild Wars: they seem useless at first, but they end up providing a very elegant solution to all sorts of weird problems. And yet, if they had never existed, people wouldn't realize that they were missing and ought to be there. Developers who are still using DirectX 9.0c likely don't realize that they're making things much harder on themselves than necessary.
And then, yes, the newer APIs can produce better quality graphics. But that's largely because if you can do the same thing as before while putting a lighter load on your hardware, then that frees up resources to do more than you could before.
-----
So let's talk about hardware compatibility. Tessellation requires DirectX 11 or later, or OpenGL 4.0 or later. That means video cards from roughly the last three years. But even if your game is designed around tessellation, it's not that hard to port it back to DirectX 10 or OpenGL 3.2. (Well, I only know that for OpenGL, but I'd assume it's the same for DirectX.) You can tessellate your data once on the processor before uploading it to the video card. Doing tessellation properly means that you need so little vertex data that even if tessellating it before uploading it increases the size of the vertex data buffers by a factor of 100, that's still basically a rounding error. Now maybe you need hundreds of KB instead of only several KB of video memory.
If you're using geometry shaders much, then porting that back to DirectX 9.0c is probably impractical, or at least more trouble than it's worth. But that means your game requires a video card from roughly sometime in the last 6 years. If you're starting a project today and think you'll take 3 years to finish it, then by the time it launches, it will require a video card from sometime in the last 9 years. Is that really such an onerous hardware requirement?
Even if you could port it back further to make it playable on older hardware, would the older hardware have the level of performance necessary to run your game? People who would have bought a GeForce 7900 GTX or Radeon HD 1900 XT tend not to be the sort of people who would still be using the same video card a decade later. Rather, the older hardware that might still be in use would be low end stuff like a GeForce 6150 SE that can't run most modern DirectX 9.0c games at playable frame rates, either.
As for operating systems, DirectX 10 and 11 are both compatible with Windows Vista, 7, and 8, and presumably future versions, too. You mention DirectX 11.1, but using 11.1 is just stupid for anything beyond a tech demo. I'm not advocating that for real games. DirectX 10 does mean you lose compatibility with Windows XP, but if that bothers you, you can use OpenGL 3.2 instead and get compatibility with Windows XP, Vista, 7, Linux, Mac OS X, and likely soon Google Android.
Maybe being incompatible with Windows XP is a problem today. Think it will still be a problem three years from now, when you hope to finish your game if you start today?
Would my game engine be better than the Hero Engine for making Greed Monger, The Repopulation, or any other game besides my own? Nope. Not even close. They probably wouldn't be able to do much with my game engine. Artists would be completely stuck, as mine produces artwork by typing out source code without any sort of graphical interface. Even with full access to the source code, comments that assume that the reader understands, say, homology groups make perfect sense to me but would be completely baffling to anyone who doesn't. Furthermore, some capabilities that most games need are completely missing from my game engine, and not going to be added. For example, loading textures from a hard drive.
But you know what my game engine is better than Hero Engine for? My game. Because that's exactly what it's designed for. I want to do a bunch of stuff that Hero Engine won't allow, but my game engine does. This isn't peculiar to me, either; an awful lot of game engines that are designed for some particular game are the best engine for that particular game, or at least, far better than any off-the-shelf game engine that doesn't give you full access to modify the source code.
There's a lot of truth to that, though it depends on what you're doing. If you're trying to create a purely text-based game, then no, creating the art assets isn't that hard. But that's not what you meant.
It's kind of like being asked to evaluate NFL draft picks on draft day, and saying that you'd rather wait about a decade to see how good the players end up being in the NFL before you give teams their draft grades.
Have there ever been any good games that used the Hero Engine without having full access to the source code? As far as I'm aware, no. Will there ever be? Find me in a decade and I can give you a better answer than I can today.
i really dont know how you would define indie or non indie but the people who were making eve and darkfall seem to have a fairly strong engineering base. eve and darkfall use their own game engines and server technnologies. eve was an extremely well thought out practial and feasible mmorpg -->
- very few assets ( art, models, textures) due to space theme
- no story cutscenes and cgi stuff
- simple, slow paced combat
- no/minimal animation requirements
i cant say is wrong for you to make greed monger or above and beyond studios to make repopulation but i dont recommend tackling huge projects such as mmorpgs unless you have the knowledge and experience and business mindset required. i have no idea what you guys are capable of but choosing an unproven engine, surrendering server control to a third party and the forfeiting 35% of your profits makes it hard for me to take you guys seriously. dont get me wrong, i would like for you guys to succeed, i really hope everything works out fine for you guys and i harbor no ill wishes i just dont find the whole idea of herocloud and small groups of inexperienced dudes making mmorpgs very promising. having fun and learning lessons is great but it means nothing to an end user.
just a few questions:
- what are you doing about hackers?
- herocloud servers have provide enough bandwidth, how are you ensuring they will?
- do you know what server hardware is being used and can you prove that it is good enough?
- have you tested your servers with the targetted number of concurrent users?
- how are you dealing with latency (do u have client side prediction or something in place)
You have to have an awful lot of stuff in place before you can meaningfully test server capabilities. If you're launching a game next week, then yeah, you need to have answers to all of those questions by now. But if the first beta is a year or two away, the game could easily be in a state where the most you can do is have plans that you're fully aware are unlikely to survive contact with reality.
Don't read too much into criticism with pre-made engines. They are industry standard for a very good reason. Namely, building a game engine up from scratch is very time intensive, which means expensive to do. I offered criticism for Hero Cloud not because I think building it yourself is a better choice, just that a lot of people go into this not realize just how monumental a task creating a game really is, regardless of the engine used.
