giuliomoro ahh that's awesome to hear -- the instrument is coming along !! 🙂

a few years back (when I first got the Bela) I ported the Pd code for the instrument from the Mac to the Bela. The problem was that Heavy doesn't support some objects; FFT was especially an issue (you had helped me to understand the scenario here: https://forum.bela.io/d/1698-optimizing-pd-patch-heavy-compatibility-other-ideas). Because I was caught up in hardware design stuff at that time, I decided best route was to keep the software running on the Mac, stick with the Arduino I was using for ADC, and later, down the road, work to re-code everything in C++ for the Bela. I'm at the point where a lot of the HW stuff is fixed and am now switching back to the Bela for the ADC (better conversion in comparison to Arduino). I'm still kicking the can down the road for the C++ port lol. Anyway, that's the backstory 🙂

    9 months later

    violapwr wanted to share an update on this in case useful to anyone; I finally got all the bugs out in this Bela analog data -> PD function I was building.
    -data is being transmitted with minimal errors and audio signals (over analog) actually sound good (as opposed to crunchy lol) when received on the PD end
    -uses OSC to send & sync sample rate, block size, and number of channels to PD

    below is the Bela code and here's link to the PD files:
    https://github.com/brianlindgren/Bela-to-PD-over-OSC

    #include <Bela.h>
    #include <cmath>
    #include <iostream>
    #include <libraries/UdpClient/UdpClient.h>
    #include <libraries/OscSender/OscSender.h>
    #include <libraries/OscReceiver/OscReceiver.h>
    #include <libraries/Pipe/Pipe.h>
    
    Pipe oscPipe;
    
    OscReceiver oscReceiver;
    OscSender oscSender;
    int localPort = 7562;
    int remotePort = 5000;
    const char* remoteIp = "192.168.7.1";
    
    constexpr int kBufferSize = 1024; // Total size of the circular buffer
    int kAnalogInChannels;
    int gAnalogFrames;
    int gAnalogSampleRate;
    
    float input; //from Bela->context
    uint16_t circularBuffer[kBufferSize]; // Circular buffer
    UdpClient udpClient(4567, "192.168.7.1");
    AuxiliaryTask serialCommsTask;
    
    int readIndex = 0;  // Current read index
    int writeIndex = 0; // Current write index
    
    
    void ioLoop(void* arg) {
        while (!Bela_stopRequested()) {
            while (readIndex != writeIndex) { //send 1 block at time until caught up
            
                uint16_t buffer[kAnalogInChannels * gAnalogFrames]; // init transfer buffer
                
                //copy analog frames to transfer buffer
                for (int frame = 0; frame < gAnalogFrames; ++frame) {
                	//cycle through channels and copy content from circular buffer
                    for (int channel = 0; channel < kAnalogInChannels; ++channel) {
                        buffer[(frame * kAnalogInChannels) + channel] = circularBuffer[readIndex + ((frame * kAnalogInChannels) + channel)];
                    }
                }
        
                readIndex = (readIndex + (gAnalogFrames * kAnalogInChannels)); // Move the read index to the next block
                
                if (readIndex >= (kBufferSize - 1))  //wrap read pointer
        			readIndex = 0;
                
    			udpClient.send(buffer, sizeof(buffer)); //send to PD
            }
            usleep(250);
        }
    }
    
    void on_receive(oscpkt::Message* msg, void*)
    {
    	// we make a copy of the incoming message and we send it down the pipe to the real-time thread
    	oscpkt::Message* incomingMsg = new oscpkt::Message(msg);
    	oscPipe.writeNonRt(incomingMsg);
    
    	// the real-time thread sends back to us the pointer once it is done with it
    	oscpkt::Message* returnedMsg;
    	while(oscPipe.readNonRt(returnedMsg) > 0)
    	{
    		delete returnedMsg;
    	}
    }
    
    bool setup(BelaContext *context, void *userData) {
    	/// from Bela example
    	oscPipe.setup("incomingOsc");
    	oscReceiver.setup(localPort, on_receive);
    	oscSender.setup(remotePort, remoteIp);
    
