[sdiy] Audio Weaver for ST Discovery boards -Free!

[rsdio at audiobanshee.com]

Texas Instruments includes CODEC electronics on their TMS320 DSP boards, priced as low as $49. One example is a small USB stick with headphones and microphone jacks. Not what you want for your final platform, but very convenient and portable for the development phase of your project.

I couldn't tell from a quick scan of a few Audio Weaver web pages, but I assume that TMS320 is supported. If not, Texas Instruments has an open source DSP library that you can use instead.

Texas Instruments also has ARM boards, but I am not aware of whether any of the ARM boards have audio CODEC on-board.

Brian


On May 11, 2017, at 9:11 AM, Scott Gravenhorst wrote:
> I agree about the selection of CODEC and attention to audio details, but at least any CODEC will
> allow "practice" coding I2S so that when you make your own board you have those chops, even if it's
> a different CODEC.
> 
> "Richie Burnett" wrote:
>> The inclusion of built-in CODECs on dev boards is a bit of a mixed bag. 
>> It's useful for getting things up and running quickly if there is some type 
>> of CODEC built in.  However, many of the DSP boards I've used either have 
>> crappy low-resolution CODECs and/or pay little attention to proper 
>> analogue/digital segregation and ground-plane practices, so you get 
>> relatively poor audio quality and quite a lot of digital noise present at 
>> the analogue outputs.  At least if you have to make your own CODEC "shield" 
>> you can make sure that you choose a decent CODEC and lay out the board 
>> properly to realise its potential performance.
>> 
>> -----Original Message----- 
>> From: Scott Gravenhorst
>> Sent: Thursday, May 11, 2017 2:46 PM
>> 
>> Ah, thanks for that.  One thing that pisses me off a bit is that it's still 
>> too uncommon for dev
>> boards (like this one) to have an on board audio codec.  Thankfully, the 
>> Discovery board does.  I
>> would think in these days where every product out there has a screen and 
>> sound that it would be
>> imperative to put at least an audio codec on the board.  The advantage for 
>> me is that when
>> developing driver code for a codec, it's comforting to know that the codec 
>> being placed there by
>> the mfr would seem to indicate that it _should_ work and that if there's a 
>> problem - it's your code.
>> 
>> I may purchase the SAM E70 board after looking at the datasheet, but I can 
>> see I'll have to
>> jerry-rig a codec to some pins on the board - which will be SMD and having 
>> myopia plus presbyopia,
>> that's a problem.  Perhaps there's a plug in board... (not with my luck) 
>> I've not yet had luck
>> coding I2S, but I can sharpen my chops on the Discovery board.  At least the 
>> E70 board isn't
>> expensive at about $30usa.  I'll need to determine what the difference is 
>> between S70 and E70.  The
>> S70 chip is the one used for the Tsunami Super WAV Trigger board.
>> 
>> On 2017-05-11 12:30, Scott Gravenhorst wrote:
>>> Well, to be honest, I started looking at Microchip's (Atmel)
>>> ATSAMS70N20 which includes
>>> almost identical internal features.  It has a bit more internal RAM,
>>> but the datasheet
>>> says it runs up to 300 MHz.  The ARM on the 32F746GDISCOVERY board
>>> runs up to 216 MHz.
>>> However, I've not been able to find a dev board for the ATSAMS70N20.
>>> 
>>> So I have to say that this Discovery board is my second choice and
>>> will be a super-major
>>> step up from a dsPIC.  As for performance, I'll just have to see how
>>> much synth I can
>>> cram into it.  One interesting thing about both of the ARM ICs is that
>>> they have a 2
>>> channel 12 bit DAC that runs up to 1 MHz sample rate which might be
>>> fun with naive
>>> waveforms since the Nyquist limit is 500 kHz.  I've worked with a 12
>>> bit DAC at that
>>> sample rate using an FPGA and I noticed no alias problems with naive
>>> waveforms.
>>> However, such a high sample rate will limit voice and feature count.
>>> 
>>> Anyway, I'll post about it once I've got a good feel for what it can
>>> do.  All I have
>>> right now is a bunch of PDF files.
>>> 
>>> Michael Zacherl wrote:
>>>> 
>>>> On 11.May 2017, at 3:44 , Scott Gravenhorst wrote:
>>>>> I got some birthday money and bought this: 32F746GDISCOVERY
>>>>> =
>>>> 
>>>>> It looks pretty nice.  I will try Audio Weaver when the board comes
>>>>> here.
>>>> 
>>>> at this price ($49 really?) it looks like a steal!
>>>> How you=92d rank the cpu performance-wise?
>>>> Additional I/O via SPI? (still thinking dc-coupled audio rate A/D D/A)