Print

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

[v_erich]

Hi,
a short try for the DLNA DSP feature, but didn't work for me.
I tried to upsample on JRiver 20.0.27 on my Win2012 Server, control point was a android BubbleUPnP and Renderer was JRiver 20.0.26 on an other PC.
This JRiver showed me the original samplerate of the file, also on BubbleUPnP it shows the original samplerate.
The option "Specified Output Format" is active, I renamed the DLNA Server and the new name was shown right in BubbleUPnP.

Is it not planned to use the DSP feature when you control a 2nd JRiver from a JRiver on e.g. a Server via remote desktop?

BR
Erich

[Matt]

It's possible the sample rate is a special case.  Could you test other EQ like the Equalizer, Parametric EQ, etc.?

[AndrewFG]

I am guessing that due to the nature of UPnP the DSP effects would be limited only to ones that deliver the same media format, the same sample rate, the same bit depth, and the same channel count as the original...

[bob]

I am guessing that due to the nature of UPnP the DSP effects would be limited only to ones that deliver the same media format, the same sample rate, the same bit bepth, and the same channel count as the original...

It's also possible that the sample rate is being changed but not marked as such properly in the DIDL.

[daveman]

what exactly does "DSP Studio to DLNA" mean?

why are people so happy about this?

Please provide info for us naive folks.

dave

[bob]

what exactly does "DSP Studio to DLNA" mean?

why are people so happy about this?

Please provide info for us naive folks.

dave

It means you can apply any DSP studio effect, i.e. equalization to audio streams served by a MC DLNA server to a renderer.
Since you can also associate a particular zone with a particular server this means you can have zone based DSP to rendering devices.

[fitbrit]

Before this feature, MC was able to send the bit-perfect stream to the audio renderer, for example a DLNA receiver connected to one or more amplifiers. Now, however, MC's DSP effects, including room correction can be sent too.

[v_erich]

Hi,
will try something different on the evening (I'm in europe).

BR
Erich

[csimon]

Quick question - under what circumstances is the DLNA server required when playing on a renderer?

For example, using an instance of MC, I can enable DLNA Controller without enabling Server so that all DLNA renderers appear in Playing Now and I can select one to play to.  But in this case, Server isn't enabled so presumably DSP processing won't work?  The similar case is using a remote such as Gizmo to browse a JR library server and send to a DLNA renderer - library server is enabled but DLNA server isn't.

I guess what I'm saying is that DSP processing via DLNA only works when you're using a DLNA player to browse a JR DLNA server, or you enable a redundant DLNA server for situations when you're not playing from a JR DLNA library (redundant because the library is being served by JR Library Server) but want to use DSP processing.  Is that all correct?

[v_erich]

Hi,

with an EQ as DSP function everything is working.

BR
Erich

[bob]

Quick question - under what circumstances is the DLNA server required when playing on a renderer?

For example, using an instance of MC, I can enable DLNA Controller without enabling Server so that all DLNA renderers appear in Playing Now and I can select one to play to.  But in this case, Server isn't enabled so presumably DSP processing won't work?  The similar case is using a remote such as Gizmo to browse a JR library server and send to a DLNA renderer - library server is enabled but DLNA server isn't.

I guess what I'm saying is that DSP processing via DLNA only works when you're using a DLNA player to browse a JR DLNA server, or you enable a redundant DLNA server for situations when you're not playing from a JR DLNA library (redundant because the library is being served by JR Library Server) but want to use DSP processing.  Is that all correct?

Correct.
The controller by itself is a very simple concept.
The server is what does most of the work. In this case you'd have to enable the MC DLNA server to get DSP effects.

[brucemck2]

11. NEW: Added dsp studio to DLNA server audio advanced options. REQUIRES the output format to be set to "Specified Output Format".

Where in the  setup pages do I accomplish the instruction given above ... I see two choices ...

First I navigate to:
Media Network
... Add or Configure Servers
Advanced

Do I perform the instructions up in 11 above in the "Sample Rate" which is currently set to "Same as source" , or, in the dsp Studio page where sample rate outputs are set separately for each specific sample rate input?  I have done the latter.  Assuming that is correct, can I set different (but specific) output sample rates for each input sample rate, or, do I have to choose one output sample rate and use that for all input sample rates.

