News:

SMF for DIYStompboxes.com!

Main Menu

Consensus for DSP?

Started by varialbender, February 28, 2006, 01:03:55 PM

Previous topic - Next topic

varialbender

I think we've reached a consensus for learning uCs with the Atmel AVR STK500.
I'm going to dive right into that and learn as much as I can.

Is there a consensus for DSPs now? I'd like to learn DSPs and it would be nice to learn 'with the group'.
What's a good chip to learn DSP on?

I took a course in ARM, learned about the architecture and assembly language and such. Would this be any good? I can probably con some teachers to get a development kit in. Even if this isn't the chip we choose, would it be helpful for me to learn with the ARM as well?

Thanks a lot

And thanks to everyone involved with the forum, and those who will be writing articles and such, it's great help  :)

Peter Snowberg

I think we're going to play with the Wavefront chips first; using an AVR as the host processor.

The Wavefront chips are really cool devices and the guys of Axoris have made some WONDERFUL tools for drag and drop algorithm development.

The first full-on DSP project will involve a Wavefront AL3102 DSP with 2, 4, or 8 channels of analog in and out.

We probably won't be using the ARM too soon around here. They're wonderful chips but we're going to start with 8 bit parts.
Eschew paradigm obfuscation

R.G.

QuoteI think we've reached a consensus for learning uCs with the Atmel AVR STK500.
I didn't see the results of that vote... where was that?
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

Peter Snowberg

#3
There is a consensus that several people will be focusing on the STK500, but we're not going to forget the PIC!  :icon_biggrin:

We already have PIC example code posted courtesy of Transmogrifox for his PIC operated TBP switching controller.
Eschew paradigm obfuscation

varialbender

It wasn't so much a vote as it was The Tone God's 11th commandment. Peter had also recommended it. So it was written, and so it shall be.

I'll try not to neglect PICs, but I'll be focusing on the Atmel AVR stuff for a little while until I'm comfortable taking on another chip.

I'll see what I can do about the Wavefront stuff. I've read a couple posts on the Femtoverb and such, and I'm interested in seeing what can be done.

Thanks

varialbender

Oops, wrong wavefront chip. I thought perhaps the reverb was done on the 3102, but nope, forgot it's its own chip. Either way, can't wait to give it a go.

danngreen

So, what kind of development environment is good for the wavefront chips? Wavefront doesn't make the eval boards anymore (is that true?). Has anyone tried working with the Axoris Miss Parker board? They provide the gerber files so if we wanted to make up a batch of pcbs, it would be possible. The neat thing about that is the user interface is on a separate board, so for knob-crazy builders like myself, there could be different UI boards with different knob/switch configurations, but the same main board.

ps yes, im glad we're going with Atmel AVR!

Peter Snowberg

The Wavefront chips can be built into a working circuit for less cost than the development boards Alesis-Semi offered. I would actually suggest getting an Alesis Ineko and hacking into that with the aid of the STK500 if you must hack now.  :icon_biggrin: It contains an 1K DSP and a DRE.

The Axoris software and the Wavefront assembler are about all you'll find out there for the 1K DSP. Last I checked, the only thing for the DRE was the Wavefront software.

There will be some Wavefront stuff going on here after the first group is up to speed on basic microcontrollers.  :icon_wink: I'll lay out some boards with STK500 compatible I/O connections when we get to that point. No time these days.

Remoting the user interface is something I've been thinking about for a long time. There are a number of details in making equipment talk efficiently. For control panel style settings the interface to the UI doesn't need much speed and a simple I2C interface becomes a good candidate. If you want to modulate effects in real time, use an expression pedal or LDR, you're probably going to want to sample it a bit more often than a dumb I2C interface will give you. The real trick is making this stuff modular:icon_wink:   


More on communications really soon.  :icon_biggrin:

.
Eschew paradigm obfuscation

DavidS

QuoteMore on communications really soon.  icon_biggrin

Great! I'm looking forward to info on shift registers, multiplexers and such. I don't yet have a solid feel for just what can be accomplished with just a certain number of I/O pins...