Page 1 of 2
Hacking a Nord OS
Posted: 10 Aug 2024, 00:03
by Mac06
I was reading this article:
http://blog.petersobot.com/patching-the-k2500 about "hacking" the Kurzweil K25 OS using GHIDRA (a reverse-engineering tool developed by the NSA).
Wondering if anyone has ever attempted something similar on any Nord instrument?
Re: Hacking a Nord OS
Posted: 10 Aug 2024, 00:56
by FZiegler
Wondering if this was legal.
Re: Hacking a Nord OS
Posted: 10 Aug 2024, 02:37
by tsss27
I don't know remotely enough about this stuff to even attempt it but it sure would be interesting if someone did. Honestly I'd be more interested in someone reverse engineering the .npno files so we can figure out how to make our own piano samples...
Re: Hacking a Nord OS
Posted: 19 Aug 2024, 09:35
by cookie
tsss27 wrote: 10 Aug 2024, 02:37
I don't know remotely enough about this stuff to even attempt it but it sure would be interesting if someone did. Honestly I'd be more interested in someone reverse engineering the .npno files so we can figure out how to make our own piano samples...
In terms of reverse engineering, I would be most interested in having file information on nsmp, nsmp3 and nsmp4 (including different versions) in order to convert (with restrictions) any format to any other. NPNO is also interesting for sure, but from Nord perspective, it should be much more sensitive.
Fred
Re: Hacking a Nord OS
Posted: 27 Aug 2024, 07:23
by dmamfmgm
I made some good progress on the .nsmp and .nsmp3 file formats. However nothing close to fully understanding the format.
That's when I realized any tool I might create out of curiosity? I think it would quickly be applied to hijacking the Nord Sample Library. For my appreciation of the Nord Piano library, I decided to pause my efforts.
I expect this is a controversial topic. Keep calm and keep playing!
P.S. I'm totally convinced the field of reverse engineering will be swept by machine-generated reverse engineering soon, so for me, doing nothing is the smart move. Any executable code will likely become extremely easy to automatically decompile once the tools for it are published. There are so many things much more valuable than Nord Sound Manager where the secrets are hidden in the executable code.
Re: Hacking a Nord OS
Posted: 27 Aug 2024, 07:54
by tsss27
I suspect the main reason Nord has their formats protected in this manner is due to the fact that the vast majority of the samples are supplied by other companies. For example all acoustic pianos come from SampleTekk, created exclusively for Clavia. They supply various other sounds as well, along with ProjectSAM, Spitfire, Kirk Hunter, Sonic Reality, PrecisionSound and probably others I've forgotten. So the samples cannot be easily accessible as they're not even the property of Clavia themselves.
That doesn't explain why it's not possible for users to create .npno files, though...
Re: Hacking a Nord OS
Posted: 27 Aug 2024, 08:49
by maxpiano
tsss27 wrote: 27 Aug 2024, 07:54
I suspect the main reason Nord has their formats protected in this manner is due to the fact that the vast majority of the samples are supplied by other companies. For example all acoustic pianos come from SampleTekk, created exclusively for Clavia. They supply various other sounds as well, along with ProjectSAM, Spitfire, Kirk Hunter, Sonic Reality, PrecisionSound and probably others I've forgotten. So the samples cannot be easily accessible as they're not even the property of Clavia themselves.
That doesn't explain why it's not possible for users to create .npno files, though...
Maybe because that would reveal too much about how Nord Piano engine works (amount of layers, rules to interpolate them, etc...) and it is something that probably Nord is not willing to disclose, while instead for .nsmp this is not a problem being it just a mono layer wavetable with a small bunch of collateral parameters, nothing really special.
Re: Hacking a Nord OS
Posted: 27 Aug 2024, 15:45
by metalfingers
In terms of "hacking" the OS I think there is more value in adding back little features which were inexplicably stripped away between models.
For example I relied heavily upon the MIDI channel assignment per-engine in the Stage 2. The Stage 3 only allows MIDI channels to be assigned to a Panel, Dual KB, or Global. I've gotten by, but it was not ideal.
I have also been yearning for the NS3 to be a MIDI clock master. I have an NL3, downstream of the NS3 and its clock is about 0.15% off... which ends up being a full beat off after about 4min at 140bpm...
Re: Hacking a Nord OS
Posted: 28 Aug 2024, 08:24
by tsss27
maxpiano wrote: 27 Aug 2024, 08:49Maybe because that would reveal too much about how Nord Piano engine works (amount of layers, rules to interpolate them, etc...) and it is something that probably Nord is not willing to disclose, while instead for .nsmp this is not a problem being it just a mono layer wavetable with a small bunch of collateral parameters, nothing really special.
It would be rather odd that Nord would specifically not want this information out there, since the ability to create a format does not mean we would be able to deconstruct existing files, just as it is with nsmp. Furthermore it'd be quite easy to figure out how many velocity layers an existing piano has via MIDI, unless samples are velocity crossfaded and then it'd be more difficult. But I am quite sure none of this technology is that special either, it is just a multi-velocity nsmp with additional layers for release, resonance and perhaps pedal noise... So I don't think there is much for them to hide. But none of us know for sure anyway...
Re: Hacking a Nord OS
Posted: 29 Aug 2024, 13:44
by analogika
They have good reason not to make their proprietary Nord Piano format public.
For one, it gives them full control over the Piano Library, which is arguably their biggest asset next to their user interface.
For another, it might allow people to reverse engineer it, which could allow others to extract the piano samples and repackage them.
In any case, it’s their decision.