[sdiy] Nord Modular DIY

Roman Sowa modular at go2.pl
Thu Mar 3 10:36:57 CET 2016


Looking at the picture I made while repairing someone's Nord Modular 12 
years ago, I see 4 (four!) Motorola DSPs, and one bigger Motorola chip, 
that obviously takes care of all the interfacing things, knobs, keyboard 
and DSP code upload. Aprat from that, 2 Analog Devices DACs, but I can't 
read numbers on them, plus very popular at my times AD1877 stereo ADC. 
8Mbit flash, 2 RAM chips (unreadable), socketed DIP EPROM with firmware 
and some glue logic and discrete stuff.

I have Axoloti and my first impression was - it's poor man's Nord 
Modular! But as I never really used Nord, I could not tell if it really 
is. Don't get me wrong, poor man's in terms of price only, not quality 
and performance which is very good.

As for group effort to make one big uber synth, it pops up here every 
now and then, and I would love to see that happening, but it's so 
difficult to make complex advanced project even when all designers work 
in the same room, hard to imagine doing it when they communicate over 
emails only. IMHO.

Roman


W dniu 2016-03-03 o 00:45, Tim Ressel pisze:
> Hi Brian,
>
> I hear you about the power. But procs like the Sharc have seriously
> expensive compilers. Its true the STM32 doesn't have bit reversing or
> other niceties, but for this application I'm not sure how much it gains
> you. As long as you have single cycle MAC and high clock rates, I think
> we're okay.
>
> As for wifi interference, its a very good point. I'd be afraid of
> interfering with other equipment as well. On the other hand I was an EMC
> engineer in a past life and know a bit about radiated and conducted
> issues. Still, a challenge.
>
> Um, which Sharc chips are in the Nord?
>
> --tr
>
>
> On 3/2/2016 12:25 PM, rsdio at audiobanshee.com wrote:
>> Hi Tim,
>>
>> I was going to suggest that an actual DSP chip, such as TMS320 or
>> SHARC, would be a much better choice for a large modular. It's true
>> that a generic processor like the STM32 can do quite a lot of signal
>> processing, but DSP chips are designed to accomplish an order of
>> magnitude more in the same amount of time and power. Seeing Rutger's
>> comment that the Nord runs on multiple SHARC chips sort of makes my
>> point.
>>
>> By the way, I'd recommend against combining WiFi (radio) and audio
>> signal generation in the same product. It's doable, but seriously
>> increases the challenges. Digital is already difficult to get right
>> when it comes to generating analog signals. Having radio interference
>> generated on the same board just seems to be asking for trouble.
>>
>> Brian
>>
>>
>> On Mar 2, 2016, at 8:58 AM, Rutger Vlek <rutgervlek at gmail.com> wrote:
>>> I've been thinking along the same lines. However, I've seen many
>>> similar projects get stranded half-way because it really really takes
>>> a lot of coordination, good UI design, debugging, testing, etc to get
>>> something that is as pleasant to work with as the Nord G2.
>>> Additionally, doing that in VST rather than hardware also solves many
>>> initial problems (though sacrificing playability IMHO).
>>>
>>> I'm also a big fan of the Nord! I wish Nord would make the system
>>> open-source such that we can further improve it, add addition
>>> modules, but benefit from the huge amount of work already done. I
>>> believe it runs on 4 (or 8) Sharc DSPs by the way.
>>>
>>> The most important thing about your idea to me is: let's combine
>>> forces and make a monster of a synth! Something no one could ever
>>> pull off alone. Something that can grow for about 5 or 10 years (I'm
>>> thinking like the MidiBox platform did for quite a while).
>>> Personally, I'm more into the hybrid stuff, combining a modular DSP
>>> environment with analog voice-cards and effect chain options.
>>>
>>> While where at it, why don't we make a platform that also allows
>>> direct WiFi connection for patch sharing and rating (much like an App
>>> store on a mobile phone). Oh, and when we do something big and long
>>> term, it also pays of to put a decent keyboard in it with aftertouch,
>>> rather than the cheap mini-keys we get from the big corp's these days.
>>>
>>> On 2 mrt 2016, at 17:17, Tim Ressel wrote:
>>>> I've always been interested in a software based synth that was
>>>> configured through an app. I took a whack at it many years ago but
>>>> never got very far. Then I found the Nord Modular and that seemed to
>>>> fill the need nicely. Why re-invent the wheel?  But then they get
>>>> discontinued, and used G2s go for $1000.
>>>>
>>>> So maybe its time for the DIY thing after all.  Anybody interested?
>>>>
>>>> I did some research. Someone out there has a clone for the app that
>>>> configures the synth. If that code is workable then we just need the
>>>> synth side.  On that front there are several excellent choices for a
>>>> processor. I'm thinking a high-end STM32. We could start with a
>>>> Nucleo board with a daughter board; nice and cheap. The code will be
>>>> intense, but many hands makes work light.
>>>>
>>>> This would be an open source project for hardware and software. That
>>>> way nobody's toes get stepped on. Or everybody's does, depending on
>>>> your viewpoint.
>>>>
>>>> Thoughts?
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>
>



More information about the Synth-diy mailing list