I’ll use the DT for some of the answers because is the machine that inspires the idea for me.
So you can have every track on different channels to pan, eq, volume, send/return, etc.
Yes, some kind of routing to internal channels.
I don’t know hahaha
Maybe just set the same for every input (?) hopefully.
It depends on the implementation I guess. I’ve never listen too much difference between recording analog outs from the DT to the recordings through Overbridge, but it is a subjective thing I guess.
The time and money Elektron invested into Overbridge could be useful for a hardware solution mixer that allows to connect DN, DT and all the others and run them either stereo or individual outs. They would all sync to the box and provide some stereo outs also via USB. Analog Heat could be run as send efx or in the master track. I would totally buy that. Today. Maybe different versions for small or large setups with the corresponding price tag…
I think with more and more Groovebox which are able to transfer multi track sound with USB there is something here.
Usb cable can be really long, and if you have a Groovebox with 8 internal track it let you export each track separately on the same cable.
The beauty of a thing like this is you can divide by 8 the number of cable in the end with the best quality possible.
You loose a bit of latency doing that, but if you have only ´one hop’, it won’t bother anyone.
And you can process huge number of channel with FX/panning/EQ.
On my side I won’t try to use that, but I would hope the industry with take this way in the future.
An over bridge enabled octatrack would be the best selling thing elektron could make I reckon haha.
I own a mk2 but would buy a mk3 instantly if it was ob enabled.
Now imagine if it could even be an ob host!? Would be the greatest
The idea of integrating different devices with a single USB connection is nice, but there are some inherent problems with this.
Two of those were already mentioned in the form of a question by @Stone262:
The answer to the first question is simple: It probably won’t.
If a device is not class compliant, there is no easy way to access the audio streams without drivers.
Your best bet to make an integrated product for this is Linux.
Since most companies don’t provide Linux drivers, integration seems unlikely unless it’s all from the same company.
Maybe Windows IOT?
The answer to the second question is a bit more difficult, but simply said:
The same way as Overbridge, resampling.
To sync one audio device to another without an audio clock like Wordclock, ADAT or S/PDIF, variable rate resampling is used. There are several papers on this subject if you google it.
Things like aggregate device on mac use the same technique, same for listen to this device on windows and software like VoiceMeeter, Overbridge and Virus TI.
The idea is simple:
Figure out the sample rate of the incoming device compared to the outgoing device.
This can be done for instance by monitoring how fast the incoming device fills a buffer, and the outgoing device empties a buffer of the same size.
When you know the difference, you can resample the input to match the output.
In practice this is not so easy and difficult to get stable, as you’ve noticed by all the Overbridge delays. Some of you might have a Virus TI, which is completely unpredictable on what system the TI function will actually work.
And this whole process introduces latency which is yet another downside.
There is this project though:
Someone like that could make it possible but it still introduces latency.
I would go Windows + VoiceMeeter and call it a day, it also takes care of the driver problem.
Mini pc’s are cheap and portable.
Thanks for the deep answer to those questions. I don’t know much from the implementation details (like the ones you mentioned) and is interesting to read how this “simple” ideas get very comolicated very quickly hahaha
