[sdiy] Andromeda A6 hardware questions

[cheater cheater]

Tom,
every task seems daunting at first but once you start working on it,
if you already have a lot of experience in synth-diy and software
engineering (necessarily at the same time..!) then a lot of the stuff
happening there can become easier.

Even easier if an opensource approach is taken on and other people can
take over where you're failing.

Or even if you terminally fail at delivering the code, at least the
code is there, on sourceforge or somewhere else, waiting for someone
else who's interested enough. For an eternity. And an eternity is a
long time. In fact, it's such a long time, that it just might be
possible to mathematically prove the probability of the project
succeeding will approach 100% as t->+infinity ;-)

Bear in mind that a lot of the problems you have mentioned are removed
or mitigated. Such as getting a reliable hardware platform so that you
*know* the only thing that could have gone wrong is the code.

D.

On Wed, Mar 3, 2010 at 13:55, Tom Wiltshire <tom at electricdruid.net> wrote:
> I think we could go round and round with "that's loads" and  "that's not
> that much", so let me put some numbers on it.
>
> I've been working on a digital/analog hybrid monosynth project for some
> three years now. It's taken that long because I've had to learn everything I
> needed to know as I went along. This includes:
>
> Programming PIC/dsPIC processors
> How to generate envelopes
> How to generate LFOs and other modulation sources
> How to talk SPI
> How to get DACs running from a uP
> How to talk to an external EEPROM
> How to scan a keyboard
> How to scan a panel full of knobs
> How to talk to an LCD
> How to design PCBs
>
> As yet, I haven't written any MIDI handling code, done the patch storage,
> dealt with a sequencer, or written a menu system.
>
> So, after three years, I've done about half the things on your list. I'm not
> claiming to be the most efficient person in the world (like most synthheads,
> I get wildly sidetracked by other projects) but I bet I go about as quickly
> as the next guy. I learned programming when I was a kid (including 6502
> assembly language on a BBC micro, not a million miles from a modern PIC) ,
> so at least I pick up new programming stuff pretty fast by most people's
> standards.
>
> I don't want to sound like I'm poohpoohing the idea, because I'm not, but my
> personal experience leads me to believe that such a task would take some
> number of years (plural) and that anyone who is thinking of taking the job
> on had better have that in mind or they will simply get discouraged long
> before they could ever realistically hope to finish. Perhaps others can
> chime in with how long their own synth projects have taken to help us get an
> overview.
> Every journey does start with a first step, but it helps to have some idea
> how far you're likely to be going before you set off so that you're better
> equipped for the trip! If it's a marathon, getting in training would be a
> good idea, or you're going to finish up collapsed by the roadside.
>
> On a related note, Bruce Duncan and I took several months to design and
> program some of the new Modcan hybrid synth modules, and that was with two
> people (Bruce on hardware design, me doing software) and a single function.
> Multiply up by the number of functional units, then add a bit for all the
> integration.
>
> Regards,
> Tom
>
>
> On 3 Mar 2010, at 11:24, cheater cheater wrote:
>
>> Tom,
>> as they say, every journey starts with the first step.
>>
>> On the other hand, I think you are overestimating the complexity of
>> the andromeda's digital part. It's just some menus, midi, patch
>> storage, a simple sequencer, voice assignment, and modulation. That's
>> not *that* much.
>>
>> D.
>