Print

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

[PapiChulo]

I'm running MC21. I LOVE all of its features, especially the DSP, but what gives regarding lack of gapless?

I pull files from a NAS, use BubbleUpnp to stream to my Chromecast audio. Audio quality is EXCELLENT, as is MC's file handling from the NAS.

BUT...I can't play gapless. THAT SUCKS.

People have said that its impossible to play gapless over DLNA. I can tell you that statement is INNACURATE on MY setup. I can play gapless using Foobar 2000 on the exact same setup...with zero problems.

Foobar works fine but it has a crummy interface and no DSP like MC.

So...a simple question to the technical people writing code for MC - why can Foobar do DLNA gapless to my Chromecast, while MC cannot?

[AndrewFG]

Three things..

1) Anyone who says DLNA can’t do gapless, doesn’t know what they are talking about

2) MC does do gapless playback to DLNA renderers that support it. In MC right click on the renderer and make sure “Ignore SetNext” is NOT checked. But note that not all renderers support it.

3) As far as I know, Chromecast is not a DLNA renderer.

[Scobie]

3) As far as I know, Chromecast is not a DLNA renderer.

It’s not, you need BubbleUPnP for it to be picked up as a renderer

[DJLegba]

My understanding is that the application needs to send a continuous stream to the CCA device if you want gapless. This requires custom programming, as it's not the way DLNA works.

You can get MC to see your CCA via BubbleUPnP, but as of now this doesn't give you gapless playback. Android remotes such as JRemote and MO4Media will both cast to your CCA without BubbleUPnP, but again, not with gapless playback. You can achieve gapless playback if you configure your Android remote to play to "this device", and then use Android's cast feature to send to the CCA. But in this case the chain is MC-->Phone-->CCA and I believe MC's DSP doesn't kick in when you tell MC to play to your phone. I think Android does some funky converting as well. I never need to set up playback this way though, so I could be wrong. You might have some fun testing it yourself.

[PapiChulo]

Three things..

3) As far as I know, Chromecast is not a DLNA renderer.

Note I mentioned that I'm using Bubble UPnP that converts Chromecast into a DLNA renderer.

Again...this SAME setup plays gapless just fine using Foobar 2000. What is different between Foobar and MCs implementation of gapless? ?

[AndrewFG]

Again...this SAME setup plays gapless just fine using Foobar 2000. What is different between Foobar and MCs implementation of gapless? ?

Did you right click in MC on the Bubble/Chrome renderer and uncheck the Disable SetNext setting?

PS you might want to post your same question to bubbleguum too..

[PapiChulo]

Did you right click in MC on the Bubble/Chrome renderer and uncheck the Disable SetNext setting?

I tried it both ways and that setting had no effect...

[AndrewFG]

Please download the DMRA from my sig, run it against the Bubble/Chrome rendere, and post the renderer report here.

[Scobie]

I've use the same setup in my DLNA mix Andrew, MC25 streaming to CC Ultra via Bubble.

DMRA reports that Gapless is not supported.

Description=BubbleUPnP Google Cast DLNA renderer

AVT:GetDeviceCapabilities action=Supported
AVT:GetMediaInfo action=Supported
AVT:GetPositionInfo action=Supported
AVT:GetTransportInfo action=Supported
AVT:GetTransportSettings action=Supported
AVT:SetNextAVTransportURI (gapless play)=NOT Supported
AVT:SyncPlay (synchronous play)=NOT Supported
RC:GetVolume action=Supported
RC:SetVolume action=Supported
AVT:Event Subscription=Succeeded
RC:Event Subscription=Succeeded

[AndrewFG]

AVT:SetNextAVTransportURI (gapless play)=NOT Supported

Weell, that’s fairly explicit then: Bubble does not support gapless on its UPNP interface, so it is a Bubble issue, and not an MC issue..

I think you said that nevertheless when FooBar is pushing to Bubble, you do get gapless. Or?? If that is the case then either FooBar or Bubble are doing something different than normal UPNP. Perhaps FooBar is concatenating the tracks into a single push file? You would need to ask FooBar what they are doing.

[DJLegba]

Weell, that’s fairly explicit then: Bubble does not support gapless on its UPNP interface, so it is a Bubble issue, and not an MC issue..

I think you said that nevertheless when FooBar is pushing to Bubble, you do get gapless. Or?? If that is the case then either FooBar or Bubble are doing something different than normal UPNP. Perhaps FooBar is concatenating the tracks into a single push file? You would need to ask FooBar what they are doing.

