Game Studio (GS)

Engines, Libraries and Tools Tools Discussion
#1

Hi everybody,

I am proposing a do-all-in-one next-gen tool for game content creation. Here is the post I sent for the project approuval thread. I know that this is a huge project, and that’s why I want to make it become a community project … to have a chance to see it finished one day. Please tell me what you are thinking about it.

Thank you.

PS: I am serious about this project, that’s not a troll.

Hello,

I am planning this project since 3 years. It have been modified and adapted so many times that I think his design and ideas are mature enough to make them public and start to implement.

This project aims to reduce the game content creation time, specifically in the environments where there are lot of people involved in the project (many artists, programmers, managers, game designers, a boss, etc …). During many years of experience in various game companies, I analysed where were the common sources of loss of time in the development of games. My conclusion is that people are using tools that are just not good enough for their needs, and most of people don’t even realize it.

From the notes I took during 3 years, from a lot of “it would be cool if …” that looked pretty inadaptable to the existing tools, I designed the plans of a big game content creation tool that is supposed to mark the limit between the current and the next generation of tools in the game industry.

My project mainly focus on the following objectives :

  • Inhilation of the causes of loss of time and the unnecessary user tasks.
  • Distributed resources, Collaboration of resource editors and users over the network, and realtime updates.
  • Integration of all the game content tools into a big one, homogenous framework for game content resources.
  • The tool and the game share the same environment : you edit in the game, in realtime.

In practice, the project can be seems as a set of different parts :

  1. A distributed resource management system and its framework.
  2. An integrated task management system and integrated tools for user communication.
  3. A lot of resource definitions and their editors.
    Those resources are related to :
  • Text.
  • 2D.
  • 3D.
  • Music.
  • Network properties.
  • Game logic.
  • etc … just whatever you need for a game.

I am actually working on the first 2 items of the list. I think that with some help about a few details where I am not familiar I can make a prototype of them by myself, but for the resource editors I will clearly need the help of the community since it is about making a TheGimp-like and a Maya-like in java, and I cannot implement those huge parts alone.

The implementation of this project is making use of the Java Persistence API 1.0 and follows the ideas of the POJO programming.

I am sorry to not give more details about the implementation and the differents components of the project, but just to give all the details is a big task for me and might take me several weeks to write them down into a big document.

The actual name of this project is GS, like “Game Studio”. It is a temporal name since I didn’t find a good one for now. I will give this project a full name once it reachs a 1.0 version, i.e. something finished enough to let users make games with it.

I may need some assistance soon about the network, and JPA (I am not yet familiar enough with it).

KarmaGfa.

#2

I think I posted in the wrong place, there is no java tool developers here :-[

[Edit] I also had no answer about my project proposal, it was not published neither, so I guess there is no admin. I 'd better no count too much on the community, continue alone and use my own website to publish my sources.

Bye

#3

Well, personally I didn’t reply because it sounds insane. The companies I’ve worked at we had entire departments who considered what you’ve proposed above more than they could get done in 3 years. Some parts of it (for instance in-game editing) are more than they could get done in 10 years - that stuff takes years even if you do it entirely hardcoded for a single game; trying to do it generically is much harder.

#4

There is 3 kind of persons :

  • those who can.
  • those who can’t.
  • those who don’t even try.

You seems to be the third one. See you in 3 years and let’s see what I did. Probably it won’t be something completely finished, but it will be enough to make games and to gather a community to help me to finish in less than what you expect. That’s not because most people are afraid to do it that I won’t make it.

One last thing : I worked for a world famous game company, and I saw what was their main R&D department : 1 developper, using C++ like C and working during 5 years with old ideas and old technics. So … can we compare ?

#5

Now you’ve done it. :o
Blah^3 was only offering some insight on this matter.
Most of the people on these boards have years of professional experience. Please don’t presume what he has or hasn’t done.

#6

KarmaGfa you are right on the spot on your analisis. However you are not going to raise much help without first creating a demo and some design docs. Then ask for help in other boards besides this one like the generic java forums and the boards dedicated to lwjgl and jme.

My sugestion is build an idea for a basic game studio, then document it and create a demo, even a mockup demo to show your idea.

#7

somewhere along the way ppl made a wrong turn. I see why we would end up here, making faulty interpertations.

There is some official constant in planning which defines that what ever gues we make we are gonna need 2,97 times more time making bla3 response perhaps not totally off… but the reason I was replying is to avoid the pissing contest instead of continuing it.

