Difference in digital source volume, why?


I know that you can digitally control volume if you are using Spotify Connect or the like whereby the actual digital signal is altered in some way. But when using a fixed output mode, shouldnt both digital sources give the same level of volume?

Example: I have a Samsung TV Optical out using Toslink and a Bluesound Node2i using Coaxial to feed my DAC. Both are fixed outputs yet the Node is way louder than the TV. Is the difference due to Toslink vs Coaxial, or is there some other explanation?


Not sure why that is a surprise to you. Why would the volume of the output from the TV be the same as your Bluesound? Does CD have the same output level as Vinyl? It is normal for different sources to have different output levels.


I guess I thought that all digital streams in their purest form would have the same volume level. I’m not an expert in computer science or engineering or anything, but I thought bits were bits (jitter discussion notwithstanding).



1 Like

I think you are thinking of the maximum/minimum output levels of a particular format which is not necessarily the same thing. The actual volume you get would depend on other factors such as the recording volume for whatever show your are watching, the amount of compression applied, etc.


Even if you compare apples to apples though, say we run Spotify connect on the Roku vs. on the Node, same difference in volume. In fact I would say regardless of what comes out of the TV, the volume difference to the Node is about the same.


I’m not sure why that would be though. Perhaps someone more knowledgeable like Ted Smith can explain when he gets back from his trip.


Perhaps, someone on another forum suggested that the TV might be reducing the overall output volume to work better with consumer electronics. The only way I can test this is to use an HDMI audio extractor but Id like to get some consensus on this before I go rewiring everything.

1 Like

It’s possible. I’ve also noticed that different channels have vastly different volumes too. For example over here BBC channels seem to be quieter than some of the lower quality commercial channels. I always presumed it was due to greater compression on the latter but I don’t know for sure. Someone will likely know.


At the DAC all digital inputs are identical: any differences in volume come from different processing upstream. If you have big perfect paths then the volumes will all be identical. Even with no effects in the paths, sometimes it’s hard to find all of the volume controls. With video players, there is often compression in the path, there may be speaker configuration settings (e.g. crossovers to a sub, speaker size adjustment, speaker distance adjustment, etc.) Perhaps folding of 5.1 (or whatever) to stereo… With broadcasters in the path there may be a lot of processing going on, including louder (or more compressed) volume for commercials which make me turn things down on the average. Movies often have more dynamic range than TV shows so they have to be quieter on average. Satellite receivers, cable boxes, etc. often have dynamic range adjustment settings… The list goes on forever. Even with simple stereo from streaming services some have their own volumes, but some share volume control with the OS, some allow a volume “above 1” (VLC, for example) so unless you can run some sort of bit perfect test you can’t tell if you have all of the volumes set to unity.


Thanks! I knew you would have the answer. This is pretty much what I had presumed especially in regards to TV. It is still infuriating how much louder commercials are than regular TV. Ditto with ads on the radio (which is crazy since radio tends to already have so much dynamic range compression).


Is there any way to do a 44.1k/16 bit perfect test with the DSDAC? I know there is no file with CD res in the PS Audio downloads, but if the right file were created, does the DSDac support it?


The current implementation in the DS needs 24 bits, the most common kinds of failures are small changes caused by volume controls, sample rate conversion, etc. With 44.1k/16 bits those kind of changes will show up as a reading of 44.1/18, 44.1/20, … 44.1/24 instead of 44.1k16. In general if the 24/96 bit perfect test works and playing a 16 bit file shows up as 16 bits everything is fine.


Thanks, Ted.

I find the PS Audio 24 bit files extremely valuable in assuring bit perfect performance before DAC inputs are compared, different cables are compared on the same input, and in evaluating the effects of noise reduction efforts. It would have been great to have a 16 bit file to use for transport evaluation… but thank you very much for including the 24 bit feature!