Free FM Synthesizer Dexed (VST Windows and Mac)

[fmr]

But a 6-Op FM synth wouldn't sound the same as a 4-Op one, since the algorithms aren't the same. If you convert 4-Op algorithms to 6-Op ones, the sound may change anyway. So if you're looking for "authentic" 4-Op FM synthesis, wouldn't it make sense (and be easier) to just use a 4-Op synth?

The problems is not the algorithms. AFAIK, all the algorithms a 4-Op synth can have, ar also in the 6-Op line. The problem resides in the waveforms. To go over the limitation of only have 4 Operator, Yamaha introduced several waveforms in the 4-Op line, which are not present in the 6-Op line (this one only had sine waves).

Only later, with the SY77/TG77/SY99 line, did Yamaha introduce those waveforms in the 6-Op line which was then called AFM (Advanced FM).

AFAIK, DeXed only produces sine waves too, so the 4-Op presets will probably sound completely different, due to the lack of the proper waveforms.

[Failed Muso]

The problems is not the algorithms. AFAIK, all the algorithms a 4-Op synth can have, ar also in the 6-Op line. The problem resides in the waveforms. To go over the limitation of only have 4 Operator, Yamaha introduced several waveforms in the 4-Op line, which are not present in the 6-Op line (this one only had sine waves).

Not strictly true. You can approximate some of the 4-Op algorithms in a 6-Op, but none of the algorithms are freely assignable, so importing 4-Op sysex just wouldn't work. You can, however, get close if you're prepared to program the 6-Op machine from scratch. MadFame on YouTube did something like this to approximate the Solid Bass 4-Op patch on a Mk.I DX7...

https://youtu.be/4LcJsiNkpb4

AFAIK, DeXed only produces sine waves too, so the 4-Op presets will probably sound completely different, due to the lack of the proper waveforms.

And this is one of the same reasons why they wouldn't work on a DX7

[fmr]

The problems is not the algorithms. AFAIK, all the algorithms a 4-Op synth can have, ar also in the 6-Op line. The problem resides in the waveforms. To go over the limitation of only have 4 Operator, Yamaha introduced several waveforms in the 4-Op line, which are not present in the 6-Op line (this one only had sine waves).

Not strictly true. You can approximate some of the 4-Op algorithms in a 6-Op, but none of the algorithms are freely assignable, so importing 4-Op sysex just wouldn't work. You can, however, get close if you're prepared to program the 6-Op machine from scratch. MadFame on YouTube did something like this to approximate the Solid Bass 4-Op patch on a Mk.I DX7...

Well, he said it himself in the video: "I started by define an algorithm that matched the one used in the DX100."

So, the problem isn't the algorithm

[gentleclockdivider]

I re-created solid bass for dexed ( it's an fxp file ) and for zebra

You guys might also want to check out aurora fm , cheap as chips sounds great and has all the tx waveforms

[Failed Muso]

Well, he said it himself in the video: "I started by define an algorithm that matched the one used in the DX100."

So, the problem isn't the algorithm

Well, he didn't quite say that, but hey.

None of the algorithms match. Some are very close and with a bit of tweaking you can approximate, but none of them match. As he says, the DX7 doesn't have algorithm 2 from the DX100, but it does have algorithm 14 which means you have to go in and turn off operators 1 and 2 and use their settings on operators 3 and 4. All of this is manual work and you couldn't simply send over the 4-Op sysex and have the 6-Op synth remap it.

So yeah, the algorithm is kinda the problem

[fmr]

You guys might also want to check out aurora fm , cheap as chips sounds great and has all the tx waveforms

Arturia DX7 V has all the TX81Z waveforms as well, and a lot of other features. And it imports Sys-Ex perfectly, with an extraordinary fidelity, and has a great sound

It may not be as cheap as aurora, but it has a much better GUI, and offers a lot more.

[fmr]

Well, he said it himself in the video: "I started by define an algorithm that matched the one used in the DX100."

So, the problem isn't the algorithm

Well, he didn't quite say that, but hey.

None of the algorithms match. Some are very close and with a bit of tweaking you can approximate, but none of them match. As he says, the DX7 doesn't have algorithm 2 from the DX100, but it does have algorithm 14 which means you have to go in and turn off operators 1 and 2 and use their settings on operators 3 and 4. All of this is manual work and you couldn't simply send over the 4-Op sysex and have the 6-Op synth remap it.

