* MP3 Shield * - Rogue Robotics rMP3

ooops... thanks mowcius. Fixed. (I was wondering why there was a http:// lying on the floor next to my desk).

b

XMAS time...

This sketch randomly plays all the songs in a folder. You can control the volume, play/pause/stop/next, and ff/rew using serial input (from the IDE).

Throw a bunch of your favorite holiday songs in a folder named "XMAS" on your SD card, stick it in the rMP3, and run the sketch.

I haven't got around to building a tree light sequencer to run at the same time as the songs are playing. Maybe Grumpy_Mike will lend me his Tree-Sink. ;D

Happy Holidays everyone!

b

#include <RogueSD.h>
#include <RogueMP3.h>
#include <NewSoftSerial.h>

NewSoftSerial rmp3_serial(6, 7);

RogueMP3 rmp3(rmp3_serial);
RogueSD filecommands(rmp3_serial);

int numberOfSongs;
int lastSong;
char path[96];

const char *directory = "/XMAS";


void setup()
{
  Serial.begin(9600);

  Serial.println("Merry Xmas!");

  rmp3_serial.begin(9600);

  rmp3.sync();
  rmp3.stop();

  filecommands.sync();

  // mix up our random number generator
  randomSeed(analogRead(0));

  // get the number of songs available
  strcpy(path, directory);
  strcat(path, "/");
  strcat(path, "*.mp3");  // we have to do this because the IDE thinks that "/*" needs to be terminated everywhere

  numberOfSongs = filecommands.filecount(path);

  lastSong = -1;
}



void playNextSong()
{
  uint8_t i;
  char filename[80];
  char path[96];
  int nextSong = 0;

  if (numberOfSongs > 0)
  {
    // Select our next song randomly
    if (numberOfSongs > 2)
    {
      do
        nextSong = random(numberOfSongs);
      while (nextSong == lastSong);
    }
    else if (numberOfSongs == 2)
    {
      // we only have two songs
      if (lastSong == 0)
        nextSong = 1;
      else
        nextSong = 0;
    }

    // now, get our file name from file list

    filecommands.opendir(directory);

    for (i = 0; i <= nextSong; i++)
    {
      filecommands.readdir(filename, "*.mp3");
    }

    strcpy(path, directory);
    strcat(path, "/");
    strcat(path, filename);

    rmp3.playfile(path);

    Serial.print("Playing: ");
    Serial.println(path);

    lastSong = nextSong;
  }
  else
  {
    Serial.println("No files to play.");
  }
}


void loop()
{
  char c;
  uint8_t i;
  int16_t newtime;

  playbackinfo playinfo;

  char status = 'S';
  uint8_t playing = 1;
  uint8_t volume = 20;
  uint8_t boostOn = false;

  volume = rmp3.getvolume();  // this only gets the left volume
  
  playNextSong();

  while(1)
  {
    while(!Serial.available())
    {
      // we should do fancy stuff like flash lights on our Xmas tree here!
      // got lots of time!
      delay(200);

      status = rmp3.getplaybackstatus();
      playinfo = rmp3.getplaybackinfo();

      if (status == 'S' && playing)
        playNextSong();
    }
    
    // OOH!! got a character!
    c = Serial.read();

    switch(c)
    {
      case 'p':
        // pause
        if(status == 'D')
        {
          // fade in
          rmp3.playpause();
          rmp3.fade(volume, 400);
        }
        else if(status == 'P')
        {
          // fade out
          rmp3.fade(100, 400);
          rmp3.playpause();
        }
        else
        {
          // start playing
          playNextSong();
          playing = 1;
        }
        break;
      case 's':
        rmp3.stop();
        playing = 0;
        break;
      case 'n':
        playNextSong();
        playing = 1;
        break;

      case 'e':
        if(boostOn)
        {
          rmp3.setboost(0);
          boostOn = false;
        }
        else
        {
          rmp3.setboost(8, 6, 7, 3);
          boostOn = true;
        }
        break;
      
      case 'a':
        // jump back 5 seconds
        newtime = playinfo.position - 5;
        if (newtime < 0) newtime = 0;
        rmp3.jump(newtime);
        break;

      case 'd':
        // jump forward 5 seconds
        rmp3.jump(playinfo.position + 5);
        break;
        
      case 'k':
        if(volume < 254) volume++;
        if(status != 'D') rmp3.setvolume(volume);
        break;

      case 'i':
        if(volume > 0) volume--;
        if(status != 'D') rmp3.setvolume(volume);
        break;

    }
  }
}

volume, play/pause/stop/next, and ff/rew using serial input (from the IDE)

Now it's getting interesting. (to be honest I probably should have read through the page in more detail)

This is definitely on my list and I may be able to get one soon if I get my other stuff free for sparkfun free day or I have to buy it...

Mowcius

Hey Brett,

I'm currently trying to use the rMP3 shield with the DFRduino LCD Keypad shield. The problem is, while rMP3 uses digital pins 6, 7, 8, 11, 12 & 13, the LCD shield uses 5, 6, 7, 8 & 9. Therefore, I'd like to reroute pins 6, 7 & 8 to use pins 3, 4 & 10 by cutting the jumpers and wiring them up. So far, so good. The problem I have, though, is that I can't seem to work out how I'd let the C compiler know of the change. There aren't any obvious locations in the C file or associated header and I can't see any mention of this in the documentation. Could you please give me a clue as to how I would do this?

Thanks,
Lee

Does it really use all those pins?

I have not really delved into how it does what it does but I presumed that it just could use those other pins.

Isn't the communication via serial? In which case it would require 2 pins?

