Print

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

[Nabor]

I am using a Denon DNP-720AE in two modes 1)  Music Server Mode to access my flac library.  When I play Beatles  Sgt Pepper (a Gapless CD originally) by way of a JRMC18 server it plays Gaplessly like the original CD. no Problem.

Mode 2): When I invoke JRMC18 on my actual "server" and "push" or "play to" Sgt Pepper to the Denon unit by way of DNLA, it plays the flac files in sequence BUT adds at least a 6 second MegaGaps between tracks/files. Why this difference and is there a way to get it to play at least closer to Gapless?  This is DNLA mode and adjusting prebuffering times has no effect.

I have also have no problems gapless or otherwise when using JRMC18 direct to various DACs.

Thanks anyone.    

[csimon]

Please see http://yabb.jriver.com/interact/index.php?topic=69920.0 - it looks like Denon players don't support gapless playing by DLNA.

[doulos]

Please see http://yabb.jriver.com/interact/index.php?topic=69920.0 - it looks like Denon players don't support gapless playing by DLNA.

actually, I have to withdraw my statements made in that thread. In march, I received a firmaware update for my Denon DNP-F109, which did add the often mentioned SetNextAVTransportURI action to the UPnP/DLNA client. I have since tested this many times and am truly enjoying gapless playback on my live albums. I have even bought an additional Denon CEOL Piccolo device for my office. One other thing that makes them very special BTW is the fact that they keep broadcasting their presence while in standby, and automatically wake up when sent a UPnP command. Extremely nice.

I am rather confident that the DNP-720AE shares the same firmware, but cannot say for sure. In any case, there must have been an update recently. Until then, gapless was not possible via UPnP/DLNA, which was also officially confirmed.

regards,
chris

[csimon]

Well, that sounds positive!  I guess it remains for Nabor to ensure he has the latest firmware?

[Nabor]

Perhaps I should have clarified that both Mode 1 and Mode 2 described in my original post use DNLA.  (I assume that any ethernet only device that shows as a "player" device in MC uses DNLA).
Originally the DNP-720AE did not support Gapless in Mode 1 as you note,  but now it does (see original post).   Maybe the Denon design team read my review on Amazon!!

What we appear to have here is asymmetrical behavior between mode 1 (GAPLESS) and mode 2 (MEGAGAPS).
I suspect that in Mode 1, the Denon is the control point and in Mode 2 JRMC18 is the control point.

Another way to view this: Would your gapless Denon Piccolo play gapless in both what I [incorrectly?] call Mode 1 (Initiate DNLA stream FROM Piccilo box) and Mode 2 (initiate DNLA stream TO Piccolo box from MC).  Can Piccolo play flac files in both Mode 1&2 gapless on Sgt Pepper (or live albums, etc) ?   If not, why not?  Is it related to an unimplemented SetNextAVTransportURI action on MCs end?

I like the the standby wake up also, especially on Haloween.   
Thanks

[doulos]

I think your question was quite clear. Your device plays gapless when you initiate playback from the device itself (using the remote and the built-in display), but not when initiated externally by a DLNA control point. As I mentioned, this was also officially admitted as a limitation by Denon support. I also mentioned that this was fixed by a recent firmware update.

HOWEVER: I have just reread the Denon FAQ which I had previously quoted (http://denon-uk.custhelp.com/app/answers/detail/a_id/1427/~/gapless-function), and found that it has since been updated to reflect the changes in the new firmware. Unfortunately, the update seems to not have been applied to all devices. According to the FAQ, the limitation remains for the AVR-1713 and DNP-720AE. Too bad!

BTW, I also posted a furious review when I found that the gapless support that was promised for my device only applied when I was kneeling in front of it with the remote in hand. I guess (am quite sure) your device is one generation older, so you even had to wait for the basic gapless support, and now you are left behind with gapless via DLNA. Funny thing is that the DNP-720AE is still sold with a higher profile than the DNP-F109..

regards,
chris

[Cebolla]

Presumably this lack of gapless support when the streamers are used as a DLNA renderer, but present and working in playback mode following a firmware update, is related to a particular hardware deficiency, since it appears that most (if not all) streamers of a certain generation and price point across all manufacturers still have this problem, eg:
Cambridge Audio's NP30 & Stream Magic 6
Marantz NA7004
Pioneer's N-30 & N-50
and now the Denon DNP-720AE

Does anyone have a clue as to what this could be, especially relating to possible chipsets involved with DLNA?

