Forum Replies Created
- AuthorPosts
-
Thanks π
The beauty with that RJ45-ML adapter thing is that it can be made quickly and at pretty low cost.
I don’t really see this for the DIN based PL connectors using a similar approach.Indeed the BS3 is a nice unit. Only the picture output is barely usable by todays standards. Not really because of the HD-only resolution – but it does some post-processing even to the projector output that just doesn’t look right on modern OLED.
As sound processor it still does work really well. Just set up a matching source and feed it with S/PDIF (PCM or Dolby Digital) from the modern TV. Couldn’t notice lip sync issues either.
Still wondering about the best remote integration in that setup. Probably best to get rid of the ugly LG standard remote altogether. Most IR remote events the ML master receives are not directly sent into the ML bus. So maybe the most straight forward solution would be using a generic PUC in the BS3 and externally convert that to IP commands that control the LG TV accordingly. Let’s see, something for a bit later…
Thanks Guy, interesting input. Can’t validate that with the BC9500, unfortunately.
One cool addition to my setup here.
Have a modern LG OLED TV connected to a BS3 (which only handles audio) and a few BeoLabs.
Once I switch on the LG TV, my little ML adapter box notices that and automatically switches on the BS3 on the correct source as well. If the LG turn off, it will also turn off the BS3…If BV10 has a recent enough firmware installed it will forward all LIGHT commands to ML. From there you can pick it up with e.g. a MLGW and trigger commands on other network connected device. Probably Philips Hue or similar can be made controllable through the Beo 4 like that.
For good measure I just did a test with a Beogram 2000 connected to the AAL connector on a 1611 converter, and a Masterlink cable to a Beolab 3500.As expected the Beogram did not react to any audio commands sent to the Beolab 3500.Ah, great. Thanks for testing. Indeed I must have mixed up something then. So the 1611 cannot switch between the two different datalink protocols.Iβm idly curious on whether BS5 fails to broadcast this information to ML, or perhaps BLC NL/ML strips our the data, or is it the B&O app that ignores the data (if available) and simply displays an inane βPlaying from N.Musicβ, etc. TIA,
Yes, I can confirm that BS5 broadcasts track and station info to ML by default. Must be the BLC / NL side that does not take care of that.
Can confirm, currently hacking around with what I believe is the Datalink 86 protocol on my Beosystem 5500 ( so CD / TP / TP2 / PHONO ). TP and TP2 protocol are identical, except that TP2 has no support for running status (to provide the MCP5500 with live display). CD and PHONO also seems identical, but I havenβt pushed the investigation enough to confirm that.
How I understood it is that DL ’80 is used for that CD TP PHONO stuff.
Ah, true. With external CD unit of course track status info needs to be transmitted as well. Looking forward to your findings. Can’t test that here with the BC9500.The AAL port uses DL ’86 which has proper addressing and is similar to the IR commands. It is pretty well documented already I think.
True, could be.
Was testing with a BC9500 here. So no CD / TP1 sockets.
Phono and TP2 commands are looking different but could be that Phono / CD and TP1 / TP2 are the same.I could be wrong. Maybe I mixed up something regarding that.
Indeed the two protocols are pretty different. Sometimes they are also referred to Datalink ’80 and Datalink ’86.
The first generation is used for talking to the Beogram / Beocord peripherals. Just two identical bytes being sent back and forth.
Data pins of the TP2 and the Phono are always connected in parallel. I guess two or maybe even more bits are some kind of simple device address while the other ones are the control command. Haven’t investigated that in depth as it’s simple enough to record and just play back the commands when needed.Great, thanks for your input!Yes, both scenarios would be possible with such a USB connected light version.One use case for both DL and ML could be to actually control a Beogram from a Masterlink only product (BS Ouverture, BC2, BS4, BS5 and Beovisions with ML). Would probably need the use of some alternative source commands to make this work
True, thought about this one as well. Not exactly sure if I remember correctly but couldn’t you just use a 1611 converter for using a turntable in a ML setup (Phono / N.Radio source)?
That sounds interesting. Which use cases could this scaled version be used for please? Is there a web interface or app that would allow to control ML or Datalink audio systems via WiFi and to stream these audio sources via AirPlay (there will not be a network RJ45 connection?) Beyond a missing RIIA preamplifier which other functions will be missing from the light version?
Probably similar to that old PC2 interface. No direct network connection or streaming – the host system needs to take care of that.
A possible “light” version would cut away the processor including the network interface as well as the nice housing.
Maybe leaving the RIAA integrated makes a lot of sense. Probably better splitting it up in a dedicated DL and a dedicated ML version. Both at the same time isn’t necessary anyway.Technically it would still be capable of the same functions like mentioned for the “full” version. Just not fully integrated. The end user would be required soldering some connectors, putting it in a standard housing and setting up the software. Could supply some software examples how certain things can be done…
Had some more thoughts about this one.
What could probably work out is developing a light version sold as a semi-finished kit to DIY people.
