Get the latest Education e-news
  • Postmortem: Age of Chivalry

    - Tibberius Bane
  •  A cold wind whips across a grassy plain near a famed landmark known as Stoneshill, which in the land of Agatha will forever be remembered for the great battle that is soon to take place here. This wind is strong and it pulls at the leaves of the trees and the blades of the grass relentlessly, it is cold enough to have driven most of the animals in the area into whatever shelters they could find.

    This wind whistles around the mountain pass where it chills the skin of many well equipped soldiers waiting there. One man is further up the path than the others and he is crouched there, looking over the ledge towards the enemy force that is marching towards his men. This man does not shiver at the wind, for this man has fought many a battle, and this man knows composure must be kept if there is any hope to win.

    He unsheathes his blade and bellows a cry to his troops just as the enemy thunders through the pass below: “CHARRRGE!” This man is not afraid because this man is a true warrior. This man is certain of his victory, because this man loves his country. This man thirsts for battle because this man is you, and you are in the Age of Chivalry.

    This article will act as a reflection of the time myself and the team have spent creating the Half-Life 2 Mod Age of Chivalry, it touches on some of the things that went wrong, as well as some of the things that went right over the course of development. It also has some tips for aspiring mod/indie developers and I tried to make it appeal to a wide audience by including AOC-specific examples as well as general advice that anyone could apply to any project.

    What Went Wrong

    1. Movement System

    One of the biggest problems with the actual gameplay of the game even in the current public version has been the standard HL2 movement system. At first it may be difficult to understand why something so simple as movement should need to be redone -- and that’s what we thought too. Unfortunately in doing so we overlooked a major difference that is prevalent in our game that is not nearly as large of a factor in gun-based games.

    The importance of the movement is amplified when engaged in close combat and there has probably never been a better example of this than our game. Veteran players are able to exploit the lacking movement system with “Mouse-whip” techniques that allow them to turn on a dime, even at full speed. This lead to some very frustrating situations, especially for new players where their opponents felt like they were equipped with not-so-medieval rollerblades and this took away from the game for a lot of people. We made a weak attempt at addressing these issues with the “Toe-to-Toe” system, but in the end it only watered down the problems of the movement system and created other problems.Essentially what we struggled with here is something all games struggle with at some point in their development and that is the freedom of movement available for a player versus the game-play. In our case, we should have had a much more strictly defined movement system that complimented our game but still felt natural, it is “the one feature that got away” for me, being that I am also an avid fan and player of the game.

    2. Testing Process

    UGH. That should begin to describe how unfortunate our testing situation was. I think our average weekly testing session attendance was probably around 3, maybe 3 and a half members… We used Mantis, a free bug-tracking software and it was great, but as with all great tools, they must be used by people to have any benefit. Our testing team appeared to consist again of about 3 solid dedicated members who actually reported and updated bugs on the software. For those among you who are developers of other mods or games, I cannot stress enough the importance of playtesting and testing your products, it will be PAINFULLY obvious to your fanbase if your testing team was not up to par.

    Furthermore I suggest very strongly that you enforce internal testing, that is, getting your development team to actually partake in testing themselves and gain a true understanding for the game. This will actually raise productivity in the end, because they will understand the game and won’t just be making assets blindly. Basically, if you expect other people to play your game, you better make sure that it is running smoothly and fun. You wouldn’t expect to sell a house that you didn’t bother to clean before showcasing, no matter how nice the house was, the garbage and dirt would stand out far more than any of the nice features the house offered and the same applies to games.

    This article is not intended in anyway to slight the beta team for Age of Chivalry, I can admit that it was the failing of the development team ourselves to implement a proper process and recruit the right people. I urge all future and current developers to learn from our very poor example and take the time to critically think about how to reach the most valuable testing people in your community and keep the testing process interesting for them, it will save you from a landslide of negative effects that we had to deal with over the course of the development cycle. The testers that stayed with us despite our flawed structure for the beta team and lacklustre communication with them cannot be thanked enough, they are the only reason what exists today is playable!


comments powered by Disqus