nVidia Should Keep Innovating Instead of Attacking Intel

A few days ago Brooke Crothers from cNet's Nanotech blog posted some comments by an analyst from CRT Capital Group that I thought were spot on:

"Huang seems to believe that Nvidia's graphics solutions are better than Intel's because Intel simply doesn't know how to do better. But there is another element that Huang seems to overlook--Intel has not, so far, been interested in the high-end 3D-gaming market, other than as a vehicle to sell their own high-end CPUs."

From the first second I saw this whole thing starting to unfold, I just got a real bad feeling in my gut for nVidia. For so long, there's been this monstrous battle at the core of the industry: nVidia + Intel vs. ATI + AMD. It certainly hasn't been made up of official alliances, and in recent years since AMD's acquisition of ATI and their subsequent plummet, the game has changed significantly, but the fact remains that deep down Intel and nVidia have complimentary technologies. Why, at the height of success, would you want to start burning bridges? It's just downright shortsighted. This past year has seen a severe breakdown of relation between the two companies, and I'm not a big fan.

nVidia's got a long way to go to attack Intel's mainstream business, and, in the long run, they will continue to contend with ATI over the GPU, with the possible addition of Intel (via Larrabee) as a competitor. Their domination of the industry is by no means permanent. So, now, with the industry temporarily at its feet, nVidia needs to look to its future and innovate! Buying Ageia was a great move, but there needs to be more. No one knows the visual computing market better than nVidia, so I can't assume to know better than they as to what advancement of that industry would look like. But whatever it is, concentrate on that. Open up new demand, new markets. Go one step further and be just a little bit smarter. Tap potential that has yet to be tapped.

And in the mean time, no more cowboy PR fiascos, please.

----
05/09 - Brooke Crothers just this morning posted a great interview with Jen-Hsun Huang about this very topic, so I'm going to tack my reaction onto this post since it's directly related.

First, I don't think Huang's "fearless challenging of Intel" is admirable at all. I think it's a careless, emotional response, and I don't think that belongs in the business world. If Huang thinks Intel is a monopoly and is acting like a monopoly, then he should take them to court. Don't play PR games. This type of public activity makes nVidia look small and threatened and I would be very uncomfortable if I were in the upper ranks there.

Huang makes several comments about Larrabee and Pat Gelsinger's comments about ray tracing possibly replacing rasterization as the dominant graphics architecture. "They shouldn't be able to say that other peoples' businesses are going to die," Huang says.

Seriously? Gelsinger is certainly entitled to his opinion on where the industry is going, and is allowed to talk about an exciting technology that he thinks will make a big impact on the market. He said nothing about nVidia directly and did the whole thing in a professional presentation as his own IDF conference. Huang's reaction, on the other hand, is an outburst. It's galling, really. Yeah, he is right that Larrabee is nothing but a PowerPoint slide right now, but if ray tracing revolutionizes the graphics industry, then that's a good thing. Nobody is going to feel bad for the companies that have inferior technology. How bad does Huang feel that AMD/ATI is getting ground into the dirt right now?

I say stick to the facts. Does nVidia think ray tracing is bunk? OK, let's hear about it, then. Where's your PowerPoint slide? Tell us why it won't work or won't happen, and if you think Larrabee is a myth designed to just "cast a shadow" over you, then all of the PR around it that you've created has done nothing but harm you and your investors. This is how level-headed business people should approach a problem. Instead, we get "I believe that the entire world believes that what Intel does is build a factory, stuff that people don't want to buy, and then shoves it down its customer's throats."

Brooke makes the comment at the end of the interview that Huang's attitude "borders on paranoid," but that former Intel CEO Andrew Grove's credo was "only the paranoid survive" in Silicon Valley. Fine. But when Huang's mantra is "Intel cannot share the world with someone else," that goes beyond paranoid and becomes defeatist.

I say walk away from this whole thing and get back to work. Look to your own business and keep innovating, yourself. I've heard enough.