HDMI

If you've shopped for an HDTV, a PlayStation 3, or an HD-DVD
or Blu-ray pla­yer, you've probably heard about HDMI. It can
seem like just one of many connections on televisions or
home-theater receivers. But HDMI is more than a port on the
back of a TV (and the often expensive cable that fits inside).
It's a set of rules for allowing high-definition electronic
devices to communicate. ­­

Before the development of high-definition televisions, most
TVs displayed pictures in what is now known as standard
definition. The picture was roughly square -- its aspect
ratio was 4:3. Its resolution, or the number of dots that
make up the picture on the screen, was about 704 x 480 pixels.
The picture was interlaced -- each piece of the moving image
was really half a picture, but the pictures changed quickly
enough that the human brain didn't really notice. Finally,
older TVs relied on analog signals, which travel as
a constantly varying electrical current.

HDTVs, on the other hand, are digital. They use information
in the form of ones and zeros. This information travels
through cables as distinct electrical pulses. HDTVs have
an aspect ratio of 16:9, so the picture is rectangular. They
also have a higher resolution -- current HDTV standards allow
for resolutions of up to 1920 x 1080 pixels. HDTV signals can
also be progressive, meaning that the each frame of the
moving image is a whole picture rather than half of one.

So, compared to standard TVs, HDTVs have a wider screen, more
pixels and a faster refresh rate. Often, HDTVs can display
more colors than older sets. This means that HDTVs need more
data and need it a lot faster than standard-definition TVs
do. If an HDTV can receive this information digitally, it
also doesn't have to spend time or processing power
converting the signal from an analog format.

This leads us to HDMI. Created by a group of electronics
manufacturers, the HDMI standard is a set of guidelines for
creating high-bandwidth connections between digital devices.
With the right setup, HDMI can make a significant difference
in a home-theater system. The current standard can carry
1080p high-definition signals, and it supports eight channels
of uncompressed audio, enough for a 7.1 surround-sound
system. HDMI can cut down on the number of cables required to
connect components, and it can even reduce the number of
remote controls needed to watch a movie.

­But there's a catch. In order to take advantage of
everything HDMI has to offer, all of the components of a home
theater have to be compatible with them. Some of the features
HDMI touts also don't yet exist in the consumer marketplace.
In addition, there's a limit to how long an HDMI cable can
be, and some users complain that the limit is too short to
support convenient setups.

How to Build a DVR

The digital video recorder (DVR) is a wonderful piece of
modern technology. While the VCR may have freed us from the
rigid schedules of TV programming, the DVR has freed us from
having to learn how to program our VCRs, not to mention
trying to figure out what's on that pile of unlabelled video
tapes.

But DVRs have two major flaws -- you have to pay for the
privilege of using one, and you're stuck with whatever
capabilities the DVR you purchase happens to come with. If
you want to expand the space on your DVR for storing
additional programs, or want to tweak your user interface,
there isn't much you can do about it.

If you build your own DVR, you sidestep both problems. You
won't have to buy the hardware or sign up for a contract with
your satellite or cable company for the device, you won't
have to pay for the service, and you can modify and expand
your DVR all you want. And it turns out a DVR isn't very
complicated -- you can easily build one with off-the-shelf
technology. That's because a DVR is basically a hard drive
with a way to interface with your TV signal, plus some
software to control it. If you have an old desktop computer
lying around, you're already halfway there.

Laying the Foundation for Your DVR

The most basic element of a homemade DVR is a computer. You
could technically use any computer for this, but you'll be
installing some new internal components, including a new hard
drive, so a desktop is a better bet than a laptop. You could
use a Mac, but you'll generally find that you have more
options with a PC.

How powerful of a computer will you need? Well, that depends
on what you want to do with your DVR. If you're just
recording TV shows and movies in standard definition, then
you don't need anything very powerful. If you'll be recording
and playing back high-definition (HD) programming or want to
watch Blu-ray discs, then you're going to need a computer
with a little more horsepower. There are five things to watch
for when buying or setting up the computer that you'll use
for your DVR:

Processor - You don't need a top-of-the-line quad-core
processor to run a DVR. In fact, for basic DVRing, older and
slower processors will work fine, especially since the
encoding will be handled by a separate component (the TV
capture card). The bottom line here is that the processor
should be powerful enough to run a modern operating system.
If you're stuck running Windows 95 or earlier, you may have
a harder time finding compatible software.

Hard Drive - You'll be surprised how quickly hard drive space
gets eaten up by recorded TV shows. A single movie in HD can
use up several gigabytes of drive space by itself. Hard drive
prices have dropped considerably in the last five years, so
if you're repurposing an old computer, you'll want to swap
out the old drive for something larger. Remember, you can't
have too much hard drive space, so buy as much as you can
afford. You'll want a speedy drive to avoid problems with
video quality. Keep in mind that you can always expand later
by adding an additional drive.

Video Card - It's going to take a fairly powerful video card
to process the video signal and send it to your TV. If we're
talking about HD and Blu-ray, the video card will need even
more power. This is the part of the DVR that's going to
matter most to what you actually see on-screen, so if you're
going to splurge on any part of your project, this should be
it. There are many video cards to choose from, with new ones
coming out all the time, so your best bet is to check
audio/visual message boards for tips on which card is best
suited to your purpose.

Memory - A minimum of two gigabytes of RAM will help your DVR
run smoothly, although upgrading to four gigabytes won't hurt
(and, again, might be necessary if HD is involved). Memory is
inexpensive these days, and more RAM is almost always better.

Power Supply - A high-end video card and an extra hard drive
can draw a lot of power, so you'll need a power supply that
can provide it. A good way to get a rough idea of how much
power you'll need is with an online power supply calculator
like the one Journey Systems provides.

Next, we'll choose and install a TV capture card.

TV Capture Cards

A typical computer doesn't have a video input that can accept
a signal from a antenna or a set-top cable or satellite box.
You'll need a TV capture card, which has its own tuner to
receive television signals, to take the incoming signal and
read it. Better capture cards have onboard processors that
handle the encoding of the signal into the digital format the
show will be saved as. A capture card that does its own
encoding takes a lot of stress off the computer's main
processor. In fact, if you use this type of card, you won't
need a powerful main processor.

One important thing to look for in a video capture card is
the ability to accept an MPEG-2 transport stream in both DBV
and ATSC, sometimes referred to as digital hardware cards.
That technical jargon simply refers to the format that TV
shows are transmitted in over digital broadcasting networks.
With the United States and many other countries switching to
all-digital broadcasting, you'll want to avoid older capture
cards that only accept analog signals. Many cards can handle
both analog and digital signals.

A typical capture card will allow a user to record up to two
programs at the same time while watching a third. If you want
to record even more programs simultaneously, you'll need to
install an additional capture card. But remember to make sure
your power supply can handle the extra load.

Installing a video capture card isn't difficult. Most cards
are PCI or PCI-Express cards -- they fit into slots on the
motherboard of your computer. After removing the cover of
your computer, the card is installed by pressing it firmly
into the appropriate slot. The card is then secured with
a single screw. There are also video capture devices available
that are even easier to install. They plug into to any
available USB 2.0 port.

Mesmerizing Tube Bender Looks Like It's Pooping Steel

Bent tubing is key. Roll cages need it. Tube frame chassis
need it. Even the storage arrangement for my welding clamps
needed it. There are a number of ways to go about bending
tubing. Rotary draw benders like the Hossfeld Universal
bender make a tight bend at one point. Three-roll benders
create more gradual curves. But none of them compare to this
automated CNC tube bender that just spits out steel in any
shape you want. I could watch it all day.