Or am I being too kind and is it just that all these manufacturers' technical departments have coinincidentally had the same brainless post design fix strategy by providing the same half baked solution for gapless support? Or/and are all using the same third party software support, perhaps?

BTW, I also get a nice spooky DLNA controller wake up out of standby on my Pioneer N-50, so these guys do get most things right!

[JimH]

Probably firmware.

[JimH]

It's DLNA, not DNLA.

[Cebolla]

Quick double check to make sure that I hadn't also transposed the letters, easy enough to do due to same phonetic 'e' start sounds of letters involved. Know someone from childhood who's still got the aminals/animals speech impediment and whose first instinct is to write it the same way he says it!

[Cebolla]

Probably firmware.

Hopefully that's the case, so should be no excuse for not providing a 'full' fix at some point. Strange how they've all done the same thing, though.

[Nabor]

Thanks gentlemen but I'm not quite persuaded the issue is 720AE firmware.  As a former Intel sw/hw inter-company developer, I'd be real reluctant to rely on FAQs as a substitute for reality. The FAQs can be delayed, have transcription or other errors etc.  I certainly would consider FAQs, but in this case Mode 1 went from injecting a 1 sec gap to gapless,  a carefully monitored fact.

Currently, there is no evidence that the 720AE is any older/younger than the Piccolo. You may want to read my detailed description of the 720AE chipset on my (US) Amazon review.  Chris, I'd love for you to do a similar chipset breakdown of the Piccolo.  But I think that's an extra credit side issue.

One issue that should be clarified is that I've only used MC to initiate DLNA, I do not know that "any" control point fails to give gapless or gap perfect results. 'll be looking into that.

From your cited FAQ: "If you use the Windows "Play To" - or other media renderer features (Android music streaming, control by DMC, PlugPlayer, Media Connect, etc.) and use the Denon Remote App, gapless is not available."  [But otherwise its available].  Thats how the recently upgraded 720AE works.

With the above quote in mind Chis,  I'm a little unclear as to your findings with regard to your Piccolo:  What I call Mode 1 supports gapless/gap perfect play back, yes/no.   What I call Mode 2 supports gapless playback, yes/no. Feel free to correct my terminology.

My current concern is that with MC you're never quite sure if there isn't some setting somewhere that is the real culprit.  Perhaps it exists, but I'd love to see a user accessible Config file that can be exchanged between users or different setups.  

  

  

[JimH]

If you're not aware of it, gapless is one of the weak points of DLNA and most devices don't do it correctly.

[doulos]

Ok, Nabor, I feel I have wasted some of my time here. As a final suggestion, I would recommend to install the UPnP device spy software (from intel, so you should find your way there), and check if your device's AvTransport service has the SetNextAVTransportURI action. If it doesn't, then it will never do gapless over UPnP. If it does, we can talk further.

cheers

[Nabor]

JimH

If I recall correctly, the dbPowerAMP developer said that the DNLA protocol didn't support gapless.  Thus I was surprised to see gapless suddenly supported with a firmware change at least in one mode.

But keep in mind the issue isnt just gapless v gapped, its Megagaps.  Whats going on for 6-10 seconds between tracks in Mode 2.    

[Nabor]

Thanks Chris  - You assume I'm in love with the 720AE I'm not. I could be falling in love with the Piccolo.  
For me, This isnt about the hardware - that can be changed.  Does MC support gapless in mode 2 to your Piccolo or any hardware Linn etc.  
 

[JimH]

Please consider what doulos advised above.  You could search for other discussion on the same subject.

[Cebolla]

From your cited FAQ: "If you use the Windows "Play To" - or other media renderer features (Android music streaming, control by DMC, PlugPlayer, Media Connect, etc.) and use the Denon Remote App, gapless is not available."  [But otherwise its available].  Thats how the recently upgraded 720AE works.

Nabor, I believe you should have another look at your interpretation of this:

"If you use the Windows "Play To" - or other media renderer features (Android music streaming, control by DMC, PlugPlayer, Media Connect, etc.) and use the Denon Remote App, gapless is not available."

