News:

CWP2Song, public beta.
My  DAW is Reaper
YouTube channel

Main Menu

Few programming questions about the X-air18 / MR18

Started by Tomecki, April 24, 2024, 05:55:32 PM

Previous topic - Next topic

Tomecki

Hi. I would like to try to create a tool that monitors and announces with sound or speech various events from the Mr18 and probably also the X-air18 mixers. I'm not a programmer and I use GPT Chat. The idea is this: we are not always 100% sure when we press a button. A finger may slip, a hand may shake, something may be pressed accidentally, etc. In live or radio performance situations, feedback is extremely important. I found a library that supports X-Air mixers and I think a few others. It can be found at https://github.com/onyx-and-iris/xair-api-python and it seems to work. I have several problems. First of all, it seems to me that in order to monitor the status of anything in the mixer, I have to query the device for each piece of information separately, and from time to time. Is there a way for the mixer to send information about changes itself so that you don't have to ask it about it every now and then? However, if this is the only way, maybe it is possible to send a query that causes the mixer to return the state of everything, so as not to send, for example, every half a second dozens if not hundreds of queries about the state of each fader, mute, solo and several other things. I don't know if I will be able to do it at all, because, as I have already written, I am not a programmer. It is true that the first steps are already behind me. I managed to write a simple script in Python that monitors the fader from channel 1 and announces its volume in such a way that 0db is a4 and every 1db the pitch changes by a semitone. Additionally, each half-decibel is signaled by a sound that is twice as short.

azslow3

AZAOSC has build-in tracker. So it is able to report changes. That was designed for X32 mixers with physical controls, not sure that is useful for rack mixers, till someone else is operating it and you want to know what is going on.

Note that ChatGPT pretend to be "smart", but in practice it is just able to compose information which someone has published in the Internet. For programmer which can easily distinguish between garbage and something which make sense, the tool can save time needed to find original information manually and adopt it for particular case. For non programmers, the result is more or less unpredictable. The result normally somehow works, but it can do something stupid. For small things the probability of good result is high, for complex things, especially if you try to do something no one has done before or done but has never published the algorithm, the probability to get reasonable code is zero.
That is one of biggest worries some people have about ChatGPT, it may be more stupid with time. Before AI, when someone had to code something, he was asking on some forum and answers was also on the forum. Such forums is the source for AI knowledge. Now people ask AI in this case, answers are obviously have no value for AI training. But periodically published in forums, even in case they are bad. So AI get the confirmation the answer is ok, even if not, and so learn more and more stupid solutions.

Tomecki

I am aware of this. Sure, it may not work, but maybe something will come of it. It seems to me that tracking in azaosc doesn't take into account the assumptions I made, so I'm trying to create something of my own.

azslow3

Well, it is always fun to DIY something. I recommend reading X32 OSC remote protocol to understand details, the PDF is easy to find. Not sure if ChatGPT can answer such questions, but you can try. That pdf is long time available, so may be it is already learned. The command is /xremote which should be repeated periodically, then you should get notifications on changed. Meters are subscribed separately, since they constantly change. Getting complete status or just 100 parameters take significant time, so that is not an option for the purpose.