Information on this Nissin bender is a bit sparse, but I've
come across these two videos of it's impressive capabilities.
Tube to be bent is supported, lubricated, and driven forward
at a constant speed by the stationary part of the machine. As
the tube is pushed through the machine, it is forced through
the moveable die, which positions itself appropriately to
bend the tube as needed.

The bend is originally specified in terms of the geometry of
the desired part. That bend geometry is translated into a set
of instructions that control the position of the moveable
die. The die will be positioned in the right place at the
right time such that the tube is bent the right amount and in
the right direction to produce a finished piece closely
matching the original specifications.

High-Tech Glasses Beam Info Directly Onto Your Retinas

Office workers may never have to worry again about viewing
hilarious but NSFW images surreptitiously. A pair of glasses
developed by Brother Industries can project images or
documents directly onto a wearer's retinas.

The Retinal Imaging Display technology displays a small image
10 centimeters wide that appears to float about 1 meter (3.3
ft) in front of a user's eye. Images have an 800x600
resolution and refresh at 60Hz.

Each pair of glasses holds an optical scanner, eyepiece and
a light source that also contains a tiny power box. Brother
Industries derived its application from laser printing
technology, as well as piezoelectric technologies based on
inkjet printing.

Similar technologies have tried embedding heads-up displays
within the lens of glasses. Other heads-up displays have gone
for more specific applications, such as giving drivers
a safer experience on the road.

company has yet to announce how much the specs will cost when
it commercializes them next year. But reading those operation
manuals hands-free at your desk never sounded so exciting.
Those are operation manuals, right?

How Plasma Displays Work

For the past 75 years, the vast majority of televisions have
been built around the same technology: the cathode ray tube
(CRT). In a CRT television, a gun fires a beam of electrons
(negatively-charged particles) inside a large glass tube. The
electrons excite phosphor atoms along the wide end of the
tube (the screen), which causes the phosphor atoms to light
up. The television image is produced by lighting up different
areas of the phosphor coating with different colors at
different intensities.

Cathode ray tubes produce crisp, vibrant images, but they do
have a serious drawback: They are bulky. In order to increase
the screen width in a CRT set, you also have to increase the
length of the tube (to give the scanning electron gun room to
reach all parts of the screen). Consequently, any big-screen
CRT television is going to weigh a ton and take up a sizable
chunk of a room.

A new alternative has popped up on store shelves: the plasma
flat panel display. These televisions have wide screens,
comparable to the largest CRT sets, but they are only about 6
inches (15 cm) thick. In this article, we'll see how these
sets do so much in such a small space.

If you've read How Television Works, then you understand the
basic idea of a standard television or monitor. Based on the
information in a video signal, the television lights up
thousands of tiny dots (called pixels) with a high-energy
beam of electrons. In most systems, there are three pixel
colors -- red, green and blue -- which are evenly distributed
on the screen. By combining these colors in different
proportions, the television can produce the entire color
spectrum.

The basic idea of a plasma display is to illuminate tiny,
colored fluorescent lights to form an image. Each pixel is
made up of three fluorescent lights -- a red light, a green
light and a blue light. Just like a CRT television, the
plasma display varies the intensities of the different lights
to produce a full range of colors.

What is plasma?

The central element in a fluorescent light is a plasma, a gas
made up of free-flowing ions (electrically charged atoms) and
electrons (negatively charged particles). Under normal
conditions, a gas is mainly made up of uncharged particles.
That is, the individual gas atoms include equal numbers of
protons (positively charged particles in the atom's nucleus)
and electrons. The negatively charged electrons perfectly
balance the positively charged protons, so the atom has a net
charge of zero.

If you introduce many free electrons into the gas by
establishing an electrical voltage across it, the situation
changes very quickly. The free electrons collide with the
atoms, knocking loose other electrons. With a missing
electron, an atom loses its balance. It has a net positive
charge, making it an ion.

In a plasma with an electrical current running through it,
negatively charged particles are rushing toward the
positively charged area of the plasma, and positively charged
particles are rushing toward the negatively charged area.

­In this mad rush, particles are constantly bumping into each
other. These collisions excite the gas atoms in the plasma,
causing them to release photons of energy.

Xenon and neon atoms, the atoms used in plasma screens,
release light photons when they are excited. Mostly, these
atoms release ultraviolet light photons, which are invisible
to the human eye. But ultraviolet photons can be used to
excite visible light photons.

Inside a Plasma Display

The xenon and neon gas in a plasma television is contained in
hundreds of thousands of tiny cells positioned between two
plates of glass. Long electrodes are also sandwiched between
the glass plates, on both sides of the cells. The address
electrodes sit behind the cells, along the rear glass plate.
The transparent display electrodes, which are surrounded by
an insulating dielectric material and covered by a magnesium
oxide protective layer, are mounted above the cell, along the
front glass plate.

Both sets of electrodes extend across the entire screen. The
display electrodes are arranged in horizontal rows along the
screen and the address electrodes are arranged in vertical
columns. As you can see in the diagram below, the vertical
and horizontal electrodes form a basic grid.

To ionize the gas in a particular cell, the plasma display's
computer charges the electrodes that intersect at that cell.
It does this thousands of times in a small fraction of
a second, charging each cell in turn.

When the intersecting electrodes are charged (with a voltage
difference between them), an electric current flows through
the gas in the cell. As we saw in the last section, the
current creates a rapid flow of charged particles, which
stimulates the gas atoms to release ultraviolet photons.

The released ultraviolet photons interact with phosphor
material coated on the inside wall of the cell. Phosphors are
substances that give off light when they are exposed to other
light. When an ultraviolet photon hits a phosphor atom in the
cell, one of the phosphor's electrons jumps to a higher
energy level and the atom heats up. When the electron falls
back to its normal level, it releases energy in the form of
a visible light photon.

The phosphors in a plasma display give off colored light when
they are excited. Every pixel is made up of three separate
subpixel cells, each with different colored phosphors. One
subpixel has a red light phosphor, one subpixel has a green
light phosphor and one subpixel has a blue light phosphor.
These colors blend together to create the overall color of
the pixel.

By varying the pulses of current flowing through the
different cells, the control system can increase or decrease
the intensity of each subpixel color to create hundreds of
different combinations of red, green and blue. In this way,
the control system can produce colors across the entire
spectrum.

The main advantage of plasma display technology is that you
can produce a very wide screen using extremely thin materials.
And because each pixel is lit individually, the image is very
bright and looks good from almost every angle. The image
quality isn't quite up to the standards of the best cathode
ray tube sets, but it certainly meets most people's
expectations.

The biggest drawback of this technology has been the price.
However, falling prices and advances in technology mean that
the plasma display may soon edge out the old CRT sets.

Building a Digital Thermometer

Now that you understand a little bit about your Stamp and the
LCD, we can add another component and create a digital
thermometer. To create a thermometer, we will use a chip
called the DS1620. This chip contains:

* A temperature-sensing device
* An analog-to-digital (A/D) converter for the
temperature-sensing device
* A shift register to read the data out of the A/D
converter
* A little EEPROM (electrically erasable programmable
read-only memory) to remember settings

The DS1620 has two modes: In one mode, it acts as
a stand-alone thermostat chip, and in the other mode you hook
it up to a computer and use it as a thermometer. The EEPROM
remembers the current mode as well as the set temperatures
for the thermostat.

