Postmortem: Little Racers
Note: This is a personal post written by me. I’m not talking in the name of the Milkstone Studios team, their opinions about the project could be different.
I’ve had a lot of fun developing and marketing Little Racers, but it’s time to move on and focus on our next project. Before doing it, I have written this Postmortem about the game. A Postmortem is a document written once the game has been completed, to point out the things done good and the things done wrong, so they can be improved for the next product.
What went Right
Although Little Racers will not be remembered for using top-notch technology on the Xbox 360, I think the technical quality achieved is above average on Xbox Live Indie Games. Moreover, our creative use of shader techniques allowed us to give a 3D look to some graphics to the point that some people believed they really were 3D!. Also, the use of Farseer physics engine was the right thing to do, most things we’ve needed to do worked like a charm, allowing us to focus on the handling schemes instead of collision detection etc.
- Online mode
Networking for games has never been an easy task. It usually takes a lot of time to develop and test, due to the complexity of the topic (the need of more than one machine for testing, latency issues, different server/client behavior…). We managed to get a fairly good online mode in one month, which works fine even in the worst lag conditions.
Also, the online mode enabled us to play with other people over the world, so we could know how people performed in terms of handling.
This topic had also some things to note that could be improved, they’re treated in a further point.
We’ve done 5 updates for the game after its release in Indie Games (The last one is currently in review process at the moment of writing this). 3 of them were mostly patches, and the other two included new features and improvements.
As an example of our dedication with these patches, the number of tracks available in the full version increased from 16 tracks in v1.0 to 24 tracks in v1.5. I think we owe that much to the users that reported the bugs in the earlier versions of the game, and we wanted to thank their patience with new content.
All these fixes of course were also included in the demo version of the game, which also saw increased the number of tracks from 3 to 6, so they helped a bit with sales.
I think 400 points was the right point tier for the game when the game came out. Maybe the value for the game was a little below that point, I’d say about 350 points, but more than 200 for sure. When the online mode was added in v1.2, for sure the game’s value was 400 points.
When the ratings system was implemented in Xbox Live (what we could name as “The second coming” for Indie Games), we decided to drop the price to the new 240 points tier, in order to get a little more popularity and also to compete with the new 80 points games and apps. It really didn’t meant an important boost in sales, but it helped a bit, that’s for sure.
Receiving a mail in your inbox with an user thoughts about your game is one of the best things of being a developer, even when the mail is just to let you know of a crash issue. Also website reviews are welcome, even when the reviewer thinks the game isn’t that good.
Many users wanted to share their thoughts with us when the game came out, and we’ve read all of them. Some of their suggestions have been added as updates to the current game. Many other have been added as possible features for a future game sequel.
Also, since my gamertag can be seen on one of the loading tips, some people have sent me messages and friend requests over the Xbox messaging system. That’s another good thing, you can see people that bought your game and still play it from time to time.
Little Racers was originally developed as a multiplayer-only experience with a single player time trial mode. The reason for that was the technical challenge of getting a competitive AI working. The knowledge an AI could get of the track was very limited, and it didn’t seem easy to get an AI that could take corners precisely at both high and low speeds.
In the end, the core system for AI was surprisingly done in a single afternoon, some of our ideas for the AI performed rather fine.
In fact, it was so good that some people thought the AI cheats by driving faster or getting more grip: Well it’s not, it’s just accelerating, braking and turning as every player does!
What went wrong
We’ve not received much feedback regarding the usability of the menu, but I think it’s still one of the places where there’s much room for improvement. Since the game appearance, we’ve redesigned the GUI system on our game engine, but it would take too much time to design a new interface for Little Racers.
Anyway, we’ve learned from our errors and next time we’ll make a more console-like and user-friendly menu interface.
Car handling has been a tricky thing since the first point. The main difference between Little Racers and other 2D racing games is in the handling. Since we decided not to add gameplay elements such as power-ups or weapons, it had to be really enjoyable.
This is were a problem for having too few people developing a game arises. On the first versions, the game had a too unforgiving handling scheme: You had to be really good to complete a clean lap. We were so good at the game that we just didn’t notice.
We’ve been making the control scheme easier and easier update after update, trying to balance speed and grip to allow players to take corners more easily, without making it too boring.
Although I like the current scheme (as I’ve liked all the previous ones), it still is one of the most hated things of the game. Maybe the problem is more related to track design than to handling.
