11/26/2023 0 Comments Sonos my library best rip softwareThe observable differences between these two graphs, such as the fact that the noise floor is flat in fig.7 but slopes up with frequency in fig.2, are due to the different analyzers used: a swept analog 1/3-octave bandpass filter and a logarithmic frequency scale in fig.2, FFT analysis and a linear frequency scale in fig.7.įig.6 Sonos ZP80, spectrum of 1kHz sinewave at 0dBFS into 5k ohms, DC–10kHz, 16-bit data (linear frequency scale).įig.7 Sonos ZP80, spectrum of 1kHz sinewave at –90dBFS into 5k ohms, DC–10kHz, 16-bit data (linear frequency scale). This graph, by the way, was taken under the same measurement conditions as fig.2. Dropping the signal level to –90dBFS resulted in all the harmonic components lying at or below the ZP80 DAC's noise floor (fig.7). Reducing the load to 600 ohms increased the level of the third harmonic to –80dB (0.01%), but did not affect the levels of the higher harmonics (not shown). The ZP80's analog output offered very low levels of harmonic distortion, though at maximum level (fig.6) some low-level, higher-order harmonics could be seen in addition to the third harmonic at –84dBFS (0.006%). The ZP80's reproduction of an undithered 1kHz sinewave at exactly –90.31dBFS, which consists of just three voltage levels, showed excellent waveform symmetry (fig.5).įig.4 Sonos ZP80, left-channel departure from linearity, 16-bit data (2dB/vertical div.).įig.5 Sonos ZP80, waveform of undithered 1kHz sinewave at –90.31dBFS, 16-bit CD data. Linearity error was low to below –100dBFS (fig.4), with almost all the apparent error actually resulting from the dither used to encode the 500Hz test signal. Extending the measurement bandwidth to 200kHz and repeating the spectral analysis, playing a file comprising a –1LSB DC offset revealed a rising level of ultrasonic noise well above the audioband, due to the noiseshaping used in the DAC to achieve 16-bit performance (fig.3).įig.2 Sonos ZP80, 1/3-octave spectrum with noise and spuriae of dithered 1kHz tone at –90dBFS, 16-bit data (right channel dashed).įig.3 Sonos ZP80, 1/3-octave spectrum with noise and spuriae of –1LSB, 16-bit data (right channel dashed). Spectral analysis of a dithered 1kHz tone at –90dBFS was free from AC supply artifacts, though the noise level was a little higher than the best 16-bit D/As (fig.2). However, if you do rip a pre-emphasized CD (they're rare), you need to equalize the audio files with a program like that available from (My thanks to reader and denizen of our website forums Jeff Wong for alerting me to the existence of this program.) I understand that iTunes also compensates for pre-emphasis, if present, during ripping.Ĭhannel separation (not shown) was superb, at better than 110dB below 3kHz, and still 96dB at 20kHz. I didn't test the Sonos with pre-emphasized data, as this will be irrelevant. (Right channel dashed, 0.5dB/vertical div.) At the other end of the spectrum the output is down 0.5dB at 20kHz, which will not be audible.įig.1 Sonos ZP80, frequency response at –12dBFS into 100k ohms (top below 200Hz) and 600 ohms (bottom below 200Hz). Into 100k ohms (fig.1 top traces), the LF response was flat to 30Hz or so. This rise will be inconsequential with typical preamplifiers having an input impedance of 10k ohms or more, but with the low, "torture test" load of 600 ohms it does result in the response starting to roll off at low-bass frequencies, reaching –3dB at 14Hz (fig.1, bottom pair of traces). The output preserved absolute polarity- ie, it was noninverting-and was sourced from a usefully low impedance of around 93 ohms across most the audioband, this rising to 224 ohms at 20Hz. Looking first at the ZP80's analog output set to Fixed, the maximum level at 1kHz was 2.12V, 0.5dB higher than the CD standard's 2V RMS. There were no differences, showing that Apple Lossless, the Sonosnet encoding and encryption, and the ZP80's data processing are all transparent to the original data. These were then translated to AIF files, which were compared, bit for bit, with the original AIFs that had been used to generate the ALC versions. I recorded the ZP80's digital output on my laptop, using a Metric Halo FireWire interface to turn the S/PDIF data into SD2 files. (I found no differences doing spot-check comparisons over the wireless link.) I did check the integrity of the data transmitted over the wireless link, selecting a variety of Apple Lossless files in my iTunes library and played these on the ZP80 in my listening room. All measurements of the Sonos Zone Players were made using uncompressed 16-bit AIF files stored on my Mac mini's internal drive and accessed via a hardwired Ethernet connection.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |