[sdiy] In praise of the ATM STM32F303

[Eric Brombaugh]

A couple guesses why:

1) for high-volume production such as Waldorf usually does, every cent makes a difference. In large quantity F3 parts like the 303 are a few bucks cheaper than F4 parts, so that adds up.

2) I've used both F3 and F4 parts in products so I've got some experience with them. The F3 parts analog sections are considerably higher quality than the F4. Noise on F3 ADCs and DACs are a few lsbs lower amplitude than F4 parts.

Eric

On Feb 12, 2016, at 8:29 PM, Declare Update wrote:

> Really loving this thread! I'm curious though: why do you guys think Waldorf chose this chip? It's not particularly cheap, they could have had an F4 running much faster with more ram for very similar cost. Were F4 parts with built in DACs less common at the time, maybe? 
> 
> A few months ago, I had a great time trying out reverbs on a Teensy 3.1. I had the main feedback loop going from the dac out to a spare adc channel, which gave me a few free samples of delay and the chance for simple passive analog filtering. made a huge difference on simple algorithms.
> 
> Playing with the STM32F446 now, and it's a real treat in comparison. 
> 
> 
> Chris
> 
> Sent from my iPhone
> 
>> On Feb 12, 2016, at 4:00 PM, Gordonjcp <gordonjcp at gjcp.net> wrote:
>> 
>>> On Fri, Feb 12, 2016 at 09:01:50PM -0000, Richie Burnett wrote:
>>> Did you base it on the Jon Datorro "Lexicon Plate" reverb algorithm
>>> too?  Or one of the Spin Semiconductor FV-1 algorithms?  Or
>>> something you designed yourself?
>> 
>> I based it on https://ccrma.stanford.edu/~jos/pasp/Schroeder_Reverberators.html
>> and a lot of fiddling and guesswork
>> 
>> This is the floating point version - I can't seem to find the fixed-point one, which may be on my spare laptop.  It uses longer periods but the fixed-point one uses 2048-word delays.
>> 
>> https://github.com/gordonjcp/reverb
>> 
>>> I know the early Reverb pioneers had to be really clever in making
>>> their algorithms make the best use of what limited (expensive!)
>>> memory they had available to them back then.  With only a small
>>> amount of delay memory you have to use lots of diffusion to scatter
>>> the energy around each time it passes around the loop otherwise you
>>> hear audible repetitions.
>> 
>> Mine has a distinctive "rattle" at certain settings, and certainly sounds a lot better on sounds with a relatively slow attack and decay.  That being said, the "bittiness" seems to be lost with busy drum patterns.
>> 
>> -- 
>> Gordonjcp MM0YEQ
>> 
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy