GTA 5 art director: "Let what's best for the game dictate the way forward
GTA 6 studio head Aaron Garbutt said that the most valuable lesson Rockstar learned ahead of GTA 5 was to kill its darlings, and "let what's best for the game dictate the way forward."Speaking to Edge magazine back in…

GTA 6 studio head Aaron Garbutt said that the most valuable lesson Rockstar learned ahead of GTA 5 was to kill its darlings, and "let what's best for the game dictate the way forward."
Speaking to Edge magazine back in 2013, Aaron Garbutt, who served as art director on GTA 5, was asked what the most valuable lessons Rockstar had learned from earlier Grand Theft Auto games had been. In response, he said that it was "mostly not to get locked down to anything, whether it's a process, a plan, a goal, an asset, an idea - anything."
Garbutt said the key was "to keep completely open-minded, adapt to the situation and always look for a better way forward, rather than stick with what's been decided previously. I don't mean chop and change for the sake of it, but let what's best for the game dictate the way forward. With a game that's as large and complex as this, the ideas need to be fluid, and you need to be paying as much attention as you can to every aspect to make sure they're working together."
His philosophy isn't exactly a brand-new one. The idea of killing your darlings - abandoning even your very favorite ideas if you find they no longer actually serve the project you're working on - has long been touted across the entire artistic industry. But applied to the complexity of a game like GTA 5, and its then-industry-leading open-world, the idea of knowing when to move on takes on increased importance.
Even back in 2013, Garbutt acknowledged that Rockstar was lucky to have an art team "that is extremely experienced at building open worlds," and was rarely stumped by brand-new problems. Even so, "there's always new things that come along [...] and the nature of how we approach development means that we need to be ready to adapt." For Rockstar, that's a layering approach, where the world is built first, and then missions, story, and game structure are added on top. "Each of these things influences the others massively," Garbutt said, "so each needs to adapt to the others. We are always playing and refining, adjusting each element of the game to play better with the others."
That approach puts Rockstar in a spot where even when it does do away with ideas that aren't working "we're rarely throwing away content." Instead, plans are adapted, and the studio is "constantly looking ahead with a clean slate, rather than sticking stubornly to old decisions."
It sounds basic, but Garbutt's idea here is a pivotal one in a medium as complex as game design. There's a though that few, if any, other forms of art are trying as hard to not get made as video games are, and when some ideas work well with certain features but clash with others, or take ages to actually implement (often at the expense of other systems), reworking them, or even abandoning them entirely, can be an important way to actually get projects finished on time. Sometimes that means that the grand ambitions of the developers don't get entirely delivered on, but if that means getting the game into players hands - or even avoiding having your studio shuttered around you - it's an important discipline to hone.