I have a plan... or so ;)

[Supra2k0]

Hi Folks,

I like to hear some opinions on my plan.
I recently thrown away my NAD C658. A Topping D70 Pro Octo took its place. Now I need Dirac Live or REW back So, to avoid up/downsampling of audio in the miniDSP Flex or somehow, I like to complicate things...

N130 -> USB-Out -> Topping D10B or E50 -> TRS to TRS -> miniDSP Flex balanced -> USB-Out -> Topping D70 Pro Octo -> XLR-Y-Cable to Apollon Amp -> BiAmp to Speakers

As stated in my signature N130, D70 Pro Octo and Apollon Amp with Speakers are in place. My ideas are not to use the "bad" DAC in N130, avoid ASRC in miniDSP and not using unbalanced connections.

Description of the miniDSP Flex stated that USB is useable also as output but maybe just on a PC - is someone using this output on a DAC?


BR
Ricardo

 

[staticV3]

Several issues:

-to apply Dirac Live, the miniDSP Flex must first resample all analog and digital inputs to 48kHz.
You're not gaining anything by going Streamer->DAC->Flex->DAC.
In reality, the extra D->A and A->D conversions in that chain will degrade the signal more than if you stuck to a fully digital signal chain until the final DAC.

-Unlike the Node N130, the miniDSP Flex cannot output audio to external DACs via USB. It can only output audio via USB to USB Host devices (PCs etc).

If you want to add Dirac Live to your existing components, then the best way would be:
N130->USB/Toslink/Coax->Flex Digital->Toslink/Coax->D70 Pro Octo

And for a completely fresh setup, just a WiiM Mini->Flex Balanced would achieve the exact same audio quality for about half the price.

 

[Supra2k0]

Thanks @staticV3 for the reply

I think I read too much about (might) worst Upsampling in miniDSP...
Thought it is better to let the upsampling/downsampling happen on an analog signal cause there is no sampling rate mismatch (especially 44.1khz to 48khz or 96khz) (and D10b or E50 measured really good on DAC-Performance and Flex RCA/balanced also), or am I misleaded with the mismatch issues?

But yeah, one of my first ideas was just with a Flex digital in the chain.

WiiM Mini or Pro looks good but I am missing HDMI eARC for TV

 

[staticV3]

Hmm I googled "miniDSP ASRC issues", but can only find discussions on the theoretical.

Could you maybe post some links to the things that you read? Or simply elaborate on this?

 

[DVDdoug]

Thought it is better to let the upsampling/downsampling happen on an analog signal cause there is no sampling rate mismatch

No... Trust the "math" (DSP).

Analog is almost always worse than digital, although there probably won't be an audible difference in a proper, blind, level-matched, ABX test.

 

[voodooless]

N130 -> USB-Out -> Topping D10B or E50 -> TRS to TRS -> miniDSP Flex balanced -> USB-Out -> Topping D70 Pro Octo -> XLR-Y-Cable to Apollon Amp -> BiAmp to Speakers

This is impossible. The Flex cannot drive a USB DAC.

Just get the Digital Flex. The ASRC is fine.

 

[Supra2k0]

I just better had to wrote "miniSHARC" instead of "miniDSP"... I read in some german forums and diy-audio.com but somehow older threads (2018) - but I think this was just an intiator to think over it, althought it is just a theoretical discussion. I googled too much these days - my history is a mess ^^
There is still "just" a miniSHARC with some ADC/DAC Chips in current miniDSP devices or am I wrong?

This and some threads after that were in my readings:
https://www.diyaudio.com/community/threads/mini-dsp-quality.302256/post-5443644

@voodooless I could use also coax there Do you have a Flex Digital in use in a similar constellation?

 

[voodooless]

Do you have a Flex Digital in use in a similar constellation?

No, I would just skip the extra DAC nonsense and use one of the analog Flexes.

This and some threads after that were in my readings:
https://www.diyaudio.com/community/threads/mini-dsp-quality.302256/post-5443644

What of it?

In my opinion ASCR (asynchronous resampling) is very bad for sound quality because data samples are created out of thin air.

Hitchens's razor: "what can be asserted without evidence can also be dismissed without evidence."

 

[Supra2k0]

OK - thanks.

Think I will test the digital way with Flex digital first.
And maybe I test than (for fun) with analog output from N130 to Flex Digital in comparison - I can let you know than from my perspective.

There was some more but i can´t find it at the moment.
From the thread mentioned:

The question is how the signal changes.
If you perform convolution (ie digital signal processing) on the original 16 bit data stream, 32bit floating point calculations will deal nicely with rounding errors. In the end you will have to round the numbers (again) to convert back to 16 bits, but you will get "one degree of error".
Going from 44.1Khz to 48Khz means guessing the second sample which is at a 0.91875 distance between the first two (44.1Khz) samples. (Good luck with that being correctly done). So in the end you will have "two degrees of error".
We could argue that conversion algorithms are very good at it but my idea is that it is best to remove what is not necessary.

Hitchens's razor: "what can be asserted without evidence can also be dismissed without evidence."

 

[voodooless]

Going from 44.1Khz to 48Khz means guessing the second sample which is at a 0.91875 distance between the first two (44.1Khz) samples. (Good luck with that being correctly done). So in the end you will have "two degrees of error".

This is nonsense. There is no guessing involved. Clearly, the person who wrote this has no understanding of the sampling theorem.

The ASRCs in the Sharc DSP are equivalent to this one:

https://www.analog.com/media/en/technical-documentation/data-sheets/ad1896.pdf

THD is -133 dB, SNR 142 dB. I think it's safe to say that it is utterly transparent.