Head in the Cloud

Confounding many of its detractors - including this columnist, I admit - the OnLive service is now up and running. Early users seem impressed by certain aspects, such as the community features, but nonplussed by others - most notably, the performance of the service in fast-moving, graphically intensive scenes, which has drawn criticism from most quarters.

As a result, OnLive presently seems like a pretty bad deal as a replacement for a games console or high-spec PC. However, the very fact that the service has launched at all and works in some contexts has struck a blow for the idea of cloud gaming. Internet speeds will continue to rise rapidly in the coming years, while the cost of powerful video decoding hardware for the client side will continue to fall, helping to eliminate two of the key bottlenecks which OnLive-style services presently face.

The extent to which those obstacles will block the popularity of cloud gaming in the coming years should not, however, be underestimated. Internet infrastructure by its very nature is plagued by bottlenecks. There are whole swathes of the planet which don't provide connections fast enough to operate OnLive even in its present state - and not just in poorer countries, either, with many developed nations (including large parts of the UK) still regarding 2Mb as a top-end connection.

Even in places where connections are fast enough, contention between users at peak times or congested pipelines between ISPs can drastically reduce speeds, making cloud games unplayable. More unreliable still is the situation in individuals' homes, which lies entirely beyond the control of companies such as OnLive.

The company would argue, for example, that its deal with BT in the UK will bypass many of the network issues that could hinder the service - but it is powerless to stop your gaming from being wrecked by a flatmate starting to watch a HD video on iPlayer or downloading something large from iTunes or Xbox Live. Many homes also use WiFi connections to hook up their various devices, opening themselves up to a host of interference problems from appliances or other networks - problems which most services are robust enough to survive, but which would seriously impair an OnLive session.

These are not insurmountable problems, but they are not minor issues either, and it is naive to make assumptions about a timescale on which they may be resolved. These technical issues mean that the audience for cloud gaming is heavily restricted at present, and it's not unfair to suggest that the growth of the addressable audience will be quite slow.

There is, however, an even bigger question facing OnLive - one which cannot be answered by reference to the march of technology. That is the question of the business model for the service. At present, it allows consumers to play a lower-quality version of existing games, for which privilege they pay full price for the game plus a monthly subscription fee. It is a singularly unattractive proposition, one which reduces all of OnLive's technical achievements into being an answer to a question nobody has asked.

Actually, that's not entirely fair - there's a select group of people who are asking questions to which OnLive is the logical answer. They are, of course, games industry execs, for whom the concept of a system which takes ownership of games entirely out of the hands of consumers is something of a holy grail.

It's not merely that this could, in theory, be a silver bullet against the demon of piracy, either. In common with other media industries, games companies have shown great enthusiasm in recent years for moving away from the idea of selling products towards a new idea - selling licenses. Under this system, you never actually own a piece of software, you have merely paid for a license to use it in certain restricted ways.

The widespread adoption of cloud gaming would be the ultimate triumph of this approach to business - finally allowing publishers to bypass those pesky consumers who seem loth to give up the old-fashioned idea of actually owning things.

This ointment, however, is full of flies. Firstly, there are the huge technical challenges which must be tackled before cloud gaming's audience can grow to encompass a large proportion of the audience, let alone a majority. Then there's the even bigger question - how on earth do you persuade consumers to actually use this thing?

Bluntly, game publishers right now find something like OnLive sufficiently attractive that many of them are blinded to just how unattractive it is to consumers. The sole proposition of the service as it stands is that, rather than buying a games console or a new graphics card for your PC, you can pay $15 each month for access to the service.

For that $15, the consumer loses any concept of ownership of the games he buys, including the right to lend them to friends or sell them on to fund further game purchases. He loses the ability to shop around for better game prices, to hunt out a bargain in a sale or on eBay, and is locked into the incredibly optimistically high prices publishers continually lumber digital distribution services with.

Despite paying over the odds for his games, and a subscription fee to add insult to injury, the consumer doesn't get graphical performance or low-latency gameplay on a par with playing on a local system. Within ten months, he'll have pumped $150 into OnLive, plus plenty more dollars thanks to the premium on game prices - easily enough to buy a great graphics card or a new console - but, as a final slap in the face, he will lose access to his library of games if he ever stops paying up.

Needless to say, it's hard to see consumers leaping at this opportunity - and until the world is ready to move en masse to a cloud gaming service, the advantages which publishing executives dream of simply won't materialise. While code is still resident on consumers' machines in any form - be it downloads or discs - there will still be piracy, and while physical products are still available, consumers will still choose them simply in order to be able to exercise their age-old right to lend or sell on the products they buy.

If the industry wants cloud gaming to work, it will need to work exceptionally hard to convince consumers of its advantages - and that will involve opening wallets, not just marketing budgets. Prices must fall - drastically so. The present pricing model may simply not be appropriate, in fact, since a system whereby you buy a game but can be denied access to it when you stop paying a subscription fee at a later date is not only deeply unattractive to consumers, but may also attract regulatory attention, given time.

Despite all of this, there's no question but that the technology being used to provide OnLive is impressive, and will improve over time. However, I suspect that its uses will be rather different from those envisaged by OnLive's executives - and by the publishers who have become enamoured of their vision. Streaming live gameplay over a home network, for example, is an appealing possibility, as is the potential for logging into persistent world games on your home machine remotely while "on the go", or even providing playable demos of retail titles on a publisher's website, a market which the rival Gaikai service has targeted.

These are scenarios in which the technical limitations of cloud gaming will be more acceptable, and which will add value to existing games rather than attempting to move consumers to a radically different business model which costs them more, provides them with a poorer experience and disenfranchises them of their consumer rights. Until the business model can be substantially changed, it's unlikely that cloud gaming will gain much traction as an alternative to PCs and consoles - but the impact of the technology itself could still be impressive.