Platformer Starter Kit - New feature incoming!

It has been nearly a week since the TGB Platformer Starter Kit was launched and I couldn't be happier with the reception it received from the community. I am ecstatic that I have been able to get a product created and out onto the market. I hope that this will be the first of many, not only starter kits, but games down the track.

I would just like to give a small introduction about my experience with GG and a brief description of my time developing the kit. If you want to get straight to my announcement, just scroll down.

My Introduction to Garage Games

I first stumbled upon GarageGames and TGB when I was wondering how much an XBLA game made. Google bought me to the Make It Big in Games blog written by Jeff Tunnell and Josh Dallman. Once I had stumbled upon this website, I was hooked.

It didn't take long for me to understand that, for the most part, this is a great community with a great passion for developing games. So many people were out there helping others, donating their own time and energy to help accomplish something new.

One of the first posts I had read on the TGB forum was related to creating a Platform game in TGB, containing all of the pitfalls and hurdles that one would have to avoid in development. I opened up TGB 1.3.1 and hoped over to the TDN to start creating the Mini Platformer. The users in that post were right and many problems mentioned were frequently displaying, one of which was the good ol' walk up wall trick.

I played around with everything a bit more and bought it upon myself to improve its current state by making my own platformer from a new direction. It turned out okay, way more buggy than the Mini Platformer, but I had learnt a lot and elements from that tutorial would go on to help in the PSK.

I had already purchased the TX PSK but had very little experience in C#, which deterred me from actually using it. In my opinion, the TX PSK is a great example of what a few guys can do with a bit of free time. They had laid an excellent point of reference for anybody looking to create a platformer game in TGB, but nobody had done it. Being the budding game developer and part time entrepreneur that I am, I saw an excellent way for me to create my own game and learn TGB and TS. Since most programming languages are easy enough to read through and understand, I decided to use that as a foundation to my own platformer.

The Development

The first, what some would call a *build*, was just a black cube running around a white background. I had to implement a method of movement first. This was a critical aspect that added an element of difficulty in production, but made everything worthwhile.

When you come across difficulties in game development, you have to sit down and ask yourself "Is it worth it?" I bet that in many indie and commercial games there are elements which should have been scrapped due to the input cost in relation to benefit for the user. I believe that critical aspects of your game should be reviewed if you spend weeks of developing prototypes with little improvement. I am one of those people who need to walk away when I encounter a problem. I know that if I sit there and spend a few more hours on a stubborn bug that I can fix it, but I also know that if I walk away and come back to it later, I will fix it in minutes. I learnt quickly that you need to take breaks.

I upgraded from the little black box to the ninja art from the Mini Platformer. At this point I had started to ponder the use of source modifications in order to get One-Way collisions working as well as sloped surfaces. I messed around with a few mods and got both elements to work.

I added the ability to ignore collisions from specific objects. It was just a vector of scene objects stored away in an object that was checked upon collision. It did its job and gave me the ability to jump through platforms.

My method of detecting sloped surfaces was a glorified cast-collision (a function I strongly disliked at the time). There wasn't much too it really. I created a function that simulated a collision between the ground poly and a shape that I could create. It would return collision details including the contact normal and the position, two critical elements for sloped surfaces.

It was at this point that I started work on my animation FSM. I had been using a rather large quantity of if statements which got really, really confusing and prompted me to upgrade. I had never used FSM's of even understood the concept, but I was able to get the manager working in a few days. It turned out to be really simple system and I am really happy with the results. The FSM was written in TS and I have outlined its framework in this post. Carpenter Software went on to develop a larger, more detailed FSM here.

It was at this point that I had started to add a lot of the TX PSK's features and I decided that there might be some commercial gain in this for me. I contacted Thomas (the GG employee who created the TX PSK) threw over a build and enquired about the possibility of a starter kit. A few emails were sent back and forth and when I had it to a stage where critical elements were working (relatively) bug free, I got in contact with Derek Bronson.

Derek manages 3rd party content at GG is someone who I get along with quite well. He was incredibly supportive (as was the whole GG team) and helped me get everything prepped and ready for release. I had never done anything like this before and it was fantastic to have someone like Derek there to help me package the kit.

Derek prompted me find a method of one-way platforms and sloped surfaces without using source modifications. I saw the obvious potential in this requirement, but did not fancy the idea of actually doing it. I argued with Dan Maruschak and boldly stated that it was not possible to do either of these two things in TS. A week later, I had been proved wrong.

These two elements were the biggest pain in the ass, period.

From this point it would take a few more months of refining and adding features to get it to the point where it works just as well (if not better) as the TX PSK. The next phase in my scheme for world domination would be the announcement.

The Announcement

Derek approached me and asked if he could display my work at the GG booth in this year's GDC. I was over the moon! The feeling was incredible. I had achieved what many developers dream of, a product that is shown to the world and that people take interest in.

I whipped up a snapshot to submit and wrote a small blog highlighting some of the finer points of the kit. Personally, I was surprised that I didn't receive any comments about my "Object Manager". I thought that this little system was awesome and it certainly saved me the trouble of remembering object groups. Anyway, everything had been announced and I was one step closer to release.

Derek and I sat down and discussed price.

I know that a lot of people were disappointed in the price difference between the TX PSK and my own, but there were reasons for it.

I don't think some people fully appreciate the amount of work that goes into developing something like this. Platform games have been absent from the TGB scene since its introduction a few years back. This is because it is a difficult system to implement. 4,000 lines of code and comments went into the kit and it has taken me 6 months of work to get it to this stage. If it wasn't a difficult system to implement, then I am sure that someone else would have done it.

I also think that quite a few people thought that I was a GG employee. I am just another one of you guys, no different. I worked hard and managed to create something great. I am sure that you have the potential too.

There have been quite a few people who have purchased the kit and made the comment that it was well worth the investment. I hope that it continues to be the case. In saying that, I would like to make an announcement.

Something New!

I wanted to add something to the kit that most of the TGB community would have thought it impossible to do in its current setup. Network support.

For the past few weeks I have been working on a network layer for the PSK. I have successfully linked the majority of the PSK's features between two clients and am thoroughly testing my work. I am smoothing out the movement to get everything the two clients as synchronized as possible.

In order to add smooth movement I created a basic pathing system to the kit. It is like the moveTo function in TGB, except it uses the x-axis only and tells the player to run to the point. I might try to improve it later on, but at the moment I am focusing on more important elements.

I don't have an ETA on the upgrade, but it is coming, promise!

In my opinion, a networkable version of the PSK will justify the $50 price tag. Just remember that all future upgrades will be free to all TGB PSK users!