It's a CCA issue. Bubble does gapless just fine when the renderer supports gapless.

[AndrewFG]

It's a CCA issue. Bubble does gapless just fine when the renderer supports gapless.

Hmm. The OP is (I think) saying..

a) MC push to Bubble push to Chromecast => no gapless
b) FooBar push to Bubble push to Chromecast => gapless

..and we have ascertained from the DMRA test that in case a) the “no gapless” result is not the fault of MC...

..but it would sure be interesting, (not necessarily for JRiver), to know how case b) manages, counterintuitively, to produce the result claimed by the OP..

[DJLegba]

Hmm. The OP is (I think) saying..

a) MC push to Bubble push to Chromecast => no gapless
b) FooBar push to Bubble push to Chromecast => gapless

..and we have ascertained from the DMRA test that in case a) the “no gapless” result is not the fault of MC...

..but it would sure be interesting, (not necessarily for JRiver), to know how case b) manages, counterintuitively, to produce the result claimed by the OP..

Stream of foobarness to the CCA. Bubble does gapless to every gapless-supporting renderer that I've tried.

[PapiChulo]

Hmm. The OP is (I think) saying..

a) MC push to Bubble push to Chromecast => no gapless
b) FooBar push to Bubble push to Chromecast => gapless

..and we have ascertained from the DMRA test that in case a) the “no gapless” result is not the fault of MC...

..but it would sure be interesting, (not necessarily for JRiver), to know how case b) manages, counterintuitively, to produce the result claimed by the OP..

Exactly! Foobar DOES support gapless with my CC Audio - this I am not imagining! There are also a couple of android apps that can pull music from my computer server and push it gapless to the CC Audio, but that is a different operating system I know.

Whatever Foobar is doing to make it's gapless system function with the practically universal Chromecast...it's a great ability. CC is resident in virtually every consumer audio device being sold today, and can be retrofitted very cheaply to the rest with a $35 dongle. I personally run mine with an optical link and iFi jitter reduction "purifier" into a 24-bit DAC of my choosing. No need to pay big $$$ for overpriced "streaming devices" that do nothing different. Sound is excellent, too especially with 24/96 sources.

OK, it looks like I'm SOL with regards to MC and CC. Darn...I like the MC interface far better than Foobar's. Plus, MC does all media, not just audio... 

[AndrewFG]

Bubble does gapless to every gapless-supporting renderer that I've tried.

Ok, so Bubble can push gapless to Chromecast. But if that is the case then why does Bubble report to MC that it does NOT support gapless?

[Scobie]

Remember that was my report not the Op’s, Andrew. Maybe a setting in Bubble? I’ll check.

In any case, when Papichulo runs DMRA we’ll know for sure.

[DJLegba]

Ok, so Bubble can push gapless to Chromecast. But if that is the case then why does Bubble report to MC that it does NOT support gapless?

Because Chromecast doesn't support gapless. For gapless playback the application needs to push a continuous stream to Chromecast, not separate tracks.

[AndrewFG]

Remember that was my report not the Op’s, Andrew. Maybe a setting in Bubble? I’ll check.

In any case, when Papichulo runs DMRA we’ll know for sure.

Hi Scobie, ah yes, sorry for confusing your report with Papichulo.

Hi Papichulo, can you please also run the DMRA (from my sig) to see if in your setup it produces a better result than in Scobie’s case?

[Scobie]

Exactly! Foobar DOES support gapless with my CC Audio

Foobar has an extension called foo_upnp_out that enables it to play gapless directly to the Chromecast via Bubble.

From another forum: https://audiophilestyle.com/forums/topic/26450-need-help-with-chromecast-audio-gapless-playback-if-possible/

The foo_upnp_out plugin diverts the Foobar2000 player's output so that it can be streamed by the nominated UPnP/DLNA renderer. So it's the Foobar2000 player that's handling gapless playback support, not the Chromcast Audio device being used as a UPnP/DLNA renderer. The Chromcast Audio will be streaming the continuous output being produced by Fb2k playback of its current playlist (the Chromecast Audio is not actually streaming the individual file tracks contained in that playlist)

That forum thread is a good CCA/BubbleUpNP discussion and worth reading.

[PapiChulo]

That forum thread is a good CCA/BubbleUpNP discussion and worth reading.

Indeed it is. Thank you.

So now we know that CCA has no intrinsic gapless capability. Programs like Foobar are actually combining adjacent music files into a continuous stream, essentially handling the "gapless" situation in the server software itself.

