Post Mortem - R.T.S Environment Pack

The Goal
To develop a content pack designed specifically for use with the Torque Game Engine and the RTS Starter Kit, which delivered quality content for the Garage Games community.

What I wanted to achieve with this project was to promote the RTS Starter kit, which seemed to be lacking any sort of community buzz, at least in comparison to T2d aka TGB. Considering the rich history and popularity of RTS games, I am a bit puzzled by this.

I also wanted to demonstrate that TGE was capable of delivering visuals that were comparable to RTS games currently on the market. And in the process, learn my way around the TGE development environment.

I don't know if I have managed to create any excitement about RTS development, time will tell. But I am definitely excited about it, more now than I was before.

As for demonstrating the potential of Torque, I would have to say I was partially successful. I learned a lot in the process of doing the pack, but mostly what I learned is that Torque has a lot more potential than what I demonstrated.

Measure twice, cut once...
Once the decision had been made to commit to the project and the subject matter was selected, I began to plan the development schedule.

Over the years I have formed a habit of doing most of my concept work in small note books. I carry them with me and sketch ideas as they come to mind. This has proven to be a very efficient method for me. I am able to workout all of the design details for various elements and dive right into development when I set down in front of the PC. Occasionally, this is even carried as far as working out the exact poly count of an object on paper, but that is the exception, not the rule.



In the weeks leading up to the first official day of the project, I had already roughed out a lot of material on paper, as a result of my exploring various subject matter for the content pack. I had things pretty well detailed, content list for each environment, development time estimates, texture counts, etc. The schedule was well defined, and I even had a 10 day buffer built in to allow for unexpected delays. So When the day came, I jumped in head first.

The plan was to progress through the environment pack in 2 stages. Stage 1 involved the initial development of all content and the rudimentary demo map development.

Stage 2 was the polishing pass, intended to tighten everything up and catch anything that slipped through the cracks in stage 1.

But as they say no, "No plan survives contact with the enemy" ...And whoever they are, they know what they're talking about.

What Went Right? "I love it when a plan comes together"

Winter Wonderland ...I wanted to create a winter environment that was a little different than what is typically seen in video games ...pine trees and snow. I was strongly inspired by one of my all time favorite paintings "Arcane Summons" by the late great Keith Parkinson. I couldn't resist a little Norse influence while I was at it.



I think I held pretty close to my original inspiration for the winter environment, although I didn't actually try to recreate Parkinson's painting. I will have to go back and make a stab at that some where down the line.

The trees came out nice and the roots on them turned out better than I expected. The rune stones were fun to develop and add a lot of personality to the environment, as well as being a very versatile element of the pack in general. I am actually kind of surprised no one has mentioned the "Torque" rune that is included.



The sky box was a learning process, but came out pretty solid.

Aztec Swamp...The swamp environment was a compromise. Age of Empires 3 had a similar environment, which I some how was unaware of until I had already selected it as an environment setting. I wasn't willing to drop it, so that forced me to take a slightly different approach, for the sake of originality. I turned to south east asian influence for some of the elements and am pretty happy with the result.

The various temples and structures that I made for this environment came together well. Combined with the tall tress in the environment, they help sculpt the 3D space very nicely.

The Lilly pad/swamp muck experiment worked out well. This was nothing more than the water block shore texture effect used in a new way.



The swamp sky was an improvement over the winter sky, in my opinion, and the way I used the sun fx helped tied it all together. I almost shipped this map with only the animated cloud layers turned on. A very nice sky can be achieved without the additional rendered sky box textures, if you plan it well.

The American South West Desert...The deserts of the south west are extremely diverse. As a subject, there was just too much material to cover, so I ended up focusing on a few elements. This is an environment I must revisit for future development.

One of the things that went right, was cliff textures. Creating and texturing terrain with the type of map tech used in TGE (height displacement) can be a challenge, but it can also be very rewarding. I have done a LOT of it, Metal Fatigue being where I cut my teeth on this sort of environment creation. The desert cliffs are very versatile, but require a map maker to give them due attention to get professional results.

I have made tileable cliff objects in the past, for other games in other engines. They can add a lot to the visual quality of an environment, but require a "snap to grid feature" in order to make their placement feasible. I would not be surprised if Torque already has this feature, but if it does, I am unaware of it. If any one reading this has any insights, or wants to look into doing some code work to support this let me know....Now that I think of it, I should make .plan with a bunch of ideas for things like this which could assist the community and see if there are any takers.

I am also very happy with the desert sky, it was another step in the right direction, and is the best one in the content pack.



Lava Land...This environment was a bit of a love hate relationship. I loved the result, but hated releasing it without developing all of the content originally conceived for it.

I am pretty happy with the lava (water) and the lava fx. The Trees came out much better than I thought they would. The stark but attractive nature of the environment makes me imagine all sorts of possibilities.



What Went Wrong? "The Best Laid Plans..."

I planned for a 60 day development cycle for 5 environments, a pace I am no stranger to. However, I ran into some problems, primarily health related (which I won't go into here) and a few small technical issues mentioned below. Combined, these hurdles dramatically extended the project from 60 to a little over 90 days. Extrapolate that to a development schedule for a big project, and its a delay of project killing proportions. Thankfully, I wasn't working for "The Man". Lets hear if for the freedom of Indi development :O)

I didn't get falling snow precipitation developed. It (precipitation) apparently is not working in the RTS kit, and I didn't get a chance in the 2nd development pass to address it. This is almost certainly a simple bug and I will try to address this in my next pack.

I ran into a problem with the collision volumes for the temple structures, but David Wyand saved my butt on that one with an update to his LW2dts exporter.

For the South West Desert environment I wanted to develop giant monument rock structures and such, as .dts shapes, but time just didn't allow for it, and pretty solid plateaus were possible with the cliff textures, so I was forced to move on.

I had big plans for this LavaLand environment , but I ran into a C code bug that held me up for about 5 days. Turns out that the fix was available all along, but I was not aware of it. None the less, this delay really killed my plans for a full development pass on this environment.

I also spent WAY too much time trying to get "shore texture" aka the crusty drying surface lava just right. Ultimately I included several versions with the content pack, which are not pictured in any of the screen shots.

I also ended up spending a good chunk of time developing multiple versions of most structures, so that there were simply and complex versions of the collision volumes available to the end user, allowing them to select whatever best fit their needs. It was unplanned and extended the dev cycle, but added value to the product and better addressed the differing needs of RTS and the FPS developers out there.

Hindsight is 20 20 20 ...get it? 3D

One of the reasons I decided to do a pack that covered multiple environments was because there was a limited amount of that sort of content available to the community, and none design specifically for RTS developers. The down side of this is that I did not devote the amount of attention to each element of the pack that I would have liked.

On the other hand, I learned a hell of a lot about Torque. And if I didn't have enough project ideas on my wish list before I started, I sure do now.

So what is next? I am not sure yet, but there is something scratching at the back on my thoughts, trying to get out. I keep seeing glimpses of an environment, like some one is flicking on and off a light, allowing brief snap shots, but no chance to grasp the whole picture.

There are only so many hours in the day. I hear that relentless tick tick tick of the clock, but I think I am going to go run a few miles before I start the hunt for my great white whale.

The community response has been phenomenal! And I can't begin to say how rewarding that feels. Thank you, everyone.

-Todd

If you are interested, you can find the content pack details and more images at the following link.
www.garagegames.com/products/82