That's precisely your mode 2, pushing files from the DLNA server (the DMS in MC) to the 720AE, via the DLNA controller (the DMC in MC or the Denon DMC app or any other DMC for that matter), also known as renderer mode;

[But otherwise its available]

and that's your Mode 1, pulling files from the DLNA server (the DMS in MC) directly by the 720AE for playback, using the remote control or the buttons on the box, also known as player mode.

To summarise: Mode 1 = control directly by the 720AE, 720AE is 'master'; Mode 2 = control by the DLNA controller, 720AE is 'slave'

[Nabor]

Cebolla  - That's exactly right and that's been my interpration through out this thread.

That doesn't answer the basic question of this thread 'Does MC support gapless "Play to"/Mode 2 on any hardware?

...Another way to view this: Would your gapless Denon Piccolo play gapless in both what I [incorrectly?] call Mode 1 (Initiate DNLA stream FROM Piccilo box) and Mode 2 (initiate DNLA stream TO Piccolo box from MC).  Can Piccolo play flac files in both Mode 1&2 gapless on Sgt Pepper (or live albums, etc) ?   If not, why not?  Is it related to an unimplemented SetNextAVTransportURI action on MCs end? ...

I was already aware of his solution to an issue that is doulos', not mine. We''ve all wasted too much time on speculating whether the 720AE supports "gapless" in unspecified modes, but have not answered my basic question.   Doulos, or anyone, would be more helpful if they answered the questions asked.    
 
Anyway as a new customer, this was an educational exercise.

[doulos]

I'll take a last stab:

1. MC supports gapless via UPnP only to devices that implement the SetNextAVTransporURI action
2. the Denon FAQ (which I have found to be absoultely reliable so far) says that the 720AE firmware does not implement SetNextAVTransporURI
3. therefore, no matter what UPnP control point you use: no gapless playback over UPnP on a 720AE

It remains for you to double check point 2. If you find there is a SetNextAVTransporURI action, then MC might be doing something wrong. Until then, the statement from the FAQ holds, which says, I repeat: the 720AE does not support gapless via UPnP

this is my favorite subject, which you can find when searching the other related threads. That may also explain why I dont mind repeating myself so often.

BTW: I am in no way affiliated to JRiver, other than being a customer and having been admitted to their beta testing program. The only official here is JimH

[Nabor]

doulos

I appreciate the last stab, but the question you are attempting to answer is not my question or a question I care about.  

You think the question is "Does my Denon DNP-720AE support gapless playback (mode unspecified)?"  

My actual question is: Does MC Support gapless,  DLNA playback from MC  ("Play to"/Mode2) function for ANY hardware on planet earth (e.g. Piccolo, Linn)?
You don't need to know a thing about SetNextAVTransporURI action to answer the real question.

You own a Piccolo. What happens when you "play to"  Gapless flac files by way of DNLA FROM MC?   6 sec Gap added or not?

The diagnostic tools will be useful for other questions I have, but not needed for this issue.

You can hop a plane and come to my house and listen to the DNP-720AE play gapless in Mode 1 anytime.     





    
 

[csimon]

I think your question is ambiguous or meaningless, as the answer is both Yes and No.  The use of gapless in "mode 2" implies support for the SetNextAVTransportURI protocol.  Let's break the question down a bit:

Does MC support gapless via DLNA push?  Yes.
Will gapless via DLNA push work with hardware that doesn't support gapless via DLNA push? No.

Does that make it clearer?  MC's ability to support gapless isn't affected by the renderer's ability to do it.

Does MC support sending MP3 to any hardware?  Yes.
Will any/all hardware play MP3? Not necessarily.

Does my PC support a HDMI/DVI connection to a TV?  Yes.
Will it work with a TV that doesn't have HDMI/DVI? No.

Do I have the ability to speak Welsh to any person on the planet?  Yes.
Will I make myself understood if the other person doesn't understand Welsh? No.

My actual question is: Does MC Support gapless,  DLNA playback from MC  ("Play to"/Mode2) function for ANY hardware on planet earth (e.g. Piccolo, Linn)?
You don't need to know a thing about SetNextAVTransporURI action to answer the real question.

You do, because gapless DLNA playback requires SetNextAVTransportURI. MC does it, but not all renderers do it. MC either supports DLNA protocol or it doesn't.  It does do it, unconditionally, regardess of the renderer.  It doesn't drop support for the protocol if the renderer doesn't support it.

