The move to a new generation of development platforms has, we all know, brought a set of new challenges to the scheduling of any project. Currently, the key difference between developing a game at the start of a console lifespan and one towards the back-end is the amount of unknowns involved – something we’ve become aware of at Swordfish as we moved from PS2 game Cold Winter to 360 game World in Conflict.
When we were scheduling our PS2 and Xbox titles we were working from a tried and tested base where we could concentrate more on gameplay rather than getting to know the technology. But when you come up against a new console the gloves are off and you have to learn a lot of things afresh.
Below I’ve outlined the five specific challenges our production faced…
1. PEOPLE MANAGEMENT
The first hurdle is the one we all know: the number of people needed to complete a multi-SKU next-gen first person shooter is over three times the size of the team we needed to develop Brian Lara Cricket.
To enable us to grow so rapidly we needed to solidify the management structure of the team first. Low turnover of staff at Swordfish over the years has meant we already have strong working relationships in place, but we supplemented this with a lot of new hires to bring areas of expertise to the team and to allow us to grow in a controlled manner. The emphasis of this growth was to make sure we got the best out of every member of the team.
But throwing too many people at the problem can actually start to become unproductive. I’ve heard of projects where they have needed to remove large amounts of staff to actually increase productivity. So the lesson is to grow, but ensure you maximise the staff you have in place. To do this you need to make sure you have your scheduling well prepared and constantly communicated to the staff.
During development the main thrust of your project doesn’t change. You base your schedule around hitting milestones with tangible results and gradually form the base of your project building it up in layers.
When facing new technology, though, this means the early part of your schedule is going to involve a lot of R&D which is notoriously hard to apply time estimates to for scheduling. To combat this we started looking into agile development methodologies to help us and we decided we’d like to implement SCRUM.
Communication is essential during game development and is one element of SCRUM that is really beneficial to a growing team. Enforcing channels of communication and visibility allows developers a chance to expose where they are and what problems they may be facing, and allows the project management to be aware of how the game is progressing and to guide the team accordingly.
We scheduled our longest ever pre-production period for World in Conflict. The amount of work needed to get us ready to enter production on a next-gen platform was larger than anything we’d ever encountered before.
SCRUM works really well for pre-production as it allows the team to focus on major gameplay areas early and gives them the facility to iterate upon these areas until something fun and playable comes out the other end.
Whilst that is going on the other emphasis for the team is to be preparing and stress testing your tools and pipelines to deliver content. With so many developers primed ready to be creating assets at the same time once you hit production, these tools and pipelines need to be user-friendly and rock-solid.
If they don’t prove to be up to the job once production starts the burn-rate you would incur from not being able to utilise your content team optimally will add greatly to the cost of your project.
4. BEING AGILE
We moved over to using SCRUM shortly after joining the Vivendi family. There is nothing like a bit of healthy competition going against our other Vivendi studios to get the best out of your team.
To give your developers the chance to produce the best game possible you need to give them the platform on which they can excel. Being agile with your schedule on a constant basis allows you to react to changes of priority, new feedback from the publisher, or a ground-breaking idea that although not on the original schedule could make your game triple-A.
Knowing when to change and when not to change is almost an art. Having an agile mindset allows you to react to it correctly when you know it is the correct time to make a change.
The SCRUM methodology pushes teams to always be thinking in this way, but in truth a lot of their teachings are a mindset that the best project managers have been using for a long time.
Not everything in SCRUM suited us and how the company worked. We’ve been running under SCRUM now for over two years. We took six months to run it ‘out of the box’ and learn from how it developed within our studio. Once we understood it fully we made revisions to the process that fit best with the culture of the studio to make ourselves more efficient.
Another consideration to make for next-gen is the use of outsourcing to enable us to create the vast amounts of content to the high standards of a triple-A title.
We’d had limited experience of outsourcing through our sports titles, but the task of building an outsourcing strategy to enable us to complete work for a new gen game was a different proposition.
One of the keys to ensuring work gets done to schedule is to build an open two-way relationship with the outsourcing vendor. Just throwing work over the wall to a company and hoping it comes back as you want it is a lot less likely to succeed than to invest time into an outsourcing company in view of building a partnership. Allowing them a certain level of ownership of the assets they are creating for you will in turn allow them to become more invested in your game and increase the quality of results you get back.
Once you have managed your outsourcing pipelines and internal teams to be creating the best assets possible, and made your technology unknowns known, you are on the brink of your game being realised and ready for debut on the new generation of console platforms - which is as exciting a time for the developer as it is for the consumer.
Ian Flatt is project manager at Swordfish Studios and has overseen the production of a number of the studio’s games