What is baudline?
Mystery Signal
Griffin iMic v3.00

vendor Griffin Technology
product iMic v3.00
interface USB 1.1
duplex full
channels 2
resolution 16 bits
max rate 48 Ksample/sec
codec Philips UDA1325H
operating system Mandrake 9.0 Linux x86 2.4.19
driver usb-audio OSS 3.8.2
buffer size input 128 KB, output 128 KB
test date Sep 23 2005
notes The iMic v3.00 is not usable with either Fedora core3 or Ubuntu 5.04 because the ALSA USB audio driver locks up in a matter of minutes.

The iMic has poor EMF shielding and is literally "microphonic."  Physically tapping the device causes bass impulses on the line level inputs.  Avoid placement near AC power lines since the iMic is very susceptible to 60 Hz bleed in.

Vendor=077d ProdID=0223 Rev=3.00

This card is part of the Full Duplex DAQ comparison survey.

Sample Rate
The sample rate on DAQ cards is not a fixed absolute constant.  Like time, it fluctuates, and it is difficult to measure accurately.  Sometimes there are relationships between the input and output sample rates that can reveal interesting details about the inner working machinery.

The following table of measurements use a technique described in the sample rate stability application note. The rate column is the sample rate value that the collection hardware is programmed to.  The in/out rate and in/out error columns are absolute measurements of the ADC / DAC clock.  The loop error column uses a tone generator loopback method for a high accuracy measurement of the relative difference between the ADC and the DAC clocks.  The three error PPM columns are theoretically related by the formula: "in_error - out_error = loop_error"

rate in rate out rate in error out error loop error
5510 5512.785 5511.091 +505.445 PPM +198.004 PPM +321.184 PPM
8000 8000.414 8001.535 +51.7500 PPM +191.875 PPM -130.707 PPM
11025 11025.57 11027.12 +51.7007 PPM +192.290 PPM -132.979 PPM
12000 12000.61 12002.25 +50.8333 PPM +187.500 PPM -130.857 PPM
16000 16000.79 16003.06 +49.3750 PPM +191.250 PPM -130.901 PPM
22050 22051.10 22054.14 +49.8866 PPM +187.755 PPM -130.271 PPM
24000 24001.19 24004.61 +49.5833 PPM +192.083 PPM -130.949 PPM
32000 32001.60 32006.04 +50.0000 PPM +188.750 PPM -130.962 PPM
44100 44102.24 44108.13 +50.7937 PPM +184.354 PPM -130.270 PPM
48000  48002.45  48008.97  +51.0417 PPM  +186.875 PPM  -130.653 PPM 

The theoretical "in_error - out_error = loop_error" formula is true for all of the same rates.  All sample rates except for 5510 have a -131 PPM loop error measurement.  This machine was not NTP disciplined so the absolute rates could have the typical system clock drift.  The relative values, on the other hand, can be trusted to a high degree of accuracy.

Note that the 4000 sample rate was not operational with the iMic v3.00 device.

The -debugrate command line option was used to create the sample rate estimate plot shown below.

This smooth looking convergence has the output approach inverted.  Usually both the input and output paths have an upward convergence decay.  The reason for this flipped abnormality with the iMic v3.00 sample rate data could have to do with the 2.4 kernel and/or the older OSS audio drivers.

The delta between the input and output channels matches the -131 PPM loop error measurement.  The stability of the rate tracking curves signify that the ADC and DAC are operating at different sample rates and that some other dropping mechanism is not at work.

All of the sample rates except 5510 share a similar looking smooth convergence plot.  There is something special about the 5510 rate and it's -debugrate sample rate plot is shown below.

The 5510 rate has the same inverted convergence plot but the added caveat of the input and output paths crossing over.  It is interesting that the 5512 rate is directly in between the input and output curves.

The 48000 rate, and to a lesser extent the 44100 rate, exhibit random spectral glitches that occur every couple minutes.  Below is an example of what the wideband spectral glitch looks like.

The waveform display shows a 1 ms zero filled gap with standard RC attack-decay.  The zero gap is always one millisecond and this duration is independent of sample rate.  The zero gap has a slight DC offset that is different than the DC offset of the noise floor.  The combination of the RC ringing and the DC offset suggests that this sampling flaw is occurring on the DAC output side.

Frequency Domain
The sound card's input and output jacks are connected with a short external cable and run in full duplex mode.  This is a loopback test and baudline's tone generator is the signal source.  Distortion, noise floor, filter response, and inter channel crosstalk are the frequency domain measurements of interest in this section. 

The signal test sources are a pure sine wave, a linear sine sweep, and WGN.  The sine wave is used for the distortions and crosstalk measurements.  The linear sine sweep and WGN are used for the filter characterization measurement.  Both are an application of the swept sine vs. WGN technique and are equivalent measures of the frequency response. 

Since spectral performance is a function of sample rate, each of the sound card's native rates will be tested.  The highest sample rate is usually the cleanest and this is advantageous because it allows the isolated testing of the ADC and the DAC.  The matched, source, and sink sample rate combinations are described below.

The input and output sample rates are the same.  This combination tests the performance of both the ADC and the DAC in a matched mode of operation.  The linear sine sweep signal in the left spectrogram display and the WGN (orange) in the Average window characterize the in-band filter response.  The sine wave (green) in the Average window is used for distortion and crosstalk measurements.  The sine leakage (purple) is used for crosstalk measurement

The sample rate of the input (sink) is the card's highest clean rate.  This combination tests the performance of the DAC.  The linear sine sweep signal in the middle spectrogram display characterizes the DAC filter response.  The position of the pass-band and the stop-band filter transition is defined by the Nyquist frequency of the DAC.  The noise floor (purple) is the Average collection of a silent channel.

The sample rate of the output (source) is the card's highest clean rate.  This combination tests the performance of the ADC.  The linear sine sweep signal in the rightmost spectrogram display and the orange curve in the Average window below it characterize the ADC filter response.  The position of both the pass-band and the stop-band filter transition is defined by time in the spectrogram and by folded frequency in the Average window.  The orange Average curve represents the pass-band while the cyan curve is a folded representation of the stop-band ADC filter response.  The noise floor (purple) is the Average collection of a silent channel.

The naming convention for the columns below is (DAC -> ADC) where DAC represents the source sample rate and ADC represents the sink sample rate. 

source (DAC)
sink (ADC)
5510 -> 5510 5510 -> 48000 48000 -> 5510

8000 -> 8000 8000 -> 48000 48000 -> 8000

11025 -> 11025 11025 -> 48000 48000 -> 11025

12000 -> 12000 12000 -> 48000 48000 -> 12000

16000 -> 16000 16000 -> 48000 48000 -> 16000

22050 -> 22050 22050 -> 48000 48000 -> 22050

24000 -> 24000 24000 -> 48000 48000 -> 24000

32000 -> 32000 32000 -> 48000 48000 -> 32000

44100 -> 44100 44100 -> 48000 48000 -> 44100

48000 -> 48000  

The ADC and DAC filters both have sharp roll-offs with about -60 dB of attenuation.  The source DAC filters at 32000 look like scaled versions from the previous sample rate.  The sink ADC filters have identical stop-band shapes at all sample rates 24000 and lower which is very unusual.

Also of interest is that the passband for the source column has a slight negative slope at the 22050 sample rate and below.  This coupled with the above observation suggest a crude low tap interpolation by 2 filter is used.

A 1000 Hz tone plus it's harmonics can be seen in the noise floor of the source and sink column Average spectrum plots.  This is signal contaminates all the sample rates and it is believed to be an artifact from the internal DSP engine.  Another interesting spectrogram artifact seen at the 22050 rate and below is a strong vertical tone at twice the Nyquist frequency.

The following table of measurements were made using the technique described in the sine distortion application note.  It is a full duplex test that uses a loopback of the tone generator to measure the various distortion parameters.  The stereo crosstalk column is a measure of channel leakage that uses a sine wave channel and a silent channel as the signal sources.

rate SNR THD SINAD ENOB SFDR crosstalk
5510 +78.15 dB -81.53 dB +76.51 dB +12.415 bits +79.36 dB -80.58 dB
8000 +77.40 dB -81.65 dB +76.01 dB +12.333 bits +78.47 dB -79.02 dB
11025 +77.83 dB -81.43 dB +76.25 dB +12.373 bits +79.26 dB -76.39 dB
12000 +76.722 dB -81.138 dB +75.38 dB +12.228 bits +78.01 dB -75.61 dB
16000 +76.20 dB -80.80 dB +74.91 dB +12.150 bits +77.52 dB -73.31 dB
22050 +76.21 dB -80.43 dB +74.82 dB +12.135 bits +77.95 dB -70.74 dB
24000 +75.39 dB -80.17 dB +74.14 dB +12.022 bits +76.77 dB -70.08 dB
32000 +79.97 dB -79.51 dB +76.72 dB +12.451 bits +80.77 dB -67.88 dB
44100 +78.95 dB -78.05 dB +75.47 dB +12.242 bits +79.31 dB -65.37 dB
48000  +79.18 dB  -77.60 dB  +75.31 dB  +12.216 bits  +78.74 dB  -64.76 dB 
48000 / 1024  +79.57 dB -inf.00 dB  +79.57 dB +12.924 bits +90.46 dB -64.71 dB

The performance is average and the distortion measurements are fairly constant at all of the sample rates.  The crosstalk measurement is poor and it could be related to the sensitivity of the iMic's "microphonic" nature.

deep zoom
The deep spectral zoom concept was used to discover some strange behavior with the iMic v0.06 so it is being used again here for comparison purposes.  The iMic v3.00 had none of the loop error instability problems that the iMic v0.06 had so this should be educational.

The down mixer feature in the Input Devices window is used to perform a deep zoom into the frequency domain.  This DDC increases the frequency resolution and allows the observation of finely spaced spectral details.  The sample rate is 48000 with a 1024 decimation ratio for an effective 46.875 sample rate.  The 1024 decimation factor translates to the frequency resolution of an equivalent 1M point FFT.  Below are the standard sine wave, WGN, and crosstalk spectrum plots. 

48000 / 512

The two modulation lobes left and right of the main tone are offset by about 1.5 Hz.  This amount of deep spectral zoom usually increases the ENOB by significantly more than 1 bit but the existence of the modulation lobes lowers this improvement to only 0.8 bits.  This deep spectral zoom is somewhat clean and none of the problems that were seen with the iMic v0.06 are present. 

A white Gaussian noise signal source was generated and captured in full duplex loopback fashion at each of the standard sample rates.  The Histogram plots below show a unique sample distribution that is dependent on sample rate.

4000 ... 48000

All sample rates have nice clean Gaussian shaped histogram curves.

Channel Delay
A sine wave signal was generated and captured in full duplex loopback mode.  The time domain response was observed with the Waveform window where the green curve represents the left channel and the purple curve represents the right channel. 

Zero sample inter channel delay at all sample rates.

The only global analysis correlation that can be made is that the 5510 rate had an odd loop error PPM measurement and it has an odd source DAC filter shape.

It is interesting to compare the iMic v3.00 filter shapes with those of the iMic v0.06.  Both USB devices use different revisions of the same Philips codec chip.  The filter shapes are mostly the same but the 0.06 version has about 15 - 20 dB more of attenuation which results in slightly better distortion measurements.

The iMic v3.00 has good ADC / DAC filters like it's iMic v0.06 predecessor.  Some of the old flaws have been fixed while some new ones have been added (filter attenuation for example).

Other than the occasional zero gap spectral glitch at the 48000 and 44100 sample rates the performance of the iMic v3.00 is fairly good.  The unusual -131 PPM loop error being the only exception.  If absolute or full duplex sample rate error is important then it is best to avoid the 5510 rate.

Copyright © 2005 - group - blog - site map