UPSAMPLE

upsample -n<x> <DIF input file> <DIF output file>
-n<x> upsample by x

SIMADC



simadc <-n> <DIF error file> <DIF input file> <DIF output file>
-n<nn> output to nn more bits than input
-d truncate output word with nn bit binary offsets after ADC
default is code values with float offsets
-o<n> offset INL error by n codes
-i ignore data type of DIF input file
-q<n> use n bits to requantise before modelling
-s add INL offsets
default is to subtract

PSRCORRECT

Power series correction using dB or linear coeffients

psrcorrect -n<x> <DIF input file> <DIF output file>
-d<offset> convert coefficients from dB and subtract dB offset from polynomial coefficient
-x<amp> amp is test signal amplitude as coefficient
-s(,t,f,F,S,v,e,n,T)<x> select second to tenth order coefficient

OSPROC

Time averaging interpolation of a DAC or AWG

osproc <-n> <DIF input file> <DIF output file>
-o<number> oversampling ratio in octaves/bits
-i<string> table ident string is the internal index in the look up table file
-n<bits> DAC output word length in bits
-g<file> DIF file for interpolation result
-d<file> Oversampling matrix data:look-up table file
-r keep original FSR
-s<number> Interpolation amplitude
default is 1.0 LSB

IFFT



ifft <flags> <DIF input names> <DIF output names>
NB first two files must be input (real) (imaginary)
next files are output real and imaginary
output files created are .ffr (real) and .ffi (imaginary) or .ffc(power spectrum)
-p output magnitude in Real file\n
-s< – Scale result to no bits\n

GENWAVE

genwave [flags] <DIF file>
-f<frequency> Waveform frequency kHz,
-q<frequency> Sampling frequency kHz
-w<wavetype number>
Wavetype number
0, sine,
1, triangle
2, square
3, DC
4, Ramp up
5, Ramp down
6, Pulse,
7 rand11,
8 rand12

-a<amplitude> Maximum amplitude
-l<units> Y units
-L<label> Y label
-U Unipolar wave
-P<degrees> Starting phase
-s<number> Total number of samples
-m<number> Number of MARK samples
-c<cycles> Number of cycles: valid with -q if -f not specified
with PRBS and -k sets base frequency to -q/cycles
-n<resolution> Resolution in bits of output
-r<fsr> Fix full scale range to fsr bits
-O<nn> oversampling factor
-e This allows the waveform to have a non-integer number of cycles(useful for adding ‘leakage’)
-d simulated ADC code output
-o<nn> offset in code value (integer)
-z<n> adjust amplitude by n dB N.B. default is -0.1dB
-k force odd number of cycles
also changes PRBS sampling frequency for synchronous ADC test
-u<n> update system default test frequency n
-1;

HISTGEN
histgen <-n> <DIF input file> <DIF output file>
number of ADC bits is in input file or use
-n<bits> – force number of ADC bits to this value
-s use max an min in data rather than header

HISTCON

histcon <-flag> <DIF input file> <DIF output file>
-v output table of results in output file.tbl
-a include histogram ends
-A<n> assume n : 1 resampling
-e use average of mid range codes to estimate inl confidence limits: default uses estimated pdf of all codes)
-m dnl relative to code centres (IEEE 746-1984)

-r assume ramp test signal: default is sine wave
-k assume locked histogram
-Snn report sampling frequency as nn kHz

HISTACC
histacc <-x> <DIF input file> <DIF output file>
-x – ignore missing number of bits
-R – reset accumulations to 1

PROCFFT

fftproc <-n> [DIF FFT input file]
-v provide results table
-b<n> + – bin range to use for power
default is 4 for Blackman-Harris and Rosen, 2 for Hann and Hamming and 7 for custom
-f<nn> use fundamental of nn kHz
-i<nn> do 3rd order intermod test with higher tone at nn kHz (-f must also be specified)
-I<nn> use second order difference frequency of nn Hz

-C assume complex FFT data
-e include effective bits in table
-r calculate new power values for DIF
-o<n> scale results by n dB
-w do not give aliasing warning
-d<n> zero first n bins in power calculation default is auto-zero of DC\n
-D<string> file description to use
-l<n> use up to the nth harmonic in THD
default is up to 10th harmonic\n
-c<n> ignore harmonics less than the nth one in finding THD
-s<n> ignore harmonics which are
-A<n> assume data is resampled by n:1 // also sets aliased flag for funamdntal
-H<n> n = 1 find even THD, n = 2 find odd THD
-S assume data is from a spectrum analyser more than n dB below FSR(default is 100 dB)
-x print difference between total power and fundamental power
-u<nn> limit calculations to DC to nn kHz
-U<nn> limit calculations to nn kHz to upper limit
-T<nn> remove tone at nn KHz from calculations
-p<nn> use nn dB as total power reference
-a<nn> set baseline of display to nn dB
-g give advice on noise level of ideal ADC