    	// the following code sends an OSC message to address /osc-setup
    	oscSender.newMessage("/osc-setup").send();
    
    	printf("Waiting for handshake ....\n");
    	// we want to stop our program and wait for a new message to come in.
    	// therefore, we set the pipe to blocking mode.
    	oscPipe.setBlockingNonRt(false);
    	oscPipe.setBlockingRt(true);
    	oscPipe.setTimeoutMsRt(1000);
    	oscpkt::Message* msg = nullptr;
    	int ret = oscPipe.readRt(msg);
    	bool ok = false;
    	if(ret > 0) {
    		if(msg && msg->match("/osc-setup-reply"))
    		{
    			printf("handshake received!\n");
    			ok = true;
    		}
    		delete msg;
    	}
    	if(!ok) {
    		fprintf(stderr, "No handshake received: %d\n", ret);
    		return false;
    	}
    	// in the remainder of the program, we will be calling readRt() from render(), and we want it
    	// to return immediately if there are no new messages available. We therefore set the
    	// pipe to non-blocking mode
    	oscPipe.setBlockingRt(false);
    	
    	AuxiliaryTask serialCommsTask = Bela_createAuxiliaryTask(ioLoop, 90, "serial-thread", NULL);
    	Bela_scheduleAuxiliaryTask(serialCommsTask);
    	printf("AnalogFrames: %d\n", context->analogFrames);
    	printf("AnalogInChannels: %d\n", context->analogInChannels);
    	printf("AnalogSampleRate: %d\n", (int)context->analogSampleRate);
    	gAnalogFrames = context->analogFrames; //# of analog frames
    	kAnalogInChannels = context->analogInChannels;
    	gAnalogSampleRate = (int)context->analogSampleRate;
    	oscSender.newMessage("/osc-settings").add(gAnalogSampleRate).add(gAnalogFrames).add(kAnalogInChannels).send();
    	oscPipe.writeRt(msg);
    	return true;
    }
    
    void render(BelaContext *context, void *userData)
    {
    	//write a message if read head couldn't catch up to write head last cycle
    	if (readIndex != writeIndex) {
            rt_printf("udp catchup! >> readIndex: %d, writeIndex: %d\n", readIndex, writeIndex);
    	}
    
    	
    	//cycle through frames
    	for(unsigned int n = 0; n < gAnalogFrames; n++) {
    	
    		//cycle through channels and copy content to circular buffer
    		for(unsigned int ar = 0; ar < kAnalogInChannels; ar++) {
    			
    			input = analogRead(context, n, ar); //read from ADC/context
    			circularBuffer[writeIndex + ar] = input * 65535; //convert from float to int
    	    }
    	    // Increment write index every time a new frame is written and reset to 0 if too high
    	    writeIndex = (writeIndex + kAnalogInChannels); 
    	    
    	    if (writeIndex >= (kBufferSize - 1))
    	    	writeIndex = 0;
        }
    }
    
    void cleanup(BelaContext *context, void *userData) {}
    a year later

    very strange problem I'm having...

    for some reason the Bela all of a sudden had trouble communicating with PD over OSC. I wiped and reflashed the ssd but still not working (with the code above). I'm getting this error: no viable conversion from 'void (oscpkt::Message *, void *)' to 'std::function<void (oscpkt::Message *, const char *, void *)>' column: 31, line: 73

    if i run \example Communication/OSC-pipe/render.cpp the code runs & i'm able to receive 'osc-setup' in PD, but the 'osc-setup-reply' is not received on the Bela, with PD reporting this error send: No route to host (65)

    come to think of it, a few days ago, git (on Bela) was giving me a seg fault, so i tried to update/reinstall it using apt (which did not work)... hopefully I did not screw up anything .... 😳

    any ideas?

      Probably your code needs to be amended tobe compatible with the code on the board. There was an API change which is what makes the example work but your project doesn't

      It should be as easy as replacing

      void on_receive(oscpkt::Message* msg, void*)

      with

      void on_receive(oscpkt::Message* msg, const char*, void*)

      @giuliomoro that worked, the program now compiles! thanks!

      however, i'm still getting this error (both my program and the example project) when PD responds to Bela with osc-setup-reply: send: No route to host (65). when i ping 192.168.7.2 from my computer, the Bela responds, so the IP seems to be correct...

      any other ideas?