You can hop a plane and come to my house and listen to the DNP-720AE play gapless in Mode 1 anytime.

I don't think anyone is disputing that, it's stated in the FAQs that you are suspicious of, and you want people to believe you when you prove that the FAQ was correct even though you're suspicious of it! It's also stated that the 720AE does not support Mode 2. And MC does support it.  It just aint gonna work.  You need a device that supports gapless via DLNA push, MC's support for the protocol is not in question.

You own a Piccolo. What happens when you "play to"  Gapless flac files by way of DNLA FROM MC?   6 sec Gap added or not?

I think he's already answered this: " I have since tested this many times and am truly enjoying gapless playback on my live albums. I have even bought an additional Denon CEOL Piccolo device for my office."

[doulos]

You own a Piccolo. What happens when you "play to"  Gapless flac files by way of DNLA FROM MC?   6 sec Gap added or not?

ok, since this was actually a test scenario I hadn't gone through previously, I checked it also. I normally only test with my MediaSteersman android app, because that is what I really care about  .

The quick answer to your "6 sec Gap added or not?" question is "NOT". However, I am getting a clearly distinguishable 1/2 second gap. I just checked with my app again, using the same album and track (Eric Clapton's "One More Car, One More Rider" Live album), and here I have absolutely no gaps. The difference is very clear. When controlled from MC I can also see that the display of the device flickers several times during track changes, while when controlled from my app, the display smoothly moves to the next song. So I think MC is doing something strange (like sending intermittent STOP commands) on track changes. In any case, gapless via DLNA from MC to my Denons does not work, even in the presence of the SetNextAVTransportURI action.

Now, with regard to your 720AE, I have to repeat that there is at most a 1/2 second gap between songs when controlled from MC. Moreover, I am still quite sure that the 720AE firmware does not implement the SetNextAVTransportURI action, and thus will never do true gapless over UPnP. You should be able to get down to bespoke 1/2 second, however.

As a conclusion: it seems to me that MC is doing something that is not compatible with Denon  firmware, be it the non-gapless one or the new gapless one. It might be that this incompatibility leads to the 6 seconds gap on the old firmware, and blocks gapless on the new firmware. I might mention that I myself had to adapt my app to cater for Denon's rather specific requirements (mostly in regard to timing and sequence of commands). I thought I had my gapless support ready (tested on a Sony CMT-GNiP) when the new firmware for my DNP-F109 and Piccolo arrived. I joyously ran it, only to find that my app didnt even advance to the next song at all. Fixed that and now have a mode that works on all gapless devices I know (fingers crossed for the other ones). UPnP/DLNA is true fun! I already mentioned most of this on the other threads as well.

cheers,

[csimon]

UPnP/DLNA is true fun!

He's lying!

[Nabor]

Csimon.

There is nothing ambiguous abut the question, unless you incorrectly assume mode 1 and mode 2 are interchangeable.  Its very, very clear in my question that the answer is both device and MC dependent (excluding the welsh language.).

The dialog to date on this thread has assumed that JRMC18 has perfectly implemented the "Play to"/Mode 2 scenario.  As Doulos has nominally demonstrated,  we now see that is an invalid assumption and what I was getting at all along.

My general sense is that the scenario that I was inquiring about has not been thoroughly investigated or reported by the current posters to this thread.  They were answering the question they had an answer to, not the question posed.

You will be happy to know that the Denon DNP-720AE also now supports flac file seeks again only from Mode 1 or as Denon describes it in "Music Server" mode.   

[csimon]

As Doulos has nominally demonstrated,  we now see that is an invalid assumption and what I was getting at all along.

No, you were asking whether MC supports gapless with DLNA push with any renderer.  The answer is still yes, provided the renderer also supports it.

Your first question was if there was any way to get the 720AE to perform gapless by DLNA push, and the answer is still no.

[Nabor]

Doulos

Sounds like we are making progress.   As far as I am concerned if its .5 second it might as well be 10 seconds. I certainly would not want to upgrade to a Linn until that was resolved.

From remote residential locations, as a work around,  I can always use JRMC18 in Waspi mode via an NAD Wireless DAC 1 to achieve "gapless" which I really like to think of as "gap perfect" as you get the original producers CD gap.

