Print

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

[DarkSpace]

I originally considered posting this in the Too Easy II thread, but after consideration, I'm no longer certain it belongs there. Also, posting it in its own thread allows me to give a detailed reason.

Very recently, in addition to JRiver's own audio dithering mechanism, a TPDF dither mechanism has been introduced. The dither mechanism that will be used can be chosen from the Zone's audio options page.
Currently, though, the Parametric Equalizer (2)'s Bitdepth Simulator only has a checkbox to switch dithering on or off. I presume that this means that it will use the Zone's dither mechanism if it's enabled, which should be fine for most use-cases. However, I'd like to switch the dither mechanism of the Bitdepth Simulator independently from the Zone's, so I'd appreciate a way to choose the Simulator's dither.

I realize that this sounds strange, but my reason for wanting this is the following:
I have a USB-to-SPDIF converter with an ASIO interface. The converter supports up to 24-bit data, which is fine because I can tell JRiver to output only 24 bits of meaningful information in its ASIO settings. However, I know that the headphones I have connected to that converter only use 16 bits of information, but I have found no option to make JRiver output 16-bit data via ASIO.
For that reason, I have set a Parametric Equalizer 2 with a Bitdepth Simulator (16 bit with Dither) as the last DSP step. This works for my headphones*, but after reading that effectively, TPDF dither becomes better as the difference between input and output bitdepth increases, I would like to try it, paired with JRiver bitexact dithering to 24-bit for ASIO output. I hope that makes some sense. If it doesn't, I'm willing to explain what I mean.
I've been thinking that maybe, instead of a checkbox that specifies whether to dither or not, a drop-down menu like in the Audio options page for the dither mechanism could be used instead?

* In fact, it even works for my headphones with pure 24-bit output, but I believe (no blind tests done) that I hear a difference between 16-bit with dither and 24-bit. I'd usually attribute this to placebo, but some time ago, I tried the same thing sometime using my TV's speakers, expecting to hear a difference, but I could only barely hear a difference at around 8 bit without dither. So I definitely didn't expect this... make of that what you will.

[Hendrik]

Just change the dithering in the global settings, it'll apply immediately and you can test the two alternatives. The choice is global and passing through an option through 3 layers in crucial code just for the simulator doesn't sound like a thing I would want to do.
TPDF should generally be superior - it just gets more obvious as the bitdepth decreases, as any imperfections get amplified.

[mojave]

How do headphones use only 16-bits of data?

If you really need 16 bits, then create a second zone for the headphones, use WASAPI output, and set the bitdepth to 16-bits in Options > Audio > Audio Device > Device Settings. You can use the same audio device as the speaker zone. Its also helpful to use a separate zone for headphones since you can use the headphone crossfeed DSP on that zone.

[DarkSpace]

Just change the dithering in the global settings, it'll apply immediately and you can test the two alternatives.

I did already, thank you.

The choice is global and passing through an option through 3 layers in crucial code just for the simulator doesn't sound like a thing I would want to do.

I see. In that case, it's obviously not as simple as I thought it would be... well, I can live with the current situation.
In fact, I hadn't even bothered to verify if the dither settings are stored per-Zone or globally, so here I learned something new.

TPDF should generally be superior - it just gets more obvious as the bitdepth decreases, as any imperfections get amplified.

Thanks for the confirmation. That makes sense.

How do headphones use only 16-bits of data?

They are wireless and transmit digital data. I only know that they transmit 16-bit 48 kHz from reading the spec sheet, so I don't think I would have noticed just by listening. It's no big deal, and in fact, those headphones are at least an order of magnitude better than anything else I've got, so I won't complain (though I know that better things exist).

If you really need 16 bits, then create a second zone for the headphones, use WASAPI output, and set the bitdepth to 16-bits in Options > Audio > Audio Device > Device Settings. You can use the same audio device as the speaker zone. Its also helpful to use a separate zone for headphones since you can use the headphone crossfeed DSP on that zone.

I considered using WASAPI, but since I only have those headphones connected to the USB/SPDIF converter anyway, I only need one Zone for it (in which I enabled the headphone crossfeed, thank you). From reading, though, I gathered that using ASIO should be preferable to using WASAPI in this situation, so I chose to use that, and inserted a Bitdepth Simulator at the end. It works for me, and aside from the easily-fixable bitdepth (as I said, even when sending 24-bit data, the headphones would play sound), it works beautifully.
Maybe I should reconsider, though...