My dissertation project was to take an existing Unreal Engine 3 PC game and port it to the iOS platform, in order to explore optimization techniques, evaluate tangible and notable platform differences, and outline different porting strategies. Despite the age of the engine, resources for porting an Unreal Engine 3 game to iOS are limited, so I undertook this project to expand the knowledge base.
The final result was that, despite it being close, ultimately the iOS port reviewed lower than the original PC version, even with considerations for the platform and the barriers therein. Despite this, the information gathered from this dissertation should prove useful to anyone who is engaged in trying to port a Unreal Engine 3 PC game to iOS without source engine access.
The game I used for this porting was Mendel's Farm, due to the full access I have to the source code, and because of my already existing expertise with the programming of the title. For both Mendel's Farm and this specific project all Unreal Engine 3 usage was through the Unreal Development Kit, and scripting was done with UnrealScript.
I began by conducting a study of existing Unreal Engine 3 titles that exist on both PC and iOS, and using their differences as a set of compartmentalized guidelines for the Mendel's Farm port. Component by component, I optimized and fixed the iOS port to look as high quality and run as smoothly as possible, to reflect the PC version. Where the decision had to be made, I favoured performance over aesthetics, but generally it was a good balance between the two.
To determine the subjective quality of the port, I used a modified gameplay experience testing questionnaire based on Mark Parnell's gameplay experience scales (2009). The modifications allowed the testing to have a greater focus on individually scoring areas of the game directly affected during the porting process, like input and performance.
The questionnaire results showed that the iOS port was not perceived to match the quality of the PC version, especially in areas of usability like input and interface. Now whilst the aesthetics of the iOS version were rated lower than the PC version, it is worth noting that the difference was smaller than anticipated, which could be linked to expectations of visual quality for games on each platform.
Despite the failure of the port, I anticipate the final results as being useful. For further research on this topic, it would be recommended that more time and effort be made available to the porting process, and that user testing occurs on an incremental basis throughout each stages of the porting, so that aesthetic and design choices can be judged individually as they occur.
Below is a series of screenshots showing the progress of the port at each stage. The full dissertation is available for download here, and if you have any further questions please get in contact with me here.