I like the Sgt Pepper CD as a gapless test because it is widely available world wide even though it has gone "out of style".  It also has a mix of gapless and gapped tracks.    A good validation test for ones flac file capture.

I do have one last question, if you have the time.  

While I'm sure there are significant blog discussions on the topic what are the authoritative sources of your belief that "Play To"/Mode 2 requires the SetNextAVTransporURI action for "Play to"/ Mode 2"?
Clearly that action is not required for Mode 1 gapless as the Denon 720AE I have plays gapless in mode 1 but does not currently have the SetNextAVTransporURI action. I get your point on "Play 2"

[Nabor]

CSimon

Your first question was if there was any way to get the 720AE to perform gapless by DLNA push, and the answer is still no.

Disagree. Thats the question you wanted to answer.  I don't say that anywhere.  And even if I did originally, I certainly clarified it numerous times.
Keep in mind the easiest thing to do was to blow the whole issue off as MY Denon hardware problem.

[doulos]

While I'm sure there are significant blog discussions on the topic what are the authoritative sources of your belief that "Play To"/Mode 2 requires the SetNextAVTransporURI action for "Play to"/ Mode 2"?
Clearly that action is not required for Mode 1 gapless as the Denon 720AE I have plays gapless in mode 1 but does not currently have the SetNextAVTransporURI action. I get your point on "Play 2"

this is not a matter of blog discussions, as little as it is a "belief".  If you really want to get into it (as I have, given that I have implemented a UPnP control point for android that handles gapless well, to Denons at least), the authoritative reference are the UPnP specification documents available at upnp.org, specifically the documents at http://upnp.org/specs/av/av1/.

The 720AE can do gapless in Mode 1 because that does not involve a UPnP control point - only the device accessing the server via UPnP. Dont ask me why this makes such a difference. In my mind, a device that can play gapless when pulling media from a server should also easily be able to do the same when controlled via UPnP. But as it stands, the manufacturers seem to have difficulties here. I am glad they at least soved it for my 2 Denon babies here..

[Nabor]

Thanks Doulos    A little bumpy ride but I think we are on the same wavelength now.  I'll check out the  UPnP spec.

I AGREE strongly with your observation that you dont see why where you initiate an interaction between the two "intelligent" entities matters.  
When I first sensed asymmetry my reaction was:

   Is there some way to use MC as a substitute interface for the Denon interface.  I doubt it but its a thought.      

On that .5 second issue in "Mode2".  As I note in one of my posts, you never know with MC and kind if its an implementation error/choice  OR a user setting oversight (e.g. stream from memory or stream from disk).  I'll be checking to see if the issues is resolved with  settings or implementation fixes.    

What I'd like to see is a Denon Network Player that is Made in Japan.   please let me know if any of your Denon babies are Made in Japan.

[bob]

I don't know why we'd be having any trouble with devices that support SetNextAVTransportURI. We send the next URL as soon as the renderer will accept it.
The only thing I can think of is some sort of conflict with Transport Events. You could try disabling them for the zone and see if that makes a difference.

[doulos]

I don't know why we'd be having any trouble with devices that support SetNextAVTransportURI. We send the next URL as soon as the renderer will accept it.
The only thing I can think of is some sort of conflict with Transport Events. You could try disabling them for the zone and see if that makes a difference.

as I said, I also thought I had it all working, until I ran into the new Denon firmware. To get it working I had to change the point where SetNext was sent after the initial Set. I dont think this has anything to do with events. What kind of conflict would that be?? It would be rather pitiful if one had to perform all kinds of tweaks to get something that basic to work. Anyway, I'll try to tweak a little tomorrow an report.

[doulos]

You could try disabling them for the zone and see if that makes a difference.

ok, I've run the test again, this time with eventing turned off (SetNext was turned ON).  Same result: about 1-2 seconds gap (not only .5, which would have been hardly recognizable). With Denons own app, as well as with mine, gapless is flawless  . For the fun of it, I also tested with BubbleUPnP, which also worked perfectly. Finally, I can  assure you that SetNextAVTransportURI is present, as it is detected by my app and also by the UPnP Universal Control Point tool.

