The DTS Pipeline for Torque

Summer is here and that means that theres a whole new round of bitching, whining and complaining about Torque and the DTS pipeline that it uses for its 3D models.

Now, it's pretty common for people to hilite weak areas in something, I know that I do it sometimes with Torque because I hold it to a high standard for what I need from it (after all it is one of the tools that we use that allows us to pay our bills at Max Gaming), but the DTS pipeline is definately not one of Torque's weak spots. If anything its one of the strongest features that the engine has. As such I figured I would take a few moments to hilite what makes the DTS format so great and dispell many of the myths that people have about it...



What makes the DTS pipeline so great:

1. Its a Powerful format.

Lets face it, there has been almost 0 attention put towards forwarding the pipeline and its features/workflow since we got access to the tools 6+ years ago. In all that time, we have primarily just gotten recompiled plugins for our 3D tools from helpful community members. Yet the DTS pipeline has never faultered or had issues with changes to work flow and productivity that have come with our ever evolving 3D applications.

For example when the IK constraints were introduced in 3DS Max 4, the DTS exporters never once complained or had to be modified to use them, it simply used them. If you want to use a rigging system other then biped you can, you can utilize stock bones, or the plethora of 3rd party rigging solutions such as Puppetshop or CAT, you can even download and use someone elses highly complex skeletal rig from a movie and it would still export as is with the DTS.

Why is this? The simple aswer is that the pipeline and the format were built to be solid and rugged to handle whatever was thrown at it and look at it in the simplest of ways. Where is the bones position? How has its transforms changed? Where is this vertex and what data does it hold?

2. It allows for multiple team members to work in the art pipeline.

How so? All model geometry is exported as a propriety model format called a DTS. The DTS file points to the needed textures that it looks for, allowing for a texture artist to work on a custom bitmap texture without ever needing access to the models source. Animations can be included in the file, but for the most part are best left off as a DSQ, this allows an animator to create animations for the model that can then be used elsewhere on other models with a similar skeleton. Lastly you can have a effects artist working on pixel shaders for the model, they do this by doing some simple script editing and point the effects to various textures or materials that have been applied to the model.

All of this work can be done in a non-destructive manner working simply from the geometric data stored in the DTS.

3. It doesn't let the crap through.

There was initially a time when the pipeline let everything through and didnt bother error checking, this resulted in a lot of wasted resources because it made it difficult to track down where errors were coming from. Artists would blame the code, programmers would blame the art, people would run around like headless chickens. Ya it sounds funny but to the people that had to use the export at this time it was hell.

As such it was decided that the best way to resolve this problem and speed up productivity was to add in error checking. If there were problems, don't export the model. When the model is exported, check it in the show tool to ensure its working as it should. Sounds pretty simple, but yet this was and still is a novel approach for a lot of game engines and studios even today!

4. Additive blending animations & animation sharing.

Torque has, and in many cases still has, a strong competetive advantage with its additive animation system, (aka blend animations). To sum things up what this means is that animations can be exported with only their transform offset data included. So what happens is that it adds onto whatever is occuring to the model because the animation data is stored as absolute positions.

This handy little solution means that artists have to make less animations since the engine allows them to stack common effects onto each other. It also means that you can reuse and share animations across multiple models, another significant positive.

To me, animation in Torque is a lot of fun, you tackle animations one small step at a time and use the showtool to test out how it will all work together. Sometimes I will have to build whole new sets of animations, other tims I can just use the blending system to adjust the default stock animations that come with the engine to suit our needs.



Dispelling the Myths:

1. Its a difficult pipeline.

Actually the DTS pipeline is pretty simple, it just requires that the end user be a) competent at using their 3D application and b) has read all the documentation on how the export works and what it can do.

2. The documentation for it sucks.

The documentation is literally what it says it is, documenation. Its not just a tutorial that holds your hand like and walks you through some sample steps it tells you exactly what everything does, how it works, why it works and in the process gives you a better understanding of how the game engine itself works.

The documentation however is verbose and multiple pages. It presents a number of novel concepts that users new to game devleopment won't quite yet understand. But with that said it doesnt take long to read it, and its not hard for people to quickly get up to speed with it if the same amount of effort required to make game art is put into reading it.

