On April 2, 2018 2:29:36 PM HST, Brent Busby <brent(a)keycorner.org> wrote: I think my hearing dies at about 12-15K and has for decades now. Bats and dolphins ignore me entirely. -- David W. Jones gnome(a)hawaii.rr.com authenticity, honesty, community http://dancingtreefrog.com Sent from my Android device with F/LOSS K-9 Mail. _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org https://lists.linuxaudio.org/listinfo/linux-audio-user

On Tue, 3 Apr 2018 00:03:08 -0400, Chris Chronopoulos wrote: There are far better ways to get 16 channels at 48 KHz via USB ;). Consider to get one of those 8 channel USB devices + additional 8 channels by their ADAT and BNC sync features. Btw. "With the Monolith USB DAC, you can turn any computer or laptop into an audiophile playback device!" This could be done by 2 channels @ 48 KHz and for doing this not only the DAC is important, also very important is the analog headphone amp. "Improve your playback sound quality from YouTube™", this says it all. YouTube uploads are known for their high end audio quality ;). I don't know how good YouTube audio could be, but the majority of uploads sounds like two supermarket microphones were in front of the kitchen radio's speakers, to record audio. -- pacman -Q linux{,-rt{-securityink,-cornflower,,-pussytoes}}|cut -d\ -f2 4.15.15-1 4.14.29_rt25-1 4.14.28_rt23-1 4.14.24_rt19-1 4.14.8_rt9-2 _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org https://lists.linuxaudio.org/listinfo/linux-audio-user

On Tue, 03 Apr 2018 06:54:25 +0000, Moshe Werner wrote: http://www.lavryengineering.com/pdfs/lavry-white-paper-the_optimal_sample_r… I've got a few concerns regarding this paper, however, I like the following claim a lot "It is always unwise and potentially harmful to include signals that are not needed". Indeed, the noise flour of my RME card does increase rapidly, if I switch from 48 to 192 KHz, without gaining better quality for the hearable audio signal. "However, even after adding an extra 10 KHz, we are talking about bandwidth no higher than 40 KHz. That is why 88.2 or 96 KHz are preferred rates for audio quality." I stopped reading here. We had this already a few times on this list. What we need is 48 KHz, but some audio interfaces sound better at 96 KHz, regarding the used filters. _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org https://lists.linuxaudio.org/listinfo/linux-audio-user

In not everyone read it already: http://www.lavryengineering.com/pdfs/lavry-white-paper-the_optimal_sample_r…

controversial claims by audiophiles notwithstanding... i'm somewhat interested in high-samplerate DAC's for the possibility of time domain multiplexing of multiple channels. 2 channels @ 384 kHz = 16 channels @ 48 kHz if you had a driver + external hardware to do the muxing/demuxing. of course you'd need some sort of "word select" embedded in the signal - perhaps you could do this with GPIO, or maybe use part of the 32 bits samples as metadata. wouldn't be easy, but i wonder if it's possible. anyway the first step would be getting a DAC like this to work with the host at these high sample rates, so if anyone has experience with that, i'd be curious to know. On Mon, Apr 2, 2018 at 10:10 PM, David W. Jones <gnome(a)hawaii.rr.com> wrote: _______________________________________________ Linux-audio-user mailing list Linux-audio-user(a)lists.linuxaudio.org https://lists.linuxaudio.org/listinfo/linux-audio-user

On Tue, 3 Apr 2018 14:34:54 -0400, Chris Chronopoulos wrote: How about the Focusrite Scarlett 18i20? I own one, it provides 18 input and 20 output channels for less than USD 1000.00, including an additional required ADAT deviceand and S/PDIF device. It works OTTB with Linux.

ralf.mardorf at alice-dsl.net (Ralf Mardorf) writes: If you got 1€ of satisfaction out of every mention of your Hammerfall DSP, you probably got back its price threefold. If not its weight in gold. RME does not develop their own audio ADCs, they use similar converters as everybody else. Their analog circuitry around it is good and durable, and they are pretty good with continued driver support and also delivered good info in more active ALSA development times. In the mean time, converter development has moved on. The audio quality of quite a few setups can definitely hold a candle to that of older RME gear by now. And for now. In 10 years, gear might have died of hardware failure and/or bit rot while the RME keeps running as long as you can connect it to your computer, assuming you replaced the Asian power brick which has likely given up its ghost long ago but can be replaced by a wagonload of different choices. There have been listening tests for different ADC converters ranging from very affordable to very expensive. It's already rather hard to distinguish boutique preamps from run-of-the-mill preamps, and it's much harder to actually tell apart different converters in double-blind listening tests. It's really more the reliability and robustness and product value preservation that sets apart products like those from RME. The actual conversion quality, due to relying on the same kind of parts, is not as much an issue these days.

