Indie 2.0 - Part 1
by Tony Richards · 03/13/2008 (7:06 pm) · 10 comments
Long time no blog, but recently I've been having to choose between doing or talking about doing and I've been choosing the "doing" path.
So what's been taking so much of my time?
Around this time last year I had just finished Fractured Universe when I had a revelation.
Indie tool chains and game engines have been created by commercial game developers for commercial game developers.
The question that started this revelation:
"What tools and customizations do Blender, Milkshape, ZBrush, 3d Studio Max, Maya have that are specifically designed for Indie Game Developers?"
The same question applies to 3d World Studio, Quark, and even Constructor as well as Codeweaver, Torsion, L3DT and the list goes on...
So, what is it that Indies need that our counterparts don't need? What is it that these tools aren't providing?
First and foremost: We need content packs
Garage Games and a few others have done an excellent job with current technology publishing great content packs.
The problem is with that key phrase current technology.
What help do existing products give in producing content packs?
The answer: nothing
Note that I have nothing against these tools and I own and have enjoyed using most of them... they simply don't meet my needs as an Indie Game Developer.
None of these existing tools provide a way for you to create content packs that could easily be used by Indie game developers to customize and create their own content without significant modeling and/or coding experience.
Wouldn't it be better if a content pack could have sliders, checkboxes and other options to allow you to pick and choose what you would like to export into your game?
What about a tool that would allow you to eazily manipulate the base mesh like ZBrush and Blender, but still retain LOD, UV Mapping, animations and generate a normal map for lower levels of detail?
How about a tool that would allow you to merge multiple animations, bake the results and export to your favorite game engine?
The Quest
Just because the tools don't exist today doesn't mean someone can't create the tools to revolutionize the industry... revolutionize the Indie game development industry.
I'm a great coder with over 30 years of experience and I felt up to this huge challenge, so I accepted this quest.
IndieZen Game Studio
Nearly a year ago in April 2007 I started my quest to revolutionize the Indie Game Development industry.
At first I was coding 20-30 hours a week since I still had my day job, but about a month ago I dumped my day job in order to complete my quest sooner. That's how dedicated I am to completing this.
Over the past year I've been concentrating on what I'm calling IndieZen Game Studio.
This is a complete IDE for creating games without being dedicated to any specific game engine.
First, the IndieZen Game Studio has built-in editors for scripting languages such as TorqueScript, Python, Lua, JavaScript / ECMA Script, etc. Each of these editors are context sensitive and provide many language specific details, including a full interactive debuggers, call graphs and class heircarchy browsers.
Another nice feature of the IndieZen Game Studio is it's Cg, GLSL, and HLSL editors. These help you put together great shaders and test them interactively, not just on a sphere, teapot or plane, but on your own models to see exactly how they will look in-game.
And then that takes me to the modeling plugins.
The IndieZen ZSpace modeling plugin is perfect for creating or importing models, skinning, UV mapping, animating, and exporting into your required game format.
Additional features include advanced sculpting, 3d painting directly on the mesh, multiple skins, multiple skeletons, etc.
And that's only the beginning. With the IndieZen ZSpace plugin, it's easy to create morph targets, texture layers and animation blends.
As a creative content provider, this allows you to export a content pack that your users can then pick and choose exactly what they want in their game.
As a game developer, this provides you with a wealth of content from which you can export fully customized content. No longer will your content look just like everyone else's. You will have access to unique artwork at about the same price as today's content packs that only provide for a few options.
Indie 2.0 content packs
That leads me to my next question. As I'm nearing completion of the IndieZen Game Studio ZSpace plugin, what kinds of content packs would most interest you?
I see a lot of requests for fantasy warriors and commoners with both male and female models.
I'd like to create a Tech Armor pack for a game along the lines of Tribes / Ascension to provide some of the original Tribes game modders with new content.
What else? A zombie pack? A Vamps and Vixens pack? A dragon pack? A racing vehicles pack?
I'm putting together a store with Indie 2.0 content packs over the next few months and I'd like to know more about what the community desires as far as styles and models.
Cast your votes here and let me know your target price. I'm looking for a $50 - $250 range for packs that have enough options to export hundreds if not thousands of completely unique models.
More about the Indie 2.0 Revolution
I'll be giving a presentation at IMGDC at the end of this month covering more of the Indie 2.0 Revolution.
There's a whole lot more to come, including some revolutions in terrain generation and editing, BSP modeling, game engine and game development frameworks, a revolutionary MMO framework and much more.
If you can't make it then I'll be sure to post blogs afterwards, but I hope to see you there!
So what's been taking so much of my time?
