Integrated Tools
by Kyle Carter · in Torque Game Engine · 12/29/2003 (1:55 pm) · 17 replies
So, here's a hypothetical...
Suppose someone were developing a toolkit for Torque, to replace all the existing tools. I can see a few different alternatives...
A GNU/POSIX style toolchain that focuses mostly on converting from existing formats like 3d studio and .map to Torque's formats? Pros - easy to automate and port and maintain, cons - requires external applications for editing tasks.
Tools integrated into Torque? This is nice in that you get a complete editing suite for free with Torque. The downside is the tools won't be as nice as standalone versions of the same thing, and the GUI toolkit will probably be clunky for a while. Also, unless someone writes a very funky platform layer, you'll be doing all your editing in an OpenGL window. So no native GUI elements. This isn't a big deal for, say, buttons, but for file open dialogs it could be a pain.
Tools integrated into a 3rd party suite or suites? For instance, I think that getting Eclipse to edit TorqueScript would be incredibly useful. But for other things like 3d work - CSG and the like - there aren't any really good frameworks that aren't costly. The plus here is that the 3rd party suite will probably be well maintained, and is liable to use niceties like the OS's file opener.
Of course, I've probably missed something... and not all of these approaches will work. So what would work best for you? What needs are you finding that the existing tools don't meet? What neat things do you want to do that you can't - or conversely, what neat things do the current approaches let you do?
Suppose someone were developing a toolkit for Torque, to replace all the existing tools. I can see a few different alternatives...
A GNU/POSIX style toolchain that focuses mostly on converting from existing formats like 3d studio and .map to Torque's formats? Pros - easy to automate and port and maintain, cons - requires external applications for editing tasks.
Tools integrated into Torque? This is nice in that you get a complete editing suite for free with Torque. The downside is the tools won't be as nice as standalone versions of the same thing, and the GUI toolkit will probably be clunky for a while. Also, unless someone writes a very funky platform layer, you'll be doing all your editing in an OpenGL window. So no native GUI elements. This isn't a big deal for, say, buttons, but for file open dialogs it could be a pain.
Tools integrated into a 3rd party suite or suites? For instance, I think that getting Eclipse to edit TorqueScript would be incredibly useful. But for other things like 3d work - CSG and the like - there aren't any really good frameworks that aren't costly. The plus here is that the 3rd party suite will probably be well maintained, and is liable to use niceties like the OS's file opener.
Of course, I've probably missed something... and not all of these approaches will work. So what would work best for you? What needs are you finding that the existing tools don't meet? What neat things do you want to do that you can't - or conversely, what neat things do the current approaches let you do?
#2
As for a list of what tools I think are needed, what tools I could make use of, etc... do you REALLY want that list? :-) There's a limit of 4k for posts on here... ;-) I've got a whole lotta "Damn it would make life easier if..."'s on my personal list of ToDo's for TorqueIDE (and I'm slowly building up information for doin' 'em too.)
I also don't think there's a single solution to the questions you've presented - what I do with TorqueIDE, for instance, isn't going to be the solution that someone wants for thier development project. And, I'm all for multiple ways of addressing the same problems - while it replicates efforts, Choice Is Good!!
12/29/2003 (3:03 pm)
Same goes for other things, like using MilkShape 3D for editing (it's decent at it's job) and using an exporter to get it into Torque - it works, it enables us to get the job done, and it's already there. Same goes for the 3DS exporter. They aren't as efficent as they could be, workflow wise.As for a list of what tools I think are needed, what tools I could make use of, etc... do you REALLY want that list? :-) There's a limit of 4k for posts on here... ;-) I've got a whole lotta "Damn it would make life easier if..."'s on my personal list of ToDo's for TorqueIDE (and I'm slowly building up information for doin' 'em too.)
I also don't think there's a single solution to the questions you've presented - what I do with TorqueIDE, for instance, isn't going to be the solution that someone wants for thier development project. And, I'm all for multiple ways of addressing the same problems - while it replicates efforts, Choice Is Good!!
#3
I would rather have a plugin to say Lightwave for example that allowed me to "preview" my work in the render engine portion of the game engine seemlessly and test it right then and there. Then just like a "print preview" have a button that says "save" and you have your "exported" model.
Same with a level/world/mission editor. Click one button and preview it in the game like Worldcraft/Hammer lets you do in Half-Life. It is not the "external"ness of the tools it is the lack of automation and integration.