- Track design
Due to design decisions related with the local multiplayer mode (cars were too fast to make split screen useful), all the tracks had to fit in a single screen. The first version of the game included many tracks that tried to deal with this by using all the space available, with lot of corners and crosses.
After some feedback, the new tracks added were a lot more wide and open, giving more room to players in a crowded race. Also these new tracks made braking a lot less necessary (Braking doesn’t fit well with arcade racers), and have more high speed crosses, a really fun thing when 12 players are racing at the same time.
Still, the old tracks are still available, to ensure nobody asks to bring them back. Maybe the right decision would have been to remove them, some tracks such as SouthPeak could be removed to improve the overall quality.
Despite our best efforts to show the game graphics in a different and credible way, they were still programmer art. That means many in-game icons, the tracks, etc. have a graphic quality that could be much better.
Since we didn’t have the resources to get a proper artist to do the work, there was no other way, but it’s still a thing to improve for our next projects.
- Long-term replayability
When playing alone, Little Racers is usually a good game for short sessions, of around 15-20 minutes. It just doesn’t offer enough content to warrant fun for much more time than that.
The game was developed with multiplayer in mind, and that’s why we didn’t really care about it. After all, this problem is inherent to almost any racing game: The fun you can get from racing alone the same track again and again is limited, the only way to get around it is offering some kind of campaign mode, or racing against other people.
We went with the second option. The game really shines when three or more players are competing. Sadly, this isn’t the usual case with Xbox Live Indie games, that you usually try when you’re bored at home and have some extra points on your account.
Although the networking system went out pretty well, not everything was as good as it should. The network mode was reduced to make development and use easier: Only the championship game mode can be used, and it’s mostly automatic.
Also, I’m not sure the time invested in adding the online mode was really worth it: since an Indie game is played by very few people, the chances of finding other people playing at the same moment are reduced.
The whole system could have been better, that’s for sure.
The game was designed to work on HD displays, and it works at a native resolution of 1280×720, with the Xbox 360 doing the scaling when it doesn’t fit the display resolution.
It seems that lots of people still have a non-HD display, and while Little Racers can be played on one of those, it sure isn’t the same experience: Cars are too tiny to see the colors and details well, and some text and icons are just to small to read (the GUI font doesn’t help). It also happens on low-quality HD displays.
We’ve learned the lesson and we’ll try to do bigger icons and clearer texts on our next games.
How the game did
- In sales terms, I can say the game success was over our expectatives, but not for much. The initial download count was really incredible, getting 100 sales per day during the first week, but after the game dropped of the new releases tab, downloads rapidly dropped to the point of 0-5 sales per day.
This number increased again in two points: When Xbox Live indie games were made available in Germany and Japan, and when the rating system was added to the dashboard. The price change didn’t have much effect on sales numbers, they increased but overall the earned amount was about the same.
In terms of reception, the game could have done much better. Aside from reviews that we directly asked for, the game hasn’t gotten much public reception.
We really wished more people talked about the game on forums and blogs, since we think the thing the game needs now is players available to compete online.
It also seems the same on the Xbox dashboard, the number of ratings of the game is around the 180 mark, when other games on the same level got about twice that amount.
Well, that’s all of it. I’d like to thank everyone who has tested and rated the game on the Xbox dashboard, and also the XNA Team and Creators Club members. Without their help, Little Racers would have never been come out.
I’ve followed the development of LR very closely, and I must say that not everybody is able of analyzing the strengths and weaknesses of his (or her) work in such an insightful way as this.
I hope your game gets better sales, as the technical aspects are mostly covered 😛
Farewell Little Racers! 😀
I’m actually making a game for Android -maybe an Xbox Live Indie version when it’s finished- and I had similar problems. I uploaded a demo version to Android Market some days ago and a lot of people was complaining with controls. I had to modify the controls I had in my mind because of problems with Android platform and I didn’t notice controls could be that hard for other people so now I’m working on that.
And it’s interesting what you said about the few amount of users playing online with Indie Games. I was thinking if I make an Xbox Indie version it could be a waste of time making an online mode and that’s a good help to take a decision -I will also consider the amount of time to make it-.
PS. I also studied in EUITIO 🙂
Reading this type of feedback is always helpful for me to see what others have tried or their thoughts on improvement.
Did you ever provide an estimate of the number of hours you put into making this game? How many people were involved in the development?
We haven’t made a record of hours invested into the game, so I can’t provide any accurate numbers, but our estimations are around 350-400 hours of development.