Until a few days ago, the IDF Spring 2008 didnt appear to be the MID-fest that I was hoping for but after returning from a few days away at The Next Web conference and having time to read through all of the presentation slides, it's clear that the MID juice was definately flowing. There has been progress in the technology, with Moblin, with software vendowrs and hardware vendors but I'm still left with question marks next to some of my more detailed queries.

The main point to come out of IDF and one that probably affected the detailed MID news was that the launch of Atom has been pushed out to June. This is two months later than previous Intel estimates and extends device availability way into summer. In most cases, customers waiting for the first hands-on reviews with retail devices are going to have to wait until July or August to finalise even early decisions. If there's another slip of just a month or two we might even see marketing departments waiting until Autumn as they take advantage of the Christmas buying wave. The Olympics will also push a few manufacturers to get devices out in time for Asian marketing campaigns. Lenovo will be leading this as an official Olympic sponsor and the manufacture of the Ideapad U8. [More after the break...]

When the Ultra Mobile PC market got it's big boost from the Origami project nearly 2 years ago, motherboard and layout designers had a fairly slim choice of processor technologies to choose from. VIA made their C7 processor available in Ultra Low Voltage (ULV) format, Intel did the same with their low-end Celeron and Pentium range and AMD offered their Geode LX CPU and chipset. We saw mid-range devices like the Amtek T700 and Samsung Q1 being built on the C7 and Celeron devices, low-end devices like the Pepper Pad and Vega being built on the Geode and eventually, higher end devices like the Q1P and i7210 being built on the Pentium. All four CPU's presented challenges to size, heat and cost calculations and resulted in devices that, while amazing in terms of size, were restricted due to component size and heat dissipation issues.

On the 1st anniversary of the Origami launch, Intel announced that they would make a new optimised platform and were going to build, from the ground up, a brand new processor architecture for portable devices focusing on the 4-5" screen size and consumer designs. Since then, the processor landscape has changed somewhat and we're now in a position where there are more than 7 different processor architectures and platforms for designers to choose from making optimal designs easier and cheaper to achieve.

We hear almost weekly about new battery technology that's going to bring week-long battery life or battery tech that's going to bring higher safety and eco-friendly characteristics. The truth of the matter is rather boring though. The whole consumer battery and consumer device electronics industry hinges on a huge Lithium Ion Industry. The 18650 industry-standard Li-Ion cell, for example, is possibly the most common component you'll find across the whole range of consumer electronic devices and is manufactured in billion-plus quantities per year. Its a huge market and one that major investments have been made into for long-term profits. How can anyone expect this to change overnight? It won't. The process of making Li-Ion batteries is so established that it would take either a major disaster or major government intervention to change it. Well thats my opinion anyway! And that's why I think it makes sense to take not of what's being said inside the current consumer electronics industry and to put aside the dreams of the gadget blogs!

I've been looking in detail at some of the presentations given in sessions that took place at IDF Fall and there are some excellent PDF downloads available. The session on battery life - "All day and Beyond" is the one I'm looking at here...

I've talked about the 'Carrypad concept device' before. It's my ideal UMPC device specification that started the ball rolling with the Journal in Feb 2006, before the MS UMPC was launched. The details got a bit lost in the journal archives underneath all the other UMPC news so here's a re-post and summary of the Carrypad concept device.

Lets talk about mobile Internet devices of today. Mobile, consumer devices that offer Internet and media capabilities on-the-go and that are consumer-priced and marketed. I'm talking about iPhones, Nokia tablets, Smartphones, PMPs and the like. In general, these devices are based on an ARM architecture and for good reasons too because the ARM architecture has a lot of advantages. ARM-based devices have the advantage of always, or, instant-on and for certain tasks you can not deny that instant-on is faster. There's a clear advantage in battery life and even the Internet browsing compatibility issues are being cleared up. The browser on the N800 is a great example of how the experience is advancing towards desktop standards. But its not nearly there yet, and its slow! Very slow and that's a serious problem because however good the browser engine is, however cool the device looks or feels, if the cant give you your results in a reasonable amount of time, its not worth it. The customer will turn off.