Coming from an Amiga background with AREXX I am really spoiled for things like that. That is why I like Lightwave, you model in a modeler and animate in the animation tool, no overlay and they work together seemlessly.
There are a couple of game engines that use Maya for the entire game development process. The game engine is an extension to Maya and all the scripting and everything is done by attaching Python scripts to objects in Maya. Since its renderer is plugable they just plugged in their renderer which just happened to be their game engine.
12/29/2003 (3:48 pm)
I would rather flip the problem around and look at it the other way around. The content exporters should use the exact same code as the engine. I doubt you are going to create as fully a featured 3d modeler as milkshape much less Lightwave in Torque. Much less creating all the supporting documentation, tutorials and other must haves to make something useful for someone other than yourself.I would rather have a plugin to say Lightwave for example that allowed me to "preview" my work in the render engine portion of the game engine seemlessly and test it right then and there. Then just like a "print preview" have a button that says "save" and you have your "exported" model.
Same with a level/world/mission editor. Click one button and preview it in the game like Worldcraft/Hammer lets you do in Half-Life. It is not the "external"ness of the tools it is the lack of automation and integration.
Coming from an Amiga background with AREXX I am really spoiled for things like that. That is why I like Lightwave, you model in a modeler and animate in the animation tool, no overlay and they work together seemlessly.
There are a couple of game engines that use Maya for the entire game development process. The game engine is an extension to Maya and all the scripting and everything is done by attaching Python scripts to objects in Maya. Since its renderer is plugable they just plugged in their renderer which just happened to be their game engine.
#4
Couple things I have observed:
* Artists tend to be very attached to their tools and don't lightly change or learn new ones. Sure the same basic skills apply no matter the tool but they still seem to get attached and comfortable.
* Max, Maya, and the like are years ahead of you already in development and have dedicated teams of people constantly striving to improve them. You will never really be able to compete on their level.
* First rule of effective tools development is provide multiple ways of accomplishing the same thing. this allows everyone to find their own way of doing things. Having support for internal and external tools at the same time is going to make far more people happy.
As far as modelling goes I would say forget it. A reasonably featured modeller is going to be a *lot* of work and is unlikely to even reach Milkshape's level. With Blender making such advances lately you really don't have to worry about cost or cross-platform anymore.
A good CSG editor might be worthwhile. There really isn't any competition for commercial use except Quark and it isn't very friendly (interface or platform support). A CSG editor wouldn't be too bad to code...far easier than a true modeller. Torque is an ideal platform for something like this.
12/29/2003 (9:41 pm)
I have been on both sides of this at various times.Couple things I have observed:
* Artists tend to be very attached to their tools and don't lightly change or learn new ones. Sure the same basic skills apply no matter the tool but they still seem to get attached and comfortable.
* Max, Maya, and the like are years ahead of you already in development and have dedicated teams of people constantly striving to improve them. You will never really be able to compete on their level.
* First rule of effective tools development is provide multiple ways of accomplishing the same thing. this allows everyone to find their own way of doing things. Having support for internal and external tools at the same time is going to make far more people happy.
As far as modelling goes I would say forget it. A reasonably featured modeller is going to be a *lot* of work and is unlikely to even reach Milkshape's level. With Blender making such advances lately you really don't have to worry about cost or cross-platform anymore.
A good CSG editor might be worthwhile. There really isn't any competition for commercial use except Quark and it isn't very friendly (interface or platform support). A CSG editor wouldn't be too bad to code...far easier than a true modeller. Torque is an ideal platform for something like this.
#5
A little "show" tool plugin for Max/Maya/etc as Jarrod would be ideal. A plugin that made it possible to instantly output a preview - or better yet, put a Torque rendered window into the editor (don't know if that is possible even in MEL).
However, it would not be as difficult to create a fairly well constructed CSG editor - at least something on par or better than Quark - so that we can build levels in real-time, in engine, and place entities and dts objects in them as we do so. This would save a lot of time, and would not be nearly the hit to flexibility that trying to integrate a dts modeler would be.
01/02/2004 (3:15 pm)
I agree with Matthew - making a dts editor in game would be insane. In order to support a level of functionality that would make it worth the switch from Max/Maya, you'd probably be writing an spplication almost as big as the engine itself! (possibly an exaggeration, but it would be a HUGE undertaking)A little "show" tool plugin for Max/Maya/etc as Jarrod would be ideal. A plugin that made it possible to instantly output a preview - or better yet, put a Torque rendered window into the editor (don't know if that is possible even in MEL).