So yeah, the algorithm is kinda the problem

Of course you will have to shut down some operators - that's implicit

If you have six operator in one side, and only four in the other side, you definelty will need to shut two down. But you may have a MATCHED algorithm, and that process could be automated in a conversion program so no, the algorithm is not the problem. The problem resides in the waveforms. If you had those in the DX7, you could create perfectly matched presets.

[Failed Muso]

Of course you will have to shut down some operators - that's implicit

If you have six operator in one side, and only four in the other side, you definelty will need to shut two down. But you may have a MATCHED algorithm, and that process could be automated in a conversion program so no, the algorithm is not the problem. The problem resides in the waveforms. If you had those in the DX7, you could create perfectly matched presets.

But there are no directly matched algorithms between the 4-Op and 6-Op Yamaha synths and you can't do what this guy did with Solid Bass on all 4-Op patches because they are inherently different and there are not always algorithms where you can simply turn two operators off like he did here.

So it would be an utterly inconsistent approach.

As for the waveforms, these present an insurmountable issue as they were only present in the 3rd gen 4-op DX/TXs (TX81Z, DX11, V2).

Now, if you're using something like the new Essence FM from Kodamo, that features freely assignable operators and multiple waveforms, you could get far closer.

[fmr]

Now, if you're using something like the new Essence FM from Kodamo, that features freely assignable operators and multiple waveforms, you could get far closer.

Arturia DX7 V All that and more.

[Failed Muso]

Arturia DX7 V All that and more.

I must have missed DX7 V's freely assignable operators...

[fmr]

Arturia DX7 V All that and more.

I must have missed DX7 V's freely assignable operators...

They are not freely assignable, but you have feedback on ALL operators, and you have a very flexivle matrix modulation, where you can have the envelope or the keyboard scaling (but not the output) of one operator as an additional modulator to any other operator.

And Native Instruments FM8 has freely assignable operators. As well as Blue II, Rhino and many others, There is no lack of very flexible FM synths out there. And all of them are very capable synths in several other areas as well.

But in what concerns the 4-Op presets, DX 7 V is perfectly capable, while being a very close DX7 emulation. And that's what was under discussion

[skippy63]

Hello,

I tried to convert various 4 op FM sysex to DX7 format without success. I used DXconvert 3.0.4 with python.

The sysex contain DX21/DX27/DX100/FB01 sounds. I know that conversion is not perfect, but I assume that lot of presets will be ok. And I like 4 op sound, different from DX7 character.

So, if anybody could convert those sysex to DX7 format, I would be very happy.

Thank you for any help

I'm not sure if this imports Yamaha 4-Op patches, but it is a great emulation of the TX81Z/DX11 engine that featured the extra waveforms in the operators...

https://www.sonicbits.com/exakt-lite.html

Yes, it's a fine synth, but there is no way to import sysex, so the purpose to convert Yamaha 4-Op sysex to DX7 sysex format.

But a 6-Op FM synth wouldn't sound the same as a 4-Op one, since the algorithms aren't the same. If you convert 4-Op algorithms to 6-Op ones, the sound may change anyway. So if you're looking for "authentic" 4-Op FM synthesis, wouldn't it make sense (and be easier) to just use a 4-Op synth?

I'm just curious about this, and not criticizing your approach.

Steve

With the inactive operators, it seems possible to have the same schemes that the algorithms of 4-Op.

[planetearth]

With the inactive operators, it seems possible to have the same schemes that the algorithms of 4-Op.

It seems possible, yes, but as @Failed Muso pointed out, it may not be. Or, in the case of what @fmr mentioned about the waveforms (which I'd forgotten), it doesn't even matter if you have the same algorithms, since the different waveforms will determine the ultimate sound.

I used to have a link to a chart that showed which 4-OP algorithms were available in 6-OP algorithm synths, and/or what you'd have to do to create similar algorithms. I can't find it now, but if I do, I'll post it. The bottom line is that they didn't necessarily "line up" the way you'd think, where there are some 6-OP algorithms that are basically some of the "better" (more musically usable) 4-OP ones with two extra operators stuck on that you could disable/mute/remove or not use during conversion. It doesn't work that way, however; all the operators are in an algorithm for a reason, so....

Steve

[surreal]

Cool!

[Dallas_Cashin]

Has there been any progress towards a scalable GUI? Sort of cluttered and hard to use on modern resolutions.