Hooking up the DS1620 to the Stamp is very easy. The DS1620
comes in an 8-pin chip. Supply +5 volts from the Stamp to pin
8 of the DS1620. Supply ground to pin 4 of the DS1620. You
then use three I/O pins from the Stamp to drive three pins on
the DS1620:

* Pin 1 on the DS1620 is the data pin. You read and write
data bits on this pin.
* Pin 2 on the DS1620 is the clock pin. You clock data in
and out of the shift register with this pin.
* Pin 3 on the DS1620 is the reset/select pin. You set
pin 3 high to select the chip and communicate with it.

For this example code, it is assumed that:

* The data pin goes to I/O pin 2 on the Stamp.
* The clock pin goes to I/O pin 1 on the Stamp.
* The reset/select pin goes to I/O pin 0 on the Stamp.

The completed wiring looks like the picture.

You can get a DS1620 either from Jameco (part number 146456)
or Parallax (part number 27917) in an "application kit" that
includes the chip, the capacitor, some good documentation and
sample code. Or you can buy the chip on its own from Jameco
(part number 114382). I would suggest getting the application
kit the first time you try using the DS1620 because the
documentation is very useful.

You can assemble the DS1620 in the prototype area of the
Stamp carrier board or on a separate breadboard. Once you
have assembled it, hook your LCD display up to I/O pin 3 of
the Stamp, and then load and run the following program:

symbol RST = 0 ' select/reset line on 1620
symbol CLK = 1 ' clock line for shift registers on 1620
symbol DQ = 2 ' data line on 1620
symbol DQ_PIN = pin2 ' pin representation for DQ
symbol LCD = 3 ' data line for LCD

begin:
low RST ' deselect the 1620 unless talking to it
high CLK ' clock pin on 1620 should default high
pause 1000 ' wait for the thermometer and LCD to boot

setup:
high RST ' select the 1620
b0 = $0C ' $0c is the 1620 command byte
' saying "Write Config"
gosub shift_out ' send it to the 1620
b0 = %10 ' %10 is the 1620 command byte
' to set thermometer mode
gosub shift_out ' send it to the 1620
low RST ' deselect the 1620
pause 50 ' delay 50ms for EEPROM

start_convert:
b0 = $EE ' $EE is the 1620 command byte
' to start conversions
high RST ' select the 1620
gosub shift_out ' send it to the 1620
low RST ' deselect the 1620

' This is the main loop
' - reads and displays temperature every second
main_loop:
high RST ' select the 1620
b0 = $AA ' $AA is the 1620 command byte
' for reading temperature
gosub shift_out ' send it to the 1620
gosub shift_in ' read the temperature
' from the 1620
low RST ' deselect the DS1620.
gosub display ' display the temp in degrees C
pause 1000 ' wait a second
goto main_loop

' The shift_out subroutine sends whatever is in
' the b0 byte to the 1620
shift_out:
output DQ ' set the DQ pin to
' output mode
for b2 = 1 to 8
low CLK ' prepare to clock the bit
' into 1620
DQ_PIN = bit0 ' Send the data bit
high CLK ' latch data bit into 1620
b0 = b0/2 ' shift all bits right
' toward bit 0
next
return

' The shift_in subroutine gets a 9-bit
' temperature from the 1620
shift_in:
input DQ ' set the DQ pin to
' input mode
w0 = 0 ' clear w0
for b5 = 1 to 9
w0 = w0/2 ' shift input right.
low CLK ' ask 1620 for next bit
bit8 = DQ_PIN ' read the bit
high CLK ' toggle clock pin
next
return

' Displays the temperature in degrees C
display:
if bit8 = 0 then pos ' if bit8=1
' then temp is negative
b0 = b0 &/ b0 ' invert b0 by NANDing it
' with itself
b0 = b0 + 1
pos:
serout LCD, n2400, (254, 1) ' clear the LCD
serout LCD, n2400, ("Temp = ") ' display "Temp="
' on the display
bit9 = bit0 ' save the half degree
b0 = b0 / 2 ' convert to degrees
if bit8 = 1 then neg ' see if temp is negative
serout LCD, n2400, (#b0) ' display positive temp
goto half
neg:
serout LCD, n2400, ("-", #b0)' display negative temp
half:
if bit9 = 0 then even
serout LCD, n2400, (".5 C") ' display the half degree
goto done
even:
serout LCD, n2400, (".0 C") ' display the half degree
done:
return

If you run this program, you will find that it displays the
centigrade temperature with an accuracy of one-half degree.

The DS1620 measures temperatures in centigrade half-degrees.
It returns the temperature in a 9-bit 2s-complement number
with a range of -110 to 250 F (-55 to 125 C). You divide the
number you receive by 2 to get the actual temperature.
2s-complement binary numbers are a convenient way to
represent negative values. The following list shows the
values for a 4-bit 2s-complement number:

0111 : 7
0110 : 6
0101 : 5
0100 : 4
0011 : 3
0010 : 2
0001 : 1
0000 : 0
1111 : -1
1110 : -2
1101 : -3
1100 : -4
1011 : -5
1010 : -6
1001 : -7
1000 : -8

­ You can see that instead of the 4 bits representing values
from 0 to 15, the 4 bits in a 2s-complement number represent
the values -8 to 7. You can look at the left-most bit to
determine if the number is negative or positive. If the
number is negative, you can invert the bits and add 1 to get
the positive representation of the number.

Here's what goes on with the digital thermometer program
shown here:

1. It uses the symbol keyword to set up several constants
that make the program slightly easier to read (and also make
it easy for you to move the chip to different I/O pins on the
Stamp).

2. It sets the CLK and RST pins on the DS1620 to their
expected values.

3. It writes a command byte to the EEPROM on the DS1620 to
tell the chip to operate in "thermometer mode." Because the
mode is stored in EEPROM, you only have to do it once, so you
could technically take this section of the code out of the
program after you run the program once (to save program
space).

4. The program sends the command $EE ("$" means
"hexadecimal number" -- $EE is 238 in decimal) to tell the
thermometer to start up its conversion process.

The program then enters a loop. Every second, it sends
a command to the DS1620 telling the DS1620 to return the
current temperature, and then it reads the 9-bit value that
the DS1620 returns into the w0 variable. The Stamp sends and
receives data 1 bit at a time by toggling the CLK line on the
DS1620. Remember that the w0 (16-bit) variable overlays the
b0/b1 (8-bit) variables, which overlay the
bit0/bit1/.../bit15 (1-bit) variables, so when you insert
a bit from the DS1620 into bit 8 and divide w0 by 2, what you
are doing is shifting each bit to the right to store the
9-bit temperature from the DS1620 into w0. Once the
temperature has been saved in w0, the display subroutine
determines whether the number is positive or negative and
displays it appropriately on the LCD as a centigrade
temperature. The conversion from degrees C to degrees F is:

dF = dC * 9/5 + 32

At this point, we have succeeded in creating an extremely
expensive thermometer. What might you do with it? Here's one
idea. Let's say you work for a drug company and you are
shipping expensive drugs across the country that MUST remain
at a certain temperature the entire way or the drugs will
spoil. What you can do with a Stamp is create a data logging
thermometer. Both Jameco (part number 143811) and Parallax
(part number 27960) sell a device called the "RAM Pack
module." It contains a low-power 8-kilobyte (or optionally
32-kilobyte) RAM chip with a serial interface. You could
add this component (or something similar) to your Stamp and
write code that saves temperature readings to the RAM every
minute. You could then slip your Stamp into the drug
shipment, and at the other end of the trip retrieve the
Stamp. The RAM module would contain the temperature history
of the entire trip and you would know whether or not the
drugs ever thawed out.

There are all kinds of neat, useful devices like this that
you can build with a Stamp now that you know how
microcontrollers work!

