Overestimated synths?

[ghettosynth]

The DX7 was unique in the early 80s, it's simply not today unless you're willing to program it.

So it is, in fact, unique. Except when you can't be arsed. Brilliant.

The point was that in the early 80s it was unique with presets. Today those presets aren't enough to make it unique. If you're willing to program it, however, then it can be interesting again if you appreciate the subtle difference between its hardware and software emulations.

I think that you're the only person here who didn't get what I was saying, but, if it wasn't clear to others, by all means, sound off so that I know who else is having difficulty following a friendly conversation?

[ghettosynth]

Dave Bristow: Programming the first DX7 Presets

https://www.youtube.com/watch?v=JVsV6wAB5Gg

I love the statement about going through all of the voices of the orchestra and then asserting that we can't think of any more instruments.

Also, I like how defensive he is about how the DX isn't warm. Well, that's because it isn't. Statements like that are seldom absolute and interpreting them as such is where intellectual weakness lies. They are intrinsically a statistical statement, which, he himself makes in the video when he says that the cold voices are what the DX does so easily.

[jancivil]

synths are overrated

They are here. Everything is synths.

[jancivil]

I was programming DX7 extensively in '85-87.
...
And 'it's no longer unique'? Because, what, FM7, FM8? No. Those do_not_have the 32 algorithms. (If you don't get that, well you don't get something, so what.)

LOL! They're all there. You must have missed that part of the class.

Yeah, you don't even know what I'm talking about. So, find a picture of the actual DX7 membrane, compare it with FM8 and take a few seconds with that.

Never once did I think 'this would be better with a filter'.

Well, Yamaha did, that's why the FS1R has them. The technology for real time digital resonant filters just wasn't there in the early 80s so there was no practical way to incorporate analog filters while keeping the costs down. As soon as it was somewhat practical, Yamaha started incorporating filters.

I don't even know what that product is. Chowning-cum-Yamaha FM was not concerned with filters, and while I was making music all day long with a minimoog years before, 'analog filters' never occurred to me as I worked with this. So there's my_view reiterated.

Your assessment of the situation in hindsight is a different subject afaic. Fine, as I said: as far as one's estimation, ya do what yez like.

FM WAS overestimated at the time. That's exactly the point. The attitude that you are expressing was commonly held. While the theory holds that FM can synthesize any sound, in practice, this is unrealizable with any practical combination of operators and programming skill. I don't for a minute believe that Yamaha's engineers weren't smart enough to know this. Consequently, the decision to exclude filters was purely practical. It would have added significantly to the cost and would have diminished the value of FM as being this new futuristic technology.

Well, I did not believe it could synthesize any sound, and while G Leuenberger was in business with the thing & liked to say 'sample it with your mind' (and work it out within this paradigm), the band I worked with bought an Ensoniq Mirage around then. (And their composer liked working with AUDIO and so I would cut up a lot of samples. At first on a tiny LCD in hexadecimal, then eventually there was teh GUI on the 128K MacIntosh.)

Were you in on Yahama's decision-making in the early-mid 1980s?

I don't think, from my vantage point which is pretty good here, that filters were part of the thrust at all, that CHOWNING FM was. Your readiness to dismiss the whole of it as marketing notwithstanding.

The fact that some people hold onto this misunderstanding of the difference between theory and practice is a testament to the success of marketing FM, but that doesn't make it any truer now than it was then.

That's not a fact, that's a statement of your disposition to this.

However, it was held, rather arrogantly, by academics back then as well. John Strawn writes (in 1985) "Like additive synthesis, FM permits control of the audio spectrum with enough precision so that the composer has adequate control over the resulting sounds"

I guess that depends on who's defining "adequate," no?

I guess it's located in 1985.
It occurs to me that "adequate control" does not exactly mean <ultimate control unto perpetuity>.

I think 'it would have been so much more useful with analog filters' is ignorant and comfort-zone laziness, intellectually.

Again, see FS1R and perhaps ask yourself why it has filters if they are unnecessary?

That isn't the DX7. So you're able to suit your argument by such a tactic, great.
From_my_viewpoint, they are not.

If I liked the FM8 I would still look to the paradigm for the sound. It isn't subtractive synthesis. I like fast filters too, FWIW but then again you and I are just very different people.

AND, you may locate my general personal attitude with you (and what I find to be arrogant in context) with all the absolute ad hominem you tossed at me over years where I'd done nothing to you, like I owed every silly person such an ass kissing. You check your own 'tude, yeah? And here, it's as though posturing very boldly makes you know more than you know.

