Search results

Hm, please correct me if I am wrong. The analog mux is part of a voltage divider between outputs of U3A and U7A where U7A tries to keep its inverting input at virtual GND (i.e. equal to its noninverting input). To get 0V at the inv. input, a current must flow through the voltage divider, i.e...

@Frunse: Do you have any data to support those circuits would be better than the digital VC in your DAC chips? CD4051 analog switch has THD of 0.15% - 0.3% - check the datasheet. No noise figure, but I kind of do not think your circuit with so many resistors and opamps in the signal path will...

Very nice. Just a note - since the newer kernel has async UAC2 gadget, the CDSP config should enable rate adjust. CDSP will hit (sooner or later) xruns without that. If the gadget were adaptive (as in older android kernels), it would require rate adjust and async resampler.

Does your RPi airplay receiver or your airplay source have volume control? Maybe your RPi3 analog output is too weak for your amp. Easy to check e.g. with speaker-test -t sine - for details https://linux.die.net/man/1/speaker-test

You can try hw devices - hw:CARD=Headphones, DEV=0 for playback and hw:CARD=Loopback,DEV=0 or 1 for capture, depending what device you use on the other side https://henquist.github.io/0.6.2/backend_alsa.html#alsa-loopback

Fix that russian if you want my help, please.

I wonder what the default:CARD=XXX actually does. IMO it's mixing dsnoop on capture and dmix on playback + plug. IMO it should be changed to hw:CARD=XXX, or optionally plughw:CARD=XXX. Rate adjust - Loopback runs at RPi clock. The "headphones" PWM device integrated in the GPU part is most...

So do I. RK3308 is designed specifically for audio and its feature set is quite unique. But Pi S does not expose all the handy capabilities of the SoC. The low consumption which allows to power the SoC from standard USB port even when running at full 4 cores @1.3GHz is very handy too. I am...

@godmax: Making any non-RPi ARM board work properly with any reasonably new kernel is ALWAYS a major fight. Most of SoC manufacturers care about android kernel only, android is their market. Very few customers actually request/run non-android linux on non-RPi SoCs. If you are not deeper into...

That's correct behavior - the gadget alsa device provides data only when the USB host plays some audio. If not, the device is stalled. CDSP handles the stall gracefully. Or you can start CDSP with projects like https://github.com/pavhofman/gaudio_ctl or...

Robert, I am not comparing digital vs. analog volume control, but digital in the player/DSP vs. digital in the DAC. Volume control in vast majority of DACs is digital too. If your tweeters hiss at zero volume, simply your analog chain (including the DAC) is electrically too noisy. Of course...

I am afraid the latter is the case - if you arduously calculate an operation at gazilion precision and then throw away most of the precision (explicit truncation to 32 bits samples, implicit truncation to 22bit noise/signal resolution), the resultant number will be identical to making the...

@Frunse: I do not know your whole plan. But have you checked your DACs really have analog volume control? In the end it has the same spare room as 24bit volume control because first float64 -> int32 before the driver and then int32 -> int22 max by physical resolution of the DA conversion.

Since HDMI is the clock master, I am not sure how the HDMI receiver would handle the I2S slave mode. IIUC that chip has also I2S receiver for the HDMI output audio, perhaps that one is slave. Also the block diagram of EP92A3E (I could not find any datasheet) suggests the I2S transmitter is...

IMO the HDMI I2S is 3.3V (most likely), RPi is 3.3V, maybe 5V tolerant, I do not know. IMO the 3.3V arduino is the correct choice.

@Frunse : congrats to your project advancement. Just a note - is it necessary to use the 3.3-> 5V the level shifters? Those small Arduino boards are available at 3.3V too, I used them a few times. Typically they run only at 8MHz but can be switched to 16MHz if needed, AFAIR, but probably not...

IMO the problem with timing is that USB mics have no second analog input to which the loopback analog signal from the DAC could be fed. As a result REW cannot measure the acoustic delay - the difference between electrical (loopback) and acoustic (microphone) signals.

@Frunse: Here is an almost exact project what you would need https://github.com/raptorlightning/I2S-Hat - measuring LRCLK samplerate with arduino, communicating the momentary samplerate to RPi over GPIOs, python script which runs CDSP with correct samplerate config triggered by GPIO change, all...

Of course my previous post should have been "Do as you want", not "as I want" :) Happens to me all the time, our house IS full of stuff not used in the end :)

@Frunse: Do as you want. That FIFO and Dual XO has zero effect (its precise output will be async-resampled two times in your chain), just major cost and issues, sorry to say that. You mention TV - did you check the FIFO latency? Adjustable 0.1, 0.2, 0.4 and 0.8s. 100ms in the FIFO + the RPi5...