The mobile Internet experience needs to be compatible and quick and if either one of those elements fails, then the customers Internet experience fails.

This is becoming a recurring theme in my UMPC discussions lately. I keep referring to the Full Internet Experience (usually while defending my decision to sideline ARM devices in the portal here) and its time I actually explained, in detail, what I mean.

Let me first separate 'productivity' UMPCs from the consumer UMPCs I'm going to talk about. I believe that for productivity, you need 7", keyboard, mouse-pointer/pad, a desktop-oriented processor platform and a full desktop operating system. Consumer UMPCs are different. Mini keyboards, low-power 'just enough' CPU's and GPUs, small solid-state disk drives and 4-6" screens bring real challenges and the operating systems that drive them need to be efficient, small-screen friendly and need to contain tailored versions of all the applications a consumer is likely to need, just as you would find on a PMP or consumer navigation device. Take the PepperPad concept as an example. Its a consumer-focused device at a low price with a tightly controlled operating system build, a tailored set of apps and it works quite well. Just like the PepperPad focuses on the Internet as its core feature, so too will the consumer UMPCs, MIDs, Carrypads and Webpads. There will, or course, be specialist applications that will differentiate different devices but through reasons of technology and marketing, UMPC's will remain fairly thin Internet-focused devices.

Before that can happens, someone needs to develop an Ultra Mobile, finger and Internet-oriented operating system. Windows XP is OK for UMPC's but Vista is taking over and its far too heavy-weight, un-optimized and finger-unfriendly for a consumer UMPC device. Apart from Vista, there really isn't any other OS out there yet. Before a new OS is developed we need to lay some ground rules down about the FIE. Without some sort of agreement between the members of a dev team, how can we expect to reach the goal of an Ultra Mobile OS and how can we hope to see the success of UMPC's in the mainstream? Read on...definition follows.

Last Friday I did something stupid. I exercised! I exercised too hard. It was too hard and too soon in my Solar UMPC training schedule and as a result I hurt my back. I spent the next three days either laying down or hobbling round like a on old man and finally today, I'm able to sit at my desk for short periods. [Excuse me while I go and get another cup of tea - I need to move about you understand.] During the time on my back I was pretty happy to have a couple of UMPCs around me. I used the Q1 (a bit too heavy to use while lying on your back) and the N800 (lighter but slower) but finally I settled on a common garden paperback book. (common garden UMPCs don't exist yet as the screens aren't bright enough!) The book is called Crossing The Chasm [aff.] by Geoffrey Moore and it will be familiar to many marketing people in fact I picked up the tip on this through 'Brown Knows', the blog of VIA's Richard Brown. Its an amazing book. Not because there's anything ground-breaking in it (I'm not formerly trained in any form of marketing but it all seems to make good sense to me) but because on every other page I'm hearing 'UMPC' in the back of my head. I'm only half way through the book but I guess I've had to stop 20-30 times to ponder on a paragraph or to make some notes related to UMPCs. I've expanded my notes and thoughts here into a fairly long post but as is often the case here, its a post that's as much for myself as it is for others. I need to get these thoughts organised and writing a report is the best way I know of doing it. You might want to hit that PDF icon there and print this one out for the train home! If not, read on...

I was reading a review of the Archos 704 WiFi today and that question came into my mind again. 'Can an Uber-PMP perform the browsing functions of a UMPC?' Could it even replace the UMPC? The answer is - not if its called '704 WiFi.' and my feeling at the moment is, to be honest, not if its RISC based. In all fairness, the 704 is not marketed as a UMPC or a web browsing device, but the form factor of this new generation of PMPs (including the Cowon and Clearview) coupled with their specifications and high-aiming marketing makes them something that would be be considered if someone was looking for a consumer-level ultra mobile computing solution - or Mobile Internet Device should I say? Or Ultra Mobile Device should I say? [Intel, VIA - Stop it!]

This review of the 704 WiFi (via SlashGear) sums up the problem though. In reference to the browser...

My first major complaint is the speed. The browser takes quite a while to connect to an access point and actually start up. Going to most websites takes a while (roughly 15 seconds in most cases, depending on the site). Obviously, sites like Google loaded up fairly quickly, but going to other sites like Slashdot or Digg took a long, long time.

