Converting Synth1 presets to Hive?

[sl23]

I have tried to convert these to Airwave, but it is proving massively difficult seeing as no-one knows how fxp/fxb are structured. I cant seem to find any info on them.

So, I thought, maybe Hive would be a better bet, it has many of the parameters, not exact, but then no plugin is, even those based on Nord synths.

I checked the h2p files and wondered if it would be a lot easier to do the conversion using Hive as a host for these converted presets instead of Airwave. But I need some info on the preset files:
1. Preset files are structured like this:

/*@meta

Author:
'ALTER BOY Sounds'

*/

#AM=Hive
#Vers=200
#Endian=little
#nm=21
#ms=none
#ms=ModWhl
#ms=PitchW
#ms=Breath
#ms=Xpress
#ms=Sequencer
#ms=Gate
#ms=Velocity
#ms=Pressure
#ms=KeyFollow
#ms=KeyFollow2
#ms=Alternate
#ms=Random
#ms=Constant
#ms=Amp Envelope 1
#ms=Mod Envelope 1
#ms=Amp Envelope 2
#ms=Mod Envelope 2
#ms=Vibrato LFO
#ms=LFO 1
#ms=LFO 2
#nv=5
#mv=Gate
#mv=Amp Envelope 1
#mv=Mod Envelope 1
#mv=Amp Envelope 2
#mv=Mod Envelope 2
#cm=main
CcOp=100.00
#cm=PCore
SBase=4
Swing=0.00
STrig=1
PSong=0
PFold=0
PFile=1
GFile=2
GScale=0
ChLay=0
SurrO=1
Rev=3304
LED=0.00
PAGE=0
PagesOn=0
CoreN=3
Slice=4
UI_op=1
MidiA=5
MidiP=6
#cm=MM1
Active=1
Source=0
Via=0
Dest1=none:assigned
Depth1=0.00
Dest2=none:assigned
Depth2=0.00
#cm=MM2
Active=1
Source=0
Via=0
Dest1=none:assigned
Depth1=0.00
Dest2=none:assigned
Depth2=0.00
#cm=MM3
Active=1
Source=0
Via=0
Dest1=none:assigned
Depth1=0.00
Dest2=none:assigned
Depth2=0.00
#cm=MM4
Active=1
Source=0
Via=0
Dest1=none:assigned
Depth1=0.00
Dest2=none:assigned
Depth2=0.00
#cm=MM5
Active=1
Source=0
Via=0
Dest1=none:assigned
Depth1=0.00
Dest2=none:assigned
Depth2=0.00
#cm=MM6
Active=1
etc...

Is there a map of all Hives presets for easier mapping? So I know what parameter on the gui saves to which line in the preset file?

2. At the end of the file, there is binary data, is this necessary and how is that generated? How would I generate that info via a conversion script? What is it for? Is it related to licensing? Are users not allowed to touch this?

I am using Github Copilot ChatGPT5.4 to create the scripts for this, which is what I have been doing for Airwave, which for the most part has been successful, but is getting really problematic even just trying to get naming added to the converter script.

Anyway, if anyone has any info on this, it'd be greatly appreciated. Once done it'll mean 10,000 presets ready-made straight from Synth1! Thanks for any help you can offer.

[jooster]

Even if a conversion were available, would you expect those presets to sound in Hive the way they were in Synth1? Also, i myself would not look forward to 10k worth of presets - would it not be better/quicker/educational, to rebuild just those few from Synth1 you like in Hive?

[sl23]

Better - not necessarily.
Quicker - not a chance.
Educational - yes, if you're into that.
They will most likely sound similar but different. Whether you are interested wasn't on the table, it was a personal request.

[Seafire Mk2]

It's not possible

[sl23]

Why? I got nearly half way there with Airwave, and that uses fxp files which are binary or hex, but naming presets using a converter script got really messy, so I abandoned that and am trying Hive which uses simple text files, so I'm betting it can be done. All thats needed is a script to read synth1 preset files and convert that to the appropriate name and value used in the Hive preset file. Why do you think it cannot be done? I'm not expecting a 1:1 conversion. Thats not the point.

[ThoughtExperiment]

So now we know what isn't the point, could you tell us what is?

[pdxindy]

Why? I got nearly half way there with Airwave, and that uses fxp files which are binary or hex, but naming presets using a converter script got really messy, so I abandoned that and am trying Hive which uses simple text files, so I'm betting it can be done. All thats needed is a script to read synth1 preset files and convert that to the appropriate name and value used in the Hive preset file. Why do you think it cannot be done? I'm not expecting a 1:1 conversion. Thats not the point.

I wouldn't know how to do it, but it sounds like a fun project. (Don't listen to the grumpy old naysayers.)

Seems possible, except if there are features in Synth1 that are not in Hive. For example, if Synth1 has Osc to Osc fm, then presets using that feature would not have an equivalent in Hive.

[DrGonzo]

You probably don't want to hear this, but what you are doing is not going to work.
It would be quicker and more accurate to convert the presets you want, manually by hand. Preset by preset.
I also experimented a lot with ChatGTP to mess around with parameters in synths and after countless lost files, hours and episodes of insane AI psychosis the most probable thing you could get out of it is a semi-random preset creator where the results will be mediocre at best.

But it's a great exercise to experience the limitations and how overhyped AI is. At least at this point in time. Maybe in a few years it will be better. But right now - a complete waste of time.

/C

[sl23]

Thanks for the input! Like I said, I do understand the conversion won't be 100% accurate, but I want to see how it turns out.

