Slaying sleeping giants
We couldn't let Hook go before pressing him on the big, burning question: Why is Intel doing all of this? PR and events aside, making a dent in the discrete graphics market is a formidable task, to say the least. Intel already has its hands full, with the decline of Moore's Law and public whiffs on process node shrinks denting the chipmaker's armor, not to mention the "ryze" of AMD as a true competitor in the CPU space over the last couple of years.
Hook fully acknowledges the great challenge that looms ahead. But he's optimistic, and he makes it seem like Intel is positively crackling with excitement over its discrete graphics endeavor.
Of course, this is precisely the time to be excited. Intel has recently looted AMD's RTG group of talent, and there's what amounts to almost a blank slate that all those engineers can work from. It's true that Nvidia and AMD have had a 25-year head start, but for hungry and ambitious workers, the notion of putting their own stamp on a discrete graphics roadmap must be tantalizing. Enthusiasts are buzzing over the prospect of a true challenger to the graphics establishment. Nvidia and AMD are surely annoyed at best and worried at worst. And there's still more than three full quarters before Intel has to show anything at all for its efforts. I imagine meeting rooms around the company are full of brainstorms and moonshots, and lots of new employees are enjoying the salary boosts that lured them from their previous employers.
Next year will be a different story. Expectations will be high, and consumers will be impatient. This is where the unbridled optimism gives way to pragmatism. "We're not going to be able to deliver everything on January 1st, 2020," said Hook. "It's going to take time to do it right."
Hook admitted that Intel needs to learn from past mistakes. He also said, bringing things back around to the original impetus for the conversation, that "I think for us to be successful, we have to look at discrete graphics as a grassroots movement." (An odd take, perhaps.) He continued, "My belief is that something like the Odyssey [...] really is the core of the marketing. And there's still a lot of other things we have to do, but this is the thing we have to do first."
Some of the practical parts, though, are themselves reason for optimism. Hook pointed to the roster of fresh talent, Intel's rich graphics IP, and the investments the company has and will make into the venture.
I stopped him and asked if he thinks that Intel can and will spend what it needs to. Answer: "I absolutely do."
He also pointed out that there may be an opportunity to catch the competition on its heels. Nvidia and AMD, Intel seems to have decided, believed that there would never be another serious entrant into the discrete GPU space. "It was clear that some of the market participants were getting really comfortable," said Hook. He added, "At the same time, there are a lot of people who are disenchanted with Nvidia, and Radeon," gracefully avoiding saying the name of his most recent former employer. "I think our opportunity to be successful is that we have to be joined at the hip with the community."
The optimism re-emerged: "Between the talent, the IP block, and just the willingness to win, I think we're going to do really well in the space. But," he added with a tongue-in-cheek flourish that probably won him a gift card or something, "It's...an odyssey."
Indeed it will be. Competition always benefits consumers, so in that sense, Intel stomping its way into the discrete graphics market will be good for all of us. But in a year, we'll see if chipzilla can begin to deliver on its promises.