Print

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

[tiggerkater]

good evening,

is there a setting in JRiver DSP (or a resample plugin?) to upsample ALL files to 24bit? (I could manage to output all files in 192khz, but 16bit files remained 16bit..)

Thanks a lot,

TIggerkater

[mojave]

Increasing bitdepth doesn't do anything unless there is processing occurring at the higher bitdepth. From the wiki:

Converting from less bits to more bits is perfectly lossless. Conceptually, imagine adding bits like adding zeroes at the end of a decimal. For example, the number "10" might become "10.0" or "10.00" if you add more bits, but all three representations are perfectly identical.
When Media Center inputs data, all audio is first converted to 64bit. This ensures that any processing like digital volume, Replay Gain, or any other DSP (if any is enabled) is done with as much precision as possible. It also puts the data into a format that is efficient for a computer to handle, and makes it so that tracks of varying bitdepths can seamlessly transition.
When outputting data to a soundcard or DAC, the 64bit data is converted back to the format required by hardware. This is often 24bit for high-end DACs.
The transition from 16bit to the output bitdepth (often 24bit) is bit-perfect. Again, it's like "10" vs "10.0".

With that said, you can right click and use Library Tools > Convert Format and in the Options change the bitdepth.

[6233638]

Media Center should be outputting the highest bit-depth that you hardware supports when you play a file.

[tiggerkater]

Thanks everybody for the fast reply,

Some more infos: My DAC is a Fireface UC, so it is capable of 24bit. (And usually a 16/44.1 (e.g.) flac file is outputted in 16 bit not 24 bit)
I do not want to convert the files and safe them as 24bit files. and yes, i know it does nothing more than adding some more "0" to the file.
I was wondering, if there is the possibility within JRiver to do a "online" conversion, as it is done for the samplerate.

Thanks a lot, with kind regards
tiggerkater

[Vincent Kars]

JRiver today set the bit depth automatically
However, you can force it to use a fixed bit depth (see attachment)

Either the automatic option is not working properly or you have a driver issue.
E.G If you use Direct Sound it might be that the bit depth is limited to 16 bits in the Windows Sound panel.
Check your driver settings

[kstuart]

You do understand that a 16-bit audio file converted to 24-bits should sound exactly the same ?

So what would be the point ?

[mwillems]

You do understand that a 16-bit audio file converted to 24-bits should sound exactly the same ?

So what would be the point ?

If one uses digital volume attenuation and were outputting 16-bit audio as 16-bit audio, you might lose some (fairly quiet) parts of the original signal, whereas outputting 16-bit as 24-bit gives you some pad for attentuation.  

I'm not sure how audible any of those losses would be in most situations, but as someone who routinely plays back at -40dB in JRiver, I'm thankful to have the option to output at 24-bit regardless of source bitdepth.

[Vincent Kars]

As 1 bit is 6 dB, at -40 you lose almost 7 bits of the 16 when using a 16 bits audio path.
I do think this a substantial loss.
I think you will profit (if doable) to lower the analog gain of your system
 

[mwillems]

As 1 bit is 6 dB, at -40 you lose almost 7 bits of the 16 when using a 16 bits audio path.
I do think this a substantial loss.
I think you will profit (if doable) to lower the analog gain of your system
  

Reducing analog gain isn't really an option: I have my DAC hooked directly to block power amps.  But your point is exactly why I output 16-bit files at 24-bits, the extra "empty" 8 bits allows for 48dB of additional space, which more than offsets the 40 dB of attenuation, so I lose nothing.  That's why the pad is significant, and I assume that's why OP wants to pad his outputs.

