Blog for Alex Swanson at GarageGames.com

Rokkitball Q&A

Alex Swanson — 2008-06-12T20:47:27+00:00

It has been a long time since I posted a blog here at the main GG site, so I though I'd share the responses to some questions that Deborah sent over regarding the rapid development of Rokkitball, our latest game on InstantAction.com.

Deborah Marshall: How did you balance the FPS shooter aspects of the game with the sports aspect of it?

Alex Swanson: This was actually a pretty difficult balance to find. When we started on the game it had a very strong FPS feel. There was not really any way to control the ball aside from scooting it away from yourself by shooting it. Ball control is pretty critical when it comes to having a sport-like strategy game, and after a while it became clear that just having rockets was not going to be sufficient.
Once the magno-beam was implemented, we spent a lot of time tweaking the balance between it and the effectiveness of the rockets to move the ball and disrupt other players. We wanted disruption and rocket ability to be important, but taken too far it made trying to control the ball up close impossible - you'd just get hammered by high speed, rapid-fire rockets. That was no fun, so we slowed the rockets down to a point where it is possible to avoid them if you are paying attention, and limited their rate of fire to a point where, even in a 4v4 game, it is possible to grab and move the ball.

This made the rocket strategy a lot more strategic - having good aim is certainly a major factor toward being good at the game, but know when and when not to use your rockets is equally important.

Not having the player die as a result of rocket impacts also helped the game develop a more "sports-like" feel. Rockets allow you to "check" a player and take momentarily reduce their ability to act by draining their energy, but it doesn't just clear them out and force them to restart in another location. This is much more like the way that a real-world sport plays, and I think adds a lot to the sports feel of the game.

Image: Rokkitball Athlete concept art by Lance Bass

DM: What was the development process like (SCRUM, etc.)? How big was the team? How long did it take to make?

AS: The development process was fairly free-form. The whole team was only 7 people, with most of them in the same room or just around the corner from one another. This made fairly tight communication possible.

We kept development fairly light on meetings and focused on clearing out lists of tasks. I think that in some ways we could have benefited from a more structured approach such as Scrum, but given the small scope of the project it may have been overkill to impose too many rules.

The game has existed in many forms over the years since its first introduction at IGC 2003 by Matt Fairfax and Tom Bampton, and last year with quite a number of experiments done by Plastic Games. Concepting and prototyping are an important part of any game development - and without all the previous editions focusing the game design I do not think that it could have been completed as quickly as it was.

We started working on the game internally only about four months before launch, and the first month was just Robert, the lead programmer, working part time developing the attractor-beam mechanic. Full production took three months for the initial release - a pretty quick turnaround for a game of any size, much less a 3d multiplayer game. Of course since then we have been doing ongoing development on the beta version - finishing up additional levels and adding other content. All of this makes it difficult to compare to a traditional development model.

Image: New Rokkitball Arena "Canyon"

DM: Why did you choose Torque? What other technologies did you consider?

AS: Well, given that all previous versions of the game were written on Torque, and we work with many of the most experienced Torque developers in the world, along with those who created the engine, Torque was the obvious solution.

We brought in several pieces of technology from our future engine R&D area, including an early version of polysoup collision that Andy Malone whipped into shape one weekend when we were still debating whether to use DIF files. That was the only major technological unknown that we introduced to the game. Using all of Torque's capabilities and stable core was a major factor in the quick development cycle.

Image: New Rokkitball Arena "S-Hook"

DM: What were the major challenges in developing the game?

AS: We were on a very short timeline, so we had to stay focused. This meant avoiding feature creep as much as possible, which can actually be quite difficult when you are building something cool. It is very tempting to throw in a ton of features without realizing the time implications of seeing them through to full polish.

We stuck to a pretty minimal spec for the game and pushed hudreds of ideas to the "for future consideration" pile just to keep it manageable. In a lot of ways this also turned into an advantage - we ended up with a fairly tightly focused play experience by just polishing the game elements that we started with.

DM: Was there anything specific challenge to make Rokkitball work with Instant Action's web plug-in?