However, it would not be as difficult to create a fairly well constructed CSG editor - at least something on par or better than Quark - so that we can build levels in real-time, in engine, and place entities and dts objects in them as we do so. This would save a lot of time, and would not be nearly the hit to flexibility that trying to integrate a dts modeler would be.
#6
There are screeshots in a thread somewhere :)
01/03/2004 (10:43 am)
Alex : Joshua Ritter has done this with Blender and his tge & python stuff for .dts models, ie tge as python module iirc that is loaded in the Blender python environment. There are screeshots in a thread somewhere :)
#7
With all these software packages and new exporters flying around, its hard to keep up with things!
01/05/2004 (12:26 pm)
@Nicolas: well cool :)With all these software packages and new exporters flying around, its hard to keep up with things!
#8
If you guys are possesing the mental skills of actually pulling something like internal DTS editing off, would it not be better to focus those skills on producing rock-hard exporters for the major tools already in public (Maya, LightWave etc.) since the ones we have at the moment are far from complete.
If I may be so bold to shamelessly plug Maya as THE tool to use, a solution described by Jarrod would be close to a godsend. I've seen examples of the integrated workflow in Maya and believe me it is streamlined beyond your wildest dreams :-)
Altough I must admit, the CSG editing is really lousy for all games but Unreal, heck I myself are having a hard time getting comfortable with Quark since I mainly have worked with Worldcraft/Hammer. I am currently about to look into Gamelevel builder for MAX since I am used to working in MAX aswell (dam these exporter issues hehe), thus falling back to a workflow I am used to as stated by Matthew Fairfax.
Ohh, I almost forgot, Torque rocks !!
01/07/2004 (11:53 am)
IMHO, external solution tends to be defacto standard in every major game-pipeline and trying to change the way of thinking alone would be a challenge.If you guys are possesing the mental skills of actually pulling something like internal DTS editing off, would it not be better to focus those skills on producing rock-hard exporters for the major tools already in public (Maya, LightWave etc.) since the ones we have at the moment are far from complete.
If I may be so bold to shamelessly plug Maya as THE tool to use, a solution described by Jarrod would be close to a godsend. I've seen examples of the integrated workflow in Maya and believe me it is streamlined beyond your wildest dreams :-)
Altough I must admit, the CSG editing is really lousy for all games but Unreal, heck I myself are having a hard time getting comfortable with Quark since I mainly have worked with Worldcraft/Hammer. I am currently about to look into Gamelevel builder for MAX since I am used to working in MAX aswell (dam these exporter issues hehe), thus falling back to a workflow I am used to as stated by Matthew Fairfax.
Ohh, I almost forgot, Torque rocks !!
#9
On enhancing the current exporters:
Making the existing exporters better would certainly help beginners, but I'm not sure how much it would help the more advanced members. By advanced, I mean those people that have worked with the exporters for "awhile" - I would imagine they've developed a process to get their objects/maps converted and into a level.
The problem for beginners is the learning curve of getting everything setup in order to do your first (and subsequent) exports. There's building WAD files, setting up directories, copying files (exporters and finished shapes), etc.
On creating built-in tools:
As pointed out, this is not going to be a trivial task. The current 3D modelers/mappers (Milkshape, 3DStudio Max, Hammer, Quark) are fairly significant projects with a lot of features. Recreating Torque specific versions is probably not worth the effort.
However, a simplified version of an editor might do a lot of people some good. I don't think you'd want to recreate the entire 3DS interface, but allowing someone to tweak their models inside Torque would be useful (maybe just tweaking vertices for instance). This doesn't come without other issues - if you tweak things for an hour in Torque and get the model the way you want, but then decide you need a few more tweaks that 3DS could do, you'd need/want a way to convert your DTS shape back into 3DS format. Again, not an easy task.
On creating a preview of Torque inside a tool:
Meaning, a way to view your current model as Torque would show it. This would not be unlike the Show Tool, but possibly built into the 3D modelling software. As pointed out, it would be nice to just press a button (or go through a dialog) that takes your currently edited model/map and shows how it would look in Torque. If nothing else, maybe a Torque mod that launches a dummy mission and loads the model/map. Combined with a streamlined exporter, it could be pretty useful.
@Jorgen:
I can only imagine (literraly) how nice Maya is - both in terms of modelling and a streamlined workflow. At my day job, I get ALL the best tools for doing my business work.
