Digital Media

1
Digital Media

• Dr. Jim Rowan
• Chapter 2

2
First, somemac software you will be using(found
in the application folder)

• Grab - used to do a screen capture
• TextEdit - used as a simple word processor
• PhotoBooth - used to take your picture with the
  built in camera and take video with the built in
  camera
• Preview Change image file formats

3
About file formats and extensions(like .au,
.doc, .ppt, .mov)

• Indication to us (the humans) what kind of file
  this is
• Some software looks at the extension
• so... some software will try to open files with
  improper extensions
• results in file corrupted error message
• try it... change the extension from .doc to .jpg

4
File formats and extensions

• Some software looks at the data in the file for
  more definitive answer (the header)
• important file-related information is encoded in
  the data of the file
• for example some image formats have color tables
  to reduce the size of the file
• some video just saves the changes from one frame
  to the next
• weve seen the header before when we used
  hexFiend to look at images image size is stored
  in the header

5
The Question

• How do you put stuff in a computer
• so that you can manipulate it
• so that you can send it
• so that someone else can see and use it?
• How do you represent the real world in a digital
  world?

6
The answer

• Represent the real world as numbers
• Store the numbers
• Transmit the numbers
• Retrieve the numbers
• Display them in a form humans understand

7
Today

• Overview of things to come
• About the real world
• About digital representation

8

• From the first days lecture we saw different
  types of real world stuff as a bunch of numbers

9
Note on paper
10
Picture
11
Song fieldsOfGold.mp3
12
Video
13
So its all just numbers, and binary numbers at
that!First we must talk about numbering
systems!
14
First, Numbering systemsDecimalBinary
Hexadecimal
15
Which statement is True?

• 5 5 10
• 1 1 10
• 7 1 10
• F 1 10

16
well it depends

• 5 5 10 (in decimal)
• 1 1 10 (in binary)
• 7 1 10 (in octal)

welcome to numbering systems!
17
Numbering systems

• Humans decimal
• Humans 10 fingers, 10 digits
• 0, 1, 2, 3, 4, 5, 6, 7, 8 9 10
• Computers binary
• Computers 1 finger, 2 digits
• 0 1

18
Hexadecimal

• Humans and Computers hexadecimal
• Hexadecimal 15 fingers, 16 digits
• 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F

19
Why Hexadecimal?

• You can use one hexadecimal instead of 4 binary
  digits
• While this seems complicated.. it is actually
  easier (after some practice!) for humans to deal
  with 16 different digits than 4 0s and 1s
• In Hex 0123456789ABCDEF
• In binary 0000 0001 0010 0011 0100 0101 0110
  0111 1000 1001 1010 1011 1100 1101 1110 1111

20
How many different things?

• In Decimal 1 digit can represent 10 different
  things
• 0 1 2 3 4 5 6 7 8 9
• In Decimal 2 digits can represent 100 different
  things
• 00 01 02 03 04 05 06 07 08 09 10 11 12 97 98 99
• In Binary 1 digit can represent 2 different
  things
• 0 and 1
• In Binary 2 digits can represent 4 different
  things
• 00 01 10 11
• In Hexadecimal 1 digit can represent 16 different
  things
• 0 1 2 3 4 5 6 7 8 9 A B C D E F
• In Hexadecimal 2 digits can represent 256
  different things
• 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F
  10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
  20 21 F9 FA FB FC FD FE FF

21
How many different things?

• So how many things can you count with 4 hex
  digits?
• USE THE FORMULA!
• number of digits in the numbering
  systemnumber of digits used
• 164 65,536
• How many things can you count with 4 decimal
  digits?
• number of digits in the numbering
  systemnumber of digits used
• 104 10,000
• How many things can you count with 4 binary
  digits?
• number of digits in the numbering
  systemnumber of digits used
• 24 16

22
Countingwith a different number of
fingers(its the same process but different
number sets)

• 10 fingers Counting in decimal
• 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 then start over
  with 0 and increment the digit to the left gt 10
• 1 finger Counting in binary
• 0, 1 then start over with 0 and increment the
  digit to the left gt 10
• 16 fingers Counting in hexadecimal
• 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F
  then start over with 0 and increment the digit
  to the left gt 10

23
Binary Coding

• Data for a computer... Binary
• zeros and ones
• Data for humans... Hex
• 1 Hexadecimal represents 4 binary digits
• Data for humans ASCII (the alphabet)
• 2 hex codes gt 1 ASCII code
• For example gt

24
Example ASCII Code

• Humans and Computers ASCII
• The ASCII code for C is
• in hexadecimal 43 (In decimal 67)
• which is binary 0100 0011

25
Stuff in the Real WorldtoStuff on a computer
HOW?

• A note
• Letters of the alphabet-gt bits (0s and 1s)
• A picture
• Reflected light -gt bits (0s and 1s)
• A song
• Pressure waves in air -gt bits (0s and 1s)
• A video
• Pressure waves in air and Reflected light -gt
• bits (0s and 1s)

26
HOW?After numbering systems we need to know a
bit about the real world, aboutDiscreteContin
uous
27
Phenomena in the Real world discrete vs
continuous

