Nord User Forum

[cookie]

Nailed it, at least for identifying the same program content !

Idea was first to strip the ns2p files and work on copies : starting at byte 12 and stripping the last 2 bytes.
I used DD.EXE with the following command :
DD bs=1 count=531 if=%i of=%i_ skip=12
(run in a batch file with a FOR %i in (*.ns2p) )
This gives you 531 bytes files which can be compared with FC /B for exact match. Seems findstr is only for text files but more interested in the "comp" internal command.

But in order to have comparison difference value per file pair, I piped a neat : find /c /v ""
Rerouting to a text file, I have now the names of the 159600 possible couples (399*400) with the byte number difference.

I used the XDF (Exact Duplicate Finder) to speed things up in finding duplicates, finding 84 duplicates out of my 378 programs !!
Will clean this first as this was my target goal.

Next will be to compare the stripped couples with only 1 byte difference and identify which function with which value at what position.

Fred

[Mr_-G-]

[I disagree.
I know it has been done for the DX-line of synths which has a propriety format as well.

Do you mean Yamaha DX series? That is well known format for storing patches.
Sure you can compare binary as I already pointed out, but you cannot tell what is being changed. You do not know whether there are several settings packed into the same byte or not, like in the DX-7 synths. Just changing the volume of one section or having a different (yet inactive sample in the piano or synth or slot section) would return a different value.

[cookie]

[quote="Mr_-G-"... but you cannot tell what is being changed. You do not know whether there are several settings packed into the same byte or not, like in the DX-7 synths. Just changing the volume of one section or having a different (yet inactive sample in the piano or synth or slot section) would return a different value.[/quote]

I believe we all agree on that.
What can be done is to find and discover minor changes based on few byte changes, say on 1 parameter to start with.

0 byte difference : same file
1 byte difference : same bank or same program #
2 bytes difference : starting to have more combinations (one being bank program # different)

I'll not spend time comparing 2 stripped ns2p that have 185 different bytes
But surely will take a look at the ones with few bytes differences and will try to understand the byte-to-parameter values.

Idea again is to clear programs that are almost equivalent say 1 is transposed, the other not.

A much wider project (would nns2p format stays the same ) would be to gather all the byte-to-paramter difference and integrate that in a CTRLR environment or just an HTML parameters export file.

Fred

[maxpiano]

Wouldn’it be easier if you choose a single Program (existing or brand new, for example starting from one of the Inits) then change one parameter at a time-> upload to PC -> compare?

[cookie]

Wouldn’it be easier if you choose a single Program (existing or brand new, for example starting from one of the Inits) then change one parameter at a time-> upload to PC -> compare?

I agree but remember my initial goal is only to identify similar programs.
Therefore not getting too deep into the byte-to-function comparison.

Thing is, once I have my "exact" duplicate programs cleared out, I will analyse 2 programs with small bytes difference and easily find the corresponding function.
We could do cross check comparison with your proposal but the main problem is that I don't have 100 enmpty program slots to make all tests with 1 single function change

Fred

[florence]

My use case: I own a NS3, and many programs here are for NS2 and I cannot use them of course. This is a bit annoying but my wish would be to read these ns2p file and display all the settings to try to reproduce it on the NS3. No automatic converter, just having the settings would help to create them from scratch.

This could be added to the Sound Manager of course but I don't think that Nord will add this one day.

For the fun I'm trying to read the ns3f files and except the CRC formula (that I don't need to understand, I don't want to change any file), the format is pretty easy to understand. I decoded the piano section, and I guess all other sections are similar. It is just very time consuming, changing each parameter, one by one, exporting the file with the Sound Manager, comparing the hex value with previous version...

But I would need that for NS2 program and having no NS2 it is impossible for me. Maybe if I ask Nord they would lend a gear

thanks, i think I'm making progress!

I know this is a very old thread but wondering if you could share your finding...

[rpossemo]

Definitely looking for a ns2p file description to be able to mimic the same on my NS3. Just what @florence said on previous post. Please make that tool/info public if you can!

[cookie]

Happy to see this thread back on top after 2 years !
Yes Florence, I can feel your frustration in not being able to convert directly or at least have value information from an NS2 program.

In this thread, my initial goal was to identify ONLY the exact same programs (.ns2p) in a backup file (.ns2b). Success, eventhough this would need a little clean programming to make it a non-geek tool.

Back to the need to extract parameters from the ns2p, that's a challenge !

First of all, if there are not that many programs, I can load them in my NS2EX and use CTRLR to share the parameters (send the files in PM). OK, doesn't answer the ns2p decyphering but provides a quick solution

Back to the topic, this is tedious work and will need lots of time to change/save/compare the hex values.
I also don't believe Nord will share the ns2p file structure and remember this could be useful for the Piano/Organ section but a bit tricker for the Synth section.
Am I correct in saying the most important parameters would be the effect section ?

Fred

[florence]

Right some frustration about this format. I’m really wondering why Nord is not providing such simple utility tool to View the parameters even in a simple list mode.

Thanks Fred for your offer in pm

[cookie]

Florence,

If OK with you, would like to share your ns3p post here : https://www.norduserforum.com/nord-stage-3-programs-ns3p-ns3pb-files-f32/ns3-program-viewer-t19939.html
Great work (https://ns3-program-viewer.herokuapp.com/) !

I wonder how we could catch the train and have the ns2p format deciphered, in order to help with the v3 transition ?
Anyone willing to give it a try (could split the tedious work into Piano/Organ/Sample/Effect sub-section) ?

Again, sample section is quite a different story between v2 and v3 but this could be useful for quick nsXp format conversion.

Fred