Gear / Technical Help > Post-Processing, Computer / Streaming / Internet Devices & Related Activity

Measuring performance of audio interfaces

(1/11) > >>


some discussion has arisen in other threads about testing the devices we commonly use for recording, by measuring actual performance to the best of our ability.

The desire to measure for me was born out of necessity. One of my main recording devices i use is the DPA MMA d:vice unit, which uses a synchronous USB connection to iOS, which puts it at the mercy of the phone's clock timing which can potentially create jitter, i needed to be able to generate an FFT plot of a 10K sinewave fed to it to look for possible jitter sidebands. Which was a little bit of a learning curve in understanding FFT plots, filtering, windowing, etc., to get usable results.

in a similar time period, i was reading many threads on the audiosciencereview forum, and gaining a better understanding of what measurements are done, and why, and what effect different types of noise have on the resulting sound. While that forum is mostly playback-centric, they do on occasion do some ADC tests

the main person at that site, user amirm, has amassed an impressive collection of measurements using his Audio Precision APx555 analyzer. The site in general is a great resource for those interested in exploring new gear. you can look at any individual review by clicking on the name of the DAC in the list on the left. You can read more about the APx555 unit here. The embedded youtube video by Dr. Tom Kite is highly informative.

Lacking the motivation and spare change to buy a $30,000 analyzer for a hobby, further reading on that forum led me to several other options for measurements.

While more economical stand-alone signal analyzers can be had, the modern day DACs and ADC combos have really come of age, and have the resolution to measuring most field recording equipment we use in excess of 110dB (human hearing is thought to have a maximum dynamic range of 116 dB, state-of-the-art (SOTA) playback systems are north of 120 dB

one ADC/DAC of choice for measurement is the RME ADI-2 PRO FS-R BE. As you can see from its loopback measurements, it has sufficient specs to measure the gear that we commonly use

However, one need not break the bank to find measurement-quality interfaces. As you can see from this link, cheaper, more common interfaces can also be effective for measuring gear as well, particularly the handheld recorders that are popular around here, which usually top out at around 100 dB of dynamic range

here are some measurements ive done on both dedicated ADCs and handhelds, the dedicated ADCs beat them handily, however, this could partially be a consequence of the higher input levels they are designed to receive

Fortuantely for our purposes, we are measuring ADC stages, where DACs are a bit more mature, and looking at those numbers, one can see that almost any decent modern DAC can generate a signal clean enough to test all the handhelds at least. All the signals for the test above were played from a Mytek Brooklyn DAC+ which has a signal-to noise of around 116 dB at its +22dBU output. Not quite the performance of the RME unit but in same class.

I could have likely gotten similar results on the handhelds with my sub-$100 Grace SDAC which measures pretty well for what it costs

I am reserving the rest of first page to allow us a space to create a wiki, which will highlight the most useful and relevant software packages, and some test methodology that we can use. My hopes are that this is an iterative document and we can contribute to advance each others' understanding and capabilities

Reserved: Rightmark Audio Analyzer

Reserved: Room EQ Wizard (REW)

Reserved: Discussion of noise relative to input level and why it is important in our context

Reserved: noise floor measurement


[0] Message Index

[#] Next page

Go to full version