I know the limits of AI all too well! I've spent the last three months on various projects, some went well, others not so. From my experience, it always starts with perfect code, everything works. Then you get so far and the AI suddenly develops alzheimers! Maybe that is the literal truth seeing how we're going through this ram crisis?! Lol

Anyway, again, as I said above, I managed to get around 25-30 parameters converted in airwaves and they work perfect even now, but fxp is a pita to work with! Hive should work due to its much simpler preset file format.

There isn't a chance in hell it would be quicker by hand! 10000 presets? Off ya go then! Let me know when you've done one full bank and tell me how long it took.

The point is that I want to do it, like you people want to create your own presets. Your opinions and understanding of why someone has a desire to do something aren't relevant. What is it with people? I just hoped to get some help with something and just keep getting told it's pointless. To YOU it's pointless, keep your negativity, I want to do it because I want to try. Ya think edison had the same negativity when trying something new for the sake of it?

All I was asking was if there were any parameter mapping available, which most likely be from the devs anyway! So, devs, please can you supply this data if it's easy enough to do so, thanks.

[tasmaniandevil]

All I was asking was if there were any parameter mapping available, which most likely be from the devs anyway! So, devs, please can you supply this data if it's easy enough to do so, thanks.

In Hive, right-click the preset save button and choose “h2p extended”, then save a preset. Open this preset in your favorite text editor. The preset will contain lots of comments explaining each parameter.

[DrGonzo]

The point is that I want to do it, like you people want to create your own presets. Your opinions and understanding of why someone has a desire to do something aren't relevant. What is it with people? I just hoped to get some help with something and just keep getting told it's pointless. To YOU it's pointless, keep your negativity, I want to do it because I want to try. Ya think edison had the same negativity when trying something new for the sake of it?

Ok Mr Edison - you have a valid point.

I absolutely can understand the exhilaration of trying to finish something like this. It can be fun and educational - almost like playing an advanced puzzle game. I've done and still love to do conceptual experiments like that.

But there is a difference between having fun trying to solve things and actually trying to accomplish what your end goal is. I mean, I can say I am going to climb Mount Everest, blindfolded and only by jumping on one leg. You can stand in your back yard practicing jumping on one leg blindfolded, but that doesn't make the actual climb possible to go through with.

To put it in different words. 15 years ago I was at the Modwiggler forum talking about how to build a Juno 60 replica using Eurorack modules. I got some replies and suggestions about modules, but the most common reply was "Why don't you just get a Juno?". I did my experiment and I learned stuff from it. But a Juno it wasn't.

/C

[sl23]

All I was asking was if there were any parameter mapping available, which most likely be from the devs anyway! So, devs, please can you supply this data if it's easy enough to do so, thanks.

In Hive, right-click the preset save button and choose “h2p extended”, then save a preset. Open this preset in your favorite text editor. The preset will contain lots of comments explaining each parameter.

Fantastic! Thanks for that

[sl23]

@DrGonzo: I appreciate your position. I am fully aware of the crap that AI pumps out! I am no programmer so am relying on it 100% to do any type of coding. I know "vibe-coding" is looked down on, but if you don’t have the knowledge to do something and there is a tool to do it for you, what are you going to do? Fro simple scripts, which is what this will likely be, then I am fairly confident it will work. But rather than try convince you, I will get it working and post back in a week or two.

Like I said, I got many parameters mapped perfectly in Adam Szabo's Airwave if you care to test them? But it seemed to get overly complex when naming was introduced into the converter. For some reason, it doesnt just change one code block, it changes several other blocks, even those assigned to parameters themselves, thus breaking them! Meaning you get Cutoff assigned correctly to a block of Hex/Binary, but that block then becomes void when naming is introduced as the act of adding a name changes how the Cutoff is saved. I actually got quite a way into it and only because of this complexity, I stopped. That is where the AI fails... Not for basic plain text to plain text conversion, that’s simple to do, I could do that by hand, but for 96 parameters over 10000 presets, that’s a lot! Id be dead before I finished it!

Check my Github page for projects Ive made so far with AI. Not perfect, most of it is still a work in progress, but some works as is. So yeah, I get the limits of AI, but just something to do! My favourite so far has to be PolyHost! I quite like that one, and its functional for my purposes. I am hoping to maybe add support for older plugins, but that went terribly first time around, so not looking good. I also think it would make a great VST3 plugin, but again, I think that may be too much, although I haven’t tried that yet. I'm trying to get the app itself up and running with everything first. It's not been easy, but I've learnt AI is really good at first, then you have to start a new chat, explain what’s been done and sometimes you can then continue with a good refresh, other times the shite sticks to yer shoe and follows you to the next chat! But, it's just for fun. I cant stick to anything for long... I have ADHD! So I skip from one interesting thing to another! I don’t want to be like that, but nothing I can do, what Ive been dealt with! So, this is the latest of crazy interesting things (to me at least) that I am trying out. Let me know when you finish that Synth1-Hive bank btw

[DrGonzo]

Maybe try to make a Synth1-compatible plugin instead? I am sure lots of people wouldn't mind if it was Vibe coded or not, as long as it worked. Just a thought. In either case, good luck - I'd be happy to see you succeed.

[sl23]

Well, I found Synth1 to be a buggy plugin, hence the decision to try this out. Thanks for the sentiment!
another reason I wanted to convert was simply to cut down the amount of plugins I have. I was just trying this out for starters to see if it could be done. But the Airwave incident proved not. Because if it could be done, I would’ve tried converting more synths, mainly older less complex ones, but it failed in that respect. Hive should be much easier??? lol