So just the ML and DL interface with a single USB connection for power, audio and data.Anybody interested in that?
Happy to donate a certain amount to beoworld for every unity sold or something similar.Open sourcing the whole thing would not be that much useful. For the average DIY people it will be much too difficult to build – mostly because of the processor part.
Given the risk of low sales numbers – making a regular product out of that would result in being somewhat on the “expensive” side. Probably more than the devices it can control are worth on the 2nd hand market nowadays.
Probably not a huge issue if it would be an original beo product. Nevertheless in the end it would be a 3rd party unit and as matador wrote a few days ago I also see this as a critical point.Stunning result you have here, congratulations ! I am very curious about that enclosure ! Is that something you designed ? Or is it an off the shelf part ? Full disclosure here, I have an ongoing project that bears some resemblance with yours, but I have already settled on an more conventional enclosure.
Thanks! Cool that you are working on something similar. Looking forward to your project. π
Yes, the enclosure was entirely custom made for this one. Two CNC machined and anodised aluminium shells combined with two 3D printed parts. The one on the front is a nice SLA print. On the inside there are two indentations for the wireless flex antennas. The one on the backside currently is an FDM print – will probably update it to a metal one some time in future.
This is a thing I never understood: why nobody made a little black box to stupidly airplay broadcast any analogue signal, even under the coat. The same way Iβve never understood why nobody made the same thing for light control. Anyway, I wish you good luck with your project and in any case it made that forum shiver again and only this is a beautiful achievement. Welcome to Beoworld B3OHACK3R.
I really don’t know as well. There are already several tinkering projects for Datalink and B&O IR stuff on the web. Why nobody made the airplay thing before, no clue.
Sure, thanks for that warm welcome. π
@MM as I said I’m not looking into start making them and offer them for sale here.
The list why this would be not a good idea is long. One could probably write books about it…Yes, probably a good idea not turning this into a “political” discussion. π
A few good pro/cons have been given already.
Otherwise such things have the potential filling up multiple pages without any meaningful new info.
From the commercial side I’m into such things already for too long to ignore the many small and easily overlooked details that have great potential killing your product in the end.Hi,
I would not go so far and say that I reverse engineered the ML protocol. It was done (probably even multiple times) before. For example have a look here: https://github.com/giachello/mlgw
It uses the capability of the MLGW to send and receive raw ML commands.What is probably more or less new is an independent receiver / sender. The only trick here was to understand what the parity bit does. For receiving this can be ignored but for sending you have to use it. A mark parity bit signals start and end of an telegram while space is used for anything in-between. Apart from that it’s just a differential serial bus running at only +/- 0.25 V.
The addressing and also that lockmaster stuff (which also is already documented on the web) can mostly be ignored if you are just “simulating” a link node. Pretty fault tolerant I would say.Since you have the experience, could you tell use how difficult it would be and if you have any insight on why this is something that never came up?
Happy that you like it π
I think there are commercial software solutions available that can act as a airplay sender. For example the one for Windows that can send audio to any airplay enabled speaker. It probably is neither approved nor an official implementation. Also that open source application I’m using here is around for a pretty long time already.
Why such a product was never a thing – I really don’t know. Of course a “big player” can’t use that at all in an official product. But why niche businesses never came up with that…How difficult it would be? Technical wise it entirely depends on the point of view. What I consider easy could be difficult for others and visa verse. So hard to make a statement on that one.
As with every product the difficulty lies mostly one the financial risk. Nowadays the usual go-to-market duties besides the development itself can leave a huge hole in your pockets. Especially when sold internationally it can easily screw up your whole product pricing if you are not selling it in large numbers. Not even to mention customer support and marketing.It is hard for me to say anything to the presented projectβ¦.other than that it is cool! Lots of respect for that!
Thanks for that, MM π
Also see the explanation above. Me speaking as a company – we are not set up for selling products to end-customers. Been there already, got the t-shirt, didn’t make much sense.
I really wanted to build that thing for a long time and the project was then released on the website to show an example of how we are working. So to at least mentally justify a bit of the effort that went into it.
So for sure the project itself has some kind of marketing idea attached – but not for anything happening here. Just wanted to share it because I think it’s cool to show.Happy to answer any technical questions.
Glitch,
thanks for the praise π
Audio wise the “non-B&O world” currently can be served either by the AirPlay protocol or by a html webstream. For both I’m using a really nice open source software called owntone.
For the control part there is a simple html5 webapp that can be easily added to the homescreen of your smartphone. One can select all ML or DL sources from there and direct it to the desired output device.
There are many scenarios that are tested in hardware but not implemented in software yet. It could e.g. act as a “source center” and handle N.Radio and N.Music requests on the ML bus. So for example on an incoming AirPlay stream from your smartphone it could automatically switch on the audio master and inject the music stream accordingly.
Also it can map many functions to custom buttons on a BR Halo. So that one does not depend on the webapp only. It was implemented using the public available Halo python library.
- AuthorPosts