i’m now quite confused. does “recording from TAPE” mean that:
- you played from some other device, into the Fates inputs,
- enabled monitoring through norns mixer (stereo mode, one hopes),
- used TAPE record function,
- and that
Recording_from_tape_volume_OK.wav is the resulting file?
and just to be clear, ORIGINAL_SAMPLE_VOLUME_LOW.wav is what?
the actual original sample?- capture of playback from TAPE into some other device?
- something else?
so, in fact, the TAPE playback mechanism is not used at all here? or am i missing something?
(on another reading, i think you are actually describing the second case, but i could be mistaken.)
in Recording_from_tape..., the waveform is clearly being clipped at some point in the chain; not at 0dB as it would be if the clipping took place at the end of TAPE capture, but at -1.6 dB :
[Parsed_volumedetect_0 @ 0x55fbd6fd30c0] n_samples: 512512
[Parsed_volumedetect_0 @ 0x55fbd6fd30c0] mean_volume: -10.3 dB
[Parsed_volumedetect_0 @ 0x55fbd6fd30c0] max_volume: -1.3 dB
[Parsed_volumedetect_0 @ 0x55fbd6fd30c0] histogram_1db: 33277
whereas ORIGINAL_SAMPLE... is normalized to 0dB, but overall quieter and no sign of clipping:
[Parsed_volumedetect_0 @ 0x55f20a90f0c0] n_samples: 360000
[Parsed_volumedetect_0 @ 0x55f20a90f0c0] mean_volume: -17.5 dB
[Parsed_volumedetect_0 @ 0x55f20a90f0c0] max_volume: 0.0 dB
[Parsed_volumedetect_0 @ 0x55f20a90f0c0] histogram_0db: 225
[Parsed_volumedetect_0 @ 0x55f20a90f0c0] histogram_1db: 358
assuming my description of the capture chain for the test recording is correct, i’d suspect the clipping to be in the ADC driver, and the way to address it might be to use alsamixer to change the soundcard capture levels (you can search this thread for more info if you’re unfamiliar with ALSA configuration tools.) that’s assuming you can’t, or don’t wish to change the level of the analog signal coming in to Fates.
also, this is obviously not the same file as the last one. that had a lot of sub-bass components that would clearly demonstrate phase cancellation (and it sounded kind of like that - sub-bass being quiet and distorted, with noisy transients poking through - but it’s hard for me to say for sure by listening to soundcloud.)
both experiments make me wonder if you are perhaps doing something with a balanced output from some device, and something odd with the wiring between that and the (unbalanced) Fates inputs. (in the first case: something like connecting negative / positive to L/R, and then recording with mono monitor mode. in the second case… i’m not sure. isolating parts of the audio that are “bassy”, it seems that low fequencies are more prominent in the “tape” audiofile, but the clipping makes it hard to be scientific about this.)
but i really don’t know. it seems like you are expecting that a given signal coming in to Fates, captured by Tape, should produce a file identical to the original. but this is of course dependent on the signal path between the playback device and the Fates ADC. you should be able to find some setting where a given voltage coming into Fates, captured to TAPE, and played back from TAPE, should produce the same voltage (given that it is below the electrical limit of 0dbV or so). but “calibrating” this will probably require a little tweaking of the audio capture levels; i’m not sure if @okyeron explicitly designed ADC/DAC circuits to be 1:1 “out of the box.”
if you still believe there is a software problem with TAPE (which i doubt), we need a clearer reproduction case. if the problem is elsewhere i’m not going to be able to help any further, but maybe others on this thread will. if you do want to follow up with me about a possible software issue, let’s do it by DM so that we don’t clog this thread any further.
