Print

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

[windowsx]

Is it possible to make prebuffering having smaller size? Minimum at 2000ms is quite high comparing to foobar2000 that can set down to 50ms if its software buffering is similar to jriver's prebuffering?

[Matt]

I believe you are comparing apples and oranges.

But what problem are you experiencing that you're trying to solve?

[windowsx]

I believe being capable to operate on smaller buffer playback can lead to improvements of low latency playback. I love solid performance like lower loopback/buffering. Just saying it would be nice if that could be improved some ways like no DSP chain mode for lower DSP latency.

[BryanC]

I believe being capable to operate on smaller buffer playback can lead to improvements of low latency playback. I love solid performance like lower loopback/buffering. Just saying it would be nice if that could be improved some ways like no DSP chain mode for lower DSP latency.

Hardware buffering is set in the output mode settings and can go as low as your hardware allows (<10m).

[windowsx]

I'm not referring to hardware buffering but secondary prebuffering in J River option having default at 6000ms. I can set down to 2000ms. I believe this is for DSP pre-processing part before feeding to output.

[Matt]

I believe being capable to operate on smaller buffer playback can lead to improvements of low latency playback. I love solid performance like lower loopback/buffering. Just saying it would be nice if that could be improved some ways like no DSP chain mode for lower DSP latency.

Bryan gave good advice.

The loopback latency is directly related to the primary buffer size, set in:
Options > Audio > Output mode settings...

The secondary buffer size in Options > Audio > Prebuffering has no impact on latency, and the setting is essentially ignored during loopback.

[windowsx]

What is secondary buffer being used for? For processing DSP?

[Matt]

DSP is processed just-in-time (JIT) before data reaches the primary/hardware buffer.

This was one of the major architecture shifts in the JRiver audio engine compared to Winamp (or similar), and it's really smart in my opinion.  It allows low latency _and_ flexible pre-buffering, gapless, cross-fading, etc.

[windowsx]

So it works like buffer pool for pre-processing some output effects like what you mentioned. Is it possible to make test features like 'pure direct' mode bypassing those and optimize secondary buffering to its minimalist. I think it could easily be adjusted to 50ms-100ms with effects bypassed. It's alright if this is too hard for burden to test but I want to confirm something against other audiophile apps. They claim direct mode/low buffer for purer sound.

[Matt]

They claim direct mode/low buffer for purer sound.

The hardware buffer (the only buffer that could possibly matter) is fully configurable.

If the arrangement of general memory in your computer affected sound quality, a typical computer with 4GB of memory would have 256^4294967296 possible memory configurations (each byte of 8 bits has 256 arrangements).

My calculator exploded trying to do this math, but I believe that's more atoms than the universe contains (around 10^82 atoms).

So how could we begin searching for the best arrangement of memory to optimize sound quality?

[windowsx]

lol. but still nice if things could be optimized for lower buffer pool (100ms buffer pool would be nice if possible) so those who favor others can test this ones. If there's no changes in performance, that'd be a laughter.