Tag Archives: intel


Intel WiDi Wireless Display LogoI missed the release of a whole new acronym  – Widi – which is, broadly speaking, wireless display technology that will see PCs hook up with displays (that have special network adapters connected to them) without the need for any sort of cabling.

This sort of thing has been tried before but thanks to the fact that Intel is backing this particular standard it likely stands a chance where it hasn’t before. Huzzah – anyone with any IT skill who works in an office knows that helping people hook up a laptop to an external display – trivial though it is in theory – is inevitably a cause for some tedium.

With Windows 8 and mobile computing, CPU manufacturers are interesting again

One of the bizarrely fascinating things about the next generation of computing hardware and software is the fact that microprocessor manufacturers suddenly have an interesting stake in the game. Over the last few years, the Intel/AMD speed race turned into a tediously uninteresting one-horse race, and the mobile processor guys just kept quietly plugging away in the background… and now; ARM and Qualcomm are suddenly a threat to Intel – which – other than its Atom line – hasn’t made significant inroads into the mobile computing market at all.

In a bizarre twist, Microsoft is trying to tie chip vendors to a single hardware manufacturer for Windows 8 (I have no idea how this would work). I suspect this is its ham-fisted attempt to get some consistency of experience established – so that one Windows tablet is very like another – but like the Acer CEO JT Wang, I’m extremely doubtful about the effectiveness of this scheme. The theory makes sense; Qualcomm chips are suited to different form factors to Intel chips (right now, anyway) so it’s not like the Intel/AMD battles of old – where one chip was interchangeable with another.

Still, it will be interesting to see if this new wave of competition will spark some interesting form factors. Perhaps a shoe-tablet

Moore’s law, suffering

In the 90s, as a PC-gaming geek, I bemoaned Moore’s law; Intel’s founder’s dictum that the processing power of computers would double every two years or so (technically, the dictum might be that the number of transistors on an integrated circuit would double, but the former sounds more comprehensible and comes to much the same thing).

Why did I care? Because every new piece of software required a faster computer, and that meant, when I was trying to play Dune 2 on an ageing Amstrad PC2286 (clocked at a staggering 12 megahertz, costing the best part of £2k) that I had to do some considerable Macgyvery hackery to free enough system resources to allow the game to load – and it dragged agonisingly slowly when I’d built too many Harkonnen Devastator tanks (we didn’t have the money or inclination to upgrade every two years, certainly not so I could game more).

Today, over 40 years on from the original prediction, material and manufacturing limits are slowing the pace at which hardware becomes defunct. That is to say – the processes by which CPUs are created are reaching the point where we simply will not be able to make the processing elements any smaller, and so Moore’s law is slowing. Intel’s website explains, and gives us this handy graph:

Source: Intel

By 2020, Moore’s law will not apply to traditional CPU development. So one of two things will happen; first, companies will continue to shift multiple cores onto CPUs (dual, quad-, octo-core machines already exist) to allow for even more parallel processing. Second; we will move onto a new substrate for computer processing – potentially optical based computing or some such. I don’t think the technology for the latter is quite ready yet.

But the net impact for me, as a consumer and Deputy CIO for my company – lack of processing power no longer fuels hardware refresh in the way that it once did. A server I put in three years ago is showing no significant signs of performance degradation and the only real point of concern I have is that a hard drive will fail. Cost of replacement hard drive? A couple of hundred quid, if that. Cost of new server? In the thousands.

Of course, ideologically speaking, I’m looking to outsource all of this stuff (personally and professionally) to the cloud. But the UK’s internet infrastructure isn’t quite there yet – and neither are the services. But they move 10% closer every 6-12 months…*




* Yes, this is nonsense. I need to come up with some better laws.

Augmented reality all over the shop

@tim has a big piece in today’s FT about Augmented Reality and its potential in advertising. In addition to trying out the FT’s AR app, you could also try Intel’s thin laptop app and check out the AR magician on YouTube. The former two require flash, a webcam, printer and miscellaneous plugin. The latter requires flash, headphones and some imagination.

Augmented reality is awesome and full of practical potential – in design, medicine, gaming and no doubt a dozen other fields – but I’m not surprised its kicking off in marketing first. We seem to be a good test bed for this kind of stuff and have a few people willing to take chances on reasonably nascent technologies.

Charlie Stross’ Halting State has a number of interesting applications (Google Maps overlays, WoW overlays, police grid overlays amongst them) and is well worth a read if you’re into the topic (and Sci-Fi).