audio and midi protocols Flashcards
analogue is ….
digital is ….
continuous time signals
discrete signals, snapshots
midi is a … process
serial
what was the first analogue recorder?
Thomas Edison, tin foil cylinder 1877
made of a horn, a diaphragm a needle and a tinfoil cylinder into which the needle impresses the analogue signal from vibrations
what is tape made of and what are its characteristics in terms of sound?
made of polyester or iron oxide or ferric oxide, something that can store a pattern of magnetisation
the strength of magnetisation varies in strength and direction to represent the original waveform.
tape sound has dynamic and harmonic distortion that gives it colour compared to digital.
how does digital recording work
converts analogue signal into numbers and stores them
analogue wave played back from digital is the exact same unless it has been corrupted
what is ADC and DAC
analogue to digital converter
digital to analogue converter
define sample rate and quantisation resolution
as these two factors increase…
sample rate - samples taken per second or snapshots
quantisation resolution - the number of possible gradients available when taking those snapshots, discrete amplitude resolution
…fidelity increases
what do 1 and 0 represent in binary?
1 is on
0 is off
how many midi numbers are there
128
what are the minimum bytes needed to be sent over for midi to work, what do these 3 bytes contain
3 bytes, one status byte and two data bytes
SB has
- action - whether note is on or off
- midi channel - ch 1 is 0000, ch2 is 0001… ch 16 is 1111
first data byte has
pitch
second data byte has
velocity - from 1 to 127 and 0 is off
in the status byte under action, what denotes note on and note off?
status byte box - 1
on - 001
off - 000
what are 2 ways you can represent a note that is turned off in the midi table?
in action have 000 and max velocity
or
in action have 001 and 0 velocity
how many bits in a byte
8
give limitations of an analogue system
copies are never as good as original
analogue recording always adds noise
how many samples would be taken with sample rate 50kHz of 1 minute of stereo audio?
2 x 50000 x 60 = 6,000,000 samples
times by2 because its stereo
give the Nyquist criterion
the sampling frequency must be at least twice as fast as the highest frequency in the signal
this is to capture the peaks and troughs
define critical sampling
when samples are taken at the peaks and troughs to produce the most accurate representation possible of the wave form
what is aliasing?
what would happen to a pitch frequency of 30kHz running thru a 48k sample rate?
when the sample rate is not high enough for the pitch frequency of whatever it is sampling.
results in a wave to be produced that is lower than the pitch frequency, an aliased frequency, you cannot reconstruct the original higher freq if aliasing has occurred
the 30kHz frequency would become an 18kHz frequency because 48-30=18
what does an antialiasing or low pass filter do and where is it found?
found between the preamp and the ADC
it makes sure no frequencies are occurring that are more than half that of the sample rate
what does a smoothing filter do and where is it found?
found between the DAC and the amplifier
it changes the samples taken into a smooth representation of the original
what was the problem with early anti aliasing filters?
they were tight brick wall filters that created resonant peaks at those cut offs
now they are smooth
how do you calculate dynamic range
amount of bits x 6
what is dithering?
adding white noise to a signal on purpose to remove quantisation distortion
quantisation distortion refers to the signal noise and error that is created when lowering bit depth of a signal, quantisation values are reduced when lowering bit depth so forces some values into values that do not represent them well.
dithering turns the quantisation distortions into low level hiss sounds, getting rid of unwanted harmonics
retains the audio quality of the original sound wave
