Nord User Forum

mjbrands wrote: If those contacts are 'disabled', it seems (to me anyway) more logical they did it to improve reliability - you wouldn't want some freak defect to cause those 'ghost' keys to generate key presses.

Yes, that's exactly why they did it. There were some occurrences in a previous model where something would put pressure on the hidden contact, and someone would suddenly hear a note they couldn't shut off, so as unlikely as it was, this prevents that from occurring. An unfortunate side effect is that, previously, some people rigged up a key to get that low E working (and possibly the high notes as well), and now those mods don't work. So it would be interesting to know if they could be made to work just by manipulating these jumpers.

anotherscott wrote:

mjbrands wrote: If those contacts are 'disabled', it seems (to me anyway) more logical they did it to improve reliability - you wouldn't want some freak defect to cause those 'ghost' keys to generate key presses.

Yes, that's exactly why they did it. There were some occurrences in a previous model where something would put pressure on the hidden contact, and someone would suddenly hear a note they couldn't shut off, so as unlikely as it was, this prevents that from occurring. An unfortunate side effect is that, previously, some people rigged up a key to get that low E working (and possibly the high notes as well), and now those mods don't work. So it would be interesting to know if they could be made to work just by manipulating these jumpers.

Okay, you've convinced me to give it a try. Seems like a legitimate enough reason (for me anyway).

I don't know when I'll actually get around to it, but here's to hoping I don't break it.

I decided, having owned it for nearly 10 years, it was high time my Nord Electro needed a bath (well, a clean out), so I stripped it naked. Haven't a clue what any of the chips do, though.

Thank you mjbrands, I was really wondering what's inside the NS2 and did not dare to open it.
I'm trying to understand why this architecture remains the same since 1995, as you noticed,since Nord Lead 1.

Wikipedia says that Freescale was Motorola, big name, not new to DSPs ...
Their Symphony DSPs family seems quite modern, with some dual cores. The 56367 seems not too old (2007 for the latest release).
Freescale seems up to date too with their development environment too : Eclipse plugin, various C libraries (encodings/decodings ... , probably using hardwired instructions ?). Actually they also build the "Power" processors for big IBM servers.

Korg kronos uses Intel Atom processor, but what DSP next to it ...
Roland is known to design theirs own processors ( I have read that the Jupiter 80 - 256 voices polyphony - processor is very powerfulll, don't know what this means exaclty ).

6 DSPs into a NS2, quite impressive. The good thing is that it stays cool.
I wonder if Nord redesigned the board when they added some memory from the NS to the NS-EX.

Cheers.

Whoops: this is an original Electro, not an NE3. Move along, nothing to see here

Ritchie333 wrote:Haven't a clue what any of the chips do, though.

I think this is roughly what is what.
Note: There seems to be extra room for Flash chips. I don't think they (Clavia) ever meant the NE to have more than 256 MB Flash; instead, they may have designed it to be able to use eight 256 mbit instead of four 512 mbit Flash chips.

The NE3 has two DSP chips. It seems (roughly speaking) one is in charge of effects, the other of instrument sound (samples, piano or organ emulation). Since the NE is only mono-timbral, it needs fewer DSPs than, say, a Stage.

frantzkb wrote:I'm trying to understand why this architecture remains the same since 1995, as you noticed,since Nord Lead 1.

The simplest answer would be... it works, and redesigns are expensive!

frantzkb wrote:I'm trying to understand why this architecture remains the same since 1995, as you noticed,since Nord Lead 1.

Don't fix it if it ain't broken? It makes sense to stick to technology you've already invested a lot in, to leverage every penny put into it.

Clavia's crown jewels are probably their DSP algorithms. They've been using Motorola Symphony DSPs (now sold by Freescale) since the original Nord Lead. Switching to a different type/brand of DSP might involve quite a lot of work; it might also slightly change the sound. Even if the change in sound is only perceived (and not actually measurable), they'll have users complaining.

I don't think they haven't innovated on the hardware side: they switched to a different micro controller (from a Renesas M16C to a Freescale ColdFire, maybe even others), they added the use of samples to the Electro products and later their Lead (Wave) and Stage product lines and they seem to have changed opamps and DACs for the analog section quite a few times. Their Modular line of products were quite interesting, as they opened up access to the functionality of their DSP algorithms much more than with any other product, but this in turn also made them much more difficult to get to grips with (what a shame there's no Modular G3).

I see a lot of parallels with Access and their Virus. Both Clavia and Access make high quality (and expensive) instruments based on the same DSPs. However, Access keeps cramming new functionality into their Virus and the stability sometimes suffers (though the Virus TI is much more stable now than it was a few years ago). In a sense both Clavia and Access keep adding new sounds for free to their products. I think the Clavia products are quite easy to use (a lot of people would probably disagree), but the Virus has too much functionality squeezed into it; you really need the (excellent, in my opinion) Total Integration software to keep it user friendly and manageable, even though the Virus itself is laden with knobs most of which have a dedicated function like on the Nords.

The Nords are really player's instruments, while the Virusses are more sound modules, sometimes with an (excellent semi-weighted, IMHO) keyboard. The Virus is meant to tightly integrate into a DAW, the Nords not so much. That is one gripe I have with the Nords: their PC/Mac software is limited to updating/backupping their instruments. I would really love it if they would (have a third party) develop a VST plugin to control their instruments. Sound Tower develops and supports a commercial ($60?) plugin for Moog and Dave Smith Instruments. Granted, integrating them into a DAW isn't really what the Nords are meant for, I can see that, but it would be nice for basement tinkerers like me. Whatever exists now (including the commercial 3rd party plugin for the NL2X) is based on reverse engineering.

frantzkb wrote:Korg kronos uses Intel Atom processor, but what DSP next to it ...

I haven't been able to find out if it actually has one. Modern CPUs are quite good at DSP tasks and the Kronos might not need one and might be able to run those DSP algorithms on the Intel Atom CPU instead. The Kronos' predecessor (in a way), the Korg OASYS, probably had quite a few DSPs in it.

frantzkb wrote:Roland is known to design theirs own processors ( I have read that the Jupiter 80 - 256 voices polyphony - processor is very powerfulll, don't know what this means exaclty ).

From what I've read, the Jupiter 80 is quite a powerful beast indeed. I still haven't forgiven Roland for stopping (software) development on the Fantom G however.

I do wonder about that 256 voice polyphony. That is likely a best-case scenario. The Virus TI2 maxes out at 90 voices polyphony, but once you start adding effects, more complex voice types, etc. you quickly go down to maybe 15-25 voice polyphony. Suddenly the 18 voice polyphony on the Nord Wave doesn't look so bad, since no matter what effects, etc. I use I am still able to hit that 18 voice polyphony.

Once the smaller Jupiter 50 comes down in price, I might look at getting one. I've little patience for poorly functioning 'vintage' equipment, so it doesn't bother me the only thing the Jupiter 8 and Jupiter 80/50 have in common is their position in the market compared to competitors.

Anyway, enough bla bla and speculation.

mjbrands - Thanks, but the photo is of an Electro 1. The NE2 hasn't been around that long, let alone the NE3! Look at the date on it - 4th June, 2002.

I'll have to admit, being a Monty Python fan, the title of this thread invoked some disturbing images. I'm relieved to see I was wrong.
Thanks for the post.
Cheers, Hanon

Ritchie333 wrote:mjbrands - Thanks, but the photo is of an Electro 1. The NE2 hasn't been around that long, let alone the NE3! Look at the date on it - 4th June, 2002.

I'd seen the date, but read it as 'June 2nd, 2004'. I hadn't really given it much thought