WHAT I'VE FOUND SINCE POSTING THE ABOVE: AT THIS POINT, THE ABILITY TO CHANGE SAMPLE RATES IN THE DSP STUDIO PAGE DOES NOT WORK. (IT WILL ALWAYS OUTPUT THE SAMPLE RATE SET PRIOR TO ENGAGING THE DSP STUDIO PAGE, SO IN MY CASE, IT WILL ALWAYS OUTPUT THE SOURCE SAMPLE RATE.)  BUT, THE PARAMETRIC EQUALIZER IN DSP STUDIO DOES WORK. I WILL TEST TOMORROW IF THE CONVOLVER FUNCTIONALITY WORKS FOR ALL SAMPLE RATES.

Finally, if the dsp Studio is active should I see that on the bottom of the dsp Studio page, which I do see when I am not using dlna, or not?  I currently see "na" in the "Source" and "Internal" panels at the bottom of the dsp Studio page.

[bob]

11. NEW: Added dsp studio to DLNA server audio advanced options. REQUIRES the output format to be set to "Specified Output Format".

Where in the  setup pages do I accomplish the instruction given above ... I see two choices ...

First I navigate to:
Media Network
... Add or Configure Servers
Advanced

Do I perform the instructions up in 11 above in the "Sample Rate" which is currently set to "Same as source" , or, in the dsp Studio page where sample rate outputs are set separately for each specific sample rate input?  I have done the latter.  Assuming that is correct, can I set different (but specific) output sample rates for each input sample rate, or, do I have to choose one output sample rate and use that for all input sample rates.

WHAT I'VE FOUND SINCE POSTING THE ABOVE: AT THIS POINT, THE ABILITY TO CHANGE SAMPLE RATES IN THE DSP STUDIO PAGE DOES NOT WORK. (IT WILL ALWAYS OUTPUT THE SAMPLE RATE SET PRIOR TO ENGAGING THE DSP STUDIO PAGE, SO IN MY CASE, IT WILL ALWAYS OUTPUT THE SOURCE SAMPLE RATE.)  BUT, THE PARAMETRIC EQUALIZER IN DSP STUDIO DOES WORK. I WILL TEST TOMORROW IF THE CONVOLVER FUNCTIONALITY WORKS FOR ALL SAMPLE RATES.

Finally, if the dsp Studio is active should I see that on the bottom of the dsp Studio page, which I do see when I am not using dlna, or not?  I currently see "na" in the "Source" and "Internal" panels at the bottom of the dsp Studio page.

Currently, the sample rate, volume leveling and stereo downmix from above the DSP studio selection override what's set in DSP studio.

[K_C_]

Currently, the sample rate, volume leveling and stereo downmix from above the DSP studio selection override what's set in DSP studio.

Then will "Output Encoding" eventually work?

Consider this case: I would like to have gapless playback, so I need to have the audio files pushed from the MC DLNA server. Meanwhile, I would like to have all sort of audio files transcoded into 2xDSD. However, as the SACD/DSD files will be sent as DoP files, if I enable the transcode process in the DLNA client (MC too), even the SACD/DSD files will be transcoded as they are recognized as PCM files by the client (i.e. DSD files will be treated as PCM files and then transcoded in to 2xDSD again, noise as a result). I mentioned this problem before and no suggestion provided so far.

http://yabb.jriver.com/interact/index.php?topic=91337.0

Therefore the only way to make it work is to transcode in the DLNA server side first, then just push the files as DoP to the client. So I hope that the DSP studio over DLNA server will be fully functional asap.

[brucemck2]

Two questions:
(1) How soon after changing the settings should they take effect? It seems as if sometimes they take effect after I stop and then restart a music track, while other times it requires that I restart MediaCenter
(2) Is the auto sample rate feature of the convolver working? (I suspect not.) Without any way to observe the audio path it's hard to tell.

[drmimosa]

Currently, the sample rate, volume leveling and stereo downmix from above the DSP studio selection override what's set in DSP studio.

It might be a good idea to revisit this in the future and default to DSP studio for all functions. I notice the that the when Volume Leveling is checked, loud sections of tracks with big dynamic range will suddenly drop in volume, as if the Volume Leveling is using "Adaptive Volume: Night Mode" as a default. Unchecking Volume Leveling solves this and the dynamic range sounds correct.

Bob, thank you for developing this feature. DSP over DLNA is a great addition to the MC20!