AS: As one of the very early games to use the plugin, there were definitely some rough spots. Actually getting the game up and running in the test version of the plugin turned out to be quite easy, but full site integration was a major challenge. Working with the web team involved a lot of communication challenges - they are even busier than we are!

Image: New Rokkitball Arena "Cathedral"

DM: How is it different making a game meant to be played in a browser?

AS: Well in many ways there is not a lot of difference from what we did at GG before. There is a lot of focus on size off assets and speed of loading - we want players to get their action in as instant a manner as possible, but we also want to make a great looking game with high quality textures. It is an interesting balancing act trying to match the two, but it is something that we already had to learn to deal with because all our games have been downloadable in the past.

One really awesome thing is that the IA site team has built the lobby and game UI for us. This is usually a much more substantial part of game development than people realize, and not having to worry about it shaved even more time off the development process.

DM: What can we expect to see next with the live development of Rokkitball?

AS: There are a lot of bug fixes still to come, in addition to ongoing new content. I can't say too much about that yet.

Highlighting and rewarding player skill and letting people view their records is also quite important to us, but there is a lot of work yet to be done there! Team and individual stats are really the missing piece of the equation for Rokkitball - we feel it as much as the community no doubt does!

We also want to continue to tweak and streamline the gameplay to emphasize team strategy and skillful play... there's been a bit of contention on this issue in the community already, but I am confident that there are a lot of tweaks that will make the game even more fun and encourage a great level of team vs. team competition.

Image: The yellow team goes for a goal, cyan moves to defend.

New TGB Demo Art Preview

Alex Swanson — 2006-10-04T21:41:18+00:00

As many of you who have been working with Torque Game Builder know, the art included with the base "TGB" package has not changed for quite some time! As a result, most of our tutorials and a lot of new developers' prototypes end up including a whole lot of crates and GG logos. It is time that we gave you something more exciting to look at!

The primary goal is to create a set of basic art assets in a simple, clean style that will be easy for people to imitate and extend with their own assets.

Finally the base art will include a good tile-set for side-on games, in addition to improved scrolling backgrounds - allowing for some more advanced tile mapping and game building tutorials in the future!

The space shooter art will be getting another pass as well - to bring it up to par with the other assets, and give it a cleaner style.

It has been a while since my last post, and I have been busy with quite a few projects of varying scope (some meeting with more success than others.) One thing that has been in the works for a very long time is Minigolf Mania - a game by Kevin Ryan that I was privileged to do art direction and art production for. This game has been a long time in the works - having been put on hold for Marble Blast Ultra and Puzzle Poker development. Look for more announcements regarding this title soon!

This should be an exciting weekend at the Garage, as we will be hosting many of our associates and community members for Associates weekend. Here's to hanging out, drinking beer, checking out new games and demos, and discussing the many super-cool things that we have in the works!

DTS & DIF Exporter Matrices Added to TDN

Alex Swanson — 2006-02-16T01:52:19+00:00

Just a quick update to point out the move of the Exporter Matrices from my personal site to TDN.

They have been renamed "Exporter Feature Charts" to be more clear about their purpose. Hopefully we will see some links from features on the charts to docs on how to implement those features in the various 3D applications as those docs crop up in TDN. Also, you'll notice that the charts took a bit of a visual hit in the transition. If any Wiki Wizards want to pretty them up, please feel free!

For those of you currently linking to my site for the charts, please redirect your links to these pages, as these will be kept up to date, and mine will not (though I will link to these pages from there.)

DTS Exporter Feature Chart
DIF Exporter Feature Chart

Marble Blast Ultra... Finally Complete

Alex Swanson — 2006-01-26T03:07:46+00:00

The journey to Marble Blast on the 360 has finally come to an end. Today at around 2:30pm PST, the game went up on XBox Live Arcade servers. Since then, over 1000 people have posted high scores on the leaderboards, and the online forums have lit up with discussion of the game.

As we'd hoped, it has been really well received so far. People are already posting time challenges for the single player levels and having a blast in multiplayer. I watched over Mark and Pat's shoulders as they layed down some smack on upstart live players. Even after just a few hours though, some users were starting to be able to give them a run for their money.