Again: there is also a visible difference, as with MC the track display on the device flickers twice while changing tracks, which it does not when controlled from any of the other apps. In my mind, there is no other way than to admit: MC does not do gapless with a Denon CEOL Piccolo or DNP-F109. I was almost tempted to capture the network traffic going on between MC and the Denon, which I am sure would have uncovered the issue. But that would take quite a while, and its not my business (need to fix my app, dont care about MC as a controller). I am willing to test, however

[bob]

as I said, I also thought I had it all working, until I ran into the new Denon firmware. To get it working I had to change the point where SetNext was sent after the initial Set. I dont think this has anything to do with events. What kind of conflict would that be?? It would be rather pitiful if one had to perform all kinds of tweaks to get something that basic to work. Anyway, I'll try to tweak a little tomorrow an report.

Plenty of these devices are rather pitiful!  
Anyway, thanks for checking!
If someone has one, wants it to work gapplessly and would like to send me a wireshark trace of the interaction, that'd be great.

[bob]

doulos, I've tried to install Media Steersman on a few different devices without success (like a Galaxy Nexus which has the same resolution as many tablets and a generic Android HDMI stick). I really would like to get it going so I can use it for testing. What is the installer looking for? Thanks...

[doulos]

I really would like to get it going so I can use it for testing. What is the installer looking for? Thanks...

it requires at least a 7" tablet. Resolution does not matter.

I can pull you a wireshark trace and send it. I just dont want to do the analysis part.

[bob]

it requires at least a 7" tablet. Resolution does not matter.

I can pull you a wireshark trace and send it. I just dont want to do the analysis part.

If you wouldn't mind, that would be great, thanks!
bob (at) jriver (dot) com. Our email should be able to handle just about any size attachment.

[JimH]

Bob,
There is a Google Nexus on my desk.

Jim

[Nabor]

Its an issue that can be deferred for now, but as a reminder the thread title and original post indicate, this is also about MEGAGAPS.

Fine, some magic command is missing from the renderer/device that prevents it from doing Gapless from MC.   But why the 4-6 second gap between tracks in my situation?

If I had reported 5 minutes between tracks, that would have raised red flags.  From my perspective, 4-6 seconds in the high speed digital world is equivalent to 5 minutes.

At a very high level of abstraction, it feels like the Renderer is waiting for MC and MC is waiting for the Render and so somebody times out and we then move on to the next track.  That explanation could be dead wrong but its historically a common error.  

Bobs comment "We send the next URL as soon as the renderer will accept it".  Raises the high level question, how do you know its ready to accept? Signal from Renderer or query from MC?
But I'll use that as an exercise to getup to speed on UPnP and Wireshark.  

OBSERVATIONS (UNRELATED?):  These need more work and unfortunately I upgraded Denon Firmware and MC at about the same time.  1) Now MC initially will only play a single track to the Renderer until its "trained" with a "previous track" and "next track" action.  [See Subject: Bug? DLNA AVTransport event shows previous track] 2) The Denon will occasionally appear as a DLNA resource but disappear when you try to use it within MC. Rebooting the Denon resolves the issue.        
    

[bob]

Bob,
There is a Google Nexus on my desk.

Jim

Tablet??
Isn't that a phone like Matts? Google has the absolutely most stupid naming "protocol".

[bob]

...
Bobs comment "We send the next URL as soon as the renderer will accept it".  Raises the high level question, how do you know its ready to accept? Signal from Renderer or query from MC?
But I'll use that as an exercise to getup to speed on UPnP and Wireshark.  

This point was meant to address Gapless playback. It means that as we are polling the device while playing the current track, as soon as it's able to accept a SetNext command, we send it.

The logic for playing the next track on a system that doesn't support SetNext is totally different. This requires a STOP event or Poll to transition through the state necessary to send the next track.

[Nabor]

This point was meant to address Gapless playback. It means that as we are polling the device while playing the current track, as soon as it's able to accept a SetNext command, we send it.

The logic for playing the next track on a system that doesn't support SetNext is totally different. This requires a STOP event or Poll to transition through the state necessary to send the next track.

Thanks Bob, that's  informative.  So many device/renderer interfaces so little uniformity.  Sounds like a committee was involved.  In the NOT SetNext scenario polling "sounds" slower than an event, but what do I know about UPnP.

Only if you have time,  is there a way to characterize WireShark for DLNA/UPnP chat only?  I've got a few filters in mind but was wondering if those in the know use a specific widely available filter. 
 

