With Intel's focus on the new Atom-brand processors being described at the Intel Developer Forum this week, "Atom-powered" is the obvious description of the mobile Internet devices (MIDs) these chips will go into... and it seems like half the IDF stories on the Internet this week are using that phrase.
Intel's Atom processor (on the right) and its companion System Controller Hub code-named Poulsbo.
(Credit: Intel Corp.)Intel, however, seems to want even more hyperbole-- it expects people to believe that Atom will recharge the whole company. CEO Paul Otellini reportedly said "This is as important to Intel as the launch of the Pentium in the mid-1990s"-- but that's ridiculous.
The original Pentium processor and its descendants were responsible for nearly all of Intel's revenue. Atom will be merely a blip on Intel's financial reports.
The problem with Atom, especially these early models, is that the niche they occupy is a no-man's land between truly mobile devices like cellphones and MP3 players, and truly powerful devices such as laptop computers.
Atom consumes ten times as much power as cellphone processors and one-tenth the power of laptop processors. This power consumption makes for a device that has to be larger than a cellphone, and has to be smaller than a laptop because it can't provide comparable functionality. And there simply aren't enough applications that fit naturally into devices in that size range.
Look, I have as much experience with MIDs as anyone. I used an Apple Newton for seven years (as I've written about here several times). The Newton had roughly the same form factor and somewhat lower power consumption than today's MIDs. For a while I had a Metricom Ricochet wireless modem that gave me wireless Internet access.
But the simple fact is that the Newton wasn't useful enough to make me carry it around all the time. I loved mine because I had one critical application for which it was perfect. It was my electronic reporter's notebook, and no small-screen device could ever substitute for it. But most people don't need one of those.
And most people don't need a 5" to 7" display for basic Web browsing... at least, not enough to actually carry one around. And if you can carry something too large for a pocket, you can carry a small notebook PC that can handle a traditional notebook CPU.
Even after the Atom family evolves to the point that it can fit into cellphones-- which is the only way it's going to achieve significant sales volumes-- profits from these chips will never be very high. Intel's never going to achieve a monopoly in cellphone processors, and the competition from ARM-based cellphone chips will keep the value of a CPU core under a few bucks.
What Intel doesn't want people to think about very much right now is that in a cellphone, the CPU core is about the least-valuable part of the system. Even in a MID with an Atom processor, the CPU is just a tiny part of the whole package. Look at that picture up there-- the Atom processor is small compared with its companion system controller. In a cellphone, there's even more circuitry required for the radios.
And that's why Intel's never going to be an Atom-powered company, and I'm sure Otellini knows that in spite of everything he's been saying. But when your stock price has been trending downward for seven years in spite of the fact that you're running the world's largest semiconductor company with a stranglehold on the world's largest semiconductor market, I suppose you have to try to drum up as much excitement as possible for every new product that comes along.
I used an Apple Newton MessagePad 2100 pretty much every day for about seven years (see my previous blog entry about my Newton experiences).
What was so great about it? What kept me using it?
Peter's MessagePad 2100
(Credit: Peter N. Glaskowsky)Well, I used the Newton to keep track of the good things about the Newton--and the bad things too. So to write this article, I just powered up the unit (which still works fine, along with a Farallon Ethernet PC Card, a 24MB linear flash card, and two of the three nickel metal hydride battery packs, plus another pack for AA alkaline batteries) and reviewed the "Newton Notes" entry.
Updated 2007-08-02, 1500 PDT-- That's it in the photo. It isn't as pretty as an official Apple product shot, but that's how it looks today. If you click here, you can see a higher-resolution version of the photo. The stylus in the picture is a custom model I turned from a solid bar of titanium. It sits nicely in the stylus rest (shown extended) and locks properly into the internal storage slot.
The note shown is the last one I entered, when I took the machine to the Worldwide Newton Conference on January 14, 2006. Note that there are 2,777 notes in the Newton. There are a few hundred more that were created in the months before the Ricochet wireless modem problem I described in that earlier blog entry, and these were recovered from the device's memory to a disk file. So all told, I averaged well over one note per day for about seven years.
Here are some of the high points from my Newton experience:
• Never having to save files. Files are always saved. In fact, there were a few times over the years when the Newton rebooted or the battery died while I was in the middle of writing a note. After restarting, the note was usually only a word or two behind what I'd been writing.
• Ultrahigh reliability. Other than the Ricochet problem, the Newton rarely crashed (just a few times per year, not counting those caused by software testing, which I did a lot of), and almost never lost data. In fact, it was so reliable that I became quite blase about backups, and I never really worried about installing third-party software.