The nail in the coffin was the lack of flash support. Being able to view flash videos on the go would have been awesome, but there's nothing to be seen here. Flash 7 support could have been possible, but even that wasn't implemented.

Its not only the 704 that has this problem either, many of the Windows CE-based devices, even those with the very good Opera browser, have issues with either processing power, plug-in support or simply screen size. Client-side processing requirement for browsing are increasing at a pace, browser real-estate requirements are also going up and when you think about the number of plug-ins and add-ons that are being written for x86-based browsers you wonder if the RISC-based app's are ever going to keep up. Even the browser-focused Nokia N800 isn't up to scratch. Flash video overloads it and Ajax or JavaScript-rich pages either cant be rendered or grind the device to a halt. There really needs to be a step-change in the processing power of these low-end devices before the complete Internet experience is possible.

When I started thinking about UMPCs a year and a half ago, I focused on a consumer level device. It was the device that I needed and I defined it on paper and eventually called it the Carrypad. The following blog was called 'Time for the Carrypad?' [defined in Net-print here.] It was, effectively, a MID, a mobile Internet device, a low-end device with full browsing capabilities and a efficient software suite. Here's the summary of the Carrypad concept. I got the screen aspect ratio wrong (it needs to be wide screen to get the keyboard to fit!) but the rest is pretty much what Intel define as a MID.

• 5-7" 800x640 Touch Screen
• Detachable Mini Keyboard
• Wifi, BT2.0, SD slot, USB2.0
• 600mhz advanced processor
• 3 hour battery
• Min 10GB storage
• Reduced OS
• Price$600

I originally thought that a RISC/ARM architecture might have been good enough as an 'advanced' processor but in the last year I've seen a lot of new and demanding Internet applications. Glide, Google App's, Speedtest, Picasso, Ustream, Blip.tv and I've seen where people will be going with their Internet experience. I think my Carrypad concept, while good enough for me a year and a half ago, might not be good enough for the 2008 market. The processor definitely needs to be x86.

Pepper had the right idea a few years back. They moved their software development onto the x86 architecture and now produce one of the most capable browser-focused internet devices on the market. The PepperPad 3 might not be MID sized but its got the MID heart and shows what can be done with low-power processing and tailored software. MIDs aren't far away now and look like they will hit the Uber-PMP's head on and if the MIDs combine the ease-of-use and portability of an Uber PMP with the style and pricing that's required, the Archos 704, Cowon Q5 and Clearview will have some tough competition. x86 is where its at now.

As a result of all this focus on x86, I'm considering taking all the RISC-based devices out of the portal. Its a harsh move but I think its the right move for the future. If you're looking for a mobile internet experience, I can't honestly say that the E90, The N800, the HTC Advantage, the Sharp Em-One (I'll miss you) or the Uber-PMPs are going to bring you complete Internet happiness. What do you think? RISC out, x86 in? If you've got reason to believe that the RISC browsers are going to pull through, let me know now because, for the interests of the UMPC-focused visitors here, I'm very close to pulling all the RISC/ARM devices out of the database. N800, that means you too. I'll give it one more month and then, unless something comes along to prove me wrong, they're OUT!

Since the launch of 'Origami' in March 2006, there have been many varying reports about what an Ultra Mobile PC is and why someone might buy one. This guide has been written to help those looking at ultra-mobile PC's with a view to buying one. Its a long report but we guess that if you're considering spending over $500 on a UMPC, you'll want to read and learn as much as you can.

Battery life continuously tops the list of complaints that people have about Ultra mobile PC's and over the last 48 hours there has been a huge amount of talk from Intel, blogs and potential customers about Intels new UMPC platforms and how battery life is going to improve. I thought it might be worth giving a little time and thought to Intel's Ultra Mobile platform (UMP) 2007, putting some figures down and trying to work out what UMP 2007 really means in terms of battery life. Its not an easy thing to do when there are no devices available but we've seen some figures now and luckily, I have some real-life figures to use from testing and playing with the Q1 Ultra at CeBIT in. Read on for details about UMP 2007 and battery life.