But, as this is "Garage Games", I think a lot of us here don't have the resources to buy Maya. 3D Studio is still a good chunk of change for many of us independent, weekend warrior developers. I don't feel I have the "right" to ask anyone to build me a cool, free, integrated 3D modelling tool into Torque. I also know my budget won't allow a *nice* 3D modeller, so I'll take what I can get. But, I would that when GG gets enough money in to start developing the Tools a little more, that they don't JUST concentrate on integrated tools for the bigger, better - and expensive - 3D tools. Don't forget about us small frys :)
-Ner
01/07/2004 (3:42 pm)
I have a few thoughts to add.On enhancing the current exporters:
Making the existing exporters better would certainly help beginners, but I'm not sure how much it would help the more advanced members. By advanced, I mean those people that have worked with the exporters for "awhile" - I would imagine they've developed a process to get their objects/maps converted and into a level.
The problem for beginners is the learning curve of getting everything setup in order to do your first (and subsequent) exports. There's building WAD files, setting up directories, copying files (exporters and finished shapes), etc.
On creating built-in tools:
As pointed out, this is not going to be a trivial task. The current 3D modelers/mappers (Milkshape, 3DStudio Max, Hammer, Quark) are fairly significant projects with a lot of features. Recreating Torque specific versions is probably not worth the effort.
However, a simplified version of an editor might do a lot of people some good. I don't think you'd want to recreate the entire 3DS interface, but allowing someone to tweak their models inside Torque would be useful (maybe just tweaking vertices for instance). This doesn't come without other issues - if you tweak things for an hour in Torque and get the model the way you want, but then decide you need a few more tweaks that 3DS could do, you'd need/want a way to convert your DTS shape back into 3DS format. Again, not an easy task.
On creating a preview of Torque inside a tool:
Meaning, a way to view your current model as Torque would show it. This would not be unlike the Show Tool, but possibly built into the 3D modelling software. As pointed out, it would be nice to just press a button (or go through a dialog) that takes your currently edited model/map and shows how it would look in Torque. If nothing else, maybe a Torque mod that launches a dummy mission and loads the model/map. Combined with a streamlined exporter, it could be pretty useful.
@Jorgen:
I can only imagine (literraly) how nice Maya is - both in terms of modelling and a streamlined workflow. At my day job, I get ALL the best tools for doing my business work.
But, as this is "Garage Games", I think a lot of us here don't have the resources to buy Maya. 3D Studio is still a good chunk of change for many of us independent, weekend warrior developers. I don't feel I have the "right" to ask anyone to build me a cool, free, integrated 3D modelling tool into Torque. I also know my budget won't allow a *nice* 3D modeller, so I'll take what I can get. But, I would that when GG gets enough money in to start developing the Tools a little more, that they don't JUST concentrate on integrated tools for the bigger, better - and expensive - 3D tools. Don't forget about us small frys :)
-Ner
#10
The more that gets integrated into Torque for editor use, is more I have to rip out of the codebase for my release builds of the game. I dont need an extra meg in my exe for code I'm not gonna use in my game after its released.
01/07/2004 (4:05 pm)
Just like to point something that some may not be thinking about...The more that gets integrated into Torque for editor use, is more I have to rip out of the codebase for my release builds of the game. I dont need an extra meg in my exe for code I'm not gonna use in my game after its released.
#11
Look at the realistic scenario. People using Torque want to build commercial games using a professional engine.
That means that you want top notch gfx and models
Meaning you get a hold of "real" modellers - and they use Max, Maya or similar tools.
So these MUST be kept in their environment. You cannot tell a modeller that he has to learn some ingame tool that is way inferior (or just different) to what he usually uses.
For the casual "I want to try out a few things" a ingame modeller might be good enough, but face it. For any real project you will want to use real tools.
Other engines have plugins for the modellers for preview - thats what we should have too. A revamped show tool might be good enough.
01/08/2004 (2:25 am)
I also favour the external tool situation.Look at the realistic scenario. People using Torque want to build commercial games using a professional engine.
That means that you want top notch gfx and models
Meaning you get a hold of "real" modellers - and they use Max, Maya or similar tools.
So these MUST be kept in their environment. You cannot tell a modeller that he has to learn some ingame tool that is way inferior (or just different) to what he usually uses.
For the casual "I want to try out a few things" a ingame modeller might be good enough, but face it. For any real project you will want to use real tools.
Other engines have plugins for the modellers for preview - thats what we should have too. A revamped show tool might be good enough.
#12
I am sorry, got carried away there for a while since I am fortunate to enough to be working with Maya. I understand fully that a indie-game-creator lacks the fund to buy even the cheapest working version for $1999 for a project that may not even see the light of day.
