Post number 6
Before I've even had time to barely blink life here on AltDev has moved along at a merry pace and, already, it's 90 days since I first posted here. Which feels like it might now be time to start tackling the biggest problem that I currently believe is holding the games industry back:
The tools we use
We have code teams that can create amazing effects in real time, at 60fps. We have artists that can recreate Rome, build Alien planets and animators that can make the stars dance. But, at any point that we decide we need a bigger hole, we just get more men with spades. It's very rare for anyone to take the leap and get a JCB along to do the job for us.
So why not? Why is it so often perceived that it's cheaper to hire in some more people and train them up rather than improve the tools that we use to make the games with? Unfortunately, because it is. But why?
This is the off-the-shelf stuff that every team uses: Photoshop, 3D Studio Max, mudbox, Dev Studio, through to Microsoft Office and version control software. They're pretty standard across the world, though there'll be the usual fight as to whether the 3D artists use Max or Maya, and possibly mudbox or zbrush. But the concept is the same - several thousand pounds / dollars / euros gets you a fully featured piece of software designed to do something.
As an example, here's a video someone made of how to sculpt a head inside mud box:
That looks pretty quick yeah?
Let me explain. While it is fairly quick to create a head like this, this head is nowhere near ready to be put into a game at this stage. Just going over the basics it would need several textures to create the skin, eyes and hair (possibly all on one page, but unlikely in today's world). Then it would need a rig putting inside it - a simplified structure that all the vertices inside the model get assigned (and weighted) to, thus allowing for it to be animated. The list goes on and, with today's tools, it takes a talented character artist between 4 to 8 weeks to make a character ready for a game.
Ten years ago, that same process took roughly 3 days.
For sure, that's a sweeping generalisation. And initially making characters on PSOne took far longer but, by about the same time in the PSOne's lifecycle that we're currently at with the xbox360 and PS3, a good character artist could knock out characters at a merry pace. It all comes down to the complexity of the model and what it is for - a background character shouldn't have as much time spent on it as the player character does. And the visual quality of the art created today is, without doubt, far greater than ever before. And our tools are better. But they could be even better.
Every game developer has software unique to them, be it a simple batch process to copy data over from the server onto the development kits through to a fully featured engine and tools suite. It depends greatly upon the size of the company, its age and the general consensus towards developing internal software. There's a balancing point between reinventing the wheel and cutting edge research project that is very difficult, and costly, to find. Realtime Worlds created this amazing character creator for All Points Bulletin:
For whatever internal reasons the company as a whole didn't work, but whoever worked on this editor deserves a lot of credit and it's things like these that I hope we'll see a lot more of in the coming years.
I've picked on one very focussed area of games development for this post, but it's equally possible to use similar examples across the board of content creation, from level design to cinematics. The problems facing the traditional industry are very real: we're under a lot of pressure to bring costs down and create games faster because, traditionally, we've been able to do that on the previous rounds of hardware by this point. There is little doubt that we need to improve our tools and methods for creating content, but the difficulty is finding out where the tipping point lies: where the cost barrier of more men Vs better tools comes into play. It's not easy, but if you've got a great idea for a tool that could be of help to you, write it down and get someone to make it for you. It should pay dividends in the end.