JRiver optimum setup.

[Vincent Kars]

It is basically about buffer management.
In case of WASAPI the driver pushes the data to the audio device.
In case of event style, the audio device pulls the data from the buffer.
Event style often works better with async USB DACs

http://wiki.jriver.com/index.php/WASAPI

[Gazjam]

Have a look here, set your Jriver up this way...
http://www.computeraudiophile.com/co...ter-17-detail/

Its a great "baseline" to get you started before you delve in to deeper settings.

[Ammonite Audio]

It is basically about buffer management.
In case of WASAPI the driver pushes the data to the audio device.
In case of event style, the audio device pulls the data from the buffer.
Event style often works better with async USB DACs

http://wiki.jriver.com/index.php/WASAPI

I know that. The HiFace Evo is async, so WASAPI Event Style should be better, but ordinary WASAPI still results (in my system and to my ears) in more enjoyable and musically satisfying reproduction.

[Martinh]

Have a look here, set your Jriver up this way...
http://www.computeraudiophile.com/co...ter-17-detail/

Its a great "baseline" to get you started before you delve in to deeper settings.

Thanks for that Gary,

I think I already had most of the items in the guide covered, except the tip to auto load only JRiver server when used without a display.

Have changed this and it still seems to function correctly.

Cheers,

[laozi]

I've been using foobar2000 for years and I never thought that there could be any better program for music...untill now. Now I am using JRiver.

Thanks...

[Rothchild]

But if you are playing 16 bit audio files and you enable 24 bit output in Jriver settings, you can use the extra bits for digital volume control (I think) without any loss of signal.. as long as you don't go too low on the slider.

I'm pretty confident this is simply not true.

A 16bit file is a 16bit file, adding any 'extra bits' is worthless as they'll be empty!

You can't just invent the 8bits that weren't captured in the first place, all you'll do is make the file larger and the computer therefore more likely to choke on it (not that this is very likely these days either given that we're talking about playing a stereo (2 channel) file.

(My (not so up-to-date) studio computer regularly copes with 50 or channels of audio with processor munching realtime effects such as convolution reverbs quite happlily.)

Potentially, where you might be getting confused is that a lot of audio software now works with 'floating point' maths, which enables gain to be manipulated beyond the range of the converter, this is handy if you're mixing as you can 'clip' a channel in the daw but pull the master down (at some point before it reaches the converters) and there's no distortion. However, if that (over-peak) level makes it through your converters (where it is returned to a 'fixed point' (ie 16bit)) they will distort/clip and things will sound properly horrible.

I'm sure J-River is a very competent audio player but some of the stuff written about it reads like the kind of stuff that I thought had been debunked for quite some time - adding in dead bits for headroom is a prime example!

As for the volume issue, the best solution IMO is to calibrate your room to a standardised SPL (83dB C Weighted Slow at the listening position, with a -20dBFS (RMS) Pink noise source is the Dolby cinema standard) mark this gain setting on your amplifier (with the rest of your gain staging at Unity). Depending on what you listen to it might be worth setting a quieter point too (83dB -20dB is perfect for classical, if you listen to pop and rock you might go for 77dB with the -20dB test tone to compensate for the greater overall loudness).

After that if you play something and it's too loud, (most modern recordings) and you have to turn the amplifier down, write an angry letter to the record label pointing out that it's stupid and uneccesary and insist that they should stop with the overmastering nonsense! ;-)

[NRG]

Padding to 24bits is exactly how the Logitech squeezebox works, in fact they implemented it from the SB Mk1 but only at 20 bits, from the SB2 onwards it's been 24 bits so it ccould be true if JRiver works the same way.

[Rothchild]

Moving the file in to floating point maths will probably allow the software volume control to work with finer resolution but this will not improve the resolution of the file. I will concede that if you want to feel sure that there's no 'bit reduction' going on you should use an analogue volume control (ie the knob on your amplifier) but there is no value to be gained from simply encoding a 16bit fixed file in to a 24bit fixed file, regardless of whether the software volume knob is working in fixed or floating point.

[Vincent Kars]

Maybe I’m stating the obvious but if your DAC is 24 bits, you have to feed it 24 bits otherwise it won’t play.
If you are playing a 16 bit file on a 24 bit DAC, eight zero bytes are added.
In case of JRiver/WASAPI you have to tell this in the DSP studio/output format.
These 8 zero bytes are added on the fly

If we play 16 bits and apply volume control
MSB - LSB
1111111111111111
000000001111111111111111
Yes we do loose resolution

If we do the same on a 24 bit DAC with 16 bit audio
111111111111111100000000
000000001111111111111111
We could indeed chop those 8 zero bits of without loosing resolution
This is 6x8= 48 dB reduction.

However, this is a bit theoretical no DAC is able to resolve 24 bits.
A bit more on volume control can be found here:
http://thewelltemperedcomputer.com/I...umeControl.htm

[NRG]

Moving the file in to floating point maths will probably allow the software volume control to work with finer resolution but this will not improve the resolution of the file. I will concede that if you want to feel sure that there's no 'bit reduction' going on you should use an analogue volume control (ie the knob on your amplifier) but there is no value to be gained from simply encoding a 16bit fixed file in to a 24bit fixed file, regardless of whether the software volume knob is working in fixed or floating point.

Well yes thats why its done to provide a digital volume control, it does not meant to improve the resolution. You get about -35db attenuation on the SB before resolution suffers. Personally I never use it and attenuate at the pre-amp.