In my case I've been working on "related" projects for some time and since I already posses the software, denying that would be plain dumb.
That is where I agree with Thomas Lund, it is my understanding that there are alot of people in these forums that already have a professional 3d-software at hand, may it be a old version of 3DsMax or even something as "cheap" as Caligari GameSpace and starting development on in-game tools as described above in favor of pro-exporters would be a shoot in the foot.
I don't mean for this to be another whiny-post, I am actually quite happy with the completeness of Torque. It has allowed me (the artist) to actually complete stuff on my own without resorting to C++ gurus or what-have-u. Even tough this is supposed to be Garage, the product is amazing (especially compared to other garage-enginges).
Extending the possibilites to debug your game is ok, as described by Davis Ray Sickmon, but going so far as to have a "3D Modeler" ingame, naaah that is to bloat a perfectly fine piece of software. My basic programming knowledge says: if there is software available that is as good or better than what can achieve, don't start another project.
01/08/2004 (4:40 am)
@Nerseus:I am sorry, got carried away there for a while since I am fortunate to enough to be working with Maya. I understand fully that a indie-game-creator lacks the fund to buy even the cheapest working version for $1999 for a project that may not even see the light of day.
In my case I've been working on "related" projects for some time and since I already posses the software, denying that would be plain dumb.
That is where I agree with Thomas Lund, it is my understanding that there are alot of people in these forums that already have a professional 3d-software at hand, may it be a old version of 3DsMax or even something as "cheap" as Caligari GameSpace and starting development on in-game tools as described above in favor of pro-exporters would be a shoot in the foot.
I don't mean for this to be another whiny-post, I am actually quite happy with the completeness of Torque. It has allowed me (the artist) to actually complete stuff on my own without resorting to C++ gurus or what-have-u. Even tough this is supposed to be Garage, the product is amazing (especially compared to other garage-enginges).
Extending the possibilites to debug your game is ok, as described by Davis Ray Sickmon, but going so far as to have a "3D Modeler" ingame, naaah that is to bloat a perfectly fine piece of software. My basic programming knowledge says: if there is software available that is as good or better than what can achieve, don't start another project.
#13
Based on what I've seen here, it seems like the best approach is to have good integration with the strong existing tools like Max and Maya via plugins, and then for weaker areas, like CSG, to integrate it into Torque.
Personally, I don't think that I like having all my text editing in Torque - there are a lot of apps out there that do text editing extremely well, that would be better to leverage, just as we do with Max et al. But that's just my opinion on the matter. I use vi willingly, for pete's sake, so don't take me too seriously. ;)
In general, I also like the idea of assets flowing almost exclusively "into" the engine. A sort of creative control thing, I guess - like with films, it should be much simpler to watch than to edit. Not to say editing should be impossible...
01/08/2004 (12:20 pm)
Hey guys, thanks for all the thoughts - keep them coming!Based on what I've seen here, it seems like the best approach is to have good integration with the strong existing tools like Max and Maya via plugins, and then for weaker areas, like CSG, to integrate it into Torque.
Personally, I don't think that I like having all my text editing in Torque - there are a lot of apps out there that do text editing extremely well, that would be better to leverage, just as we do with Max et al. But that's just my opinion on the matter. I use vi willingly, for pete's sake, so don't take me too seriously. ;)
In general, I also like the idea of assets flowing almost exclusively "into" the engine. A sort of creative control thing, I guess - like with films, it should be much simpler to watch than to edit. Not to say editing should be impossible...
#14
I've come to the conclusion by reading the forums that most developers seem to aim their use of TGE for commercial game-production.
While there is nothing wrong with this I have been looking for something that I could use to prototype game-ideas very quickly.
I've been meaning to have a look at Virtools Dev but never had the time..
As I've already stated it takes good exporters but it also needs a very sweet interface with something like presets for the most common actions you'd like in different types of games.
I've personally have been sketching on something resembling a flowchart where you could insert or move around pieces as necessary, where every piece contains code, making it very simple changing certain aspects of your prototype.
This could be a completly new marketshare for GarageGames since rapid prototyping are quickly becoming essential for every entertainment-producer.
I could probably discuss details for a long time but this is enough to state the general idea.. don't you think ?
01/10/2004 (11:33 am)
Hey Ben, something that escaped me during my initial response:I've come to the conclusion by reading the forums that most developers seem to aim their use of TGE for commercial game-production.
