This release is an exploration of corrupting sound presets on a Nord Lead. Installing a new battery in the memory card results in random data being interpreted as presets. The aim of this project is to create an audio work that catalogues these sounds.
Each preset plays the same MIDI sequence. This is not only to show how each preset reacts to the same notes, but also to allow for layering if desired. The sounds presented are unmodified, apart from occasional normalisation.
Keeping the sounds unmodified is an intentional aesthetic decision; as such, there are occasional harsh, high frequency tones that may cause damage when played at high volumes. This release is intended primarily as an installation, rather than a traditional album.
The cover artwork, and individual track artwork, was generated using selected bytes from the sysex data for each patch.
Also included with this release is a full sysex dump from the memory card, compatible with the Nord Lead/Rack 1. Please note that the TXT file included must be renamed to ZIP in order to open; this is due to Bandcamp’s upload restrictions.
This release is presented under CC-BY license; please feel free to sample, remix, or otherwise use these sounds in your own work, but please include acknowledgement of the source (with a link to this album page).
Yes, cool idea, and hats off for going to the extra effort to make it all available, I look forward to more listening and sampling.
I am curious, what happens when you turn a knob on the nord, does it immediately snap back to a virtual analogue sound or can you affect digital weirdness?
A bit of both! Some parameters are weirdly out of range, mostly things like octave, or arp destination. Those parameters snap back when trying to modify them. Others act the same as on a normal patch, in that it’ll snap to the knob value if you move it. I have a feeling that the sysex dump I included doesn’t tell the whole story (ie. it might have snapped some values to their 8-bit limits), but the only way I can analyse that is if I dump it back to my Nord and compare, which I may do soon, especially now that I’ve documented the patches.
Technically I can’t see how you couldn’t? Each parameter will have a random value stored in memory. for example let’s say the resolution of the filter cutoff is 8 bits (the actual resolution is irrelevant) then it will have a value between 0 and 255 stored in memory. That same range of values is available via the cutoff knob. Likewise if you’ve got a button with 4 options then in memory there will 2 bits associated with that value. In other words that value can only be one of 4 possible values so even tho the memory is full of random values they will still only relate to the options possible for that control. Make sense?
Recreating those random patches is another story tho! Something that will get harder to replicate as the number of controls and their individual resolutions increases.
So for the 4 button modes the memory location associated with that button can only be one of 4 binary values,
00
01
10
11
The ‘random’ value after changing the battery can still only be programmed with one of those 4 values.
Each parameter is stored as a full byte in the sysex dump, which I guess is standard. That said, there are some parameters that went out of range, most commonly octave, or arp mode. Octave definitely goes very high when out of the standard settings, but I’m not sure what out of range arp modes mean. It’s an opportunity for future research, for sure.
Ah yeah I didn’t think about sysex. Wouldn’t the Nord ignore any values beyond what it need for that parameter? So just look at the LSB’s of that byte? Genuine question as I’m not well up on sysex.
Oh no, I didn’t mean to be rude.
Just saying that many things could have happen and it might result in something else than just randomizing the parameters.
It would be interesting to know !
Having listened to some of the original patches on Nord Lead 2 I actually think it is possible… there are some incredible patches in the factory banks, one imitating birds and wind at the same time for instance.