Around this time last year I had just finished Fractured Universe when I had a revelation.
Indie tool chains and game engines have been created by commercial game developers for commercial game developers.
The question that started this revelation:
"What tools and customizations do Blender, Milkshape, ZBrush, 3d Studio Max, Maya have that are specifically designed for Indie Game Developers?"
The same question applies to 3d World Studio, Quark, and even Constructor as well as Codeweaver, Torsion, L3DT and the list goes on...
So, what is it that Indies need that our counterparts don't need? What is it that these tools aren't providing?
First and foremost: We need content packs
Garage Games and a few others have done an excellent job with current technology publishing great content packs.
The problem is with that key phrase current technology.
What help do existing products give in producing content packs?
The answer: nothing
Note that I have nothing against these tools and I own and have enjoyed using most of them... they simply don't meet my needs as an Indie Game Developer.
None of these existing tools provide a way for you to create content packs that could easily be used by Indie game developers to customize and create their own content without significant modeling and/or coding experience.
Wouldn't it be better if a content pack could have sliders, checkboxes and other options to allow you to pick and choose what you would like to export into your game?
What about a tool that would allow you to eazily manipulate the base mesh like ZBrush and Blender, but still retain LOD, UV Mapping, animations and generate a normal map for lower levels of detail?
How about a tool that would allow you to merge multiple animations, bake the results and export to your favorite game engine?
The Quest
Just because the tools don't exist today doesn't mean someone can't create the tools to revolutionize the industry... revolutionize the Indie game development industry.
I'm a great coder with over 30 years of experience and I felt up to this huge challenge, so I accepted this quest.
IndieZen Game Studio
Nearly a year ago in April 2007 I started my quest to revolutionize the Indie Game Development industry.
At first I was coding 20-30 hours a week since I still had my day job, but about a month ago I dumped my day job in order to complete my quest sooner. That's how dedicated I am to completing this.
Over the past year I've been concentrating on what I'm calling IndieZen Game Studio.
This is a complete IDE for creating games without being dedicated to any specific game engine.
First, the IndieZen Game Studio has built-in editors for scripting languages such as TorqueScript, Python, Lua, JavaScript / ECMA Script, etc. Each of these editors are context sensitive and provide many language specific details, including a full interactive debuggers, call graphs and class heircarchy browsers.
Another nice feature of the IndieZen Game Studio is it's Cg, GLSL, and HLSL editors. These help you put together great shaders and test them interactively, not just on a sphere, teapot or plane, but on your own models to see exactly how they will look in-game.
And then that takes me to the modeling plugins.
The IndieZen ZSpace modeling plugin is perfect for creating or importing models, skinning, UV mapping, animating, and exporting into your required game format.
Additional features include advanced sculpting, 3d painting directly on the mesh, multiple skins, multiple skeletons, etc.
And that's only the beginning. With the IndieZen ZSpace plugin, it's easy to create morph targets, texture layers and animation blends.
As a creative content provider, this allows you to export a content pack that your users can then pick and choose exactly what they want in their game.
As a game developer, this provides you with a wealth of content from which you can export fully customized content. No longer will your content look just like everyone else's. You will have access to unique artwork at about the same price as today's content packs that only provide for a few options.
Indie 2.0 content packs
That leads me to my next question. As I'm nearing completion of the IndieZen Game Studio ZSpace plugin, what kinds of content packs would most interest you?
I see a lot of requests for fantasy warriors and commoners with both male and female models.
I'd like to create a Tech Armor pack for a game along the lines of Tribes / Ascension to provide some of the original Tribes game modders with new content.
What else? A zombie pack? A Vamps and Vixens pack? A dragon pack? A racing vehicles pack?
I'm putting together a store with Indie 2.0 content packs over the next few months and I'd like to know more about what the community desires as far as styles and models.
Cast your votes here and let me know your target price. I'm looking for a $50 - $250 range for packs that have enough options to export hundreds if not thousands of completely unique models.
More about the Indie 2.0 Revolution
I'll be giving a presentation at IMGDC at the end of this month covering more of the Indie 2.0 Revolution.
There's a whole lot more to come, including some revolutions in terrain generation and editing, BSP modeling, game engine and game development frameworks, a revolutionary MMO framework and much more.
If you can't make it then I'll be sure to post blogs afterwards, but I hope to see you there!
About the author
I am the founder of IndieZen.org, a website dedicated to the Indie 2.0 Revolution where a number of Indie game development studios and individuals collaborate and share a suite of custom built open source game development tools and middleware.
#3
03/14/2008 (5:55 am)
dang.
