Digital Music File questions

Starting point:

Music files are on a hard disk in a computer, USB drive or NAS drive
Music files were created with a quality software ( in my case Exact Audio Copy @ 16/44 )
Using Audirvana Player software ( others could be inserted here also ) on PC or Mac

Step 1 Audirvana sees the music file and reads the digital file
Step 2 Audirvana software converts/upscales the digital file to DSD 256
Step 3 Audirvana sees my USB DAC and sends the digital stream to the DAC
Step 4 The DAC senses the digital DSD stream and converts that digital stream into an analog signal
Step 5 the DAC then sends the analog signal out via XLR or RCA depending on DAC capability
Step 6 The XLR or RCA is connected to a PreAmp or directly to and amp which amplifies the analog signal to the speakers and we here music.

I think that is the process Please correct if wrong


1 the conversion of the 16/44 digital stream is processor intensive ( my belief ) therefore where the process occurs is important and why having that process done on a computer has advantages ??

2 There is much talked about a computers being ( Electronically Noisy ) and therefore adds noise to the digital stream during processing which is then sent to the DAC , while I am sure that is measurable is it significant. Also just because you have that process done in a different ( all in one streaming box )why would the process be less noisy ??


Yes, we have found that the more work the streaming computer has to do in real time the worse the sound quality is unless you’ve somehow managed to completely isolate the CPU from the output stage (almost impossible in a computer).

What we’ve done in the upcoming Octave server hardware is just that—we’v e optically isolated the busy CPU from the stage that is delivering the digital audio information. This allows a jitter free delivery of music regardless of CPU load, something
pretty much impossible with a NUC or any other type of all in. One computer.

Why would you upsample before sending to the DAC?

My old-fashioned audiophile rules of thumb include not touching anything native. Let it be. No “improvement” over the source is possible. If a DAC like the PS Audio Junior/Senior does this as part of their approach, then let them do it.

What is wrong with my old-timer approach?

Bruce in Philly

Listen both ways and upsample where ever you like the resultant sound the most, but arguments based on saving the DAC time, resources, processing load, etc. are misguided in the case of the DS and the DS Jr. The FPGA does everything every clock, it can’t skip a step, etc. The control is done by ignoring the answers you don’t need each clock, but everything is calculated whether it’s needed or not. E.g. the PCM upsampler is running and upsampling all the time, if DSD is being played the output of the PCM upsampler is ignored.

1 Like

My goal was to find out what it takes to listen to my music. I was actually surprised with all the different steps required, as today systems can make it look simple and easy.

Once you understand what is required I think you can better evaluate the available systems

Bottom line it sounds like power supply separation is a key element but not sure what that means. Ie is it just a physical separation say two different boxes or is there more to it