Was the Roland D50 the first VA Modelling Synth?

[mute]

"DCAM is a set of approaches, technologies, knowledge and experience, rather than one hard-and-fast platform. To grossly oversimplify things - we use QUCS (basically the same thing as SPICE) to analyse the circuits, then build mathematical approximations to those that will work in a realtime, discrete context. We of course refer to the actual circuit itself, as well as the schematic and the model, to make sure everything tallies. There is a certain amount of judgement involved as to which parts of the circuit, or which of its behaviours, are most important to the sound, as even with modern CPUs it's not possible to make a perfect model work in real-time at 44K. Yet."

From what I can pick up from that, that doesn't mean they have simulated all the components themselves. It still seems to rely on approximations of the overall output. It's not terribly clear either how the "most important parts" of the circuit should function correctly in the "virtual model" if you're not doing the rest as well. However, if these guys aren't doing that now - how likely is it that Arturia were doing anything of the kind with their old Minimoog / CS-80 emulations?

I think you've misinterpreted him to a degree, its not the output thats being approximated in some circumstances but the math behind some of the circuits that end up being quite resource expensive when added into the whole (such as a resistor that shows up on a board 32 times, etc) it just kills performance all together. When they've chosen to approximate whatever within a circuit its still been based on the non-realtime models and the analyzation of said circuit. That said, obviously theres judgement involved in some of the process as he stated.. and im sure that you're right that in some cases in the final output stage that signal approx. is taken into account... but since they're arent out to create recreations but new synths (so far anyways) it doesnt seem thats too important to them atm.

I read some better interviews after namm 09 with the lads where they explain their methods in greater detail but I'm having a hard time digging one up.. only thing I can really find atm is a meh SoS article and promo youtube video (which isnt actually half bad if you skip to the middle)..

[nvbrkr]

I don't think I've misinterpreted him, because I don't claim to have a proper understanding of how it actually works. It would of course help if they would give clear, more detailed descriptions on their site of what they are doing instead of simply saying that what they are doing is something exceptional. However, I wouldn't be the right person to judge the validity of any potential, more detailed descriptions either. Some others on this site, perhaps. The video doesn't tell too much, it focuses more on how detailed analyses they've done instead of how the software itself works.

It's actually more interesting to me to try to look at the comments and try to, for example, see where manufacturers are using the word "analysed" and where they are indicating that the results of the analyses have been put into execution. From the very little that I read I can't form a clear picture whether they have done approximations of circuits or approximations of "whatever there's within the circuit". There's also no information on how big a percentage of what they've analyzed has been turned into realtime implementations (and what would a non-realtime model be like? Would it sound better?). But since I don't have a proper understanding of this, I am not able to tell what this exactly means in practice and how much it all matters.

However, I still think people draw far too big conclusions from marketing speak. It's not really about whether the manufacturers are being honest or dishonest, but what type of exaggerations some people for some oblivious reason tend to spread around. This "everything can be turned into software models if you just get the calculations right" -type of thinking just doesn't seem to be founded on anything that's actually been done. There are, shall we say, "ontological" problems involved in that as well (I don't intend to sound too pretentious by introducing a philosophical angle, but it's a matter for consideration).

[Automatic Gainsay]

Part of this whole problem if synthesizer nomenclature is that so many things are given names which mean more in regard to marketing and intent than they do in regard to accurate or categorical description.

"Subtractive synthesis" means almost nothing, and yet it's thrown about as if it is a type of synthesis. It just means that the filter diminishes frequencies, and the filter is only a fraction of a synth. On top of that, the term "subtractive" doesn't indicate how this subtraction occurs, whether it is analog, digital, or even acoustic.

"Additive synthesis" is equally silly, as it only describes oscillators stacked upon each other... not what kind of oscillators (whether they are tonewheels, oscillators, or digital waveforms)

The point of the term "virtual analog" was to indicate that a synth (actually software, but I'm not going to make such a petty distinction) which isn't analog but was made to sound and operate like it was. As such, the D-50 couldn't be more DISTANT from VA.

"Analog modeling" was a term to indicate that there was some sort of actual mathematical imitation of analog sound happening on an electronic level. Again, does not remotely apply to the D-50.

All digital synths have sounds programmed to imitate traditional analog sounds, as well as many others.

Pedantic: The term VA Modeling Synth would actually suggest a synth which modeled the way VA synths sound.

[nvbrkr]

You're worse than I am.

[Automatic Gainsay]

And also:

Anyone who was alive during that time as well as interested in/playing/or owning synths will tell you: We did NOT want analog synths. Analog synths were old, they were silly, they were low on functionality, they were completely out of style. Analog synths sounded like the 70s, which all of us of a certain age found to be absolutely unacceptable. Most of the people who loved their Minis in the 80s were people who played Minis in the 70s. My generation, the first generation of people who could actually afford to own synthesizers, didn't want analog... we wanted the newest, most powerful thing. We wanted synths that sounded like what was playing on the radio... which was almost entirely digital. There was no internet. Woodgrain was anathema. It was a different time where the multiplicity of genres and decades of music didn't overlap like they do now.

An analog-emulating synthesizer would have been a grave marketing error in 1987-1989.

[aeon]

My generation, the first generation of people who could actually afford to own synthesizers, didn't want analog... we wanted the newest, most powerful thing. We wanted synths that sounded like what was playing on the radio... which was almost entirely digital. There was no internet. Woodgrain was anathema. It was a different time where the multiplicity of genres and decades of music didn't overlap like they do now.

I'm of said generation, and yea, this is largely true. Samplers and sample-based synths were the s**t in the late 80s and very early 90s.