#4
- Edit scripting (like in IDE for each scripting language supported)
- Program shaders and test test them on custom models
- Zspace: modelling, texturing, animating (bones based?), importing/exporting models (in a variety of formats?)
That would be a product you sell or will you only be selling content packs you create with IndieZen?
03/14/2008 (6:37 am)
So , IndieZen is some sort of Content Creation tool that allows the user to:- Edit scripting (like in IDE for each scripting language supported)
- Program shaders and test test them on custom models
- Zspace: modelling, texturing, animating (bones based?), importing/exporting models (in a variety of formats?)
That would be a product you sell or will you only be selling content packs you create with IndieZen?
#5
03/14/2008 (7:39 am)
I think what you're working on is very interesting, especially what you seem to be alluding to in the realm of procedural content (nice to see that I'm not the only crazy person out there working on this).
#7
Eventually, before the end of the year, we will be selling other plugins like the ZSpace plugin, the ZTerrain plugin, etc for prices that are very competitive with existing products. More expensive than Blender, but less expensive than ZBrush.
@Jonathan - I don't have any great up to date screenshots, but I do have some early work-in-progress shots if you really want to see them. Mostly these are screenshots of my customized Eclipse with the ZSpace and ZTerrain plugins as well as the PyDev, Shader and TorqueEDIT open source plugins.
I've been afraid to distribute them because as soon as I do, someone else is going to throw together a custom Eclipse distribution and try to mimic what I'm doing, but I guess it's time to start showing them off, so here goes:
This is a screenshot showing me editing TorqueScript for my MMO game Fractured Universe, along with Josh Ritter's MMO Kit written in Python.
This image shows an imported MakeHuman mesh with some custom morph targets applied, but still at the highest LOD.
This screenshot shows an untextured terrain that I generated as well as a GLSL example water shader.
03/14/2008 (10:02 am)
@Hokuto - We'll be selling content packs first as I will be using that as a way to test the ZSpace plugin.Eventually, before the end of the year, we will be selling other plugins like the ZSpace plugin, the ZTerrain plugin, etc for prices that are very competitive with existing products. More expensive than Blender, but less expensive than ZBrush.
@Jonathan - I don't have any great up to date screenshots, but I do have some early work-in-progress shots if you really want to see them. Mostly these are screenshots of my customized Eclipse with the ZSpace and ZTerrain plugins as well as the PyDev, Shader and TorqueEDIT open source plugins.
I've been afraid to distribute them because as soon as I do, someone else is going to throw together a custom Eclipse distribution and try to mimic what I'm doing, but I guess it's time to start showing them off, so here goes:
This is a screenshot showing me editing TorqueScript for my MMO game Fractured Universe, along with Josh Ritter's MMO Kit written in Python.
This image shows an imported MakeHuman mesh with some custom morph targets applied, but still at the highest LOD.
This screenshot shows an untextured terrain that I generated as well as a GLSL example water shader.
#8
03/14/2008 (10:05 am)
Oops, they were too big so I reduced them in size... refresh if they show up too big for you. 
#10
I think you probably know this already as you have great personal experience with content packs yourself. But for the record these are some issues that we encounter when we use other people's content packs, or in some cases purchase a content pack that sounds great only to discover we can't use the content.
#1) Source Models in a USEABLE format! Sorry but Blitz3D is not useable to MOST people (some yes, but not most). Many indies use Milkshape, if the source models can't be made available in Milkshape, it also imports .3DS which has become fairly industry standard. .OBJ is an option for models that don't have animation but as you know .OBJ doesn't preserve animations, so .Blend .MS3D and .3DS are preferred source formats. Sometimes a model in a content pack might be "close but not quite" and a simple tweak in an editor would transform a model from not useable to functional.
#2) LOD on EVERYTHING! Yes even the little 25 polygon plants... why? Because without LOD the game will ALWAYS render it even when the little plant is not in sight of the player, and if you have 1000 of those little plants in your game that is a LOT of needless rendering going on. PLEASE PLEASE encourage LOD on every model created, both for DTS and DIF style models. If the model has more than 1000 polies then it should have at least 3 levels of LOD. This is so important!
#3) Variety of textures. A lot of times the default texture of a model is the only thing that makes it unacceptable for use in the game. It might have LOD, great animation, and sharp next-generation graphics complete with a normal map, but the color is off, or the style is wrong for the period of the game it's to be used in. If there were more options available, perhaps by using the skinning method (the 3DDiggers horse pack is a good example of this) where the user can choose a variety of textures for a single model, and even create additional textures of their own by following the naming convention used in the skinning method, this VASTLY increases the likelihood that a particular model will be useful to a developer.