Also another question, I know that the LED on board (activity LED) flashes when music is playing and seems to brighten at corresponding points in the music. Is there any easy way to get audio values sent to the arduino so a visualiser could be done from it?

A bit more documentation would also be very nice :stuck_out_tongue:

Mowcius

I think the documentation is actually pretty good. I wish there was a support forum, tho :wink:

Lee - The rMP3 only uses 2 serial pins to communicate. The SPI port is there as an optional way to control the rMP3, but the current version of firmware (100.01) hasn't got that implemented (expect it in the next version).

You figured out how to do the serial change though. Just cut the jumper at pins 6 (rMP3 TX) and 7 (rMP3 RX), and reroute the pads to the new pins you want to use. (make sure you connect wires to the pads FURTHEST of the two from the Arduino header).

The change in the code should be really simple. Wherever you see:

NewSoftSerial rmp3_serial(6, 7);

change it to (in your case):

NewSoftSerial rmp3_serial(3, 4);

(I'm assuming that you changed pin 6 on the rMP3 to pin 3, and pin 7 to pin 4).

If that doesn't work, post your code and we can see what's going on.

mowcius - sounds like either you're listening to a VBR mp3 file (wide range one too! 32 kbps -> 320 kbps) and the music has really quiet and really loud spots, or your eyes are fooling you! The data goes directly from the card to the decoder. So, the only effect you'd see is if the MP3 frames had different bitrates.

RE: documentation, our wikidocs have always been public, unfortunately you had to register. I've changed it now so that you don't have to register. Just add any comments/questions you have on any page that has a discussion.

b

Thank you, Brett, that's awesome news. Having only two pins to worry about makes my life easier :wink:

Best,
Lee

mowcius - sounds like either you're listening to a VBR mp3 file (wide range one too! 32 kbps -> 320 kbps) and the music has really quiet and really loud spots, or your eyes are fooling you! The data goes directly from the card to the decoder. So, the only effect you'd see is if the MP3 frames had different bitrates.

I was indeed! I hadn't noticed that it didn't change with non VBR files. It was only on the loud bits that I got brighter flashes...

Any chance you could do the Input/Output Interface page on the wiki? :slight_smile:
The rest of the documentation does indeed look good, it has been a while since I have looked and I suppose I did not have a great look through.
I have just gone through and had a thorough read! Interesting stuff!

Now back to my previous question :slight_smile:

Is there any easy way to get audio values sent to the arduino so a visualiser could be done from it?

It is not a problem if not, it would just be interesting to do it without all the extra circuitry it usually requires.

Mowcius

L and R provide accurate activity for that sort of thing :wink: Why not branch from there for both output and route to analog input in the Arduino? I don't know what the upper or lower values for those pads are (maybe in the docs?), but it's surely a valid option. Just make sure it doesn't exceed the pins impedance.

You mean speaker L & R?
That requires all the annoying circuitry in between to get that to work...
I suppose i am more asking about a frequency visualiser which is more complicated...

Mowcius

Not necessarily. The analog input in the Arduino can take any value between 0v - 1v (I believe), so, as long as you limit the voltage (using the correct resistance), then you have a simple variable source right there. The biggest issue is making sure your querying of that source doesn't use the potential that would otherwise drive your speaker output (though you could increase that with an amp).

0v - 1v (I believe)

You mean 0 - 5v (well it's more like 0 - 5.5 but that's pushing it a bit.)

Yeah, 5v... That's what I meant :wink: Well, you're okay, then, seeing as 5v is the maximum voltage anywhere past the regulator! Hook even one of those pads into the Arduino and use it as an input. Actually, you could also average the highs and lows so as to get better results, as those highs and lows will be quite similar in value.

You have to be careful when using audio signals. The output from the rMP3 on the L and R pins are actually AC signals (because of the filter capacitors on the outputs). This means that the voltage with respect to ground on those pins can be negative, if you connect them to the analog inputs on the Arduino. This can damage the ADC on the Atmel chip.

You could introduce a DC offset, but that will require extra components which I think mowcius is trying to avoid. :slight_smile:

I'd stay away from connecting the audio output to one of the analog inputs, unless you know what you're doing.

mowcius - your wish is my command. Here is the updated documentation on the input interface:

http://www.roguerobotics.com/wikidocs/rmp3/documentation/input_interface

b

Hadn't thought about it being AC (which should be as obvious as the nose on my face), but it's nothing that a handful of diodes couldn't fix. In fact, one could always route the negative edge to a separate pin as a positive feed. :stuck_out_tongue:

mowcius - your wish is my command. Here is the updated documentation on the input interface:

http://www.roguerobotics.com/wikidocs/rmp3/documentation/input_interface

Oooh, I do feel powerful :stuck_out_tongue:

I presume it is still on the list but what do all the other pads do? From the looks of it, some could be for an FTDI header?

You could introduce a DC offset, but that will require extra components which I think mowcius is trying to avoid.

Aye, I will look into it with as few components as I can get away with though :stuck_out_tongue:

Mowcius

All the connections are described in detail here:

http://www.roguerobotics.com/wikidocs/rmp3/documentation/connectors

It's got:

Arduino compatible shield connectors
FTDI/SFE-DEV-09115 cable connector
Digital trigger connector (trigger with digital inputs)
... among others

b

I was wondering if you guys plan on coming up with a cheaper version of your MP3 shield. Like Sparkfuns?

Sparkfun's MP3 breakout is far less functional. I really had to search high and low for a "high quality" and flexible MP3 board for my JukeBox project I'm building. There just aren't that many out there. It was pure luck that this one was already arduino compatible, because I was quite prepared to fashion one into an arduino ready product.