[v_erich]

For me sampling rate (e.g. some devices cannot playback 88,2kHz or >96kHz), EQ, room correction for smaller speaker (e.g. subsonic filter) and convolving is essential to give each device in different rooms the sound it needs.
Hope I can do this with DLNA at the end.
Haven't played much because of workload :-(

BR
Erich

[bob]

Two questions:
(1) How soon after changing the settings should they take effect? It seems as if sometimes they take effect after I stop and then restart a music track, while other times it requires that I restart MediaCenter

On a playing state change. If the file is cached it will probably get re-used until the next file is played.

(2) Is the auto sample rate feature of the convolver working? (I suspect not.) Without any way to observe the audio path it's hard to tell.

I would think anything that changes sample rate is probably not working at this time.

[AndrewFG]

I would think anything that changes sample rate is probably not working at this time.

Also (probably) nothing that changes channel count or bit depth, I would say..

[brucemck2]

With a bit of luck we'll get visibility into the audio path.  That would make it a LOT easier to see what's working, and not.

[fitbrit]

With a bit of luck we'll get visibility into the audio path.  That would make it a LOT easier to see what's working, and not.

+1

[v_erich]

*Bump*

Inv 20.0.33 still no resampling in DLNA DSP.

BR
Erich

[brucemck2]

I have been able to confirm that convolution works for alternative (varying) sample rates.

I created convolution filters for 44k, 88k, 176k, and 192k. Each had a unique shape so that I could use an RTA to see which filter was actually being used.

The settings that worked were:
(1) Setting "Sample Rate Same As Source" on the Advanced page of the DLNA Servers Page
(2) Setting Output Format to 44.1 for all input sample rates on the DSP Studio Page, OR, setting each input sample rate to have an identical output sample rate (and checking the Output Format box) ... note using "no change" also worked, but only after using one of the prior settings
(3) Setting Convolution to "on" on the DSP Studio Page and loading all the convolution files

This is a HUGE step forward for users of JRiver. Now if we can just get the audio path to display what's going on so that convoluted test routines aren't required.

[ogs]

I can not get DLNA to play if I set Audio-> Mode:Specified output format. When I select Audio->  Mode: Original the server works as expected. I can use Gizmo for remote control or, for example Kinsky and it all works.
I use upnpcli and Mpd for playback. It seems the file stream the player receives gets changed by Specified output format so it is not recognised. I only have a EQ setting enabled in DSP for testing, but I see the same symptoms if I also enable convolution.

[bob]

I can not get DLNA to play if I set Audio-> Mode:Specified output format. When I select Audio->  Mode: Original the server works as expected. I can use Gizmo for remote control or, for example Kinsky and it all works.
I use upnpcli and Mpd for playback. It seems the file stream the player receives gets changed by Specified output format so it is not recognised. I only have a EQ setting enabled in DSP for testing, but I see the same symptoms if I also enable convolution.

You could be limited by what your formats your device accepts.
When DSP is applied, the file has to be transcoded (or decoded in the case of PCM).
Al renderers are required to support headerless PCM (L16) but there are devices that don't.
The only other universal format (though not required) is MP3.
The wave formats are usually supported, especially the 16 bit one (PCM).
The 24 bit one may not be again depending on the device.

[ogs]

You could be limited by what your formats your device accepts.
When DSP is applied, the file has to be transcoded (or decoded in the case of PCM).
Al renderers are required to support headerless PCM (L16) but there are devices that don't.
The only other universal format (though not required) is MP3.
The wave formats are usually supported, especially the 16 bit one (PCM).
The 24 bit one may not be again depending on the device.

All files are flac. Some are 16/44.1 CD rips, others 24/48, 96 or 192. They all play nicely at native sample rates/bit depths direct from JR DLNA server via upmpdcli and Mpd. Device is the USB input of a Devialet 250 which is XMOS based. Is the output after DSP different from what is sent when files are played directly? I've tried both PCM 24 and L24, but not the 16bit versions. For 16bit, is there reduction taking place for 24bit?
I'll try 16 to see what happens.

EDIT: the two 16 variants have the same problem. I have some log output from upmpdcli if you'd like to take a look.

