Print

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

[NW_Landing]

Long time user of JRemote on my iPhone but having problems.  Hoping for some insight and suggestions to address my playback issue.  I'm using a EarMen TR-Amp as a headphone amp to stream from my JRiver library to my iPhone.  This last 1-1/2 years I've been remastering my FLAC files using Remasteros' PGGB software. Depending on the sample rate of the FLAC source files my library now consists of 24 bit 705.6/768 wave or wave pack files that can be larger than a gigabyte or more in size (my whole library is 42TB).  From JRiver to my Chord DAVE DAC it sounds amazing and performs flawlessly. 

My problem statement is that playback when using JRemote on my iPhone (using the server access key) to my TRAMP results in inconsistent quality; ranging from fine to unlistenable.   By unlistenable I mean terrible distortion.  I've tested the TRAMP directly on my JRiver desktop rig, using the EartMen ASIO driver, and it plays flawlessly.  I've also tested using Neutron Play with my 24-bit 705.8/768 PGGB library with a perfect outcome. This is with no change on JRIver Option configs. So I've concluded that the problem isn't the source material, cabling or the headphone amp. Either I'm not configuring JRiver options correctly or perhaps there is a problem with the app.

My current oAudioption configurations are:
Audio:
Auris EarMen ASIO.  I've tried various levels of buffing, highest buffering works well with Neutron Play. no effect on JRemote I've also tried ASIO4ALL, no better

Everything in DSP Studio is disabled

I prefer everything decoded into memory on the server (i have 128GB memory)

Volume is disabled, I want to control that from my HP amp.

Sometimes it's the most obvious thing overlooked or misunderstood.  I'd appreciate other perspectives and questions.

Edit: I forgot to mention that Playing the PGGB library through my iPhone using Panel works. .

[JimH]

You can't really "re-master" a FLAC file.  You've upsampled it, but you haven't added any quality.  Sound differences may be volume differences.

If you're playing GB files over a network connection or USB, you may be limited by bandwidth.

This might also be a bottleneck:

I prefer everything decoded into memory on the server (i have 128GB memory)

You may also have JRemote set to do a conversion.  Try "original".

[NW_Landing]

How do you explain Panel and Neutron Play playing the same tracks flawlessly on my iPhone over the same network with the same settings in placa?  And And yes I have JRiver set to play original. 

[JimH]

Memory playback?

The way to find the problem is to start changing things and see if you can get it to work at all.   Then backtrack.

[NW_Landing]

Memory playback?

The way to find the problem is to start changing things and see if you can get it to work at all.   Then backtrack.

I've probably spent about 24 hours over the last few days iteratively, incrementally and methodically adjusting settings to no effect.  I'm using memory playback. I've tried all three types of memory playback, and no memory playback. Still distorted.  I've tried swapping cables too.  It's clear to me that it's an issue with JRemote.  Especially since Panel and Neutron Player work flawlessly. 

I'm happy to dropbox you a sample PGGB file if you have any interest in exploring the issue, just PM me.

To me JRemote is a more elegant solution compared to Panel but at the end of the day I'm more focused on an outcome that works for me when I want to sit outdoors and stream.  All other times I'm sitting inside listing to my Chord DAVE with my Focal Utopia or Audeze HPs.


To your earlier comments about PGGB not remastering the tracks, you're correct. Perhaps I should have said 'remastering' or just said upsampling.  Considering your speculation that all I'm hearing is a volume change, wouldn't the dynamic range be different between the source FLAC and upsampled wav files?

PGGB stays true to Nyquist–Shannon sampling theorem, PGGB's algorithms uses extremely long windowed Sinc filters (i.e., millions or billions of taps), with filter lengths only limited to processing hardware and track length.  The longer the filter, the higher the reconstruction accuracy and the more transparent the sound. To my ears this means  better depth and layering, improved resolution, a cleaner leading edge, and more accurate timbre.  Before PGGB I was using Chord M Scaler which is also based on Nyquist–Shannon sampling theorem, and a million taps.  During my evaluation I performed a significant amount of critical listening comparing M Scaler vs PGGB. PGGB won hands down.  And I had every reason for it not to be the outcome.  The hardware cost on the processing machine and additional NAS space was significant. With blind A/B testing with my wife, who has better ears than me, PGGB won out.

Thanks Jim

[JimH]

Start simple, add complexity.  Use ordinary FLAC, test local files only.  Then add complexity, one thing at a  time.

Anything else is shooting in the dark.

We're not going to pursue problems with such files.  Sorry.