3. People have to buy content pack art to learn how stuff works in the pipeline.

This is one of my big pet peeve myths about the DTS pipeline that makes me want to walk up to the person saying it and bitch slap them around until they realize what reality is. Say it with me people "content packs don't teach you anything new with regards to how to use the DTS pipeline, they are simply content that is provided for hobbyists and for quick demo/prototypes".

Honestly I do wonder where people come up with stuff like this, if everyone had to learn from a content pack then how on Earth did anyone learn how to make the content pack in the first place?

Im sorry but the honest truth of the matter is that people learn how the format works by reading the documentation, applying it to what they know and taking baby steps from there to do bigger and more complex things. That's how I learnt it, that's how hundreds of others have learnt it and I never worked for Dynamix nor needed sample source files to understand how things should be done and anyone else can learn it that way too, its very easy.

4. The hiearchy is too complicated.

Actually the hiearachy for a DTS is rather simple. It only looks confusing to the uneducated and can actually be descriped rather quickly.

Every DTS scene has a root dummy node of some sort that holds all the other parts that need to be exported. This root node is typically refered to your base01, but it can be called anything (hence why bip01 works too). From there all the detail markers (for LOD and Collision) must be linked to this root node. Any bones, dummy nodes or geometry is also linked off of the root node as well, but can be hierarchicly organized to make it visually easy to understand (hence the node commonly reffered to as "start01", but can be named anything) or for animation purposes.

If an item is not linked to this hierarchy then it will not export. Though the only exception to this rule are objects that have skeletal deformation (ie. skin modifier) applied to them.

That's all there is to know about the DTS hierarchy. Pretty simple huh?

World of Warcraft : Its Own Worst Enemy?

There has been a lot of talk in the gaming news for the past year just how any new MMO game is supposed to compete with the juggernaught that is World of Warcraft (WoW). I will freely admit I have read these articles because they discuss a game market that Max Gaming is actively investigating and to be honest I have to chuckle at the thoughts being thrown out there.

Some of the more common and I feel incorrect ideas include:

- More user control so that you do gestures and it influences the move or attack type that you do. This way after a five hour raid or dungeon run, you can be so mentally exhausted and sick and tired of flailing your mouse around with a Wiimote you won't want to play again for a number of days. (Sorry but I am a fan of the whole, the game does all the nitty gritty crap for you, you just tell it what to do, such as attack, or pick up an object).

- Looking East to Asian influences in order to find your own niche and avoice the beast all together. The problem that I see here is two fold. 1) Asian themes work well in Asia but not elsewhere. 2) Mixed themes just have trouble in both locations (read up on Tabula Rasa for more information, but basicly to sum it up they blew 2 years of development on trying this to find out it wasn't something that would work out well)

- Story! Story is the one word I think people throw around like its some sort of stopgap solution to every problem. Story is a double edged sword and most people haven't really got a clue how to best tame this beast. Yes story is good for creating an interest for the player to keep playing and move forward, so in that regards it is needed. How it is not needed, especially for an MMO, is to make a definative "end game" that the player is seeking to achieve. Why? Well the simple logic is that MMOs should always be evolving and changing, as such if you make the player feel as though they have completed the game, you are giving them a reason to stop playing. You want people to keep playing the core game (not the addons like PvP arenas) because the more people that play and continue to enjoy the game the more your online world feels rich and alive.

So how do you beat WoW?

IMHO you beat WoW in two seperate ways. 1) You create a great end user experiance that is fun and enjoyable by making your game something you can just fall into and play (ie. it should never be frustrating). 2) You simply let WoW continue on its course.

Creating a great game is obvious and its something that every game developer sets out to achieve, but lets look at what made WoW a success in the first place. There really is nothing unique or original to WoW, it is at its heart a rehash of ideas that exist in MMOs that have come before it, the key difference being that Blizzard made WoW an easy product to pick up and get into. The first 30 levels of play are incredibly polished, logical and fun and by that time you have either maxed out your free time and its not the game for you, or you are hooked and enjoy the simplicity and ease of use that allows you to enjoy the game and will continue to pay for another month. As such this is why I say that you have to "create a great end user experiance that is fun and enjoyable". You don't need to reinvent the wheel or even be all that innovative (innovation is a buzz word people), follow the simple lesson that WoW has taught by making a fun, polished experiance that allows the player to enjoy the game world. WoW might do this great, but its not the pinnacle or peak of the mountain, there is a lot of room to improve upon the experiance and playability that they have crafted and make it better.