[bob]

All files are flac. Some are 16/44.1 CD rips, others 24/48, 96 or 192. They all play nicely at native sample rates/bit depths direct from JR DLNA server via upmpdcli and Mpd. Device is the USB input of a Devialet 250 which is XMOS based. Is the output after DSP different from what is sent when files are played directly? I've tried both PCM 24 and L24, but not the 16bit versions. For 16bit, is there reduction taking place for 24bit?
I'll try 16 to see what happens.

EDIT: the two 16 variants have the same problem. I have some log output from upmpdcli if you'd like to take a look.

I'm not sure what is the network rendering device in this case, is it Mpd or  upmpdcli?
The output format when using the DSP is different, it's not possible to transcode to streaming flac, that is why there are a bunch of choices for output.
The simplest test would be to specify MP3 as he output format and try it with the 16/44.1 CD rips. If that works it will give us a hint as to what's going on.

[ogs]

I'm not sure what is the network rendering device in this case, is it Mpd or  upmpdcli?
The output format when using the DSP is different, it's not possible to transcode to streaming flac, that is why there are a bunch of choices for output.
The simplest test would be to specify MP3 as he output format and try it with the 16/44.1 CD rips. If that works it will give us a hint as to what's going on.

I believe upmpdcli is just a frontend passing info to Mpd so Mpd is the renderer. I'll try MP3 to see what happens. I do not think this is a problem with JRiver DLNA however. I checked with the upmpdcli developer Jean-Francois Dockes. He says that Mpd may have problems handling network streams:

"There have been quite a few issues discovered in MPD in relation to using
HTTP instead of local file access, with several decoder modules.

Specifically, the decoders were not prepared for partial reads, something
which happens routinely when the input comes from the network. The default
input routine in mpd does not compensate for this, as a decoder which
*would* be prepared could sometimes get better performance accepting
partial reads."

So the decoder that handle flac network streams in Mpd is good, while wav and aiff is not taken care of properly. I'll just wait for the Mpd team to resolve this then. Thanks

[kalston]

I haven't managed to get DSP over DLNA to work on my iPhone (4) with JRemote, has anyone had any success?

It either sends unaltered or transcoded audio, but always without any DSP effects applied.

[JimH]

Make sure you have the latest build from the top of this board.  It's 20.0.37 right now.

[bob]

I haven't managed to get DSP over DLNA to work on my iPhone (4) with JRemote, has anyone had any success?

It either sends unaltered or transcoded audio, but always without any DSP effects applied.

JRemote (and Gizmo as well) don't use DLNA, they use the Web API to MCWS commands.
MCWS doesn't have a DSP control.

[kalston]

That explains it, thanks.

[ogs]

The simplest test would be to specify MP3 as he output format and try it with the 16/44.1 CD rips. If that works it will give us a hint as to what's going on.

Confirmed. MP3 format works with Mpd. DSP enabled, performing convolution and playing via DLNA/uPnP. I will just have to wait until Mpd can handle wav http streams.

EDIT: I managed to update MPD to 0.19.1 (by temporarily adding the sid repo to apt even though my distro is weezy). All resolutions up to 24/192 now play from the JRiver DLNA server with DSP enabled. I need to use the PCM 24 bit setting. Upmpdcli depends on header info in the streamed wav file.

[TomScrut]

Hi,

I seem to be struggling to get this working, does this work if my music is based on a NAS? I can browse the MC Media Server and select files etc but I'm pretty certain the PEQ isn't having an effect. Do the files need to be stored locally rather than having the library on the NAS but MC playing from that library?

Thx in advance

Tom

[bob]

Hi,

I seem to be struggling to get this working, does this work if my music is based on a NAS? I can browse the MC Media Server and select files etc but I'm pretty certain the PEQ isn't having an effect. Do the files need to be stored locally rather than having the library on the NAS but MC playing from that library?

Thx in advance

Tom

The files need to be stored locally or on a file based NAS, not using the NAS DLNA Server.
MC's DLNA server settings need to be set to "Specified Format" and you'll need to choose one that works with your renderer.

[PappyBlue]

If followed all the instructions but everytime I try to use the Dirac Live plugin in the DLNA streamer DSP studio, JRiver crashes. I can use the Dirac Live plugin in "normal mode" (ie using it with a usb DAC). Even though I already entenred my logging/password information for the Dirac Live plugin, when I try to use it in the DLNA server DSP studio I'm asked again to enter my logging/password for Dirac and it usually crash JRiver when it does. Any idea ?