DIF Editor design documents
by FruitBatInShades · 11/12/2004 (7:38 pm) · 14 comments
Download Code File
I wanted to create a new brush editor because those that are out there are not as powerful or friendly as they could be. I started designing the basics but realised I did not have the math knowledge to code it myself. After trying to get other parties involved I decided to shelve it but thought someone else might find these docs useful.
They contain:
1. Basic application structure that enables easy building and modification of BSP models
2. 2D - 3D and back concepts
3. Basic class designs for the main operational classes.
4. Mini explanation of CSG
I wanted to create a new brush editor because those that are out there are not as powerful or friendly as they could be. I started designing the basics but realised I did not have the math knowledge to code it myself. After trying to get other parties involved I decided to shelve it but thought someone else might find these docs useful.
They contain:
1. Basic application structure that enables easy building and modification of BSP models
2. 2D - 3D and back concepts
3. Basic class designs for the main operational classes.
4. Mini explanation of CSG
About the author
#3
11/15/2004 (2:45 am)
If someone can convince me they can do it, i will provide a full detail design down to the base class structures. 
#4
and there are plenty others like gmax http://www.turbosquid.com/gmax from discreet (creators of 3d studio max).
I understand the point, but why not improve the ones that are out there? quark is open source, fix it if you think it needs improvement.
if you learn python you can add the powerful features that are missing.
maybe its just me but its sort of the same problem as why write your own graphics engine when there are so many that are free or that are inexpensive, like torque.
use what is out there as a way to jump start what your trying to accomplish. starting from scratch is so 90's lol.
11/15/2004 (7:15 am)
with so many free tools out there and others you can purchase its tough to justify spending time to write another editor. Quark might be flaky but it has 5 years worth of features in it. Hammer/Wolrdcraft are fairly well know and work fine.and there are plenty others like gmax http://www.turbosquid.com/gmax from discreet (creators of 3d studio max).
I understand the point, but why not improve the ones that are out there? quark is open source, fix it if you think it needs improvement.
if you learn python you can add the powerful features that are missing.
maybe its just me but its sort of the same problem as why write your own graphics engine when there are so many that are free or that are inexpensive, like torque.
use what is out there as a way to jump start what your trying to accomplish. starting from scratch is so 90's lol.
#5
11/15/2004 (7:27 am)
@Martian: The problem with the current editors is their architecture, that cannot be changed without a rewrite from scratch. GMax is not available for public use and the models you produce cannot be used in commercial games, plus Gmax is a poly soup editor, not CSG!
#6
you also mention they are not easy enough to use (um im a programmer not an artist or world building and I was able to figure out quark in no time and make objects and maps) so whats not easy about them and what are your suggestions to make them easier and easier for whom?
you docs dont go into this specific details they just talk about generic feature requirements. might help undestand what your trying to accomplish.
11/15/2004 (4:59 pm)
ok then how about being more specific. you mention other editors are not powerful enough yet they have been used to create dozens of games so what features are you referring to exactly?you also mention they are not easy enough to use (um im a programmer not an artist or world building and I was able to figure out quark in no time and make objects and maps) so whats not easy about them and what are your suggestions to make them easier and easier for whom?
you docs dont go into this specific details they just talk about generic feature requirements. might help undestand what your trying to accomplish.
#7
Certain things are impossible in quark and hammer without a LOT of fiddling and hacking of numbers in the brush arrays. Vertex editing is one area that neither editor can do sucessfully but should be easy to implement and would enable much more complicated geometry to be created with low tri counts.
I also envisage an editor where you can build complicated levels in less than a quarter of the time by using clever object management. The idea of intelligent brushes is new and would enable very quick model development and re-editing. Say for instance you have an Intelligent door brush that carves and entities a particluar type of door. In the other editors you have to brush subtract and thats it. done. What if you decide to swap that door brush for a different one? In the current editors you have to delete all those sections and do them all again. With the architexture I propose it would be a case of 'Swap Door Style One for Door Style Two'.
Now that is the basic premise of the editor, more power, more flexibility, easier to use and true functional libraries for day to day work.
11/15/2004 (11:41 pm)
Hammer is easy to use but lack power, quark has power but is difficult to use. Both of these editors have problems which arise from the architecture and the evolutionary way in which they have developed over their life cycle. A new editor would be a lot more powerful if its architecture was untied and objectified to a higher degree.Certain things are impossible in quark and hammer without a LOT of fiddling and hacking of numbers in the brush arrays. Vertex editing is one area that neither editor can do sucessfully but should be easy to implement and would enable much more complicated geometry to be created with low tri counts.
I also envisage an editor where you can build complicated levels in less than a quarter of the time by using clever object management. The idea of intelligent brushes is new and would enable very quick model development and re-editing. Say for instance you have an Intelligent door brush that carves and entities a particluar type of door. In the other editors you have to brush subtract and thats it. done. What if you decide to swap that door brush for a different one? In the current editors you have to delete all those sections and do them all again. With the architexture I propose it would be a case of 'Swap Door Style One for Door Style Two'.
Now that is the basic premise of the editor, more power, more flexibility, easier to use and true functional libraries for day to day work.
#8
11/16/2004 (10:07 am)
look at an application called SketchUp! that is how a CSG editor should behave!
#9
I wish that Garage Games would put out an official editor... The engine deserves one that can utilize all it's cool new features to the fullest.
11/16/2004 (10:32 am)
please keep going on this! Anything would be better than QuaRk or Hammer.... The best thing I have seen so far is DeleD.....and at this point (it's very beta) it's severely lacking in some features.I wish that Garage Games would put out an official editor... The engine deserves one that can utilize all it's cool new features to the fullest.