You make me like charity
I would use Unity or Unreal 3 rather than the hero engine, hell I prob wouldnt even bother if I had to use it. It still doesnt support muticore/threads by default, alot of processes are kinda backward because features have been slapped in badly, rather than the engine being designed with things in mind from the ground up. It's kinda like a house that was built with no doors, windows or plumbing, then had these whacked in a year or two later by bazza the tradie for a few slabs of beer.
In short, its utter crap.
any good multiplayer is built with multiplayer in consideration right from the start. from the looks of repopulation they easily have enough stuff in the game to test servers. all you need is players in an area performing actions and seeing if the server can handle relaying/processing information between clients to each other assuming authoratative server model is being used. this should really be done asap and im sure most multiplayer games are subject to such testing asap. maybe swtor wouldnt have been so crap if they taken server/multiplayer capabilities seriosulsly right from the go.
Licensing game engines may be common, but launching an MMORPG on what is basically a trial version of a game engine that doesn't let you touch the source code? Have any games done that and succeeded, ever?
-----
As for my own engine, it's not AAA graphics quality by any means. Here's a few screenshots I just took:
Looks decent enough to my eyes, especially for a game with exactly zero artists working on it. The weird cylinder in all of the screenshots is my placeholder character for use in testing collision detection, and won't be a part of the final game.
Also, at around 300 KB, each of those screenshots is larger than the entire game itself is at the moment.
opengl or directx?
also is directx free to use?
OpenGL 4.2 currently. If I ever finish it, I'll eventually port it back to OpenGL 3.2, but no further.
I have no idea about DirectX. I know that it's proprietary to Microsoft, but I think their plan on making money off of DirectX is by making it Windows-only so everyone has to buy Windows to play DirectX games. I don't think they're trying to charge developers for it directly, though you will have to pay for Visual Studio if you're using that, and I'm not sure if you can use DirectX without using Visual Studio.
OpenGL is definitely free to use, though. So is Eclipse, which is what I'm making it in. JOGL (OpenGL bindings for Java) is also free, and open source, even. And yes, I'm using Java, not C++. It was kind of a haphazard decision, as when I started, I expected the project to go basically nowhere, and was surprised to learn that it was a lot easier than I expected.
Visual Studio express is free * (corrected by bishbosh)
DirectX is free as well. And tbh youre not losing much at all by going java. At least you wont have to go through the hellish api that is the win32 api -.- god forsaken win32 api was hell the first time I tried using it.
Edit 2: dont expect to use DX11.1. You need to pay for the pro version of ual studio 2012 to be able to make anything other than metro apps. Stick to 2010 express if you were to use visual studio and want to make anything other than metro apps
''/\/\'' Posted using Iphone bunni
( o.o)
(")(")
**This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
Internet lore seems to have it that C++ is a little faster than Java on CPU-side things, but not massively so. My understanding, which could easily be wrong, is that Java has a little bit of overhead for just-in-time compilation, and C++ doesn't have that.
Regardless, the real question is whether it's fast enough. When I can get 50 frames per second on an AMD E-350 based laptop/netbook and that's GPU-bound, not CPU-bound, I think Java is fast enough. And that's outside of cities, so the game world is basically complete except for adding mobs.
A lot of processor performance is how well you can scale to more CPU cores. Give me 20 CPU cores clocked at something stupidly low like 500 MHz (so as not to immediately be GPU-bound) and I can put them all to good use. Java has some very nice threading tools, so there's no problem there. I'd assume that C++ has some nice threading tools, too, but I don't know.
so say we all
It's all about target audiences. The Hero Cloud thing is designed to be more of a hobby engine that's cheap enough for anyone to try out, but still robust enough to actually make a game with, provided you have access to the proper art assets. More importantly, it can be a stepping stone for licensing the full Hero Engine source code and migrating your project to it if you so choose.
All in all, it's a pretty good deal. That said, of course there's going to be no AAA studios trying to launch a game on Hero Cloud. That doesn't mean a smaller "studio" couldn't put it to good use, with little risk involved.
The same can be said for any of the "free" licenses of various engines, like UDK. The big selling point for Hero Cloud, in particular is that they provide a very strong set of world building tools designed around MMO workflows, rather than something like UDK, which has great tools for FPS stuff.
For it's target audience, Hero Cloud is a much better prospect for a small dev studio with little to no real funding, looking to try and create the game they have in mind, for practically nothing.
You make me like charity
The main difference between java and c++ is that you have more control over memory in c++.
And the biggest dealbreaker is if that you can not use a data driven approach in java because arrays are not necessarily stored in contiguous memory location. Not a problem considering you probably wont be using data oriented paradigm except for when you code for consoles or you really have nowhere else to optimize with the OO paradigm and you completely bottleneck on your memory speed.
And thats not me saying Java sucks at all. Java works well for 90% of the stuff. Unless you were trying to make some state of the art ID tech/ cryengine/udk - like tech, dont bother thinking "i shouldve gone c++..."
CPU speed is hardly relevant because of JIT.
''/\/\'' Posted using Iphone bunni
( o.o)
(")(")
**This bunny was cloned from bunnies belonging to Gobla and is part of the Quizzical Fanclub and the The Marvelously Meowhead Fan Club**
Whatever the case, I just hope that the Hero Engine shines on TRP, I mean who's got money to spend millions on an engine(from the Indy group I mean)?
GS
$99 / year
AND
Autodesk Max or Maya 2011 or later, x32 or x64 (According to thier own minimum specs.)
Its really the last part that kills the idea.
Where did you think the art assets were going to come from? Microsoft Paint?