#4) Normal Maps to match the textures. This is less critical, because in most cases it is possible to create a normal map in Photoshop using the Nvidia plugin when the source texture is available. But if normal maps can be created for the models it save the developer that much more time. And when you're working with a volunteer team where people only have maybe 5 or 10 hours a week to contribute to the project, time is everything!
#5) Standard BVH skeletons for animation. If models are created with a standard MoCap skeleton that will accept new BVH animations, it incereases the liklihood that the model can be used in the Developers game, even if the default animations don't meet the needs or expectations of the developer. If models in a content pack can be tweaked and modified it still saves the developer time in the creation process, rather than making everything from scratch. Animations can be found in .BVH format more plentifully than any other format, and there are converters for .BVH format also making it the most preverable for Indies. However a lot of modelers use custom skeletons, and so the .BVH animations won't import properly. If this was taken into consideration it would increase the usefulness of content packs all the more.
#6) This probably goes without saying, but Game models need to be low poly. We've encountered on occasion some REALLY NICE looking models, that were 10k to 20k polies. Obviously we can't use these in the game as-is. Most artists feel it's easier to start over than to try and poly-reduce a high count model, depending on the subject matter. So it's important to keep in mind the requirements for game models, even with LOD, shouldn't be burdened with too many polies.
Hope this input helps. Good luck with your venture!
Blessings to you,
-Sparkling
http://www.visionsgame.com
03/14/2008 (12:46 pm)
Hey Tony! Great blog!I think you probably know this already as you have great personal experience with content packs yourself. But for the record these are some issues that we encounter when we use other people's content packs, or in some cases purchase a content pack that sounds great only to discover we can't use the content.
#1) Source Models in a USEABLE format! Sorry but Blitz3D is not useable to MOST people (some yes, but not most). Many indies use Milkshape, if the source models can't be made available in Milkshape, it also imports .3DS which has become fairly industry standard. .OBJ is an option for models that don't have animation but as you know .OBJ doesn't preserve animations, so .Blend .MS3D and .3DS are preferred source formats. Sometimes a model in a content pack might be "close but not quite" and a simple tweak in an editor would transform a model from not useable to functional.
#2) LOD on EVERYTHING! Yes even the little 25 polygon plants... why? Because without LOD the game will ALWAYS render it even when the little plant is not in sight of the player, and if you have 1000 of those little plants in your game that is a LOT of needless rendering going on. PLEASE PLEASE encourage LOD on every model created, both for DTS and DIF style models. If the model has more than 1000 polies then it should have at least 3 levels of LOD. This is so important!
#3) Variety of textures. A lot of times the default texture of a model is the only thing that makes it unacceptable for use in the game. It might have LOD, great animation, and sharp next-generation graphics complete with a normal map, but the color is off, or the style is wrong for the period of the game it's to be used in. If there were more options available, perhaps by using the skinning method (the 3DDiggers horse pack is a good example of this) where the user can choose a variety of textures for a single model, and even create additional textures of their own by following the naming convention used in the skinning method, this VASTLY increases the likelihood that a particular model will be useful to a developer.
#4) Normal Maps to match the textures. This is less critical, because in most cases it is possible to create a normal map in Photoshop using the Nvidia plugin when the source texture is available. But if normal maps can be created for the models it save the developer that much more time. And when you're working with a volunteer team where people only have maybe 5 or 10 hours a week to contribute to the project, time is everything!
#5) Standard BVH skeletons for animation. If models are created with a standard MoCap skeleton that will accept new BVH animations, it incereases the liklihood that the model can be used in the Developers game, even if the default animations don't meet the needs or expectations of the developer. If models in a content pack can be tweaked and modified it still saves the developer time in the creation process, rather than making everything from scratch. Animations can be found in .BVH format more plentifully than any other format, and there are converters for .BVH format also making it the most preverable for Indies. However a lot of modelers use custom skeletons, and so the .BVH animations won't import properly. If this was taken into consideration it would increase the usefulness of content packs all the more.
#6) This probably goes without saying, but Game models need to be low poly. We've encountered on occasion some REALLY NICE looking models, that were 10k to 20k polies. Obviously we can't use these in the game as-is. Most artists feel it's easier to start over than to try and poly-reduce a high count model, depending on the subject matter. So it's important to keep in mind the requirements for game models, even with LOD, shouldn't be burdened with too many polies.
Hope this input helps. Good luck with your venture!
Blessings to you,
-Sparkling
http://www.visionsgame.com
Torque 3D Owner Rodney (OldRod) Burns