Physics 340
Audio Engineering

Final Exam Review Sheet:

Some of the things you should know for the final exam:

First, you must know everything from the previous three exams, including how to use Spectrogram and SAW. Use of the computer will be part of the final exam. Anything on the previous three exams could be on the final exam. As a start, I suggest you check out the previous three exams and the associated review sheets:

Review Sheet #1
Review Sheet #2
Review Sheet #3

In addition to the previous material, you are also responsible for the material below on digital audio recording/playback that we covered at the end of the semester:

Number Systems:

• Binary Number System:
  Only two numerals: 0 and 1
  Based on powers of 2. Base 2 system
  Each digit represented by one bit - On or Off
  8 bits per byte
  Number of different combinations for n bits is 2 to the nth power
  256 combinations (0 to 255) for 8 bits, 65536 combinations (0 to 65535) for 16 bits
  Be able to convert from binary to decimal
• Hexadecimal Number System:
  16 numerals: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F
  Based on powers of 16. Base 16 system
  Be able to convert from binary to hexadecimal and vice-versa
  Be able to convert from hexadecimal to decimal

Digital Recording and Playback:

• Generate numbers to represent the instantaneous value of the audio waveform - called sampling the audio
• Waveform is sampled at some specific rate called the sampling frequency that must be at least twice the highest audio frequency that is to be recorded (Nyquist Theorem)
  For standard CDs the sampling rate is 44.1 kHz
  To prevent aliasing the signal must be filtered prior to sampling to remove frequencies above 1/2 sampling frequency
  Aliasing consists of sampling distortion products that are below 1/2 of the sampling frequency and will be heard in the sampled file
• Sampling is done by an Analog to Digital Converter (ADC)
  Measures instantaneous signal voltage and outputs a number
  Is the most critical part of the entire process
• Bit Resolution or Bit Depth:
  Number of bits that are used to represent each sample
  Each bit represents a power of two, so the number of different numbers that represents each sample is 2 to the number of bits: 8 bits = 256 numbers, 16 bits = 65536 numbers, etc.
  1 byte = 8 bits
  Standard CD bit resolution is 16 bits or 2 bytes per sample
  Bit depth can be any number of bits, but is typically a multiple of 8
• Bit resolution and Signal to Noise (S/N) ratio:
  6 dB of S/N ratio per bit of resolution - be able to PROVE this
  8 bits - 48 dB S/N ratio - unacceptable
  CD standard: 16 bits - 96 dB S/N ratio - excellent
• Relation between actual bit depth and audio level:
  For each 6 dB below maximum that the audio is recorded, 1 bit of resolution is lost
  Causes problems with current 16 bit resolution and soft passages in the music
  Solution is to someday go to 20 or 24 bit resolution - being done already in studios
• Data Rate and relation to sampling frequency, bit resolution, and number of tracks/channels:
  Data Rate = sampling frequency x number of bytes per sample x number of channels
  For CD: approximately 11 Mbytes per minute
• Playback:
  Read back numbers from storage device at the proper rate
  Send numbers to Digital to Analog converter (DAC) that outputs a specific voltage based on the numerical input
  Second most critical part of the entire process
• Storage Media:
  Theoretically, any type of computer internal/external memory: hard disc, floppy, digital tape, ROM, RAM, CD, etc.
  Large data rate for high fidelity (44.1 kHz, 16 bit) audio requires massive amounts of storage
• Audio Compact Disc:
  Capable of storing 74 minutes of 2 channel, 44.1 kHz, 16 bit audio
  For audio, data must be a continuous, uninterrupted spiral track that starts at the center of the disc
  Data is stored by burning "pits" on the surface
  Pits are read by reflecting the light from a laser off of them
  Disc turns so that a constant linear (tangential) velocity is maintained at the read/write location
  Results in a non-constant angular velocity
  
  

Back to the Physics 340 Audio Engineering Page

Back to Dr. Greg Latta's Electrical Engineering and Amateur Radio Pages

If you have any questions or comments, you can send E-Mail to Dr. Greg Latta at glatta@frostburg.edu