Valve boss unimpressed by Next Gen consoles
Gabe Newell is the Founder and Managing Director of Valve. Prior to that, he spent 13 years at Microsoft, where he held a number of positions in the Systems, Applications, and Advanced Technology divisions. He ran program management for the first two releases of Windows, started Microsoft's multimedia division, and led the company's efforts on the Information Highway PC. This guy knows what he's talking about.
And recently, Newell commented on the upcoming consoles from Sony and Microsoft, and wasn't exactly all happiness and light. 'Technologically, I think every game developer should be terrified of the next generation of processors', said Newell. 'Your existing code, you can just throw it away. It's not going to be helpful in creating next generation game titles.'
Newell leveled most of his criticism at the theoretical performance of multi-core processors that both giants are pinning their hopes on, pointing out that, 'Most of the problems of getting these systems running on these multicore processors are not solved. They are doctoral theses, not known implementation problems. So it's not even clear that over the lifespan of these next generation systems that they will be solved problems. The amount of time it takes to get a good multicore engine running, the Xbox 360 might not even be on the market any longer. That should scare the crap out of everybody.' Oo-er.
And he sees a whole new way of thinking and training as necessary for the future. 'Really good engineers are going to be much more valuable and engineers who used to be valuable writing game code in the previous generation may end up becoming thorns in the side of key programmers who can write multi-core game code.'
It was pointed out to him that learning curves have been steep at the start of every generation. So how will the next one be any different? 'It is different,' acknowledged Newell. 'But it is much more difficult now to write code that will have predictable behavior. We have performance problems now in the out-of-order universe because we have programmers who can't figure out why the changes they made caused the system to behave the way it does.'
'So one of the people who has a deeper understanding of the overall architecture has to come in and tinker around, more or less blind, because there aren't a lot of performance tools for that. [This guy needs to] give insight into what's happening in the cache memory, where a lot of this stuff goes wrong. It goes a lot worse in a multicore world, where there's a whole bunch of stuff going on in these separate cores, can suddenly have an impact on the entire system.'
From here, Newell guaged the additional difficulty out of ten, stating that 'If writing in-order code is a one and writing out-of-order code is a four, then writing multicore code is a ten. That's going to have consequences for a lot of people in our industry. People who were marginally productive before, will now be people that you can't afford to have write engine or game code. They can't get a big enough picture of what's going on in the box so they'll be a net negative on the project.'
However, Newell kept his most scathing criticism for the console manufactures' performance claims, and their contempt for their own customers: 'Statements about 'Oh, the PS3 is going to be twice as fast as an Xbox 360' are totally meaningless. It means nothing. It's surprising that game customers don't realize how they're treated like idiots. The assumption is that you're going to swallow that kind of system, when in fact there's no code that has been run on both of those architectures that is anything close to a realistic proxy for game performance. So to make a statement like that, I'm worried for the customers. And that we view customers as complete morons that will never catch on and that we're lying to them all the time. That's a problem because in the long run, it will have an impact on our sales.'
So where does Nintendo fit in all this? He didn't say. If they're going for a powerful single core, maybe that's the right way to go, according to Newell. But until we have optimised code running on complete machines, Newell thinks all claims are bogus. We'll find out in the near future if he's right.
