Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[ajv]

I am a 66-year-old man who loves great audio but I don't have the technical expertise to answer my question. I have about 2200 CDs ripped on an external hard drive in FLAC 24/48. Using a USB cable, I send the audio to a Schiit Bifrost DAC-> Denon AVR->  an Emotiva 3 channel amp ->two PSB Imagine T2 speakers.

Would I improve the quality of my audio playback if I go to Output Format->Sample Rate->Output->and select All 192,000 Hz? That is the upper limit of the DAC.


Since my trial period ends in seven days, I would greatly appreciate any responses in the next few days.

Thank you in advance for your ideas.

ajv

[JimH]

Welcome to the forum.  Some people might encourage you to do that, but there is effectively no difference in the sound.  But there is no harm in trying it.

[toomuch40]

You sound like me except I'm only 58 years old. Probably no difference but let your ears decide. 

[stewart_pk]

If I'm not mistaken, Audyssey (which is available on most if not all Denon AVR's) works at a maximum of 48Khz so there's no point going beyond it otherwise the signal will be down-sampled.

[ajv]

Welcome to the forum.  Some people might encourage you to do that, but there is effectively no difference in the sound.  But there is no harm in trying it.

I have read some discussions on the forum about the limitations of human hearing. Is that why you say "there is effectively no difference in the sound"? If that's the case, are there other situations where those settings could be used with a noticeable effect?

And if I should ever begin to buy the hi res  recordings that are sold online from places like HDTracks, will the difference be noticeable?

Thanks. I appreciate your patience.

[ajv]

If I'm not mistaken, Audyssey (which is available on most if not all Denon AVR's) works at a maximum of 48Khz so there's no point going beyond it otherwise the signal will be down-sampled.

Thanks for the heads up regarding Audyssey.  I requested input from the Denon 2807 AVR forum regarding using the DAC for two channel playback. They did make your point. They also suggested however, that I could connect the DAC to the Denon using the analog outputs of the DAC into the analog EXT IN  inputs of the Denon. And that would bypass all processing including Audyssey.

For the Audyssey and other processing and 5.1 sound, I also still continue to process sound from my sound card to the Denon via toslink.

Are you aware of any other uses for these up up sampling settings? Thanks

[astromo]

If you've got the standard Bifrost DAC, the FAQ from Schiit tells me that it down-samples to 16 bit anyway:
http://schiit.com/products/bifrost

I've found this piece enlightening before when searching for fundamental answers to sampling frequency and bit depth questions before:
http://people.xiph.org/~xiphmont/demo/neil-young.html

What makes sense to me is that once the audio has been ripped (lossless), then the best approach for maintaining quality is to leave frequency and bit depth untouched as much as possible.

[ajv]

Thanks astromo for your prompt and thorough response. I obviously have some reading to do!!

Am I understanding correctly then that if I do buy and download hi res music from HDTracks in those higher resolution formats that it would sound better?

[blgentry]

I have about 2200 CDs ripped on an external hard drive in FLAC 24/48.

CDs are 44.1kHz, 16 bit (16/44.1).  If your ripping process has resampled these to 24/48, then it almost doesn't matter what you do, as the original information is gone.

If you actually have 16/44.1 rips, then I would try running without any conversion.  Then maybe try it at 192kHz.  I suspect you won't hear any real difference, but maybe?

If you have the standard or "uber" Bifrost, then all of this applies. If you have the multi-bit Bifrost, then you absolutely should NOT resample.  Just send the data exactly as is, with no changes at all:  No DSP of any kind.

Brian.

[kr4]

CDs are 44.1kHz, 16 bit (16/44.1).  If your ripping process has resampled these to 24/48, then it almost doesn't matter what you do, as the original information is gone.

? The original 16/44.1 information should be retained with resampling to 24/48 as well as to 24/192, separately or serially.  What would you say is "lost?"

[blgentry]

? The original 16/44.1 information should be retained with resampling to 24/48 as well as to 24/192, separately or serially.  What would you say is "lost?"

The original samples.  All of them are replaced when you resample.  How much data is left after resampling is a matter of debate.  But the fact remains that the original samples are gone.  Poof.  No more.  Maybe the resampled samples contain all of the data required for perfect playback.  Maybe not. 

Mike Moffat, who "invented" the idea of a separate audio DAC says you lose information. This is why he designed his line of multi-bit DACs at Schiit Audio, including the Bifrost Multi-Bit.  He probably knows some things we don't in terms of his understanding of digital audio.  After all outboard DACs (for audio) didn't exist before he decided to package one outside of a CD player.

Brian.

[kr4]

The original samples.  All of them are replaced when you resample.  How much data is left after resampling is a matter of debate.  But the fact remains that the original samples are gone.  Poof.  No more.  Maybe the resampled samples contain all of the data required for perfect playback.  Maybe not. 

Mike Moffat, who "invented" the idea of a separate audio DAC says you lose information. This is why he designed his line of multi-bit DACs at Schiit Audio, including the Bifrost Multi-Bit.  He probably knows some things we don't in terms of his understanding of digital audio.  After all outboard DACs (for audio) didn't exist before he decided to package one outside of a CD player.

Brian.

OK.  I understand that argument but your previous post suggested that once you have upsampled to 24/48, there is no point of upsampling to 24/192 because "then it almost doesn't matter what you do, as the original information is gone."  What I am trying to point out that is that upsampling to 24/48, to 24/192 or to 24/48 followed by 24/192 are equally affected by the "loss" you refer to. 

I find his argument a theoretical one but I could be convinced otherwise.  If one upsamples from 16/44.1 to 24/48 (or other) and then resamples back to 16/44.1, the process is such that one cannot expect a file identical to the original.  It does not prove than any information is lost.

[ajv]

Since what I'm hearing on the forum is that there is no benefit to upsampling and probably will denigrate the sound, why does JRiver offer the upsampling option?

Or is there some other use for upsampling that I'm not aware of?

Thanks in advance for any imput.

[churchmouse]

Since what I'm hearing on the forum is that there is no benefit to upsampling and probably will denigrate the sound, why does JRiver offer the upsampling option?

Or is there some other use for upsampling that I'm not aware of?

Thanks in advance for any imput.

The principle argument I have seen for upsampling within the player software is that most DACS upsample the signal in their processing of it even if they then downsample it again when sending it out. By upsampling within the computer rather than the DAC one might get a better result, either because it does a better job or because it reduces the workload on the DAC.

Since on the fly upsampling in JRiver does not harm the original file, a user can try upsampling and see if his/her ears prefer the result.

Sent from my YOGA Tablet 2-830F using Tapatalk