Nvidia's phone chips as the camcorder, console killer
Gameboy killer? No, no, bigger than that...
3GSM Sometimes the product development goes by too fast for you to notice the revolutions as they pass, and Nvidia's announcement of the GoForce 3000 and 4000 chips earlier this week was a case in point. Yeah, PC class graphics coming to handheld platforms, substantial improvements in power dissipation, cool, but so what? Well, if Nvidia is right, there are numerous so whats, and numerous pretty serious consequences. Take, for example, what happens to the camcorder and games industries - there are plenty more, but that's enough for starters.
Nvidia is aiming for a $10 part that adds the capabilities of a good specification camcorder and good 3D graphics and texturing to standard mobile phones. As The Register pointed out, $10 is maybe quite a lot of money when you're talking about a device with a build cost of $200, and these $5 and $10 parts sure as hell mount up, but Nvidia argues that the actual cost will be less than $10 because the chips will allow you to eliminate some other parts, and the power consumption advantages can also have an impact on the overall performance of the handset. And it says there are designs from several manufacturers in the works, so we'll just suppose for the moment that Nvidia's invasion of the handset market will be successful, and consider the consequences.
One of the joys of the mobile phone as a platform is that you've got it with you all the time, so if it's a pretty good camera/camcorder, then it allows you to snap things on the fly without your having to have your 'real' camera with you. It's a pretty short step from this to the phone actually being your real camera/camcorder, and you've therefore got the twin consequences of people's behaviour changing and the market for the standalone versions of the devices changing (in this case a combination of getting smaller in terms of units while pushing upwards in terms of features). With recent generation camera phones, we've got closer or maybe even moved past this point already. As the mass market tunes into the idea of camcorder phones, then demand for precisely the kind of wares Nvidia is pushing will climb.
What about games? People already play games on mobile phones, but the impact on the traditional gaming market has not been significant because we're really not much further than retreading old 80s computer games. So people play mobile phone games because they're there and they haven't got anything better with them to play. Java can exist in such a gaming market, because there isn't really much in the way of knife-edge competition to come in and kill it. Yet.
Again presuming Nvidia is getting it right, you're going to get all of the graphics and speed you'd expect from a PC or a console on a phone, in your pocket, all the time. So what happens? We had two interesting conversation threads at this point. First, we asked about form factors - no matter how good the hardware platform is, isn't a phone handset form factor too constricting for it to ever be more than a second best platform, and is there therefore scope for more diverse form factors like N-Gage, or Danger's Hiptop?
Nvidia rejects this. Alternative form factors are in general too maiming to the base telephony functions for them to bring gaming to an 'everybody' platform, and everybody, actually, plays games to some extent. If they get great games in their mobile phone, then they'll play them when they get the chance, and if you go thinking there's a gamers market and then there's a big pile of non-gamers, you're missing it. which is a plausible pitch we can just about take on board, although we'd expect designers to gently take gaming requirements into account when putting together future phones.
The second interesting thread covers the kind of development model phone games will follow. There are two patterns, PC and console. With PCs you've got a common platform to build for but that platform evolves continually. The PC games development market is much more open in that it doesn't have licensing tied to a specific platform, and because the payback mechanism differs - you don't subsidise the hardware (NB despite appearances, PC manufacturers losing money are not doing so deliberately) and make your money back on software licensing.
The games console market is characterised by two or three successful hardware platforms which are stable for a period of up to five years, and where high ticket games sold over this period provide the payback. Well, it's kind of redundant to wait for Nvidia to tell you this doesn't work in phones, because quite obviously it doesn't - there are just too many hardware players out there, and the hardware platforms move too fast. The phone games market therefore looks much more like the PC games market, and Nokia might as well dump any plans to the contrary it still has, now.
Phones are different from the PC market in that there is no common hardware platform, of course, so it's actually the market dynamics that will be similar, but addition of a layer of software tools to the phone development market brings software developers sufficient commonality for them to be getting on with. At this point, we get a disturbing rave from the grave. OpenGL ES, says Nvidia, is coming together very quickly, and is becoming the standard API for handsets. Microsoft meanwhile has its own route, Direct3D Mobile, in preparation - we have been here before, you might reckon, wondering whether reliving the DirectX v OpenGL war is a preferable option to shooting yourself.
This time around though Microsoft doesn't own all the real estate, so maybe we get a radically different outcome.
But that's yet to come, and at this juncture we can maybe squeeze in one more consequence of a take-off of the handset as a ubiquitous mobile games platform. If that happens, then can we not anticipate games played on other platforms being consigned to a similar fate to the one we've anticipated for camcorders? They become more specialist, they cease to be the revenue sweet spot for games developers, and the point of a PS2, an Xbox or a PC as a centre of gaming begins to evaporate. That could amount to a pretty substantial cull list for a $10 part, and there's surely got to be more where that came from. ®
Sponsored: Hyper-scale data management