Add second I2S output

[DGHC]

Hi, please forgive me if I've gone about this the wrong way - I've never done a pull request before.

I needed to use the 2nd I2S output with the F32 library, and noticed this hasn't been implemented yet. I've made the changes based on a review of the 1st and 2nd I2S output files in the main Teensy library.

I've tested on a Teensy 4.0 with AudioSynthWaveformSine_F32 as a source and AudioOutputI2S2_F32 as an output and it works. I haven't tested the slave code yet.

Thanks for all your work creating this float library in the first place - it's brilliant.

[boblark]

Hi - First, thanks for making this library class available. This helps the whole community.
I have no experience with using more than a single L-R output. As I understand this, it is the same as AudioOutputI2S_F32 but using the second I2S hardware port. Is this T4 only? I guess we need to indicate hardware connections on the Design Tool? Also, is your test case in shape that we could include it with the examples? Again, thanks, Bob Larkin

[DGHC]

You're very welcome. I'm very far from an expert (Python is my main thing), but it's very nice to be at a point where I can contribute something - I've benefitted greatly from communities like this over the years.

You're right - it is T4 only (as per the 2nd I2S output in the main library). I used the difference between I2S files in the first library as a guide for what needed changing. The second I2S output uses pins 2 (OUT), 3 (LRCLK) and 4 (BCLK).

I don't know how you indicate hardware connections for the Design Tool - happy to help with that if you can point me in the direction of some guidance.

I don't have a test case, but it shouldn't be hard to set that up and push an example. Let me get that together and submit...

[DGHC]

OK, just pushed the example. I've seen that I must have committed something to the wrong branch - the commit called 'AudioMerge basic framework - not working yet'. Please ignore that as it's not working (as the name suggests!). If you look at that, you'll see what I'm trying to do, which is transmit 4 channels over I2S stereo, by combining two 16bit samples into a single 32 bit sample, then do that twice and get 4 channels worth of data. I'll have to unpack it the other end, but that fine. You may see I've made a comment in the "Floating-Point Audio Library Extension" thread in the PJRC forum as I'm stuck at the moment - any wisdom is always greatly received!

[boblark]

Hi Dave - Thanks for the clarification on the usage.

I will take care of the entry for the design tool. It takes a bit of learning, and I have been down that path enough times to make it easier. I will include your pins and the T4.x restriction there. Also some comments on the usage.

The example is good, and again, if you want I can use your INO and put it into the examples directory. As far as the content, one thought is that to better illustrate what is going on, you might add a couple more generators and a regular I2S1 with all four at different frequencies? Or something of that sort.

I will hopefully get a chance tomorrow to look more at the2x16-bit packing. Bob

Oh, also, Github, as best as I can see, does not support running web pages directly. So, I have plenty of room at janbob.com and have put the web page there:
http://www.janbob.com/electron/OpenAudio_Design_Tool/index.html
I have a script that syncs the github info as well as the zip file for offline operation so that I only need to make changes one place. But it means that pull requests on the github version are not needed nor responded to directly.