#10
11/16/2004 (11:30 am)
@Thomas: "I wish that Garage Games would put out an official editor... The engine deserves one that can utilize all it
#11
What is this 2d editor document about?? 2D map editor's went out when the doom days where over.
If you want to know a little about CSG goto collective.valve-erc.com/index.php?go=map_format. It explains the format of the half-life map files read by map2dif.exe. This page is what conviced me that I could make a map editor.
11/17/2004 (2:35 pm)
I am actually coding an interior editor called tInt. Wether or not I will finish it is another issue. It will not be anywhere near as complicated as the ideas you have here. What is this 2d editor document about?? 2D map editor's went out when the doom days where over.
If you want to know a little about CSG goto collective.valve-erc.com/index.php?go=map_format. It explains the format of the half-life map files read by map2dif.exe. This page is what conviced me that I could make a map editor.
#12
[quote]What is this 2d editor document about?? 2D map editor
11/22/2004 (8:07 am)
@Micheal: I've got lots of docs on the MAP format. The problem is the maths involved with intersecting rays, vectors etc.[quote]What is this 2d editor document about?? 2D map editor
#13
while its will be released im recommending to use Cartography Shop 4.1. I hope John Kabus`s official CSM2DIF konverter will be out soon, its in beta phase.
11/24/2004 (6:51 am)
Im fully supporting the idea of making an official BSP editor for the engine. Quark and hammer ...both of them are sucks.while its will be released im recommending to use Cartography Shop 4.1. I hope John Kabus`s official CSM2DIF konverter will be out soon, its in beta phase.
#14
Also, someone asked "Why not just take an editor thats open source and work with that?" There are a few reasons, A. The open sourced ones, such as Tread 3D have serious design flaws, and Quark which, i've heard its source is hard to prot, understand and work with. B. Either way, we'd be going back on progress. Tread 3D has serious source problems, and needs to be made compatible with torque, fully. (Version 2.3 Works OK, but Version 2.4 is real bad.) Also, quark, well that thing is ahrd to work with and hard to understand, youd be wasting time just learning it.
Now, some people ask. "Why not just use hammer?" Heres a few reasons. 1. Its under-powered, sure it has power, and does a great many things. However, when it comes to time, and features it lacks in many places Torque Developers would like. For example, I know alot of developers (Well, mainly me, and my team.) would like to expand the editor by plugins. Plugins/Mods to it expand the features, power and life of the editor, they also in soem cases help with time. Another feature I know I personally would love, in a file library with prefabs (Yes Hammer does have this. I'm refering to in any editor.), and sizes that are consistant with torque. The major reason people do not use hammer is because of the license. Valve made a promise (Atleast i think it was a promise, i know they said it) That they would allow Torque Developers to use Hammer, that was how long ago? I personally do not trust valve, and would like to stay away from them.
Now, the main reasons for a new map editor is: Cross-Platform, Plugins, Sizes that are the same as Torque, ease-ability to export the map into Torque, Features, (Personally i would like it to be themeable.), a easy on the eyes UI, a easy to understand UI, and most of all, it to be free (After all the greatest things in life are free.)
I have given up any work on Tread 3D, and as we speak am looking at possibilities to create a map editor, currently i'm experimenting with Java. If someone wants to help me, my contact info is: AIM: MMKBlackKnight MSN: killer_aka@hotmail.com
FruitInBatShades: I would love to work with you on this again, I think together we can do it right.
12/05/2004 (6:07 pm)
To those of you that wonder, Whats the point of a new map editor? (However I believe Fruitinbatshades addressed this). Most people don't realize this but time means alot more then money. Theres only so long someone can stay awake, and so much they can do. Indies know that time is important in many ways, because they're independent, alot i've seen/know don't have a steady income. Any time they can save, they should. Also, someone asked "Why not just take an editor thats open source and work with that?" There are a few reasons, A. The open sourced ones, such as Tread 3D have serious design flaws, and Quark which, i've heard its source is hard to prot, understand and work with. B. Either way, we'd be going back on progress. Tread 3D has serious source problems, and needs to be made compatible with torque, fully. (Version 2.3 Works OK, but Version 2.4 is real bad.) Also, quark, well that thing is ahrd to work with and hard to understand, youd be wasting time just learning it.
Now, some people ask. "Why not just use hammer?" Heres a few reasons. 1. Its under-powered, sure it has power, and does a great many things. However, when it comes to time, and features it lacks in many places Torque Developers would like. For example, I know alot of developers (Well, mainly me, and my team.) would like to expand the editor by plugins. Plugins/Mods to it expand the features, power and life of the editor, they also in soem cases help with time. Another feature I know I personally would love, in a file library with prefabs (Yes Hammer does have this. I'm refering to in any editor.), and sizes that are consistant with torque. The major reason people do not use hammer is because of the license. Valve made a promise (Atleast i think it was a promise, i know they said it) That they would allow Torque Developers to use Hammer, that was how long ago? I personally do not trust valve, and would like to stay away from them.
Now, the main reasons for a new map editor is: Cross-Platform, Plugins, Sizes that are the same as Torque, ease-ability to export the map into Torque, Features, (Personally i would like it to be themeable.), a easy on the eyes UI, a easy to understand UI, and most of all, it to be free (After all the greatest things in life are free.)
I have given up any work on Tread 3D, and as we speak am looking at possibilities to create a map editor, currently i'm experimenting with Java. If someone wants to help me, my contact info is: AIM: MMKBlackKnight MSN: killer_aka@hotmail.com
FruitInBatShades: I would love to work with you on this again, I think together we can do it right.
Torque Owner FruitBatInShades