When we first started this project, it seemed fairly strait forward - Microsoft wanted Marble Blast with added multiplayer race functionality for the Xbox 360 launch. As it turned out, that was easier said than done, and we learned a lot of lessons along the way. There is a lot that can be said about coding, gameplay, scheduling, and management lessons that we learned from the project, but I am going to leave that to other team members and focus on the art design.

You might have noticed that MB Ultra looks a little bit different than MB Gold and other previous incarnations. The original art design by Brian Hahn of Monster Studios was starting to show its age. Because of its flat patterns and bright colors, there wasn't an easy way to show off the 360s capabilities while sticking with the same style.

While we knew that a change was needed, there was a lot of confusion early on with the art design. Unfortunately some of that confusion carried through much of the project. Spending enough time solidifying the look and getting the entire art team to buy in would probably have saved a lot of trouble in the long run. However, there was a deadline hanging over our heads, and at the time it looked like we really didn't have much time to get the look up and running.

Many of the great designs our team came up with really relied on curved surfaces to look good. We knew that we would not have time to create an all-new interior format that would allow for more organic construction than DIF was capable of, especially considering that we didn't want to rebuild all the levels from scratch. We followed many potential routes, from dark matrix-ish worlds, to mechanical mad science looks.

Ultimately, things started to get very tight time-wise, and I pulled together the first version of the look that ended up in the final game. My goal was to give the game a greater sense of realism, material-wise, but to enhance the very simple, almost mathematical feel of the levels without too much additional work. These ideas, combined with some comment Mark had made about some kind of "music of the heavens" type look contributed to the creation of the astrolabe background animation and the extensive use of circular arches that support the in-game geometry.

The circular geometry help to reinforce the strong circular profile provided by the marble, as well as creating a very diagrammatic effect, which, along with the astrolabe skybox further reiterated the mathematical precision of the physics based gameplay. As it turns out, rings and circles are some of the few "organic" shapes that are easy to build with DIF tools, so this was also very convenient considering the technical restrictions.

This pseudo-realistic look really finally came together right before we were submitting our first code-complete milestone, and evolved from then on. The key was in getting all the materials to look right - the tile had to be convincing tile, the metal convincingly metal, etc. A lot of tweaking was done on both the art and code sides to get these just right.

Ironically, the marbles were one of the last areas to get graphical love. Fortunately by the time we got to them, the project had consumed all GarageGames employees (well, nearly all) - so Brian, Ben, Tim, and Adam were able to team up to do a completely awesome job on Marble special effect. It was really the last piece of the puzzle visually. I had done a lot of tweaking on the interior shaders and color schemes for the different difficulty levels (not fun when dealing with the color vagaries of NTSC displays!), but it was not until the final marbles started getting checked in that everything really fell together.

This just underscores the importance of a consistent level of realism across a title. Having the realistic environments with flat marbles had never quite clicked, but as soon as the marbles started reflecting, refracting, and fresneling... well, you can seed the results - I hope you like them!

In conclusion, some hard learned lessons for all you aspiring Artists and Art Directors out there:

1) Give yourself enough time to really nail the look before going into full art production.

2) Remember your technical restrictions. Don't try to build something beyond the abilities of your tools or your artists.

3) Make everything consistent. Choose a level of quality and realism and BE CONSISTENT. It is just as important to make sure that you don't have objects that are too high of quality as too low.

4) Get the team to buy in - make sure you create lots of examples - or get them from google images - just make sure that the vision of the project is clear to every artist working on it. Otherwise they will spend a lot of time wondering if they are doing the right sort of thing, and you will spend a lot of time correcting things that go off-style.

Learning to do many things at once

Alex Swanson — 2005-07-01T20:06:28+00:00

A friend of mine who visited the site commented that I should perhaps update my blog, as I had not written since January.

I have been waiting to write a plan until I had something major to announce, but as many of the projects that I am working on are related to external developers with whom we have signed NDAs, there is not much that I can say.