Creating a Really Expensive Digital Clock using microcontroller

First, Playing with a BASIC Stamp

If you would like to play with a BASIC Stamp, it's very easy
to get started. What you need is a desktop computer and
a BASIC Stamp starter kit. The starter kit includes the
Stamp, a programming cable and an application that you run on
your desktop computer to download BASIC programs into the
Stamp.

You can get a starter kit either from Parallax (the
manufacturer) or from a supplier like Jameco. From Parallax,
you can order the BASIC Stamp D Starter Kit (part number
27202), or from Jameco you can order part number 140089. You
will receive the Stamp, a programming cable, software and
instructions. The kit is $79 from both suppliers.
Occasionally, Parallax runs a special called "We've Bagged
the Basics" that also includes Scott Edward's Programming
and Customizing the BASIC Stamp Computer.

Hooking up the Stamp is easy. You connect it into the
parallel port of your PC. Then you run a DOS application to
edit your BASIC program and download it to the Stamp. Here is
a screenshot of a typical editor (in this case, the one from
Scott Edward's book).

To run the program in this editor, you hit ALT-R. The editor
application checks the BASIC program and then sends it down
the wire to the EEPROM on the Stamp. The Stamp then executes
the program. In this case, the program produces a square wave
on I/O pin 3. If you hook up a logic probe or LED to pin 3,
you will see the LED flash on and off twice per second (it
changes state every 250 milliseconds because of the PAUSE
commands). This program would run for several weeks off of
a 9-volt battery. You could save power by shortening the time
that the LED is on (perhaps it is on for 50 milliseconds and
off for 450 milliseconds), and also by using the NAP
instruction instead of PAUSE.

Now, Creating a Really Expensive Digital Clock

Spending $79 to flash an LED may seem extravagant to you.
What you would probably like to do is create something useful
with your BASIC stamp. By spending about $100 more you can
create a really nice digital clock! This may seem extremely
extravagant, until you realize that the parts are reusable in
a variety of other projects that you may want to build later.

Let's say that we would like to use the I/O pins on the BASIC
Stamp to display numeric values. 7447s would work just as well
with the BASIC Stamp. You could wire four of the I/O pins
straight into a 7447 and easily display a number between 0
and 9. Since the BS-1 Stamp has eight I/O pins, it is easy to
drive two 7447s directly like this.

For a clock, we need a minimum of four digits. To drive four
7447s with eight I/O pins, we have to be slightly more
creative. The diagram shows you one approach.

In this diagram, the eight I/O lines from the Stamp enter
from the left. This approach uses four lines that run to all
four 7447s. Then the other four lines from the Stamp activate
the 7447s in sequence ("E" on the chips means "Enable" -- on
a 7447, that would be the blanking input on pin 5). To make
this arrangement work, the BASIC program in the Stamp would
output the first digit on the four data lines and activate
the first 7447 by toggling its E pin with the first control
line. Then it would send out the value for the second digit
and activate the second 7447, sequencing through all four of
the 7447s like this repeatedly. By wiring things slightly
differently, you could actually do this with only one 7447.
By using a 74154 demultiplexer chip and some drivers, you
could drive up to 16 digits using this approach.

This is, in fact, a standard way to control LED displays. For
example, if you have an old LED calculator, turn it on and
shake it while watching the display. You will actually be
able to see that only one digit is ever illuminated at once.
The approach is called multiplexing the display.

While this approach works fine for clocks and calculators, it
has two important problems:

* LEDs consume a lot of power.
* 7-segment LEDs can only display numeric values.

An alternative approach is to use an LCD screen. As it turns
out, LCDs are widely available and can be easily hooked to
a Stamp. For example, the two-line by 16-character
alphanumeric display shown below is available from both
Jameco (part number 150990) and Parallax (part number 27910).
A typical display is shown here, mounted on a breadboard for
easier interfacing.

This sort of LCD has several advantages:

* The display can be driven by a single I/O pin. The
display contains logic that lets a Stamp communicate with it
serially, so only one I/O pin is needed. In addition, the
SEROUT command in Stamp BASIC handles serial communication
easily, so talking to the display is simple.
* The LCD can display alphanumeric text: letters, numbers
and even custom characters.
* The LCD consumes very little power -- only 3 milliamps.

The only problem is that one of these displays costs $59.
Obviously, you would not embed one of these in a toaster oven.
If you were designing a toaster oven, however, you would
likely prototype with one of these displays and then create
custom chips and software to drive much cheaper LCDs in the
final product.

To drive a display like this, you simply supply it with +5
volts and ground (the Stamp supplies both from the 9-volt
battery) and then hook one of the I/O pins from the Stamp to
the display's input line. The easiest way I have found to
connect the Stamp's I/O pins to a device like an LCD is to
use a wire-wrap tool (Jameco part number 34577) and 30-gauge
wire wrap wire (Jameco part number 22541 is typical). That
way, no soldering is involved and the connections are
compact and reliable.

The following BASIC program will cause a BASIC Stamp to
behave like a clock and output the time on the LCD (assuming
the LCD is connected to I/O pin 0 on the Stamp):

pause 1000 'wait for LCD display to boot
serout 0, n2400, (254, 1) 'clear the display
serout 0, n2400, ("time:") 'Paint "time:" on the display
'preset before loading program
b0 = 0 'seconds
b1 = 27 'minutes
b2 = 6 'hours

again:
b0 = b0 + 1 'increment seconds
if b0 < 60 then minutes
b0 = 0 'if seconds=60
b1 = b1 + 1 ' then increment minutes
minutes:
if b1 < 60 then hours
b1 = 0 'if minutes=60
b2 = b2 + 1 ' then increment hours
hours:
if b2 < 13 then show
b2 = 1 'if hours=13 reset to 1

show:
serout 0, n2400, (254, 135) 'position cursor on display,
'then display time
serout 0, n2400, (#b2, ":", #b1, ":", #b0, " ")
pause 950 'pause 950 milliseconds
goto again 'repeat

In this program, the SEROUT commands send data to the LCD.
The sequence (254, 1) clears the LCD (254 is the escape
character and 1 is the command to clear the screen). The
sequence (254, 135) positions the cursor. The other two
SEROUT commands simply send text strings to the display.

This approach will create a reasonably accurate clock. By
tweaking the PAUSE statement you can get the accuracy to
within a few seconds a day. Obviously, in a real clock you
would like to wire up a push-button or two to make setting it
easier -- in this program, you preset the time before you
download the program to the Stamp.

While this approach is simple and works, it is not incredibly
accurate. If you want better accuracy, one good approach
would be to wire a real-time clock chip up to your Stamp.
Then, every second or so, you can read the time from the chip
and display it. A real-time clock chip uses a quartz crystal
to give it excellent accuracy. Clock chips also usually
contain date information and handle leap year correction
automatically.

One easy way to interface a real-time clock to a stamp is to
use a component called the Pocket Watch B.

The Pocket Watch B is available from both Jameco (part number
145630) and Parallax (part number 27962). This part is about
as big as a quarter and contains the clock chip, crystal and
a serial interface so that only one I/O pin is necessary to
communicate with it. This component costs about $30 -- again,
not something you want to embed in a toaster oven, but easy
to play with when constructing prototypes.

Microcontrollers

Microcontrollers are hidden inside a surprising number of
products these days. If your microwave oven has an LED or
LCD screen and a keypad, it contains a microcontroller. All
modern automobiles contain at least one microcontroller, and
can have as many as six or seven: The engine is controlled by
a microcontroller, as are the anti-lock brakes, the cruise
control and so on. Any device that has a remote control
almost certainly contains a microcontroller: TVs, VCRs and
high-end stereo systems all fall into this category. Nice SLR
and digital cameras, cell phones, camcorders, answering
machines, laser printers, telephones (the ones with caller
ID, 20-number memory, etc.), pagers, and feature-laden
refrigerators, dishwashers, washers and dryers (the ones with
displays and keypads)... You get the idea. Basically, any
product or device that interacts with its user has
a microcontroller buried inside.