DOWNSAMPLE
downsample -n<x> <DIF input file> <DIF output file>

-n<x> downsample by x

DISIM
disim {Flags} {Input data} {Threshold data} {time output}
-i{DIF data file} DIF user supplied input
-t{DIF threshold file} User supplied ADC thresholds (sets -bn)
NB thresholds must have INL of less than 5 lsb
or will not be recognised correctly
default is to generate ideal thresholds
-b{quant_bits} Resolution in bits (Set automatically in -t flag
-o{n} OS ratio (in bits)
-v verbose Print all information
-f{step size in LSB}
-a{Amplitude db’s} Adjust input amplitude in db’s
-A Turn off auto-gain ranging of input data
NB system auto-scales input DIF so amplitude changes ignored
-w<file> output time data file
-z operate with no interpolation

DIFYSCAL
difyscal <-x> <DIF file>
-o don’t change max and min
-d report max and min
-l<n> scale by n
-s<n> scale by n dB
-i<n> scale by 1/n
-r<nn> rescale to data starting at sample nn
-e<nn> rescale ending at sample nn

DIFYOFFS

difyoffs <-x> <DIF file>
-o<nn> offset by nn

DIFXSCAL

difxscal <-x> <DIF file>
-s<n> scale sampling frequency by n
-l<n> scale x axis by n

DIFSUB

difsub <-x> <DIF input file> <DIF input file2><DIF output file>
-x – ignore missing number of bits
-y – ignore missing Ymax or Ymin
-t<dbmin> – threshold for FFT subtraction default -30

DIFQUAN
difquan <-n> <DIF input file> <DIF output file>
Default – number of ADC bits is in input file
-n<bits> – force number of ADC bits to this value
-t – force data to be code type

DIFPSR
difpsr -n<x> <DIF input file> <DIF output file>
-c use neg coefficients
-s<x> x is second order coefficient
-t(f,F,S,v,e,n,T)<x> select third to tenth order coefficent

DIFPOWER

difpower <-flags> <DIF input file>
(s_error, ” -x – ignore missing number of bits
-y – ignore missing Ymax or Ymin
-r<nn> set power reference level to -nn dB rel FSR
-o<nn> subtract offset nn from data in power calculation
-p calculate peak rel to FSR in dB
-f calculate crest factor in dB
-m calculate offset
-c subtract header offset from data in power calculation

DIFMUX
difmux <-x> <DIF input files> <DIF output file>
-c<n> n=2 or 4 channels to mux
default is n = 2
n DIF input files ending 1,2,3,4
-x ignore missing number of bits

DIFMULT
difmult <-x> <DIF input file> <DIF input file> <DIF result file>
-x – ignore missing number of bits
-y – ignore missing Ymax or Ymin
-s – scale to main file y range

DIFMETER
difmeter <-x> <DIF input file> <DIF output file>
-R<nn> use for 0dB
-r – rms result,default is ppm result

DIFMAV
difmav <-n> <DIF input file> <DIF output file>
-n<length> moving average length
-s<length> number of stages
-r rescale output data to input DIF

DIFMAP

difmap <-n> <DIF error file> <DIF input file with non-linear distortion> <DIF output corrected file>
-n<nn> map corrected output to nn more bits than input
-d truncate output word with nn bit binary offsets after mapping
default is code values with float offsets
-o<n> offset DIF input file by n LSB
-N<m> begin using DIF error file at code m
-i ignore data type of DIF input file
-t<n> add n to segement average ILE
before tracking. Default is 0.5 LSB
if -t0 is used: lin correction using INL instead of using segment mapping
-s add map offsets
default is to subtract
-S<n> corect using n segment model(default is 16)
-M (debug) output offsets in offsets.csv

DIFLNR
diflnr <-n> <DIF input file> <DIF output file>
-s<nn> scale offset by nn
-m generate error offset file

-t<nn> use nn as decision threshold

-l limit offset to one threshold

DIFJOIN

difjoin <-x> <DIF input file> <DIF output file>
-x – ignore missing number of bits
-y – ignore missing Ymax or Ymin
-h<nn> Set maximum value to nn
-l<nn> Set minimum value to nn

DIFIIR

difiir <-n> <DIF input file> <DIF output file>
-s<number> number of stages
-n<number> fir impulse response
-r rescale data
-c<file> 5 ceoffs in file per stage

DIFEDIT
difedit <-x> <DIF input file> <DIF output file>
-s<nn> Set start sample number to nn
-n<nn> Set number of samples to extract
-e<nn> Set end number of samples nn to delete
default is rest of file

DIFDEMUX
difdemux <-x> <DIF input file> <DIF output file>
-c<n> n=2 or 4 channels to demux
default is n = 2
creates n DIF output files ending 1,2,3,4
-x ignore missing number of bits
-h<name> use header file names with attributes
.ch1, .ch2 , .ch3, .ch4

DIFCOMB

difcomb <-n> <DIF input file> <DIF output file>
-n<length> comb delay length
-p pass DC

DIFBTOA
difbtoa -h -d <Binary DIF file> [ASCII DIF file]
-h Prints header only
-d Prints data only
-s Prints binary states also
-n Number data values
-W write a WAV file

DIFAV
difav <-x> <DIF input file> <DIF output file>
-x – ignore missing number of bits
-y – ignore missing Ymax or Ymin
-R – reset averages to 1