A lot can be improven and we can be a lot more efficient then some industry giant (the other way too) so we can hardly tell you otherwise and bla3 isn’t telling you other wise he is weary to the consept because of things that he outlines in his post.

the interpertation that bla3 position on your project is that it’s impossible( “doesn’t even try”) is contradicted by a number of things. Insane is not impossible, bla3 does stuff at his work that is precieved as insane, as well as other ppl on this forum. insane is quite relative, I do not interper it as impossible in any way. but lets not start a discussion on how insane should be interpererd.

also your proposal was not rejected was it? it was not adressed(Im making a small gues here) as bla3 couldn’t justify making niether decision.

and the essence of his reply was that he wanted to set appart why you haven’t had a reply from him and to contratict the presumtion that theres no admin.

the essence of my reply (which I failed miserably at expressing) is that perhaps instead of making presumtions or pointing or getting all defensive(as your 2nd post basicly already was) perhaps you should outline/set appart why it isn’t even insane, why it can be achieved in less time(if you feel strongly about it I think length of time is not as importand as ‘in-time’) and perhaps ask bla3 what can be done to avoid ppl not repling.

on a differend note; I wouldn’t worry too much about JPA if you follow the POJO aproach as you call it then RDBMS are just one of the ways you could store them and would use JPA for(JPA’s scope might get bigger but I don’t feel it’s is now) but thats probebly not one of your main focus points. I could be wrong offcourse.

#8

It was just my personal opinion, YMMV. You seemed to want to know why people hadn’t replied.

I think the suggestion to more tightly define what you’re doing is a very good idea. Basicalyl, your proposal starts to start insane the moment you use the word “big”: off the top of my head, all the times people ni the games industry have tried a “big” tool its failed, especially for non-commercial ones.

I may be comlpetely wrong, as I said YMMV. But I’d rather know I gave you the advice and you decided what to do with that advice, than that I sat in silence and watched you possibly waste lots of time and effort. And … you may well think of a more cunning approach than the Big Tool of Everythingness :).

#9

Since I don’t work in the games industry, I don’t really know how people work in that sector on a day to day basis, but from my point of view it seems that a lot of existing tools already cover the features you’re asking for. Using basic stuff like a version control system and some build system, an instant messenger or plain old telephones, and existing tools such as blender and the gimp it seems like you should be able to set up a pretty decent working environment. I don’t see why you would want to put so much time (it really sounds like a huge project) in basically reinventing stuff that already exists. In other words, do you really think this kind of ‘integrate everything’ tool is commercially viable?

#10

I think that what he’s going after is like my project, except… even more general with 3D and project management stuff wrapped in there, which makes it exponentially more complicated than my project already is. So in essence, a professional-grade 3D game maker so to speak. I don’t know how many studios would use this if it reached completion since many develop their own in-house tools already, but it might be interesting for indie developers or studios that don’t have the resources to build their own set of tools.

#11

Personally, I think the idea would only be viable if it were specifically tailored toward a genre, say ‘Indoor First Person Shooter’; something like UnrealEd for example. Even then, it’d be a huge undertaking, even if you’d just interface with existing resource editors like 3DS or Maya (trying to recreate something like that from the bottom up as only a part of the complete project is insane imho and serves no purpose).
I don’t really believe a ‘one size fits all’ tool which allows you to be able to create any kind of game.

#12

Hello to all.

Sorry to blah^3, I recognize I over-reacted. Your advice was a good advice in a way, and I should only take it as an advice - I took it as something that is more than an advice. My fault. Apologies.

I am continuing on GS anyway. I don’t think I could handle some help before I finish the global framework, so I am preparing a prototype of the tool, trying to validate my system on a set of simple resources, for a simple 2D demo-game.

I recognize that my project is insane, but I have the feeling that this project is what the game industry will need in the long term, so I continue.

#13

Hello,

I wanted to use JPA for the persistence of the objects of my project GS. Unfortunately I ran into some its limitations and I think I expected too much from it. What I wanted was something transparent and similar to the serialization mechanism, but oriented database persistence.

What I see with JPA is that you can’t get rid of the @Id in your POJO, even if you don’t want to query them directly from their ID from the database.

Another bigger problem is that with JPA, it seems that you can only persist fields that reference entities or primary types. Entities are classes, so when you reference a class via one of its interfaces, you are stuck :-. The solution proposed on the forum of Hibernate is to transform the interface into a class, but this work around won’t work if you are using more than 1 interface on your classes :’(. JSE programmers are using many interfaces everywhere, so IMHO JPA is not for JSE’s POJO.

I feel pretty depressed to have spend so much time trying to solve the problems related to JEE and JPA. I am starting my own persistence mechanism really based on the serialization and oriented database. It won’t be optimal, won’t be multi-what like hibernate, but at least it will work for my precise case.

I recognize that my project is getting more and more insane :P, but it might be a sign that there is a big hole somewhere. ::slight_smile:

#14

Sompeting to bear in mind is that Hibernate is not JPA. Personally I am not a great fan of Hibernate (any more) - it got too complicated, too feature-rich, too many things to get wrong, too much like jumping through hoops for my taste. That was a while ago, and I’ve heard its much miproved recently, but it’s worth bearing in mind that yuo might find alternative JPA impls easier to get going with, especially if your needs are not those of a large enterprise system.

#15

I think your probebly going about it wong wenn it comes to storing.

can you tell me a typical case where you would be storing and retrieving stuff?

#16

Each time I do an edition of a resource, i.e. each time the user click on something.

#17

and what would you be storing?

#18

All kind of user created resources.