[bob]

Thanks Bob, that's  informative.  So many device/renderer interfaces so little uniformity.  Sounds like a committee was involved.  In the NOT SetNext scenario polling "sounds" slower than an event, but what do I know about UPnP.

Only if you have time,  is there a way to characterize WireShark for DLNA/UPnP chat only?  I've got a few filters in mind but was wondering if those in the know use a specific widely available filter. 
 

I basically just do the trace putting in a capture filter (not display filter) between the 2 device IP's and wade through the events and polling before and after a track is played, I don't think there is a way to actually slice out the audio file from the capture to make the capture smaller...

[doulos]

Only if you have time,  is there a way to characterize WireShark for DLNA/UPnP chat only?  I've got a few filters in mind but was wondering if those in the know use a specific widely available filter. 

would be hard to create such a filter, as UPnP uses so many different underlying protocols (nicely illustrated in this https://upload.wikimedia.org/wikipedia/de/0/08/Upnp_architecture.svg picture)

-C

[Nabor]

Nice Protocol Diagram.  Like I said, looks like a committee was involved.  I found that even inside Intel there were often small tribal wars over such things.
Think what could be done with a ground up or near ground up protocol.  But I digress.

That explains why I could not easily find any such tool.  Thinking out loud, it would seem something like a hardware/rederer emulator might do it because it has to decode all that in to basic actions at some point.  (Big job though and only considers the one endpoint).

Thanks again for the insight.     

 

[icstm]

Does MC support gapless via DLNA push?  Yes.
Will gapless via DLNA push work with hardware that doesn't support gapless via DLNA push? No.

Sorry if this is at a tangent, but is this different to wanting supporting for "synced" playback?
I have trouble getting 2 zone in MC playing in sync if connected via uPnP. I thought this was a problem with uPnP, but gapless might be related? (somehow?)

[doulos]

I have trouble getting 2 zone in MC playing in sync if connected via uPnP. I thought this was a problem with uPnP, but gapless might be related? (somehow?)

no, not related. Gapless is built into the UPnP architecture by means of the SetNextAVTransportURI action (and/or other means like playlists). Synced playback is explicitly NOT specified in UPnP. There is an extension related to syncing in the DLNA spec, but no one implements it that I know of. And there are prorietary implementations on top of UPnP, like Sony's "party mode". If you want syncing based on some implemented "standard" protocol, you'll have to go to AirPlay (hate to admit that!). I'll let JRiver comment on whether they have something proprietary built into their renderer which would enable synced playback (when using MC as renderer, that is)

[bob]

It seems from our testing that playback over DLNA is not deterministic enough between different devices to allow synchronized linking of zones. I've tried it several times (there is a mechanism to shift the delay between zones when linking) without out a lot of success.

[doulos]

With regard to gapless on the Denon CEOL and DNP-F109, heres my findings:

after taking a wireshark trace of the communication between MC and the device, and analyzing the trace, I found out that MC was sending MP3 files all the time! Everybody knows that MP3 and gapless don't go together (fixed frame sizes), and the Denon docs cearly state that gapless is only supported with lossless formats. Now, the devices do advertise correctly via UPnP that they support FLAC and other lossless formats, which is also honored by my app (MediaSteersman), and other control point apps I tested, by sending the best supported format (FLAC in this case). Only MC decides to send MP3!

So I got back to one bob to find out what this was, and I learned that MC indeed simply ignores the supported formats advertised by the device. Bob then wrote a brief note about an option I could switch off to force MC to send the native file format, but his explanations were so unclear that it took me 4 iterations to understand what he meant (NOTE: I am not interested in MC's control point capabilities. I was doing this out of courtesy. I am developing my own control point app on android, and when I once asked them to help me test it, I never got a single line of feedback).

The end result is this: you have to manually create a DLNA server config in MC that does not transcode the media. You then have to manually associate that server with the renderer in MC (look at the renderer context menu). Now MC will send the un-transcoded file format, which in my case was FLAC. And guess what: once the devices recieve lossless data, gapless works perfectly!

P.S.: bob, enjoy your hat 

[bob]

Hi doulos,
Thanks for checking the Denon.
Sorry about the feedback, until last week I wasn't able to get a tablet that would run Media Steersman. I've got one now and it's nice to have another control point to test with. Will let you know when I run into issues, the one I mentioned with the Bravia TV is the only one so far. You've got me guessing on the Hat