Archived

This topic is now archived and is closed to further replies.

dozyworm

Game prototyping

Recommended Posts

dozyworm    122
Hi, does anybody know about or use any quality game prototyping tools or any language that would serve the purpose of creating a game prototype well? I''m after something that''s very easy to use and that isn''t in C/C++ because I''m going to chuck the prototype once I''m finished with it and I don''t want to be tempted to use it in a final programme. I''d obviously like something that supported rudimentry 3D graphics and that would let me try out a number of ideas and not leave me tied to a specific genre. It does not need to be fast at all. I''d rather it was stupidly easy to use and abused its memory usage and consumed zillions of processor cycles than it was a problem to work with. I''ve been doing some scouting and although there are quite a few full scale development libraries or languages that allow you to access 3D graphics hardware like Java/Delphi there doesn''t seem to be any proper game prototyping tools or a library that is dedicated to this purpose. Preferablly I''d like to use an Open Source tool but that''s not essential. Thanks, Nick.

Share this post


Link to post
Share on other sites
dozyworm    122
"Buy a game similar to the game you want to write that comes with a level editor."

I''m sorry, I don''t think I''ve explained myself very well.
I''m looking for a set of tools that I can use to prototype full-blown game ideas before I decide if they are worthy of further work and attention. The idea is that I would create a prototype to test a game idea and to see whether the game actually worked or was in fact flawed in some way that I couldn''t see until I actually played the game.
This allows me to quickly test new game ideas and to thrash out the major algorithms before going on to the full implementation.

As I said I don''t want the tools tied to a particular genre. This makes it impossible to test new ideas because the tool will only allow you to create old ideas in new clothes which is something I''d rather not waste my time doing.
A level editor is a hopeless tool to use to develop a prototype for a game. It - by definition - doesn''t create prototypes of new games but levels for old ones.

Creating a prototype is very common in business applications and it strikes me as rather odd that the game community would be lacking a tool that would allow you to get a concept up and running in a matter of days instead of spending months on a large project until it was playable.

Anybody know of anything, a version of Basic perhaps that allows simple 3D graphics?

Share this post


Link to post
Share on other sites
LilBudyWizer    491
To put it in simplier terms game engines are what you describe. Level editors give you access to the underlying game engine to one degree or another. Level editors is really a bad term to use because they are really game editors. Often times they are pretty much what you would get if you licensed the engine. What you don''t get is the source and often lack the ability to make certain modifications. The biggest differance is really the license in many cases. You are not allowed to profit off anything you create. As you describe prototyping you don''t intend to profit off the prototype. Rather you intend to test ideas and if the ideas prove sound then you intend to develope the full game. Under that circumstance it is hard to justify tens of thousands to hundreds of thousands for licensing a game engine. There are cheaper alternatives, but the license for those imply that you are going to produce a game with the engine because they receive their money in the back end instead of upfront. If you actually intend to take the prototype, flesh it out and have the money then licensing a game engine could be cost justified assuming you can recoup the costs.

I think you fail to account for the differance in complexity between a business application and a game. Business applications are very simple. Games are more akin to scientific programming. As such what you use to prototype depends on precisely what you are prototyping. You wouldn''t use a single tool for physics, chemistry and biology and the same is true of games. A flight simulator is drastically differant than a side scrolling arcade game. It is far to complex for a single tool to do everything you might want to do in a game and have a reasonable learning curve.

Share this post


Link to post
Share on other sites
pwd    122
You could try Game Blender (Blender 2.x). It sounded kind of like a good tool for prototyping, but the interface... scares... me. Reminds me of EMACS. {shudder}

Some artists swear by it (just like some programmers swear by Vi ), so perhaps if you can find one of them, and work together, it could work. And it should have the added benefit that if you do make the game a reality, you should be able to export the art (uses some sort of Python export script plugin) in a usable manner with a little work.

Or, do it the old fashioned way - throw out your ideas to a group of people and let them rip it to shreds. If it survives, it might work, if not, back to the drawing board.