DIFATOB
difatob -h<head> <ASCII data file> <Binary DIF file>

flags -b Convert data from Unipolar to bipolar
-u Convert data from bipolar to Unipolar
-r<n> Shift data n bits towards LSB direction
-t<n> Ignore first n samples
-T<title> append to title

DIFADD

difadd <-x> <DIF input file>(<second DIF input file>) <DIF output file>
-x – ignore missing number of bits
-y – ignore missing Ymax or Ymin
-u set fundamentals from inputs
NB fund 2 not set in output if only TWO files defined
-Y set Ymax and Ymin from data
-a if two input files not equal,use shorter of two
default is to use first file

DASTODIF

DAStoDIF -h<head> <ASCII data file> <Binary DIF file>
flags -b Convert data from Unipolar to bipolar
-u Convert data from bipolar to Unipolar
-R<n> Shift data n bits towards LSB direction
-T<n> Ignore first n lines of ASCII file)
-C Convert to Complex two columns of ASCII file)
-K Convert to Complex second pair of three columns of ASCII file)
-A Convert Audio Precision .aam file)
-W Convert Microsoft .wav file)
-S Convert Microsoft .wav stereo format file)
-f<frequency> sampling frequency in KHz
-t<frequency> test frequency in kHz
-r<number> DUT serial number
-p<number> DUT part number
-w save changes to header file
-n<samples> number of samples to capture
-m<bits> number of data bits per sample
-P<nn> demultiplex bytes from 16 bit data
n= 1 : use MSB
n= 2 : use LSB
n= 3 : de-mux MSB followed by LSB
n= 4 : de-mux LSB followedby MSB
-B<nn> unpack 4 bit nibbles from byte data
NB only valid with -b option
n= 1: de-mux MS nibble followe by LS nibble
n= 2: de-mux MS nibble followed by LS nibble
-g Convert gray code to twos complement
-h<header file> DIF header file

DACPROC

dacproc <-n> <DIF input file> <DIF output file>
-s<number> DAC output level in dB rel to DAC FSR
-n<bits> DAC resolution in bits
-t<bits> min value of ADC LSB size
divided by DAC LSB size
NB default is 1.0
-r Quantise data to ADC resolution bits
default is to requantise to DAC resolution

CFFT
cfft <flags> <DIF input name> <DIF output name>
NB input files must be .tir(real) and .tii (imaginary)
or only one .tim file
output files created are .ffr (real) and .ffi (imaginary) or .ffc(power spectrum)
-w<n> – Window default is 0: no window with b=0
1 – Hanning
2 – Hamming
3 – Blackman-Harris
4 – Rosenfeld
5 – PTB
-b<n> – Number of bins either side of fundamental
to use in power computation
default is 6 for PTB
default is 4 for Blackman-Harris
3 for Rosenfeld and 1 for others
-p output power spectrum
-s<v> – Scale factor to apply to result
-n<nnn> Use first nnn samples of time file
-c adjust number of samples to a lower
value which is a multiple of 2
-C zero fill and adjust number of samples to a higher
value which is a multiple of 2
-f correct the fundament value if
Rosenfeld window in use

ADCSIM

adcsim {Flags} {Input data} {Threshold data} {Text output} {FFT output}
## Input data type ##
-s{sin_freq} Sinewave (default input)
-g{load} Gausian noise
-i{DIF data file} DIF user supplied input
## Threshold data ##
-t{DIF threshold file} User supplied ADC performance (sets -bn)
## Text output data ##
-1{Text outfile} File name of text output file
-x{DIF outfile} File name of DIF SNR output file
NB used for iterative simulations only
If osb1 = osb2, Contains SNR vs step data
else Contains SNR vs osb data
## FFT mode, output data file ##
-z{DIF spectrum file} DO FFT, file name of FFT output file
## Flags ##
-n{rec_lng} Number of samples
-b{quant_bits} Resolution in bits (Set automatically in -t flag

-p{st_points} Number of increments of P-P interpolation waveform
NB used for iterative simulations only
-d{stepinc} Starting P-P interpolation ramp amplitude in LSB’s
NB used for iterative simulations only
-o{osb1} Start OS ratio (in bits)
NB specify only -o option unless SNR vs osb is needed
-r{osb2} End OS ratio (in bits)
-S eliminate second harmonic from SNR results
-v verbose Print all information
-m monitor… use this all the time
-f{step size} Step size for single pass time or fft output
-a{Amplitude db’s} Amplitude in db’s
-l{Parameter file} Get parameters from file:<osbits p-p amp>
-w<file> put output time data here

ADCERR
adcerr <-r -d> <DIF input file> <DIF output file>
Default – Terminal based Integral non-linearity error (INL)
-a – Absolute INL using DIF header Ymax, Ymin
-l – LSB size fixed by DIF Ymax,Ymin and resolution,otherwise terminal based LSB
-d – Differential non-linearity error (DNL)
-m – use code mid points for DNL (IEEE 746-1984)N.B. the -m option is valid only when used with -d option
-n<nn> start at nn codes from start
-e<nn> finish at nn codes from nd