We will look at microcontrollers so that you can understand
what they are and how they work. Then we will go one step
further and discuss how you can start working with
microcontrollers yourself -- we will create a digital clock
with a microcontroller! We will also build a digital
thermometer. In the process, you will learn an awful lot
about how microcontrollers are used in commercial products.

What is a Microcontroller?

A microcontroller is a computer. All computers -- whether we
are talking about a personal desktop computer or a large
mainframe computer or a microcontroller -- have several
things in common:

* All computers have a CPU (central processing unit) that
executes programs. If you are sitting at a desktop computer
right now reading this article, the CPU in that machine is
executing a program that implements the Web browser that is
displaying this page.
* The CPU loads the program from somewhere. On your
desktop machine, the browser program is loaded from the hard
disk.
* The computer has some RAM (random-access memory) where
it can store "variables."
* And the computer has some input and output devices so
it can talk to people. On your desktop machine, the keyboard
and mouse are input devices and the monitor and printer are
output devices. A hard disk is an I/O device -- it handles
both input and output.

The desktop computer you are using is a "general purpose
computer" that can run any of thousands of programs.
Microcontrollers are "special purpose computers."
Microcontrollers do one thing well. There are a number of
other common characteristics that define microcontrollers. If
a computer matches a majority of these characteristics, then
you can call it a "microcontroller":

* Microcontrollers are "embedded" inside some other
device (often a consumer product) so that they can control
the features or actions of the product. Another name for
a microcontroller, therefore, is "embedded controller."

* Microcontrollers are dedicated to one task and run one
specific program. The program is stored in ROM (read-only
memory) and generally does not change.

* Microcontrollers are often low-power devices. A desktop
computer is almost always plugged into a wall socket and
might consume 50 watts of electricity. A battery-operated
microcontroller might consume 50 milliwatts.

* A microcontroller has a dedicated input device and
often (but not always) has a small LED or LCD display for
output. A microcontroller also takes input from the device it
is controlling and controls the device by sending signals to
different components in the device.

For example, the microcontroller inside a TV takes
input from the remote control and displays output on the TV
screen. The controller controls the channel selector, the
speaker system and certain adjustments on the picture tube
electronics such as tint and brightness. The engine
controller in a car takes input from sensors such as the
oxygen and knock sensors and controls things like fuel mix
and spark plug timing. A microwave oven controller takes
input from a keypad, displays output on an LCD display and
controls a relay that turns the microwave generator on and
off.

* A microcontroller is often small and low cost. The
components are chosen to minimize size and to be as
inexpensive as possible.

* A microcontroller is often, but not always, ruggedized
in some way. The microcontroller controlling a car's engine,
for example, has to work in temperature extremes that
a normal computer generally cannot handle. A car's
microcontroller in Alaska has to work fine in -30 degree F
(-34 C) weather, while the same microcontroller in Nevada
might be operating at 120 degrees F (49 C). When you add the
heat naturally generated by the engine, the temperature can
go as high as 150 or 180 degrees F (65-80 C) in the engine
compartment.

On the other hand, a microcontroller embedded inside
a VCR hasn't been ruggedized at all.

The actual processor used to implement a microcontroller can
vary widely. For example, the cell phone shown on Inside
a Digital Cell Phone contains a Z-80 processor. The Z-80 is
an 8-bit microprocessor developed in the 1970s and originally
used in home computers of the time. The Garmin GPS shown in
How GPS Receivers Work contains a low-power version of the
Intel 80386, I am told. The 80386 was originally used in
desktop computers.

In many products, such as microwave ovens, the demand on the
CPU is fairly low and price is an important consideration.
In these cases, manufacturers turn to dedicated
microcontroller chips -- chips that were originally designed
to be low-cost, small, low-power, embedded CPUs. The Motorola
6811 and Intel 8051 are both good examples of such chips.
There is also a line of popular controllers called "PIC
microcontrollers" created by a company called Microchip. By
today's standards, these CPUs are incredibly minimalistic;
but they are extremely inexpensive when purchased in large
quantities and can often meet the needs of a device's
designer with just one chip.

A typical low-end microcontroller chip might have 1,000 bytes
of ROM and 20 bytes of RAM on the chip, along with eight I/0
pins. In large quantities, the cost of these chips can
sometimes be just pennies. You certainly are never going to
run Microsoft Word on such a chip -- Microsoft Word requires
perhaps 30 megabytes of RAM and a processor that can run
millions of instructions per second. But then, you don't need
Microsoft Word to control a microwave oven, either. With
a microcontroller, you have one specific task you are trying
to accomplish, and low-cost, low-power performance is what is
important.

Using Microcontrollers

In How Electronic Gates Work, you learned about 7400-series
TTL devices, as well as where to buy them and how to assemble
them. What you found is that it can often take many gates to
implement simple devices. For example, in the digital clock
post, the clock we designed might contain 15 or 20 chips. One
of the big advantages of a microcontroller is that software
-- a small program you write and execute on the controller --
can take the place of many gates. In this article, therefore,
we will use a microcontroller to create a digital clock. This
is going to be a rather expensive digital clock (almost
$200!), but in the process you will accumulate everything you
need to play with microcontrollers for years to come. Even if
you don't actually create this digital clock, you will learn
a great deal by reading about it.

The microcontroller we will use here is a special-purpose
device designed to make life as simple as possible. The
device is called a "BASIC Stamp" and is created by a company
called Parallax. A BASIC Stamp is a PIC microcontroller that
has been customized to understand the BASIC programming
language. The use of the BASIC language makes it extremely
easy to create software for the controller. The
microcontroller chip can be purchased on a small carrier
board that accepts a 9-volt battery, and you can program it
by plugging it into one of the ports on your desktop computer.
It is unlikely that any manufacturer would use a BASIC Stamp
in an actual production device -- Stamps are expensive and
slow (relatively speaking). However, it is quite common to
use Stamps for prototyping or for one-off demo products
because they are so incredibly easy to set up and use.

They are called "Stamps," by the way, because they are about
as big as a postage stamp.

Parallax makes two versions of the BASIC Stamp: the BS-1 and
the BS-2. Here are some of the differences between the two
models:

Spec BS-1 BS-2

RAM 14 bytes 26 bytes
EEPROM 256 bytes 2 kilobytes
Max program length 75 instructions 600 instructions
Execution speed 2,000 lines/sec 4,000 lines/sec
I/O pins 8 16


The specific BASIC Stamp we will be using in the post is
called the "BASIC Stamp Revision D"

The BASIC Stamp Revision D is a BS-1 mounted on carrier board
with a 9-volt battery holder, a power regulator, a connection
for a programming cable, header pins for the I/O lines and
a small prototyping area. You could buy a BS-1 chip and wire
the other components in on a breadboard. The Revision D
simply makes life easier.

You can see from the previous table that you aren't going to
be doing anything exotic with a BASIC stamp. The 75-line
limit (the 256 bytes of EEPROM can hold a BASIC program about
75 lines long) for the BS-1 is fairly constraining. However,
you can create some pretty neat stuff, and the fact that the
Stamp is so small and battery operated means that it can go
almost anywhere.