In sum, I don't think you get the power of this paradigm. My experience with it shapes my sort of worldview. It was significant. It seems to me that your lack of experience with it combined with a strong predilection for what you do understand instead shapes yours.

[jancivil]

It isn't much and proof of nothing (except this will have to have been DX7 in '84-85), but I found this track online (hey, the vinyl fetches 250 bucks on Amazon now so there's that) weirdly enough.
(Name of patch = track name. Track 2 from PFS Illustrative Problems, Cuneiform Records)

https://www.youtube.com/watch?v=C5pdDz- ... e=youtu.be

[ghettosynth]

I was programming DX7 extensively in '85-87.
...
And 'it's no longer unique'? Because, what, FM7, FM8? No. Those do_not_have the 32 algorithms. (If you don't get that, well you don't get something, so what.)

LOL! They're all there. You must have missed that part of the class.

Yeah, you don't even know what I'm talking about. So, find a picture of the actual DX7 membrane, compare it with FM8 and take a few seconds with that.

You're not talking about anything of consequence. The statement is dumb as it stands. It is trivial to implement any of the algorithms of the DX7 in FM8.

don't even know what that product is. Chowning-cum-Yamaha FM was not concerned with filters, and while I was making music all day long with a minimoog years before, 'analog filters' never occurred to me as I worked with this. So there's my statement reiterated.

Your assessment of the situation in hindsight is a different subject afaic. Fine, as I said as far as one's estimation, you do what you like.

Not relevant. I simply said that it would have been a better synth with filters and that since it did not have filters, it was overestimated as there was a common belief, as you so clearly articulated, that they weren't necessary. Chowning's goals were academic, I understand that POV, it has jack all to do with whether or not a hybrid architecture is a better tool. It clearly is "as a synthesizer."

The fact that you aren't aware of the FS1R simply speaks to your own ignorance of the development of FM.

[jancivil]

So, I really cannot tell if the actual 32 algos are there in FM8. It gives me a headache, it's not worth doing.
So you go ahead and consider me too stupid for it, that's fine. Have a look at the actual UI for DX7 one day.
So how am I to implement it? You don't know what I mean, and I don't need to make you.

Yeah, evidently I'm ignorant of where development went. What did I say? I don't even know what that product is.

It's not me that's ignorant of the power of the paradigm; that's what interested me.


As far as me failing to get what friendly conversation is, you're a flaming hypocrite.

[jancivil]

Chowning's goals were academic,

GAHD that's arrogantly dismissive.

You're not talking about anything of consequence.

For me, it is. If you had that experience, you would get it. I would suggest you look at
fm theory and applications
algorithm by algorithm and actually compare it with the GUI of FM8, actually thinking about it.

But that seems too much to ask, because you're so sure you know better.

[ghettosynth]

So, I really cannot tell if the actual 32 algos are there in FM8. It gives me a headache, it's not worth doing.
So you go ahead and consider me too stupid for it, that's fine. Have a look at the actual UI for DX7 one day.

Don't put words into my mouth. I said your statement was dumb, and it was. If you cannot tell that all of the algorithms are there then you have a deficit in your conceptual understanding of FM. If it gives you a headache then it seems that you come at FM with a naive point of view. It should be refreshing in it's abstract openness. Thinking of FM in terms of algorithms is a product derived POV.

I've owned and still own several DX, TX, and SY series synths, so I'm not sure what you're trying to suggest about the membrane panel. I cut my teeth on a DX-21. I don't think that you're familiar enough with FM to understand the value of compromising Chowning's concept to make a better instrument. Yamaha has been doing this since the very early days starting with the the TX-81Z where they added different waveforms to compensate for fewer operators. It doesn't really do that FWIW, but it does realize different timbres. Similarly, adding filters allows you to realize timbres that are either impossible or impossibly difficult to get with pure FM.

Reducing the amplitude of a modulator to reduce harmonics has a similar effect as filtering, but in practice, it no more sounds the same as two different filters sound the same. Filters, and in fact, any per voice post operator processing, adds interest to an FM synth because you can structure your voice design to take advantage of the best aspects of each architecture.

Had Yamaha put sixteen analog filters in the DX7 along with the associated D/A converters, it probably would have doubled the cost, sold one one hundredth as much, or fewer, but I guarantee you that they would be worth an absolute mint today.

[ghettosynth]

Chowning's goals were academic,

GAHD that's arrogantly dismissive.