Realistically though, most files have a peak near digital full-scale, and a dynamic range between 6 and 20 dB.  Very few files (in my collection, anyway) have more than 20 dB of dynamic range (by either measurement standard), so I'm not sure how much loss there would be even if I output at 16 bits.  But I don't like taking chances, so I output at 24 bit  

[Vincent Kars]

Maybe I’m nitpicking but you are very close to bit 24 and most DACs don’t resolve the lower bits very well
Good DACs are linear to 18-20, only the ones with the whisper quit circuits can resolve 22 bits correctly.
A simple passive attenuator?

http://thewelltemperedcomputer.com/Intro/SQ/VolumeControl.htm

[mwillems]

Maybe I’m nitpicking but you are very close to bit 24 and most DACs don’t resolve the lower bits very well
Good DACs are linear to 18-20, only the ones with the whisper quit circuits can resolve 22 bits correctly.
A simple passive attenuator?

http://thewelltemperedcomputer.com/Intro/SQ/VolumeControl.htm

I used to use a passive line-level attenuator, and it's not a bad solution, but I prefer my current setup because it preserves the ability to get very loud if need be. 

I may be losing a few bits towards the bottom, but based on the math in my particular setup, I'm not sure it matters.  Let me lay it out; let me know if I've gotten something wrong along the way:

I use volume leveling, so the exact amount of digital attenuation varies, but on average the reduction from volume leveling and my normal internal volume settings adds up to around -40dB total.  The music is typically around 75dB RMS right in front of the speakers (based on my measurements).  The noise floor in my listening room is between 25 and 40 dB depending on the time of day (again based on my measurements).  So the difference between my noise floor (in the best case) and the average listening volume is about 50dB.  Add that to the 40 dB I'm throwing away through attenuation and you get 90 dB.  So any sounds reproduced below -90dBFS would be very challenging for me to hear, even in the best circumstances. 

That means, even with 16 bit output (96dB), I've got a little breathing room (6dB), but I agree that that would be shaving it fine as sometimes I attenuate even more than 40dB.  So I set it to 24 bits, and you're quite right, 24 bit DACs rarely fully realize the extra 8-bits.  So let's assume I'm only getting three additional bits from the 24bit output (effective 19-bit output) that would be 114dB of dynamic range . That means the quietest bit (-114dBFS) at my normal listening volume would be right around 1dB, which is near the theoretical limit of human hearing in an anechoic chamber. 

So if I'm losing sound below that, I'm not sure it matters, even in theory.  Maybe I nerfed the math somewhere? 

[Vincent Kars]

Fair enough!
Excellent reasoning
Thanks

[mojave]

I just performed Convert Audio on a 16 bit song and did -40 dB for L and R in the DSP. I then did Analyze Audio on the converted song. The Dynamic Range (R128) and (DR) stayed the same. That seems to indicate nothing was lost of the audible audio. Beyond -44 dB, the Dynamic Range (R128) started to decrease.

However, when converting the -40 dB 16 bit song back to 0 dB I do notice an increase in the noise floor. When I convert using 24 bits, I can't detect any difference.

I then converted the same song at -60 dB 24-bit and converted again at +60 dB. I still can't tell a difference and the Analyze Audio numbers are identical to the original. I think that a 16 bit file with a 24 bit DAC has quite a bit of attenuation available before it becomes audible.

[mwillems]

I just performed Convert Audio on a 16 bit song and did -40 dB for L and R in the DSP. I then did Analyze Audio on the converted song. The Dynamic Range (R128) and (DR) stayed the same. That seems to indicate nothing was lost of the audible audio. Beyond -44 dB, the Dynamic Range (R128) started to decrease.

However, when converting the -40 dB 16 bit song back to 0 dB I do notice an increase in the noise floor. When I convert using 24 bits, I can't detect any difference.

I then converted the same song at -60 dB 24-bit and converted again at +60 dB. I still can't tell a difference and the Analyze Audio numbers are identical to the original. I think that a 16 bit file with a 24 bit DAC has quite a bit of attenuation available before it becomes audible.

That was my suspicion based on the normal distribution of the DR stats I mentioned above, but thanks for thinking up (and performing) an experiment