      Can you run ifconfig on the board (in the console at the bottom of the IDE) and show the results here? Also, run ping -c 1 192.168.7.1 in there and see what comes back

        giuliomoro OK, this is what was returned:

        root@bela ~/Bela# ifconfig
        bash: line 6: ifconfig: command not found

        and

        root@bela ~/Bela# ping -c 1 192.168.7.1
        PING 192.168.7.1 (192.168.7.1) 56(84) bytes of data.
        64 bytes from 192.168.7.1: icmp_seq=1 ttl=64 time=1.09 ms
        --- 192.168.7.1 ping statistics ---
        1 packets transmitted, 1 received, 0% packet loss, time 0ms
        rtt min/avg/max/mdev = 1.099/1.099/1.099/0.000 ms

        giuliomoro

        root@bela ~/Bela# ip a
        1: lo:  mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
            link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
            inet 127.0.0.1/8 scope host lo
               valid_lft forever preferred_lft forever
            inet6 ::1/128 scope host 
               valid_lft forever preferred_lft forever
        2: eth0:  mtu 1500 qdisc mq state DOWN group default qlen 1000
            link/ether 4c:3f:d3:19:7a:91 brd ff:ff:ff:ff:ff:ff
        3: wlan0:  mtu 1500 qdisc noop state DOWN group default qlen 1000
            link/ether 1c:bf:ce:65:1a:fd brd ff:ff:ff:ff:ff:ff
        4: usb0:  mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
            link/ether be:1a:d3:19:7a:93 brd ff:ff:ff:ff:ff:ff
            inet 192.168.6.2/24 brd 192.168.6.255 scope global usb0
               valid_lft forever preferred_lft forever
            inet6 fe80::bc1a:d3ff:fe19:7a93/64 scope link 
               valid_lft forever preferred_lft forever
        5: usb1:  mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
            link/ether be:1a:d3:19:7a:96 brd ff:ff:ff:ff:ff:ff
            inet 192.168.7.2/24 brd 192.168.7.255 scope global usb1
               valid_lft forever preferred_lft forever
            inet6 fe80::bc1a:d3ff:fe19:7a96/64 scope link 
               valid_lft forever preferred_lft forever

        violapwr if i run \example Communication/OSC-pipe/render.cpp the code runs & i'm able to receive 'osc-setup' in PD, but the 'osc-setup-reply' is not received on the Bela, with PD reporting this error send: No route to host (65)

        also can you show the Pd patch?

        @"giuliomoro"#p27136

        PD:

        this code is inside pd handshake (shown in the 2nd image):

        in pd can you use "find last error" and see where it points to?

          giuliomoro PD does not seem to be able to identify the error source, strangely:

          (in the back of my mind, I keep wondering if I messed up the system when trying to fix the git segmentation fault and need to flash the emmc (since flashing the SSD has not yet fixed it).... )

          I don't think it's a bela problem though ...
          can you try a simpler patch:

          [connect 192.168.7.2 7562(
          |
          | [1 2 3 4(
          |/
          [netsend -u -b]

          do you get the same "no route to host" error?

            giuliomoro Yes, same error. I received a successful connect (1) when clicking connect 192.168.7.2 7562. Clicking 1 2 3 4 resulted in the same error and a disconnect (0)

            giuliomoro No... I also tried to turn off my macOS firewall, and received the same error in PD

            I just tried the same experiment with my Bela Mini (and the simplified PD patch), and had the same result, which leads me to agree that the issue is perhaps not with the the Bela but on the computer end... I recently updated my macOS to 15.4... could it be a macOS issue?

            Update:
            I re-tried our PD test patch on my school computer & it worked without error (screenshot below). Also, more complex program (from a year back, above) is working as well. The school machine is macOS 14.7.5. So perhaps it's a macOS thing?

              violapwr I just tried the same experiment with my Bela Mini (and the simplified PD patch), and had the same result, which leads me to agree that the issue is perhaps not with the the Bela but on the computer end... I recently updated my macOS to 15.4... could it be a macOS issue?

              I don't know what they are doing that makes this stop working ... can you access the Bela IDE at 192.168.7.2?