Programming the BASIC Stamp

You program a BASIC Stamp using the BASIC programming
language. If you already know BASIC, then you will find that
the BASIC used in a Stamp is straightforward but a little
stripped-down. If you don't know BASIC, but you do know
another language like C, Pascal or Java, then picking up
BASIC will be trivial. If you have never programmed before,
you probably want to go learn programming on a desktop
machine first. Here is a quick rundown on the instructions
available in Stamp BASIC.

Standard BASIC instructions:

* for...next - normal looping statement
* gosub - go to a subroutine
* goto - goto a label in the program (e.g. - "label:")
* if...then - normal if/then decision
* let - assignment (optional)
* return - return from a subroutine
* end - end the program and sleep

Instructions having to do with I/O pins:

* button - read a button on an input pin, with debounce
and auto-repeat
* high - set an I/O pin high
* input - set the direction of an I/O pin to input
* low - set an I/O pin low
* output - set the direction of an I/O pin to output
* pot - read a potentiometer on an I/O pin
* pulsin - read the duration of a pulse coming in on
an input pin
* pulsout - send a pulse of a specific duration out on
an output pin
* pwm - perform pulse width modulation on an output pin
* reverse - reverse the direction of an I/O pin
* serin - read serial data on an input pin
* serout - write serial data on an output pin
* sound - send a sound of a specific frequency to
an output pin
* toggle - toggle the bit on an output pin

Instructions specific to the BASIC Stamp:

* branch - read a branching table
* debug - send a debugging string to the console on the
desktop computer
* eeprom - download a program to EEPROM
* lookdown - return the index of a value in a list
* lookup - array lookup using an index
* nap - sleep for a short time
* pause - delay for the specified time
* random - pick a random number
* read - read a value from EEPROM
* sleep - power down for the specified time
* write - write data to EEPROM

Operations:

* + - addition
* - - subtraction
* * - multiplication (low-word)
* ** - multiplication (high-word)
* / - division
* // - mod
* max - return maximum of 2 values
* min - return minimum of 2 values
* & - AND
* | - OR
* ^ - XOR
* &/ - NAND
* |/ - NOR
* ^/ - XNOR

If statement logic:

* =
* <>
* <
* <=
* >
* >=
* AND
* OR

Variables

All variables in the BS-1 have pre-defined names (which you
can substitute with names of your own). Remember that there
are only 14 bytes of RAM available, so variables are
precious. Here are the standard names:

* w0, w1, w2...w6 - 16-bit word variables
* b0, b1, b2...b13 - 8-bit byte variables
* bit0, bit1, bit2...bit15 - 1-bit bit variables

Because there are only 14 bytes of memory, w0 and b0/b1 are
the same locations in RAM, and w1 and b2/b3 are the same, and
so on. Also, bit0 through bit15 reside in w0 (and therefore
b0/b1 as well).

I/O pins
You can see that 14 of the instructions in the BS-1 have to
do with the I/O pins. The reason for this emphasis is the
fact that the I/O pins are the only way for the BASIC Stamp
to talk to the world. There are eight pins on the BS-1
(numbered 0 to 7) and 16 pins on the BS-2 (numbered 0 to 15).

The pins are bi-directional, meaning that you can read input
values on them or send output values to them. The easiest way
to send a value to a pin is to use the HIGH or LOW functions.
The statement high 3 sends a 1 (+5 volts) out on pin 3. LOW
sends a 0 (Ground). Pin 3 was chosen arbitrarily here -- you
can send bits out on any pin from 0 to 7.

There are a number of interesting I/O pin instructions. For
example, POT reads the setting on a potentiometer (variable
resistor) if you wire it up with a capacitor as the POT
instruction expects. The PWM instruction sends out
pulse-width modulated signals. Instructions like these can
make it a lot easier to attach controls and motors to the
Stamp. See the documentation for the language for details.
Also, a book like Scott Edward's Programming and Customizing
the BASIC Stamp Computer can be extremely helpful because of
the example projects it contains.

Diodes and Transistors

A device that blocks current in one direction while letting
current flow in another direction is called a diode. Diodes
can be used in a number of ways. For example, a device that
uses batteries often contains a diode that protects the
device if you insert the batteries backward. The diode simply
blocks any current from leaving the battery if it is
reversed -- this protects the sensitive electronics in the
device.

A semiconductor diode's behavior is not perfect, as shown in
this graph.

When reverse-biased, an ideal diode would block all current.
A real diode lets perhaps 10 microamps through -- not a lot,
but still not perfect. And if you apply enough reverse
voltage (V), the junction breaks down and lets current
through. Usually, the breakdown voltage is a lot more voltage
than the circuit will ever see, so it is irrelevant.

When forward-biased, there is a small amount of voltage
necessary to get the diode going. In silicon, this voltage is
about 0.7 volts. This voltage is needed to start the
hole-electron combination process at the junction.

Another monumental technology that's related to the diode is
the transistor. Transistors and diodes have a lot in common.

Transistors

A transistor is created by using three layers rather than the
two layers used in a diode. You can create either an NPN or
a PNP sandwich. A transistor can act as a switch or
an amplifier.

A transistor looks like two diodes back-to-back. You'd
imagine that no current could flow through a transistor
because back-to-back diodes would block current both ways.
And this is true. However, when you apply a small current to
the center layer of the sandwich, a much larger current can
flow through the sandwich as a whole. This gives a transistor
its switching behavior. A small current can turn a larger
current on and off.

A silicon chip is a piece of silicon that can hold thousands
of transistors. With transistors acting as switches, you can
create Boolean gates, and with Boolean gates you can create
microprocessor chips.

The natural progression from silicon to doped silicon to
transistors to chips is what has made microprocessors and
other electronic devices so inexpensive and ubiquitous in
today's society. The fundamental principles are surprisingly
simple. The miracle is the constant refinement of those
principles to the point where, today, tens of millions of
transistors can be inexpensively formed onto a single chip.

Semiconductors

Semiconductors have had a monumental impact on our society.
You find semiconductors at the heart of microprocessor chips
as well as transistors. Anything that's computerized or uses
radio waves depends on semiconductors.

Today, most semiconductor chips and transistors are created
with silicon. You may have heard expressions like "Silicon
Valley" and the "silicon economy," and that's why -- silicon
is the heart of any electronic device.

A diode is the simplest possible semiconductor device, and is
therefore an excellent beginning point if you want to
understand how semiconductors work. In this article, you'll
learn what a semiconductor is, how doping works and how
a diode can be created using semiconductors. But first, let's
take a close look at silicon.

Silicon is a very common element -- for example, it is the
main element in sand and quartz. If you look "silicon" up in
the periodic table, you will find that it sits next to
aluminum, below carbon and above germanium.

Carbon, silicon and germanium (germanium, like silicon, is
also a semiconductor) have a unique property in their
electron structure -- each has four electrons in its outer
orbital. This allows them to form nice crystals. The four
electrons form perfect covalent bonds with four neighboring
atoms, creating a lattice. In carbon, we know the crystalline
form as diamond. In silicon, the crystalline form is
a silvery, metallic-looking substance.

Metals tend to be good conductors of electricity because they
usually have "free electrons" that can move easily between
atoms, and electricity involves the flow of electrons. While
silicon crystals look metallic, they are not, in fact, metals.
All of the outer electrons in a silicon crystal are involved
in perfect covalent bonds, so they can't move around. A pure
silicon crystal is nearly an insulator -- very little
electricity will flow through it.

But you can change all this through a process called doping.

Doping Silicon

You can change the behavior of silicon and turn it into
a conductor by doping it. In doping, you mix a small amount
of an impurity into the silicon crystal.