Thing is, analog was still desireable in my group, at least as it concerns a good mono that would add flavor alongside other digital sources. On the other hand, analog polys were - yesterday's news.


cheers,
Ian

[Automatic Gainsay]

Thing is, analog was still desireable in my group, at least as it concerns a good mono that would add flavor alongside other digital sources. On the other hand, analog polys were - yesterday's news.

The only analog mono I ever saw in the 80s was a second-hand Micromoog that we all laughed at. I specifically remember us wondering why anyone would want anything monophonic. :::shakes head at self:::
While the city I grew up in is a large city, there just weren't many analog synthesizers there, and since they were generally pretty worthless, music stores stopped stocking them (especially used ones).
But I have heard that people in other, more musical cities (where there was more live music, and therefore more likely to be older synths second-hand), people often bought cheap old analog monos for their first synth... of course, often because it was all they could get.
It's hard for a lot of people today to imagine what it was like before the internet. If you didn't live in a major city, there was simply no one to talk to about synthesizers, unless you happened to have friends who were into them, and even then... where did you get information about them? The only way you could was brochures from music stores (only new synths, though), morons who worked at music stores, or sending away for back issues of Keyboard magazine, which you didn't even necessarily have any idea what they contained unless you were looking for something that was mentioned on the cover. As such, people like me in my city only knew of analog synths that were mentioned in the credits of albums, or those occasionally mentioned in VERY few books on synthesizers.
There was a time when I was baffled as to why so many synth presets had the word "MINI" in them.

[cornutt]

Anyone who was alive during that time as well as interested in/playing/or owning synths will tell you: We did NOT want analog synths. Analog synths were old, they were silly, they were low on functionality, they were completely out of style. [...]
An analog-emulating synthesizer would have been a grave marketing error in 1987-1989.

Quite true. It must be remembered that in 1985, Roland was in dire trouble. They still hadn't fully recovered from the bankruptcy of their European distributor (I think that was in 1982), and then the DX7 threw them for a loop. All of a sudden, analog sounds were passe. Everyone wanted the new digital sound. I think the main reasons Roland stuck with a subtractive model for the D-50 were: (1) they needed to get a product that would respond to the DX7 on the market quickly; (2) they didn't have a lot of resources for developing anything radical, and (3) they couldn't do FM because they couldn't find a way around the Chowning patents. Plus, their marketers may have heard some of the complaints from DX7 customers about the difficulty of doing sound design with FM, and figured that they'd stick to a paradigm that everybody already knew.

Sounding like an analog synth is something they did not want the D-50 to do.

[I12]

I was able to get a quite a decent "acid" sound/riff out my d-50/programmer combo

about 10 years back.

Wish i had the track around so i could post it and

at least say "so there"

[mute]

I don't think I've misinterpreted him, because I don't claim to have a proper understanding of how it actually works.

ummn...

anyways, i'm not out to debate any of this. i find it rather pointless honestly, my point from the onset was simply that circuit modeling is indeed being done and fxpansion isn't alone.

[jupiter8]

"Analog modeling" was a term to indicate that there was some sort of actual mathematical imitation of analog sound happening on an electronic level. Again, does not remotely apply to the D-50.

Why not ? If you look at the classic papers "Alias-Free Digital Synthesis of Classic Analog Waveforms" and “Alias-free digital synthesis of classic analog waveforms” the filter is kind of based on modeling the analog equivalent while the oscillator isn't even close. While i'm not sure, that is what they used, in neither the Nord Lead or the D50 it is the most likely to me because variants of that is all you see if you look at source codes available. But since neither you or me have seen the code for the NordLead or the D50,how can you say with such authority they're completely different ?

[Primal Drive]

Sorry for going OT but I'd like to address this.

Part of this whole problem if synthesizer nomenclature is that so many things are given names which mean more in regard to marketing and intent than they do in regard to accurate or categorical description.

"Subtractive synthesis" means almost nothing, and yet it's thrown about as if it is a type of synthesis. It just means that the filter diminishes frequencies, and the filter is only a fraction of a synth. On top of that, the term "subtractive" doesn't indicate how this subtraction occurs, whether it is analog, digital, or even acoustic.

"Additive synthesis" is equally silly, as it only describes oscillators stacked upon each other... not what kind of oscillators (whether they are tonewheels, oscillators, or digital waveforms)

With all due respect, both subtractive and additive are legitimate forms of synthesis technique and are not just meaningless or silly terms.

Subtractive synthesis is just that; i.e., it is the creation of a sound by removing (subtracting) certain harmonics of a raw waveform through various circuits, and not necessarily through active or passive filters. The use of wave shaping circuits within an analog oscillator that, for example, convert a saw wave to a triangle wave is a form of subtractive synthesis. I'm sure you wouldn't consider that meaningless.

Additive synthesis is not just the combining or stacking of oscillators. As originally proposed, it was the combination of a fundamental sine wave combined with other sine waves set to various harmonic intervals in descending amplitude to create a complex waveform. Theoretically, by varying the amplitude of the harmonic waves an infinite number of timbres could be created. This may have not been viable with 70 - 80s technology, but it is far from being a silly concept.

Finally, going OOT, old man Derrik sez hey (yeah, I got all this from him).

Edit: Derrik's cool.

[b3groover]

Wow. Y'all are some serious nerds.

[braincandy]

Wow. Y'all are some serious nerds.

I was going to say that very same thing days ago (and also in the Owl City thread), but I would have inevitably had someone complain that "I can't believe a moderator is saying something like this...blah blah".

[Primal Drive]

Wow. Y'all are some serious nerds.

I just wish I was one too...