• Extraordinary battery life. My Newton would run for upward of 20 hours with the backlight off, even with the Ethernet and flash PC cards installed. I think the Newton was designed to keep the Ethernet card pretty much entirely turned off when it wasn't in use; I couldn't see a difference in battery life with it in or out. The flash card was used constantly, since it held a lot of software and data. With the backlight on, battery life was still in excess of eight hours.
This kind of battery life makes a huge difference in usability. I could sit in conferences and leave the Newton on for hours, continuously ready to take more notes. Even though the Newton turns on and off in about a second, I could set the idle timer to wait half an hour before turning the unit off automatically; battery life was just no big deal. I could also take weekend trips without taking along a power adapter. For trips up to about a week long, I'd bring along a spare battery pack and still not bother with the power adapter.
I notice that the electroluminescent backlight on my Newton is really dim now. It probably hasn't gotten any worse since 2004, but I probably just got used to it as it dimmed over time.
• The case had a nice high-friction coating. The slightly rubbery surface made it really easy to hang onto the Newton. Some people had a lot of trouble with this coating rubbing off, but it held up fine on mine. I'm sure that it helped a lot that I kept the Newton in a zippered slipcase made from a large travel wallet.
• The user interface for the notes application was basically like a continuous scroll of paper. This seemed a lot more convenient than treating each note like a separate document. On the other hand, it put a lot of pressure on the scrolling controls. By the time I had hundreds of notes in the machine, I needed something more than the standard scroll arrows. Fortunately, Newton developers noticed this need, and there were good software solutions available.
• The recognizer was an excellent match for my fairly neat printing. When I (or the recognizer) made mistakes, they were usually easy to correct by double-tapping the word to get a list of likely substitutes. I did notice that the list of suggested fixes seemed to be based exclusively on likely recognition errors; it would have been useful to get spelling corrections too.
• Direct overwrite also worked well. If I could see a single letter that was misrecognized, it was usually faster to just write the correct letter over it. Today's Tablet PCs don't respond well to this approach, and I think that's a major omission.
The default mode of operation was that text was always recognized and converted to a font-based display, as opposed to the default in Tablet PC today, which is to recognize the text but leave the ink on the display. This creates the potential for the user to be distracted by the recognition process (and errors) but leads to higher-quality recognized text. This trade-off favors experienced users at the expense of initial impressions, but I think that's the right trade-off to make.
And there's a subtle point here: the default Newton font looks like very neat handwriting--a lot like the lettering in comic books. This means that characters with similar handwritten shapes have similar font shapes too. So if the Newton misrecognized a letter here or there, the resulting word was easier to read correctly by virtue of the same skill that helps people read bad handwriting.
In reviewing this "Newton Notes" list, I realize that there were a lot of things that bothered me at one time or another about how the Newton worked, and I spent a lot of time thinking up ways to improve the software and hardware. I'll blog about these notes later.
[Click here to read part 2 of this piece, "What wasn't so great about the Newton?"]
Today the iPhone is the alpha gizmo, the one item of consumer electronics that dominates all the others.
But in 1993, the hot new gizmo was Apple's Newton, and it was a whole different thing.
Not very many people had Newtons. Apple sold fewer Newtons over the whole life of the product than it sold iPhones the evening of June 29.
Also unlike the iPhone, the first Newtons weren't even very useful. Although called "personal digital assistants" (PDAs), using a Newton was significantly more difficult than using a Day-Timer. The original MessagePad had very poor handwriting recognition, and there was no practical alternative to using it, no on-screen or slide-out keyboards. With patience, one could make notes, manage an address book and a calendar, and even send and receive faxes.
But honestly, it wasn't very good at any of these things. The return on the total investment, including the up-front costs and the time and effort of learning to use the device, was not so good.
Apple introduced several minor upgrades of the original MessagePad-- four new models in two and a half years-- addressing some of the hardware and software issues, but it wasn't until the MessagePad 2000 came out in 1997 that the Newton finally realized its full potential.
I had been watching the progress of the Newton very closely, trying to persuade IDT (Integrated Device Technology, where I was working during those early years of the Newton) to bring out a MIPS-architecture processor for this new PDA market. In 1995, I even made my own wooden prototype PDAs to show just how small a PDA could be using the technology of the day-- unaware that Palm's Jeff Hawkins had done the same thing the year before to help get the Palm Pilot project off the ground. (When the Pilot came out in 1996, I was entirely uninterested. Graffiti was a crippling defect, as far as I was concerned.)