There are two types of impurities:

* N-type - In N-type doping, phosphorus or arsenic is
added to the silicon in small quantities. Phosphorus and
arsenic each have five outer electrons, so they're out of
place when they get into the silicon lattice. The fifth
electron has nothing to bond to, so it's free to move around.
It takes only a very small quantity of the impurity to create
enough free electrons to allow an electric current to flow
through the silicon. N-type silicon is a good conductor.
Electrons have a negative charge, hence the name N-type.

* P-type - In P-type doping, boron or gallium is the
dopant. Boron and gallium each have only three outer
electrons. When mixed into the silicon lattice, they form
"holes" in the lattice where a silicon electron has nothing
to bond to. The absence of an electron creates the effect of
a positive charge, hence the name P-type. Holes can conduct
current. A hole happily accepts an electron from a neighbor,
moving the hole over a space. P-type silicon is a good
conductor.

A minute amount of either N-type or P-type doping turns
a silicon crystal from a good insulator into a viable (but
not great) conductor -- hence the name "semiconductor."

N-type and P-type silicon are not that amazing by themselves;
but when you put them together, you get some very interesting
behavior at the junction. That's what happens in a diode.

A diode is the simplest possible semiconductor device.
A diode allows current to flow in one direction but not the
other. You may have seen turnstiles at a stadium or a subway
station that let people go through in only one direction.
A diode is a one-way turnstile for electrons.

When you put N-type and P-type silicon together as shown in
this diagram, you get a very interesting phenomenon that
gives a diode its unique properties.

Even though N-type silicon by itself is a conductor, and
P-type silicon by itself is also a conductor, the combination
shown in the diagram does not conduct any electricity. The
negative electrons in the N-type silicon get attracted to the
positive terminal of the battery. The positive holes in the
P-type silicon get attracted to the negative terminal of the
battery. No current flows across the junction because the
holes and the electrons are each moving in the wrong
direction.

If you flip the battery around, the diode conducts
electricity just fine. The free electrons in the N-type
silicon are repelled by the negative terminal of the battery.
The holes in the P-type silicon are repelled by the positive
terminal. At the junction between the N-type and P-type
silicon, holes and free electrons meet. The electrons fill
the holes. Those holes and free electrons cease to exist, and
new holes and electrons spring up to take their place. The
effect is that current flows through the junction.

UMPC Battery Life

Ultra-Mobile PC (part 2)



One of the biggest issues for any portable device is how it
uses power and whether its batteries can run an entire
eight-hour day without having to be recharged. While the UMPC
is still too new to determine whether this will be the case,
there are several indications from Microsoft staff and from
analysts that it will be a very power-hungry device. Part of
the problem is that 7-inch screen, which consumes a great
deal of power. Microsoft is aiming for at least a two-hour
battery profile, but early testers say that even that amount
of time will be hard to deliver. This means that batteries
won't be able to last as long as watching your average movie
on the UMPC.

Michael Gartenberg of Jupiter Research wrote in his blog,
"What is missing? Battery life. Right now these machines
really need to be sold with a higher capacity battery if
you're going to use one as a life style device going through
the day."

There are numerous notebooks available today that offer three
or more hours of battery life, including ones from Sony,
Fujitsu and Motion Computing. Of course, these are all larger
form factors and thus able to have bigger batteries included
to last longer.

To counteract this issue, Microsoft has tried to finesse
things by offering more advanced power management on the UMPC.
Like many of today's more advanced multimedia laptops, the
unit will have a special quick boot sequence that will bring
up an embedded OS to run movies or play music files.

Dustin Hubbard at Microsoft writes in his blog, "The Samsung
unit actually has 2 boot modes on a 3 way power switch - On,
Off and AVS Multimedia (that's their name for this mode).
AVS Multimedia can do near instant on by booting into what
appears to be XP Embedded (as far as I can tell that is what
they are doing) and allows you to play movies, music and
photos without requiring you to boot into full XP. That gives
you 2 advantages; presumably better battery life by not
having so many services running in the background and fast
cold boot startup for media only consumption."

Touch Pack Enhancements

Microsoft has developed a special series of operating system
enhancements to the tablet OS to take advantage of the UMPC
platform. Called the "Touch Pack," these enhancements will
optimize the touch screen experience. It has five
applications: a program launcher, a series of touch
improvements to the normal Windows UI, a Brilliant Black skin
for Media player, a Sudoku game, and Dial Keys,
a thumb-based on-screen keyboard for inputting text.

"Currently the Touch Pack is available only to UMPC qualified
computers and only as an OEM pre-install," says Hubbard.
"The Touch Pack was designed specifically for small form
factor PCs."

Microsoft's Mitchell says, "Microsoft Touch Pack for Windows
XP software optimizes the touch screen user interface for
UMPCs to simplify navigation and ease-of-use while on the go.
The Touch Pack's customizable Program Launcher organizes
software programs into categories, and uses large buttons and
icons to make it easy to find and open your favorite
applications."

Most of the marketing materials that show screen shots of the
UMPC feature views of this application.

There are several components to the Touch Pack. The first one
is "Touch Improvements." This utility makes about 10
different settings changes to Windows such as widening the
scroll bars and enlarging the minimize and maximize buttons,
shows folders in thumbnail view. Next is a new skin for
Windows Media Player called "Brilliant Black." This skin
fills the screen on the devices with large buttons to
navigate the media controls such as play, stop and volume.

The third Touch Pack program is "DialKeys." This program,
built by Fortune Fountain Ltd., is a way to input text with
your thumbs. Dial Keys makes it easy to enter URLs, e-mail
addresses, et cetera.

"DialKeys basically takes a standard QWERTY keyboard layout
and splits it in two halves. It's a little hard to describe
the layout but there are lots of screen shots of DialKeys to
show what it looks like. The basic idea is that you hold the
device in two hands and use your thumbs on the screen to type
in text. It takes a little getting used to, but people are
always amazed once they use it a day or two how good they get
at typing with it," says Hubbard on his blog.

The final product in the Touch Pack is a Sudoku game. The
version is optimized for touch and the pen.

Manufacturing and Cost

UMPCs from Samsung, Asus, and Founder are based on Intel
microprocessors. PaceBlade Japan announced a UMPC, known as
the SmartCaddie, that uses a VIA Technology chip.

Samsung's Q1 went on sale in the second quarter of 2006 with
a manufacturer's suggested retail price of $1,099. Another
UMPC, manufactured by AMtek and available the United States
as the TabletKiosk eo v7110, retails for $899. Asus' and
Founder's UMPC should be available later this year. One
vendor that hasn't announced any UMPC yet is Motion Computing,
although the company sells a variety of tablet PCs in various
shapes and sizes.

"We see it as validation and reinforcement of where we were
headed with tablet PCs," says Bert Haskell of Motion
Computing. "We were drawn into the ultra-mobile tablets by
our customers who wanted a higher level of mobility. It is
nice to see that Microsoft is creating the same type of
vision for a consumer-oriented product."

Motion sells its tablets for the enterprise, and through the
reseller channel, largely aimed at equipping sales forces and
hospital staffs. "We have a significantly different focus in
that we are looking at high-powered enterprise computing,"
says Haskell. Motion sells their tablets with higher-end
components, such as with Pentium M CPUs and 60 GB hard disks
with 1GB of RAM. " We are set up to understand and service
the needs of the enterprise customers. We don't currently
target consumer markets. If and when we decide to make that
transition, it would be a pretty significant change for us,"
he said.

Microsoft has promised that once Vista ships that the UMPC
will be able to run some version of Vista as well, but since
the new OS isn't yet finished it is hard to make any
definitive claims in this area. "The Vista requirements
aren't out yet, but realistically if the units have the
hardware specifications that Vista requires they should be
able to run it," says Microsoft's Hubbard.

Just because the UMPC runs the tablet version of XP doesn't
mean that every tablet-based application is worth running on
it, however. "To be clear though not all tablet-enabled apps
will make sense on a UMPC and not all UMPC apps will make
sense on a traditional tablet," says Hubbard. "The small form
factor and touch interaction of UMPC means that to have
a really great user experience you need to design your UI to
really take advantage of those assets." So chances are if
UMPCs do take off, we will see differentiated applications
over time that take advantage of its touch-screen features
and enhancements that Microsoft is building into the units.

UMPCs overall are expected will cost less than most tablets
and more than most PDAs. It depends on what configurations
the various vendors will initially offer and how they will be
priced in local markets. Price is definitely an issue that
has gotten some analysts going about the new units.

"To me, one of the most important things about this, is
price," stated Bob O'Donnell, program vice president for
clients and displays at IDC. "If it's under $500, it's
a fancy gadget. People spend that much on iPods. So there's
this sense that, at a low price point, it could be very
interesting and very appealing. Think about all the free WiFi
networks that are out there now that you could use this thing
with, and it gives you a full browsing experience as opposed
to trying to browse on a two-inch cell phone, which is
a horrendously useless experience. And yet people are getting
more and more used to having information access almost
anywhere they are, at almost any time."

Scott Fulton of TGDaily.com, sums things up the best about
UMPC's shortcomings: "But what we have instead is something
that's larger than the average pocket, that can't dial out,
that doesn't have a discrete way to connect to the outside
world, is somewhat expensive, and perhaps most unanticipated
of all, is power-hungry."

So what will happen with the UMPC? Only time will tell. If
manufacturers can get several hours of battery life, if the
costs can drop way below $1000, if the touch experience works
out for many new users, and if the right distributors pick up
the products, this could be a winning product for Microsoft
and its OEMs. Hitting all of these targets won't be easy, to
be sure. But there is plenty of interest in UMPC and the size
of the devices is very appealing.

Ultra-Mobile PC (part 1)

It isn't often that Microsoft announces a completely new
concept in computing, but at CeBIT 2006 in Hanover, Germany,
the software giant finally unveiled details about its
Origami Project to a worldwide audience.

Microsoft created the Origami Project to develop (with
various hardware and software partners) a new type of
computer that falls in-between the size of a typical PDA and
tablet PC, uses a standard Windows operating system and costs
less than $1000: the Ultra-mobile PC, or UMPC.

The Microsoft team originally used "Origami" as a code name,
and they liked it so much that they kept it for the project's
public community Web site. One of the program's managers was
interested in Japanese culture and liked the name, which
refers to the art of intricate paper folded sculptures.

Clearly, there is a need for a smaller form factor than the
traditional notebook PC. PDAs don't run standard Windows OS,
and many tablets are too heavy to carry around all the time.
Gartner, an IT research Web site, posits that the UMPC space
should have the following advantages:

* An eight-hour battery life.
* A sub-$400 price.
* Low-cost, compelling content bundles from partners like
Intel and Microsoft.
* A better Microsoft shell/interface running on top of
Vista.
* Text entry options beyond thumb-typing.
* Dock-and-go synchronization, requiring minimal user
interaction.
* Sustained market momentum from Microsoft and Intel.

These are all worthy goals, but the initial UMPC units won't
deliver on all of these points at first. In this post, we'll
show you what Microsoft has planned for the UMPC and what is
still to come. We'll also talk about how these new computers
will work and how they differ from tablets and PDAs.

Ultra-mobile PC Software

The UMPC is unique in that it runs regular Windows OS -- it
currently runs XP Tablet edition and will run Vista in the
future. While there are many PDAs and PDA/phone combinations
that run Windows Mobile (such as the Palm Treo 700w) it isn't
the same as running the stock XP OS loaded onto millions of
desktop and laptop computers. Windows Mobile requires new
versions of applications to be compatible with that OS, and
the applications have to take advantage of the smaller
screens on these devices. The goal of UMPC is to
run off-the-shelf Windows applications, with no medications
whatsoever other than supporting the tablet/touch screen
features.

The UMPCs that Microsoft announced are small devices -- they
weigh less than two pounds and have seven-inch video screens.
This puts it in a new market segment not currently served by
any particular manufacturer. It is smaller than the smallest
tablet PCs that are currently available from Motion and
Fujitsu, and lighter than the Aopen MiniPC (a desktop model
that doesn't include any screen at all). It is also bigger
than the OQO PC. The other distinction is that the UMPC will
be lessexpensive than the typical tablet.

"Origami isn't an iPod killer per se, it's rather a new class
of device that will compete with other devices that cost
about the same," says Michael Gartenberg, an analyst at
Jupiter Research, using the original code name for the UMPC.
"That means portable media players, game machines, GPS units
and the like will face some new challenges. Much like there
were PCs with TV tuners long before MCE, the power of MCE was
really the 10-foot UI. The power of Origami is really in the
10" UI."

Bill Mitchell, corporate vice president of Microsoft's Windows
Mobile Platforms Division said, "While the first generation of
UMPCs will run Windows XP Tablet PC Edition 2005, future
models will run on Windows Vista."

Ultra-mobile PC Hardware

This isn't the first time that Microsoft has tried to promote
touch-screen tablets: Windows Tablet PCs have been around for
nearly a decade, and XP Tablet edition is on its second
iteration with all sorts of usability improvements. But the
UMPC is another step in this evolution.

In one report, Bill Mitchell said, "The touch-enhanced
display can be used as an on-screen QWERTY keyboard [using
Dial Keys] to navigate, or users can employ a stylus to input
handwritten information. They can also input content with
a traditional keyboard, linked either by USB port or wireless
Bluetooth connectivity."

Not all implementations of all UMPCs will have Bluetooth
ports, but the initial crop of vendors have promised this,
along with support for various USB and BT keyboards. This is
very similar to the tablet PCs that are currently on the
market: some come with keyboards and all offer support for
either pen or keyboard input devices.

Other input ports are planned for the devices, including:

* Wired Ethernet and WiFi networking connectors.
* Compact Flash card slots.
* USB v 2.0 ports.
* An external VGA monitor port.
* Various buttons and connectors for audio operations.
* Support for stereo microphones with built-in noise
reduction and echo cancellation.
* A remote control for the Media Center PC application.

There will be units with a variety of processors, including
the Intel Celeron M, Intel Pentium M and VIA C7-M. No AMD
chipsets are in any of the announced plans yet.

Microsoft's goal was to use off-the-shelf displays that are
in common consumer electronics devices to keep the costs down.
They settled on seven-inch VGA displays that had 800 x 480
native resolution and can handle 800 x 600 with some loss of
quality.

"The form factor was based around a 7-inch display panel,
a size that is currently a standard size in the electronics
industry due to broad adoption of that size display in the
portable DVD player and automotive markets," said Microsoft's
Otto Berkes in a newspaper interview. "Why use some new
custom size when a potentially good one already exists?"

These plans for the UMPC differ from the existing Motion
Computing tablets. The Motion displays use high-contrast,
non-glare, active digitizers, meaning that the pen sold with
the tablet is the only way to enter information. The UMPC
displays are general touchscreens that can work with a finger
or any other object for input. "We designed our 8.4-inch
display to meet the needs of legacy enterprise applications,
because many of these applications are designed for SVGA
(800 x 600) resolution," said Bert Haskell, a product manager
for Motion Computing. The UMPC displays are designed for
showing widescreen movies.