I know you might think that, but, IIRC, you didn't finish school, so I think that my POV is a bit more developed here than what you imagine it to be. I'll give you a hint, it's not actually dismissive, it's simply a recognition that the goals of academic publication are not the same as that of product development.

You're not talking about anything of consequence.

For me, it is. If you had that experience, you would get it. I would suggest you look at
fm theory and applications
algorithm by algorithm and actually compare it with the GUI of FM8, actually thinking about it.

But that seems too much to ask, because you're so sure you know better.

LOL, I have that book, among others. Again, it's completely lost on me why you are so dependent on the idea of an algorithm. It's a preset for the modulation matrix. If you understand FM, it's of no consequence at all. In fact, it's a hindrance because if you start with some structure and realize that you need to change it you are dependent on the instrument to preserve your intent in terms of settings while changing it. When you have a flexible matrix, this isn't true at all.

In fact, once you get past a naive point of view you realize that DX-FM is itself a hybrid architecture. Some of the algorithms aren't FM at all. Six parallel operators is really just additive synthesis. Similarly, three parallel chains of pairs that are often used for electric piano patches are also an additive combination of simple FM. Similarly, when we think about subtractive synthesis, it's really often a combination of additive techniques and even FM techniques. To think of any of these technologies in isolation as some form of sacred cow is rather myopic IMO.

BTW: FM8 isn't even the first non-yamaha synth that challenged the uniqueness of the DX7. I enjoyed FM programming on the Nord G1 where you could extensively modify the voice structure long before FM8, or even FM7 came out.

[ghettosynth]

Enjoy!

"It's a powerful combination of technologies"
"It's powerfully amplified by changes in technologies....what we can do today"
"We're lucky when we can see a day when this can come together."

https://www.youtube.com/watch?v=YB9a2lP ... e=youtu.be

[.jon]

For me, it is. If you had that experience, you would get it. I would suggest you look at
fm theory and applications
algorithm by algorithm and actually compare it with the GUI of FM8, actually thinking about it.

But that seems too much to ask, because you're so sure you know better.

You're talking about the graphic visualisations of the 32 hardwired operator arrangements printed on the plastic, without actually understanding what they represent. All of the original DX7 modulator/carrier configurations or ("algorithms" as Yamaha opted to call them) can indeed be made with the FM7/8 matrix, and quite a few more, so he really does know better.

EDIT: on topic: Repro-1.

[MadBrain]

I think that there is a big difference between talking about a synth vs a product. Even in the 80s, the DX7 WAS overestimated as a synth. The belief that it could recreate any sound, in theory, simply didn't hold up in practice. It would have been a much better synth with analog filters, although, that would most likely have limited its success as a product.

Not relevant. I simply said that it would have been a better synth with filters and that since it did not have filters, it was overestimated as there was a common belief, as you so clearly articulated, that they weren't necessary. Chowning's goals were academic, I understand that POV, it has jack all to do with whether or not a hybrid architecture is a better tool. It clearly is "as a synthesizer."

The fact that you aren't aware of the FS1R simply speaks to your own ignorance of the development of FM.

I don't think adding filters is the #1 most "effective" thing you could add to the DX7 to improve the sound... The problem is that filters are really something you use for controlling brightness, and FM already does that, so it's kinda redundant on most FM sounds (although you get resonance which is otherwise very hard to do). In fact, the whole point of how the DX7 is designed is to be able to control brightness without having to add another separate chip for each and every voice.

Things that I think make would have made more sense to add to the DX7 than filters:
- 2 x Detuned voice stacking
- Multiple waveforms
- Multiple feedback loops
- Onboard effects

[EnGee]

FM and Additive synthesis (as Wendy Carlos said) are like sewing (or painting, I can't remember now) with a fine needle and thread and you begin to 'add' till you reach what you want. It's the opposite process of Subtractive which you start 'raw' and subtracts (by filters) till you reach the desired sound.

So, in the case of FM and Additive (and PD), no filter needed. But, because people found it not as easy to program their sounds, some developers begin to use FM, PD and Additive oscillators the way the subtractive synths do (filters) like Bazzile Bazille and Razor.

I believe that the sounds can be reached with FM, PD and Additive are not possible with subtractive synthesis (with its saw, pulse, sine, ...etc waves) due to the complexity of sounds those syntheses allow.

[Auplant]

sylenth1.. yes it's possible to get a good sound quickly out of it but it lacks features and the layout sucks

How and why did that synth ever become so popular?