-pwd

Share this post


Link to post
Share on other sites
Guest Anonymous Poster   
Guest Anonymous Poster
quote:
Original post by LilBudyWizer

To put it in simplier terms game engines are what you describe. Level editors give you access to the underlying game engine to one degree or another. Level editors is really a bad term to use because they are really game editors. Often times they are pretty much what you would get if you licensed the engine. What you don''t get is the source and often lack the ability to make certain modifications. The biggest differance is really the license in many cases. You are not allowed to profit off anything you create. As you describe prototyping you don''t intend to profit off the prototype. Rather you intend to test ideas and if the ideas prove sound then you intend to develope the full game. Under that circumstance it is hard to justify tens of thousands to hundreds of thousands for licensing a game engine. There are cheaper alternatives, but the license for those imply that you are going to produce a game with the engine because they receive their money in the back end instead of upfront. If you actually intend to take the prototype, flesh it out and have the money then licensing a game engine could be cost justified assuming you can recoup the costs.

I think you fail to account for the differance in complexity between a business application and a game. Business applications are very simple. Games are more akin to scientific programming. As such what you use to prototype depends on precisely what you are prototyping. You wouldn''t use a single tool for physics, chemistry and biology and the same is true of games. A flight simulator is drastically differant than a side scrolling arcade game. It is far to complex for a single tool to do everything you might want to do in a game and have a reasonable learning curve.


I am sorry, but I have to complaint. What you just said is bullshit. Game Programming and Business Software or anything else that are not games are a;; as complex. Whereas game programming is "algoritmly" complex, business applications are standard and tools complex!

In a business applications, you need to follow strict, which means you have to know and understand a document that is somethimes 1000 pages long. You also need to learn new tools like Oracle,Sybase,SQL server etc. Business Softwares are usually huge! 3-4 hours compile time average is pretty close I think. I work at a business software company and it takes more than 2 months to get use to the system. It''s huge. Took me a week to get used to Quake 3 source...

Games are more plain, as you don''t have any standard to follow just yet. Most game still have smaller scope than business software. But some graphic algorithm are more complex. But wait! Ever worked on a routing algorithm?? Some make a bsp look like a bubble sort!!

As far as prototyping tools, yes they exists for games. I read about a bunch of them in Game Programming Architecture.

Prototype means trying idea, not doing a full featured game. What you do in a prototype is that try your idea that are differents and need to be tested for cheap before knowing if it worth developping. Tools like that exists for business, telecom, games, etc...

Developping a games is not different of developping other softwares. Granted they need different skills than business applications, but telecom also needs different skills than business so this is not an argument.

Games go through the same lifecycle as other software : Design, code, test. Also, in any kind of software, design will change, nothing is set in stone, design is just a guideline for developping. If the guidelines need to be changed then so be it.

Share this post


Link to post
Share on other sites
Guest Anonymous Poster   
Guest Anonymous Poster
try darkbasic. I''ve never used it and know nothing about it beyond that it''s basic, plus easy access to 3D graphics etc (alledgedly).

Share this post


Link to post
Share on other sites
dozyworm    122
Thanks to the anonymous poster, it seems at least some of you understood what I meant by a game prototype .

I don''t want to repeat what''s been said but I''ve got a little to say on LilBudyWizer comments about it not being possible to use a generic tool to prototype a game idea.

I feel this is just plain wrong.

The point of a prototyping tool is that you only test out the very rough outline of a game and do not completely develop it. - often prototypes are thrown away - It is perfectly possible to test the fundamental elements of a flight sim, remember only the elements that are essential to the gameplay or elements that seem unclear or difficult to implement need to be tested. Also this all doesn''t need to exist within a single application, one prototype could test the cockpit interface while another could test the A.I. and this could be a simple 2D grid application that moves points around representing planes.
Because prototyping is only concerned with simple systems I would argue that it is perfectly feasible, if not more feasible to develop a language or set of tools that are capable of creating such systems over tools for developing a complete game.

As for the comment that a tool can not be used in Physics, Chemistry and Biology, last time I checked Excel did a fairly good job with data analysis from any science and some programming languages like C are pretty simple but people use them for a wide range of tasks.

Anyway thanks to the pointers on Darkbasic and Blender, I''ll be sure to have a look into these. Oh and of course Countach is right about pencil and paper being the first port of call in the development of any prototype and any computer application for that matter.

Thanks everyone,

Nick.

Share this post


Link to post
Share on other sites
LilBudyWizer    491
Continuing with flight sims there are a number of custom cockpits that have been developed by the user community for Falcon 4.0. You could certainly mock up a cockpit without a physics system but it is just going to be pretty pictures. If you actually want to know how well that cockpit is going to work when it comes to actually playing a game the fastest way to me seems to be by putting it into an actual flight simulator. That is to me a prototype of the cockpit interface. Without the physics all you can test is that the gauges display data correctly and looks ok.