Their analog circuitry around it is good and

Ede Wolf &lt;listac(a)nebelschwaden.de&gt; writes: Mackie Onyx Firewire Card manual: Onyx FireWire Card Specifications Mic In to FireWire Card to DAW, routed back to FireWire Card to Control Room Output; all signal path gain stages set to unity Frequency Response +0, –3 dB, 10 Hz to 42 kHz @ 96 kHz sample rate +0, –1 dB, 10 Hz to 20 kHz @ 44.1 kHz sample rate Distortion (THD+N) < 0.025%, 10 Hz to 40 kHz, +4 dBu input @ 96 kHz sample rate < 0.015%, 20 Hz to 20 kHz, +4 dBu input @ 96 kHz sample rate < 0.010%, 20 Hz to 20 kHz, +4 dBu input @ 44.1 kHz sample rate Dynamic Range From the Mackie Onyx 1620 manual: Onyx 1620 Specifications Frequency Response Mic Input to Main Output (Gain @ Unity) +0, –1 dB, 20 Hz to 50 kHz +0, –3 dB, <10 Hz to >100 kHz Distortion (THD & IMD) Mic Input to Main Output (@ +4 dBu) THD: < 0.004%, 20 Hz to 20 kHz SMPTE IMD: < 0.005% (7 kHz/60 Hz, 4:1) Dynamic Range Noise Floor (20 Hz to 20 kHz Bandwidth, 150Ω source impedance) Equivalent Input Noise (E.I.N.) –129.5 dBu @ +60 dB gain Mic Output Noise: Recording Output: –100 dBu (minimum gain) Residual Output Noise: Main Out: –102 dBu (Channel and Main Mix levels off) Main Out: –88 dBu (Channel and Main Mix levels at unity) Common Mode Rejection Ratio (CMRR) Mic In: >70 dB @ 1 kHz, Gain @ maximum Crosstalk Adjacent Inputs: < –105 dB @ 1 kHz Mic Input to Main Output: < –87 dB @ 1 kHz Input Gain Control Range Mic In: 0 dB to +60 dB Line In: –20 dB to + 40 dB, mono channels Line In: –20 dB to +20 dB, stereo channels [...] Output Impedance Main: 150 Ω balanced, XLR outputs; 300 Ω TRS outputs Recording Outputs: 300 Ω Control Room: 300 Ω Alt 3/4: 300 Ω Aux Sends: 300 Ω Tape Output: 2.2 kΩ Phones Output: 25 Ω This is, of course, not primarily a "soundcard" but rather a semi-professional analog mixer with Firewire Interface. Compared to the Hammerfall DSP Multiface: AD • Auflösung AD: 24 Bit • Rauschabstand (SNR): 101 dB RMS unbewertet, 106 dBA • THD: < -100 dB, < 0,001 % • THD+N: < -92 dB, < 0,002 % • Übersprechdämpfung: > 100 dB • Frequenzgang AD @ 44,1 kHz, -0,5 dB: 5 Hz - 20,7 kHz • Frequenzgang AD @ 96 kHz, -0,5 dB: 5 Hz - 32 kHz • Eingang: 6,3 mm Stereoklinke, servosymmetrisch • Eingangsimpedanz: > 5 kOhm • Eingangsempfindlichkeit per Jumper: Lo Gain, +4 dBu, -10 dBV • Eingangspegel für 0 dBFS @ Lo Gain: +19 dBu • Eingangspegel für 0 dBFS @ +4 dBu: +13 dBu • Eingangspegel für 0 dBFS @ -10 dBV: +2 dBV [...] Stereo Monitor Ausgang • Ausgangspegel: +8 dBu @ 0 dBFS • Dynamik: 108 dB (RMS unbewertet, unmuted), 112 dBA • THD+N: -100 dB / 0,001 % • Übersprechdämpfung: > 110 dB • Frequenzgang DA, -0,1 dB: 20 Hz - 20,8 kHz (sf 44,1 kHz) • Frequenzgang DA, -0,5 dB: 10 Hz - 44 kHz (sf 96 kHz) • Ausgangsimpedanz: 75 Ohm there is admittedly little doubt that the specs here are quite narrower. But microphones will be the limiting factor either way. It's however irritating that something labelled as "headphone output" has an impedance of 75 Ohm and no pot. The Multiface II has a pot here: possibly the impedance has also been lowered. Not relevant in the recording/production chain, of course.

i'm aware of ADAT, but the motivator here is cost/scalability. most ADAT interfaces are optimized for input with preamps, etc. plus it's limited to 8 channels per link, so there's recurring overhead if you want to scale higher. imagine trying to do high channel count (e.g. spatial audio) on a budget. i'm not aware of any configuration that will get you >16 channels analog for <$1k.

On Tue, 3 Apr 2018 12:17:17 -0700 (PDT), Len Ovens wrote: It's idotic! It does cost much more, without the guarantee that it does work at all.

unfortunately the original question has still not been addressed: can this (or other similar) interfaces be run on ALSA drivers at 384 kHz?

i'm aware of ADAT, but the motivator here is cost/scalability. most ADAT interfaces are optimized for input with preamps, etc. plus it's limited to 8 channels per link, so there's recurring overhead if you want to scale higher. imagine trying to do high channel count (e.g. spatial audio) on a budget. i'm not aware of any configuration that will get you >16 channels analog for <$1k.

At my age, I fondly remember the days when I could hear up to 384kHz... Now bats and dolphins think I'm pretty ordinary.