• Things in the real world can be discrete
• They either ARE or they ARE NOT
• These things can be counted
• Examples
• The number of cars in the parking lot
• The number of beans in a jar

28
Phenomena in the Real world discrete vs
continuous

• Things in the real world can be continuous
• Continuous cant be counted, it must be measured
• Examples
• Atmospheric pressure
• Height of an ocean wave
• Frequency of a sound wave

29
The problem is... computers can only count

• Discrete data is easy for a computer
• count it and store it as a number
• Continuous data... easy? not so much
• music
• sample the music and then
• encode as a collection of numbers
• pictures
• measure the amount (intensity) and frequency of
  light (color) in a number of spots of light
  (pixels)
• encode the frequency and the intensity as a
  collection of numbers

30
Question...

• If computers only store 0s and 1s...
• How does all this continuous stuff end up in a
  computer so that we can save it and play it back?
• Answer
• Continuous data must be converted to discrete
  data through sampling

31
From the Real World and Back!

• Converting continuous phenomenon to digital data
• -You must SAMPLE to convert it to discrete
• Sampling consists of two processes
• 1) stop to take a measurement
• the number per time period is called the
  sample rate
• 2) take and store the measurement
• the number of different values each sample
  can take on is called the quantization
  level
• Digital data back to continuous phenomenon
• Display samples using sample and hold
• Play the sample for the duration of the sample
  time

32
Sampling

• more on this a bit later
• theres other stuff to consider before we jump
  into sampling

33
Before we go further with samplingWhat this
stuff means

• Bit binary digit
• Byte 8 Bits
• KB kilo byte (1000 bytes)
• MB mega byte (1,000,000 bytes)
• GB giga byte (1,000,000,000 bytes)
• TB tera byte (1,000,000,000,000 bytes)
• KBPS kilo (1,000) bits per second
• MBPS mega (1,000,000) bits per second

34
A note to the Geeks

• The rest of you guys pay no attention to this
  slide itll just upset you!
• Yes, I know that within the world of computers 1K
  is not 1000 but is 1024
• For this class lets just lie to the rest of the
  folks and keep it to ourselves!

35
Also note this!

• Communications are usually stated in bps (bits
  per second)
• File size is usually stated in bytes
• AND there are 8 bits per byte
• you will have to convert from one to the other
  when you do download/upload calculations

36
A bit about network access

• dial up connection
• ADSL
• T1
• T3

37
Network access...

• dial up connection
• phone modem
• asymmetric
• 56,000 bps (bits, not bytes) max downstream
  (internet to modem)
• 33.6 kbps upstream (modem to internet)
• rarely get these speeds

38
Network access...

• ADSL
• asymmetric digital subscriber line
• over copper phone wires
• limited to short distance from phone switch
• asymmetric
• 6.1 mbps downstream
• 640 kbps upstream

39
Network access...Commercial internet users

• T1 connection 1.544 mbps
• T3 connection 44.7 mbps
• Provide web servers for others to put websites on
• Large commercial enterprises will have their own
  web server

40
Time-to-download/upload calculations

• The Speeds
• Dial-Up
• 56,000 bps internet to modem (downstream)
• 33,600 bps modem to internet (upstream)
• ADSL
• 6.1 mbps (million bps) downstream
• 640 kbps (thousand bps) upstream
• T1
• 1.544 mbps
• T3
• 44.7 mbps

NOTE! bps is bits per second while filesize is
stated in bytes
41
And now for a little SAMPLING!
Sampling is the process of looking at stuff VERY
frequently then taking a measurement and storing
the measurement
42

• Next class sampling!

43
But... How many samples do you need?It
dependsSo lets look at sampling
44
(No Transcript)
45
(No Transcript)
46
single sample
47
single sample
48
single sample(sample and hold)
49
two samples
50
two samples
51
two samples (sample and hold)
52
three samples
53
three samples
54
three samples (sample and hold)
55
four samples
56
four samples
57
four samples (sample and hold)
58
five samples
59
five samples
60
five samples(sample and hold)
61
Taking it a lot farther
62
Taking it a lot farther
63
How frequently should I sample?

• too few
• small file size (good)
• not a faithful representation when replayed
• too many
• large file size (bad)
• excellent representation when replayed
• The Nyquist rate
• twice as many samples as the frequency being
  captured
• Results in an ok file size
• Results in faithful representation when replayed

64
CD quality is44,100 samples per second

• Why?
• Human hearing response is in the range of 20 to
  22,000 cycles per second
• Nyquist sample rate
• highest frequency to be captured 22,050 CPS
• 2 x 22,050 44,100 samples per second

65
Looking at FieldsOfGold.mp3

• 4 minutes and 59 seconds long
• 1,201,173 bytes in length
• Can this be right?
• CD quality
• 44,100 samples per second (sample rate)
• 16 bit samples (quantization level is 16 bit)
• 16 bits can store 65,536 different levels
• (216 65,536 individual levels)

66
FieldsOfGold.mp3

• 459 299 seconds long
• 299 x 44,100 samples per second
• 13,156,000 samples
• 13,185,900 x 2 bytes/sample (2 bytes 16 bits)
• 26,371,800 bytes
• Stereo 2 channels gt 52,743,600 bytes
• Should be 52 megabytes!
• Why does it show only 1.2 megabytes?
• HMMMmmm...