My second point is probably confusing anyone reading this, "just let WoW continue on its course" likely seems like the dumbest idea in the world, but hear me out on this.

The biggest problem with WoW is its content from 30 to 70. Its difficult and annoying, it runs you all over the world and lets face it, its ok but its not the cats ass of design. At this point in the game theres a huge division between what are termed "casual" and "hardcore" players. Casual players for the most part are stuck in the game world, completing the same old quests, collecting gold and maybe if they are lucky get to do a dungeon run. Unfortunately though when they hit 70, the only thing they really have to achieve is PvP arenas and creating a new character to fill up their time. Hardcore players are the guys that play way too much, they have hit 70 and they have already completed most dungeons multiple times, have the uber gear and are already in the 10 to 25 man raid dungeons.

Unfortunately hardcore players are also the ones that WoW targets far too much and to make matters worse hardcore players only make up a fraction of the subscriber population. Case in point are almost all of the so-called "end game" raids, which take place at levels 60 or 70. Only a small fraction of these raids can be done without completing some overly difficult chain questline that requires an experianced dungeoning crew, something that nearly all casual players cannot do because it requires a different line of thinking. As such casual players are locked out of the epic quests in the game simply because they cannot become "hardcore" players because the seperation between the two groups is too vast.

Another significant problem with WoW is that there is a huge design balance issue with the game which will remind most geeks of the problems Wizards of the Coast has with Magic : The Gathering. What this problem is, is that there is little or no thought put into how changes or things they introduce in expansions will reflect backward onto previous content. For example in WoW there is a saying that "green is the new purple" which refers to the fact that most of the green items, which are less common but good rewards you find from quests or on kills, are better than most 60 level purple items, which were supposed to be world epics. This simple lack of attention has as a result caused much of the old "end game" content to be undesirable, sadly much of this content was stuff that players had never experianced and likely never will with the current design attitudes.

The biggest enemy for Blizzard with WoW is themselves. Yes the game is a beast and a juggernaught right now, but it will not be foreever and the reason for this won't simply be because a better MMO comes out that dwarfs it. It will lose out because a game will come in that will continue to provide the same ease of use and playbility but also better fills the holes that WoW is incapable of filling and thus provide a better fit for "the" MMO that people have to play.

Machinations of Doom

My brother Dave and his friend Scott headed out to Tayler, Michigan this past weekend to attend the Palladium Books "Open House", which is a yearly weekend of role-playing and other events hosted by Palladium Books.

At the show Palladium was offering for sale the forthcoming "Machinations of Doom" sourcebook for RIFTS that not only collects the very cool comic strip that Ramon Perez developed (this guys artwork is simply outstanding if you haven't seen his stuff) along with a whole bunch of interesting source and adventure material that is based on the comic. My brother managed to snag me a copy and got it personalized by Ramon Perez, so all in all I am pretty pleased.

Some quick notes about the book itself:
1. It comes in at some 100+ pages in the typical perfect square bound format that all Palladium books come in. Colour cover with black & white interior on the typical heavy stock paper.
2. The comic to source/adventure material is roughly a 50/50 split. While the comic itself is well worth the price of the book, the source/adventure material is also really well done and provides a wealth of great information, campaign ideas, NPCs and other quick notes that can make both GM and player incredibly happy.
3. I think this is the first book that goes away from the traditional Palladium format of two collums. Instead the book goes to a three collum format along with a nice and clean stylized border which works outstandinly well for a black and white book. TSR and other D20 companies should take note of what it takes to make a clean and professional looking black and white book.
4. $20 USD for a book like this is a god dammed steal of a deal!

An Indoor Pool?

Got a bit of a suprise yesterday when my wife found that the basement has begun to flood due to all the heavy rain. A quick check of the sump pump revealed a dead short in the wiring leading to the Sump Pump (which triggered the breaker to blow), the end result was that the sump wasn't getting any power to move the water out.

Thankfully the issue was resolved by running an extension cord from another plug to the sump and the water began to drain. Better yet is that nothing was ruined except for a couple of cardboard boxes and some old computer game manuals I wasn't even sure why I had kept.