Tír Na HiFi

Posted: **Mon Jan 20, 2014 8:40 pm**

nige2000 wrote:i did new r2 on a usb3, phils optimiser, c++ update still in std gui
just for direct comparison between the MQn's ks vs wasapi
ill change it to core but i dont think that will do it

u using -a 2000:2000 still?

I'm using -a 2 and no -b at the moment.
I'm not really a fan of large buffers. JPlay Xtream is such a player that uses large buffers.

My DDC-driver has a streaming-buffer of 2 ms and I have kept the timerresolution equal to that value. That suited me very well with the WASAPI version.
I don't know what SqueezeLite defaults are, still have to find a description of the Squeezelite additional parameters.

Cheers

Aleg

Posted: **Mon Jan 20, 2014 8:57 pm**

lol

we shall prey for you!!!!!
may your sins be forgiven

I say Three Hail Marys before I start the session, then it's all right.

Actually, I only do a little to compensate for bloody over loud CDs (and before you say so I hate Replay Gain - I want control) and the SQ is not distinguishable compared to turning down the amp.

I don't think any equipment gets anywhere near to full 16 or 24 bit precision anyhow and so the bits 'thrown away' are likely not adding anything to SQ in the first place.

Anyhow that's what I say when kneeling at the throne of St. Gordon. Perhaps he will give it back if I pray extra long?

Posted: **Mon Jan 20, 2014 9:05 pm**

Found the possible parameters, ...... Inside Squeezelite-win itself:

Squeezelite v1.5-365, Copyright 2012-2014 Adrian Smith. See -t for license terms
Usage: squeezelite-win.exe [options]
-s <server>[:<port>] Connect to specified server, otherwise uses autodiscovery to find server
-o <output device> Specify output device, default "default", - = output to stdout
-l List output devices
-a <l> Specify Portaudio params to open output device, l = target latency in ms
-a <f> Specify sample format (16|24|32) of output file when using -o - to output samples to stdout (interleaved little endian only)
-b <stream>:<output> Specify internal Stream and Output buffer sizes in Kbytes
-c <codec1>,<codec2> Restrict codecs to those specified, otherwise load all available codecs; known codecs: flac,pcm,mp3,ogg,aac,wma,alac (mad,mpg for specific mp3 codec)
-d <log>=<level> Set logging level, logs: all|slimproto|stream|decode|output, level: info|debug|sdebug
-f <logfile> Write debug to logfile
-m <mac addr> Set mac address, format: abef:12:34:56
-n <name> Set the player name
-N <filename> Store player name in filename to allow server defined name changes to be shared between servers (not suppored with -n)
-r <rates> Specify sample rates supported by device, enables output device to be off when squeezelite is started; rates = <maxrate> | <minrate>-<maxrate> | <rate1>,<rate2>,<rate3>
-R -u [params] Resample, params = <recipe>:<flags>:<attenuation>:<precision>:<passband_end>:<stopband_start>:<phase_response>,
recipe = (v|h|m|l|q)(L|I|M)(s) [E|X], E = exception - resample only if native rate not supported, X = async - resample to max rate for device, otherwise to max sync rate
flags = num in hex,
attenuation = attenuation in dB to apply (default is -1db if not explicitly set),
precision = number of bits precision (NB. HQ = 20. VHQ = 28),
passband_end = number in percent (0dB pt. bandwidth to preserve. nyquist = 100%),
stopband_start = number in percent (Aliasing/imaging control. > passband_end),
phase_response = 0-100 (0 = minimum / 50 = linear / 100 = maximum)
-t License terms

Build options: WIN PORTAUDIO WINEVENT RESAMPLE FFMPEG

So there is not much to play with actually, and have no clou yet as what any default values are.

-a
-b
-c

Are options to try I guess

Cheers

Aleg

Posted: **Mon Jan 20, 2014 9:16 pm**

"- R" looks promising.

I found solution to play M4a files (i have load of nero aac m4a ):
Foobar>context menu>Utilities>Optimize MP4 layout and voila.

Posted: **Mon Jan 20, 2014 9:22 pm**

John Dot wrote:"- R" looks promising.

I found solution to play M4a files (i have load of nero aac m4a ):
Foobar>context menu>Utilities>Optimize MP4 layout and voila.

I don't think on-the-fly resampling is good for sound quality.
We like to reduce the amount of CPU activity and get an as perfect timing as possible. Playback and resampling will be difficult to get into a proper timing.

Cheers

Aleg

Posted: **Mon Jan 20, 2014 9:56 pm**

jrling wrote:
lol

we shall prey for you!!!!!
may your sins be forgiven
I say Three Hail Marys before I start the session, then it's all right.

Actually, I only do a little to compensate for bloody over loud CDs (and before you say so I hate Replay Gain - I want control) and the SQ is not distinguishable compared to turning down the amp.

I don't think any equipment gets anywhere near to full 16 or 24 bit precision anyhow and so the bits 'thrown away' are likely not adding anything to SQ in the first place.

Anyhow that's what I say when kneeling at the throne of St. Gordon. Perhaps he will give it back if I pray extra long?

Since I hear a clear difference between 16-bit and 24-bit music, I beg to differ.

Also because all my music is DRC-ed and dithered from 64-bit down to 24-bit or 16-bit, I know that 24-bit is definitely better than the 16-bit and wouldn't like to throw away a single bit from any of those bitdepths.

But if it doesn't make a noticeable difference for you, than why not?

Cheers

Aleg

Posted: **Mon Jan 20, 2014 10:15 pm**

-a 2 = latency in ms, so effectively your period or sample size
-b 0:2000 = output buffer - portaudio sets up a buffer, only the figure after the : is used.

resample uses sox and is not part of the build

Posted: **Mon Jan 20, 2014 11:22 pm**

Recently I was very keen on the Linux distro Daphile for Squeezelite. Did a back to back with Gordon's movsqv2 Squeezelite-win 1.5 and there was no contest. Gordon's was in a different league (better). Most impressive and in my case better than I ever had with MQn. Windows KS certainly seems to be miles better than WASAPI.

I'm back to Windows.

Posted: **Tue Jan 21, 2014 12:12 am**

Still waiting this out as enjoying 2.73 1024 raw background a lot. It seems like hard work but will get some time next week and will engage.
Lots of progress seems to have been made in a short space of time.

Posted: **Tue Jan 21, 2014 12:14 am**

Aleg wrote:Today I changed the Server 2012 R2 to Core mode, installed VS2013 C++ runtime and applied AudioPhil's Optimiser 1.23 which enables KS in Core-mode.
This was again a step up in sound quality compared to Server 2012 R2 in standard Full GUI mode.

Cheers

Aleg

Aleg, what is the purpose of the VS2013 C++ runtime?

Thanks,
Mark

Tír Na HiFi

MQN

Re: MQN

Re: MQN

Re: MQN

Re: MQN

Re: MQN

Re: MQN

Re: MQN

Re: Daphile v. Squeezelite movsqv2

Re: MQN

Re: MQN