If you know how to do it and have test instruments, its not that hard. It just not the same thing as purely analog audio, where there are books by guys like, for example, Cordell and or Self.
Not true and if you haven't compared IanCanada's fifo ii to anything, you aren't qualified to make a judgment.
I use three of his boards and have no complaints and I don't pass judgment on others that I have not tried.
Yes I know the importance of characteristic impedance to meet the signal integrity and avoid reflections & loses.
Sure I use DSD Converter, its between I2SoverUSB, and Andrea's isolated FIFO buffer. From there it goes to the DSD-only dac. However, I mostly use offline file conversion.
I have a Twisted Pear Chronus reclocker isn't a Fifo but can help?
If yes I will configure Amanero as slave clock.
If yes I will configure Amanero as slave clock.
Andrea offered me his solution:
- DRIXO oscillators at 22.5792 and 24.576 MHz I guess two units because it's only one output per board + two crystals.....$$$$$$
https://www.thewellaudio.com/twtmc-drixo/
- TWSAFB-LT FIFO buffer
https://www.thewellaudio.com/twsafb-lt/
- TWSDAC-DSD discrete DSD DAC (not yet published)
https://www.thewellaudio.com/forum/topic/the-well-segmented-dac-dsd-discrete-dac/
But this is a lot of money and I'm not sure to invest without guarantee of the best SQ for noDac + Simple DSD Modulator
- DRIXO oscillators at 22.5792 and 24.576 MHz I guess two units because it's only one output per board + two crystals.....$$$$$$
https://www.thewellaudio.com/twtmc-drixo/
- TWSAFB-LT FIFO buffer
https://www.thewellaudio.com/twsafb-lt/
- TWSDAC-DSD discrete DSD DAC (not yet published)
https://www.thewellaudio.com/forum/topic/the-well-segmented-dac-dsd-discrete-dac/
But this is a lot of money and I'm not sure to invest without guarantee of the best SQ for noDac + Simple DSD Modulator
I once paid $2,200 for a Benchmark DAC 2. Topping D90 sounded better. My AK4499 dac sounded better than D90 here on this system. Andrea's DSD dac sounds better than my AK4499 dac did. Now, I don't know how Andrea's dac compares to Chord, Holo, Mola, DCS, etc. IMHO though, I'm keeping it as my main dac. It can be amazing how much information encoded in a recording can go missing, and yet no obvious audible clue anything at all is wrong. It sounds like that's all there is supposed to be. It can measure very well, too. The illusion that you are hearing everything encoded in the recording persists until you hear the recording on a better system.
The foregoing is entirely IMHO and IME. It is an expression of opinion and of observation.
The foregoing is entirely IMHO and IME. It is an expression of opinion and of observation.
Last edited:
If I were not happy with it, then probably a fast scope would show the problem and suggest a fix.
Sure, but for the moment come back to on the fly player converter for upsample PCM to DSD files.
What you could do to verify the problem is in the DSD converter is to swap the converter outputs into your LPFs. Does the problem move with the filter or stay with the converter? if it stays with the converter then maybe that's the problem. Did you say the problem happened in passthrough mode or in conversion mode (I'm following more than one thread here 🙂 )? If in passthrough mode then swapping DATA and LRCK going into the converter might be another test to see if the problem moves from one channel to the other.
Thanks for support, good idea but for sure isn't the LPF, now I'm using only Amanero with the LPF with native DSD or upsampled by Daphile player and is silent like a tomb...
It kinda sounds like you are suggesting there is some error in the noise shaping of one channel of the DSD Converter? If so, which channel, left or right?
Do you have an RF spectrum analyzer to verify there is some difference, or do you know somebody locally that does have one of those?
IOW how are we going to verify if there is a problem, and if there is, what the problem might be?
Do you have an RF spectrum analyzer to verify there is some difference, or do you know somebody locally that does have one of those?
IOW how are we going to verify if there is a problem, and if there is, what the problem might be?