That said if you have actually experimented with a few game editors then fine. If you haven''t and simply want to dismiss the idea out of hand then that is fine as well. It is your life, your time and do what ever you please. I''ve been programming for 20 years and you are by no means the first person that has decided that a suggestion didn''t fit their preconceived notions and as such must be a bad idea.

Share this post


Link to post
Share on other sites
clonemaker    108
hey dozy. It may seem perfectly reasonable that you could use some tool to "prototype" a game. It is. But you''ll find that to create a prototype of a game to "test the ideas" is just as much work as creating the game itself, reguardless of what tools you use (Flexible tools require customization work and take time to learn). Then you throw all the code away and start over??? In my opinion, the best prototyping tool I know of is a C++ Compiler + libraries of your choice.

cmaker

- I do not make clones.

Share this post


Link to post
Share on other sites
dozyworm    122
I''ve been thinking and you''re right LilBuddyWizer, prototyping isn''t going to help with a the cockpit for a flight simulator. I''m sorry if I seemed rather dismissive .

You''re right I have been having a few preconceived ideas stuck in my head. Mainly various ideas for RPG''s, puzzle games and various strategy games, not very many more dynamic games. Anyway I do think that prototypes of some small and self contained game structures could be set up and tested using a simple programming language. Interfaces and A.I. handling and other procedures that do not rely on real time calculations are the main things that I can think of. And the idea of using game editors that already exist could be extremely cool, if you''re developing a game that shares a similar user interface and want to quickly test an idea. Of course there''s always a learning curve associated with any tool but I wouldn''t shy away from learning a tool if it could really make my life easier and I think it could broaden the mind a little by having a wide range of languages at my disposal for developing a game idea.

I am not advocating the creation of a prototype as a beall and endall and you have demonstrated examples were it would be difficult to produce such a prototype, but in some cases I feel it would be useful just to flesh out ideas that a designer is uncertain about by seeing a primitive working example of that idea.

Thanks for the heads up.

Nick.

Share this post


Link to post
Share on other sites
GrokMakeGame    122
I don''t think there''s a single instance where I''d want to use some sort of game prototyping kit. You obviously wouldn''t want such a kit for real time 3D graphics, and the couple of instances you''ve come up with in this last post are also bad.

For frontend and HUDs, there''s no better prototyping tool than your run of the mill paint program. For menu and game flow, you draw a flowchart. For controls, you play some games similar to yours, make a list of the actions you need to support, then put your hands on your keyboard or controller and hash out what makes sense. You can write a simple AI or Physics demo very quickly and tweak it, which would be the prefered method because these elements will be specific to your game.

When developing a game you get all the major elements working in some elementary fashion regardless if they are working absolutely the way you want. Once every thing is there, then you begin tuning, so you actually have a good idea of what works and what doesn''t fairly early into development. Games take a long time to make, but 3 quarters of that time is spent tuning, generating content, and debugging. If you plan thoroughly ahead of time and be objective about your design, you''ll have a good idea of what works, and what doesn''t before you even write a single line of code.

Games are a wholely different animal than business apps, thats why you don''t see rapid prototyping tools. Most games however do begin as prototypes, and any decent programmer can get something up and running and playable to some extent in a matter of a few weeks.

Oh, and you can make simple 3d graphics in Direct X and C++. You don''t need a fully lit, textured, animated model and scene to test physics or AU routines you write. I''m sorry if I''m being harsh here, but it sounds like you''re asking for Magic Good Game Maker 1.0. I''m sorry it takes work to make a game. This isn''t as simple as a business application.

Share this post


Link to post
Share on other sites
dozyworm    122
OK, OK. I''m convinced, no more prototypes.
I''m back to paper and pencil and more games playing.

Thanks everyone for helping me out on this, it''s probably saved me a lot of time.

Nick.

Share this post


Link to post
Share on other sites
Mikey_Flies    122
A prototype need not be a waste of time. You should only put in as much effort as is needed to test out the gameplay. You could even use qbasic or something equivalent to test a lot of gameplay mechanics. These are not hard to learn, and most protoypes (or should at least) use crude graphics as they are not in any way meant to make it to production. In fact, flashy graphics may actually detract from the ability to separate gameplay from the look of the game. I wouldn''t be so hasty to throw out prototypes as a waste of time. This may be true if you over develop them to the point of having a almost production game and then starting over, but this is not the point of a prototype to begin with.

Share this post


Link to post
Share on other sites