67
FieldsOfGold.mp3

• Why 52 megabytes not 1.2 megabytes?
• wait for it

68
FieldsOfGold.mp3

• Why 52 megabytes not 1.2 megabytes?
• This is an MP3!
• The data is COMPRESSED!
• If you had the song on a CD it would be 52
  megabytes long and in .aiff format

69
Two types of compression

• Lossy
• Lossless

.mp3 audio .jpeg images
run length encoding table compression
70
Further reading

• http//en.wikipedia.org/wiki/Nyquist_rate
• http//en.wikipedia.org/wiki/Sampling_28signal_pr
  ocessing29
• http//en.wikipedia.org/wiki/Mp3

71
The side effects of samplingsampling
artifactsSampling Artifacts are the negative
side effects caused by having to sample
continuous data
72
Sampling Artifacts

• Under-sampling not enough samples being taken of
  continuous data can produce undesired artifacts
• Examples might be
• Moire patterns on images
• retrograde motion on video

73
Sampling Artifacts (cont.)

• Not enough quantization levels when sampling
  continuous data can produce undesired artifacts
• Examples might be
• too few grey levels gradients become steps
• too few brightness levels posterization

74
(No Transcript)
75
(No Transcript)
76
4 samples/cycle, 2 cycles
Sampling Artifacts Retrograde Motion
2 samples/cycle, 2 cycles
77
Sampling Artifacts (cont.)

• Audio
• too few amplitude levels, quantization noise
• 8 bits (256 amplitude levels) produces
  discernable noise
• 16 bits (65,536 amplitude levels) CD quality, no
  discernable hiss
• general sound fuzziness or a flat sound

78
How can you store an image on a computer?

• Bitmapped images
• Vector graphics

79
Images, bitmapped

• Are stored as arrays of pixels
• Can be stored directly
• TIFF PNG for example
• Can have an associated color map
• JPEG for example
• Generating these pixels from the stored model is
  called rendering

80
(No Transcript)
81
(No Transcript)
82
(No Transcript)
83
(No Transcript)
84
Images, vector graphic

• Are stored as mathematical descriptions
• Often smaller than bitmapped
• Size of the file is independent of resolution or
  image size
• Not suitable for some type of images

85
Example Comparison

• Bitmapped graphics
• Defined as spots (pixels) of color
• Has problems scaling
• File size unaffected by image complexity
• File size affected by the image size
• Vector graphics
• Defined by their mathematically described parts
• File size affected by image complexity
• File size unaffected by the image size
• (scaling is easy)

86
(No Transcript)
87
(No Transcript)
88
(No Transcript)
89
Moving images

• Captured live with camera
• iMovie
• Stored as video
• Generated from animation
• Blender
• Similar to 2D vector graphics but in 3D and with
  a means of creating motion

90
2110-0102 for FridayNetwork communicationServ
ers and Clients
91
Servers Clients...

• Clients consume and display internet content
• Your browser is a client
• Clients request content from servers
• by sending a server an HTTP//URL message which
  is a request for a web page
• Servers respond to requests for internet content
• send requested web pages to Clients
• The content is sent in HTML code
• HTML sent by the server is interpreted by the
  client (browser) and displayed on your display
• Look at http//www.pondliner.com/ and view source
  

92
URL (uniform resource locator) a human-readable
name

• URL takes the form
• http//www.amazon.com/newStuff/index.html
• URL has 3 parts
• the protocol that you are using (http//)
• The domain name (www.amazon.com)
• The directory and file you want to see
  (newStuff.index.html)
• the URL maps to a number called an IP
• address

93
Servers Clients...

• servers have fixed IPs so they are easy to find
• your computer probably uses DHCP which is a
  dynamic (changing from connection to connection)
  IP
• An example my IP right now (assigned through
  dhcp) is (look it up in system preferences)
• IPv4 vs IPv6

94
requested webpages
DHCP
your browser (Safari)(client)
yahoo.com (server) 235.01.30.164
your computer
The Internet
walmart.com (server) 100.43.153.07
Domain Name System (DNS)
95
requested webpages
DHCP 135.10.34.143
your browser (Safari)(client)
yahoo.com (server) 235.01.30.164
your computer
ISP
The Internet
walmart.com (server) 100.43.153.07
Domain Name System (DNS)
96
requested webpages
DHCP 135.10.34.143
your browser (Safari)(client)
http//www.yahoo.com
yahoo.com (server) 235.01.30.164
your computer
The Internet
walmart.com (server) 100.43.153.07
Domain Name System (DNS) http//www.yahoo.com
235.01.30.164
97
requested webpages
/index.html DHCP 135.10.34.143
your browser (Safari)(client)
http//www.yahoo.com
yahoo.com (server) 235.01.30.164
your computer
The Internet
walmart.com (server) 100.43.153.07
Domain Name System (DNS) http//www.yahoo.com
235.01.30.164
98
Questions?