The question now is, would the developers of MC be interested in adding a similar feature? There are a LOT of Chromecast-enabled devices out there, and it would be a great selling point for MC to incorporate.

 

[AndrewFG]

In theory MC could be extended to convert (say) a playlist or an album into a single large file, and then push that as one “track” to the renderer. But it would have some down-sides as well..

1) one would obviously lose the Next/Previous track seek functionality; although the time slider bar might still work.

2) whenever the playlist is changed, MC would have to re-build and re-push its single file. So for example, adding tracks one by one to the end of a playing playlist, or dynamic playlists (and play doctor) would probably not work.

[Scobie]

Also what effect - if any - would this have on the audio elements of the individual files, dynamic range etc,  if they have essentially been stitched together?

Would these have been normalised by the process or would they be maintained?

[DJLegba]

Also what effect - if any - would this have on the audio elements of the individual files, dynamic range etc,  if they have essentially been stitched together?

Would these have been normalised by the process or would they be maintained?

I subscribed to Google Play Music for a while. It could always do gapless to the CCA. I don't know if it used some special mode known only to Google, or if it did the continuous stream thing, but there was definitely no normalization going on when I played a full album.

On another note, when I had trial subscriptions to GPM and Spotify I couldn't believe how bad Spotify sounded, and I was not happy that it didn't go gapless to the CCA. But eventually I discovered that Spotify (like every other streaming service, I've come to realize) had normalization on by default. Once I disabled that it was able to do gapless, and the quality improved immensely. Spotify's normalization wasn't like MC's, which doesn't attempt to normalize tracks within an album. Spotify, at that time anyway, treated everything like a playlist, and normalized every track - even if you were playing something like an opera. So if it has to send a continuous steam to achieve gapless, then my experience indicates that normalization isn't applied.

[AndrewFG]

I guess that if one was going to stitch tracks together sequentially, one would have to put all tracks through MC’s transcoder, and this could obviously include any of the DSP functions available for transcoding single tracks. So for example the Volume Levelling DSP (and/or any other DSP) could be applied, or not, at your discretion..

[PapiChulo]

1) one would obviously lose the Next/Previous track seek functionality; although the time slider bar might still work.

2) whenever the playlist is changed, MC would have to re-build and re-push its single file. So for example, adding tracks one by one to the end of a playing playlist, or dynamic playlists (and play doctor) would probably not work.

When I'm streaming gapless to my Chromecast Audio with Foobar2k, I have neither of the two issues that you brought up above! Track changes work as expected, and I can add to my playlist without any hiccups.

How on Earth they are accomplishing all this remains a mystery. All I know is...it functions perfectly!

[bubbleguuum]

The only way to play gaplessly to Chromecast is an external program that feeds it a single continuous gapless stream (usually PCM but could be FLAC or mp3 or aac).
That's what foo_out_upnp does, sending its gapless oaudio output (DSP and all) to BubbleUPnP Server (emulating a DLNA renderer for Chromecast) as a single L16 stream. MC could have an identical alternate UPNP/DLNA play mode where it decodes everything gaplessly and send the resulting stream to the renderer.

Google Music is an exception: it uses mp3 which is inherently not gapless + some trickery (possibly LAME headers) inside its custom Chromecast receiver to make it gapless
(thus not using the html audio element but custom decoding with MSE I believe).

[madbrain]

I know this is a very old thread, but I have this issue as well. I'd really like to see gapless support with the Chromecast Audio and JRiver MC. This is not inconsequential for someone who currently has 17 active Chromecast devices (14 active CCA, 2 CC with Google TV, 1 CC Ultra) in one home.

Foobar2000 is not the only program that has achieved gapless support with Google devices. I'm currently doing a free trial of Roon. It has this CC gapless support as well. Not gapless over DLNA, but DLNA isn't a must. What's important is that the audio get there uninterrupted. And I have to say Roon has done it beautifully. Roon has a lot of shortcomings compared to JRiver MC relying on cloud service for search, very minimal library editing capabilities, useless integration with Home Assistant (this works with DLNA for JRiver, cough cough) in no particular order - and of course a so-called lifetime license is a hefty $829, probably more than I spent on all my JRiver master licenses. I might be willing to deal with a few of these shortcomings, but taking them altogether, I find myself wishing JRiver would add the support for gapless on Chromecast somehow, or perhaps someone else could through a rendering plug-in, though I'm not sure the plug-in interface supports doing something like foo_upnp.