While there is nothing wrong with this I have been looking for something that I could use to prototype game-ideas very quickly.
I've been meaning to have a look at Virtools Dev but never had the time..
As I've already stated it takes good exporters but it also needs a very sweet interface with something like presets for the most common actions you'd like in different types of games.
I've personally have been sketching on something resembling a flowchart where you could insert or move around pieces as necessary, where every piece contains code, making it very simple changing certain aspects of your prototype.
This could be a completly new marketshare for GarageGames since rapid prototyping are quickly becoming essential for every entertainment-producer.
I could probably discuss details for a long time but this is enough to state the general idea.. don't you think ?
#15
But I haven't used them, so I can't say. Of course, a RAD game prototyper would be really neat. But maybe it's not really needed? It seems like the Bravetree guys get along fine prototyping in Torque... Of course, they're perhaps a special case compared to many of us. :)
It is an interesting idea, though.
01/10/2004 (12:33 pm)
Do you think that products like Conitec's engine, the various Basics and so forth, would cover this niche? That was the only real use I could envision for them ;)But I haven't used them, so I can't say. Of course, a RAD game prototyper would be really neat. But maybe it's not really needed? It seems like the Bravetree guys get along fine prototyping in Torque... Of course, they're perhaps a special case compared to many of us. :)
It is an interesting idea, though.
#16
01/12/2004 (5:04 am)
I've had no problems prototyping in TGE either ;) but that might just be experience at work.
#17
01/31/2004 (9:40 am)
I might make a Maya2Dif exporter tool pack, sell it for 1500 or so a copy - maybe I could afford maya after all. 
Torque Owner Davis Ray Sickmon, Jr
Default Studio Name
I'd rather see tools integrated into Torque rather than created externally. Why wouldn't it be as nice? To me, it's even better than an external tool or something adapted to deal with Torque. For instance: One of my long term goals is to edit .dif's and .dts files directly in Torque. No more putting up with exporters - it's right there. No more putting up with texture alignment issues (caused by the exporter) - when you align in the engine, you KNOW what the final result is going to be. Plus, then you are garanteed to always have your toolset at hand, no matter what platform you work on (Less important when dealing with stuff like Ron's Eclipse for Torque work - it's already cross-platform :-) This last point is becoming something very close to my heart, since I'm slowly trying to make my move to all cross-platform tools so I can move over to a wonderful Mac when I'm ready :-) While QuArK is quite usable, to this day the best editor I ever used was Unreal's - I KNEW, beyond a shadow of a doubt, what I just worked on was going to look EXACTLY the same in game.
As for not having native GUI elements - well, if you work with a good deal of Java based stuff, WxWindows, or other cross platform solutions, you don't nessisarily end up with native GUI Elements or requesters. Just a nature of the beast, unless someone deems it to be a high-priority item. Personally, I'd RATHER use the OSX 10.3 requester than the Windows file requester. It's much nicer. So even making it match the platform doesn't always make people happy - so, ignore that whole section of it. It's appearance, not functionality.
And of course, Torque has a lot of UI elements that are flaky, don't work like they should, etc (FOR THIS TASK - I just ammended that, since they work perfectly for writing games). Example: The ML Text Editor, at first seems quite usable - until you write an editor using it. I've had to extend the heck outta it to make it up to par with a 'real' editor (see my .plan for comments on what I did, and when it's going to be available.) Fixing UI elements up so they react and look like they should isn't that big of an issue - it just requires someone to do it. (And, gee - I'm already tweaking some of that ;-) And some day, I'm gonna re-write that bloody UI Editor from scratch as a new part for TorqueIDE. I know that it's better than what you get with pretty much any other engine, but, I was a bloody MMI (Man - Machine Interface) developer for a decade. Compaired to my normal toolset for UI development, well... it's lacking.
That's just my personal goals and beliefs (Almost to the point of "Religious Issue" to be honest about it ;-). I like integrated. I use an IDE for a reason, right? Well, to me that should continue to extend further and further for debugging - for instance, I should be able to double-click on an object in the Mission Editor, and it pops open up the datablocks attached to that object, and the functions that it makes use of. The further that stuff integrates, the more places I can get to my data from, the happier I am. I'm just funny that way :-)
BUT - there's also a lot to be said about extending existing tools like Eclipse - there you get a nice compiler, script editor, etc. all in one nice little package. And it's solid. Getting it as an all-in-one IDE for scripting and C++ work is a great thing.
(Ahem - next post. Sorry for the length.)