On these projects, however, I have been acting as Producer. A bit of a change from my normal artist role. I can say that I finally feel Jay's pain when he says that he spent an entire day doing nothing but email and phone calls. It is amazing how much time can be eaten up by communication when one is working on an externally contracted project.

I think that many indies may be unaware of what they are getting themselves into when they take on contract work, or first attempt to publish through a major channel. I know it has been said before both at IGC and throughout the forums - but having at least one full time marketing or business relations person can be a very good thing! Having your artists and programmers dealing directly with external communications can really put a damper on productivity - it can be very hard to have to constantly switch gears to take on both tasks simultaneously.

I know my art productivity has dropped significantly as a result of working as producer - fortunately I have a great team of pro artists here at GG who are also working on these projects. We can afford the time lost to business communication. In a small team, I imagine that having one of the main creative or technical minds spending all their time on external relations could really slow down the entire development process.

Being a producer has also introduced me to complexities of scheduling and external contracting of which I was previously unaware. When you are on a tightly scheduled project, it is very important to prioritize tasks and to learn how long it takes you and your team members to complete a task. Communication is also key at this stage, as a single person falling behind can impact the scheduling of the entire project.

Often small projects need to contract out some of their artwork or code to get done on time, or to achieve a style or level of concept polish that the internal team cannot reach. Communication is even more important here - making sure that the person you are contracting knows what you are expecting and when. This means talking to them on the phone or in person and laying out (in detail) what needs to be done - in the case of artwork it may also involve showing them your own rougher sketches of the design that you want them to flesh out. To avoid any hard feelings, make sure that your contract reflects what qualifies as acceptable work, and for it to have a clause for compensating the contractor for their time if what they submit is not on par with what you expected.

As a final note, I would just like to reiterate just what a great lifestyle the indie development cycle is - while working on some of these large, externally contracted and scheduled projects, I think many of us here at GG have gotten a bit of a taste of the level of stress and crunch time created by the schedules of a normal development house. While doing games on an external contract might be a great source of income and experience, the real ideal is to develop games that you want to play on your own schedule.

I apologize for the lack of pictures (I didn't really have anything to fit the subjects matter). As an amusement to those who made it this far without any visual stimulation, I present: Teen Angst Mouse!

Friday Sep 19 22:37

Alex Swanson — 2003-09-19T22:37:52+00:00

Back from Europe, looking for work, and preparing for IGC.

I have just returned from Europe. I had a wonderful time in a little village called Oira, Italy, where I took a three week class on psychogeography. We had an excellent class group there - it was fantastic to be able to spend 3 weeks with so many creative people in such an inspiring environment. I also spent some time travelling prior to the class and visited London, Edinburgh, Paris, and Bern.

As of next week I will be unemployed because I cannot continue my student job now that I have graduated. I'm in search of part time employment to supplement my royalties from Marble Blast - unfortunately Eugene does not exactly have a spectacular job market at the moment, so wish me luck.

This week I've been doing some last minute work on the Torque demo art and starting to prepare for IGC, where I will be part of a panel discussion on map building.

Thursday Jun 20 22:02

Alex Swanson — 2002-06-20T22:02:51+00:00

Internship at GarageGames!

Well, I've just started working here at GarageGames as an intern, specializing in graphics. My compatriot Rick Gorman has become a programming intern. Rick Overman set up so nice little workstations for us here at the GG office, so we'll be able to ask the experts for help whenever we get confused. I'm looking forward to unlocking the mysteries of Torque and all that, so here's the current gameplan:

* Learn how to use MilkShape (MAX is too damn expensive)
* Learn how to use QuArK
* Learn how to export to Torque from both of said products.
* Make some maps for Realm Wars.
* Figure out how to create textures in the mission editor and write a tutorial on it.

I am also working 20-30 hours at the University of Oregon doing technology consulting for faculty and staff, so unfortunately I won't be able to devote 100% of my time to GarageGames. I'll be spending as much time as I can over here, though. Nothing like trading free labor for free experience.
The really sad part is that this internship will probably do me more good in terms of getting hired in the future than my entire 5 year multimedia degree at UO :)