I bought a MessagePad 2000 in April of 1997, and it was immediately useful to me. The handwriting recognition engine had been significantly improved on the MessagePad 120, but was still constrained by that model's 20MHz ARM610 processor. On the MP2000's 162MHz StrongARM SA-110 processor, the new recognizer was nearly flawless for me after just a few days of practice. Not everyone had this kind of success, but I usually saw no more than one error per paragraph of text, and it was very easy to correct those errors.
I did have to learn to print a little more neatly than I was used to, but not much. One problem continued to dog me as long as I owned the unit-- when I print a lower-case "g", I start at the top right and draw the circle counter-clockwise, and sometimes fail to close the circle before drawing the descender, especially if I was writing quickly. The Newton's recognizer often interpreted that shape as a lower-case "s".
Anyway, the MP2000 was a great fit for me. In 1996 I had joined the staff of Microprocessor Report. Attending conferences was a big part of my job, and the Newton was the perfect device for taking notes during interviews, presentations, and while visiting exhibition booths. The Newton also helped me manage my schedule. I didn't use it as my primary address book, though. I found it more convenient to use my PowerBook for that purpose, with phone numbers in my cellphone where I could actually use them.
In 1998 I had my MP2000 upgraded to the MessagePad 2100 configuration, which basically just took the RAM configuration from 1M to 4M. That gave me enough room to fiddle around with more of the third-party Newton software that was out there. There was actually a pretty good variety, mostly from very small companies that specialized in Newton software. The Newton wasn't easy to write software for, and Apple didn't support third-party developers as well as they could have, but there was some great software on the market.
I also experimented with using the Newton for Internet access. This worked pretty well with a Farallon Ethernet adapter, but wired Ethernet on a handheld device isn't a great combination. I also tried Metricom's Ricochet wireless Internet adapter. That seemed to work, but I discovered after a few weeks of testing that it was somehow corrupting the data being written to the Newton's flash memory. Either the power draw of the adapter was too great, or the Ricochet's radio transmitter was interfering with the Newton's electronics. I hadn't been making backups of the Newton as regularly as I should have, probably because it had always been almost perfectly reliable, so this Ricochet problem caused me a lot of grief.
Ultimately I decided I didn't have any strong need for Internet access on the Newton. It couldn't substitute for a laptop anyway, so I stopped worrying about it.
By 1999 or so, I stopped experimenting with the Newton entirely; I had a good software setup, the machine did everything I wanted, and it was totally reliable. I used the Newton until 2004, when I left Microprocessor Report. I had a Palm Treo by that time, and still do. It's not a complete Newton substitute by any means, but I can use it to take short notes-- or, conveniently, voice memos-- and it's a better device for calendaring and contacts because it's always with me.
In 2005 I got a Tablet PC (a Motion Computing LE1600) to fill in that note-taking gap at conferences. The Tablet PC handwriting recognizer isn't as good as the MP2000's, but it's tolerable. Tablet PCs are also huge and heavy by comparison with the Newton, but again, that's tolerable. In exchange, a Tablet PC runs a fully-featured operating system (I now have Vista on mine), mainstream applications like Microsoft Outlook and the Firefox browser, and I bought a Sierra Wireless AirCard 850 HSDPA wireless Internet card for it.
I could go on about the design elements Microsoft should have adapted from the Newton into the Tablet PC. I suppose I will, in some future column. Some of these features would be a good fit for a future evolution of the UMPC (Ultra-Mobile PC), which today is really just a Tablet PC with a too-small screen. In time, I expect the UMPC will be adapted to be a better fit for its form factor, and some of the Newton's features would help.
So far I've felt no urge to get a UMPC, which some believe bridges the gap between the Newton and Tablet PC. The forthcoming HTC Shift is very tempting, however. I've held one, but I'll have to wait to see what the final features and price are like.
Most people believe the Newton was a huge failure for Apple. I had the chance to ask John Sculley, who was Apple's CEO when the Newton project was launched, about that. He pointed out that although the Newton never paid off its development costs as a product, Apple's early involvement in developing the product category-- particularly its investment in ARM, the company that developed the original Newton microprocessor-- paid off handsomely.
It seems to me that today's technology would support the development of a fairly Newton-like device-- about 12 ounces with a 7" screen, thin and rugged, with integrated wireless Internet or Bluetooth to borrow the connection from a nearby cellphone, good handwriting recognition, and plenty of on-board storage, selling for around $400. I'd buy one, but who else would?
If you had a Newton, or have your own ideas about this product category, why don't you add a comment? Maybe we can get some hardware company interested once again.
- prev
- 1
- next
