Has there really been much change in subtractive synthesis results in the past 30 years?

[J.F.N.]

15 hours ago, David Emm said:

 

I missed the D50 train a couple of times, but the Cloud version includes the PG-1000 programmer. I still find it a bit squirrely in use, but it does all the right tricks. It almost felt like wrestling with fidgety FM operators until I got a better handle on it. No 2-line display woe, hurray! Instead, its WYSIWYG as a softsynth. Its age acknowledged, the D50 contains pretty much every synth from the family at the time. I've been amazed at the depth. I don't feel the need for added Junos or Jupiters. I might make an exception for the TAL-Pha Juno, as my old Juno-1 loved me long time.

 

The Minimoog was my first synth in a colorful range of early analogs and grainy digitals. The change hasn't been in the method; its been in the improved execution. When it gets so good that you have to ADD instability as a vintage effect, you've pretty well arrived at relative perfection. Even a simple spring reverb or bucket brigade delay doubles your synth's power, so where's the line when you have two effects chains in a Prophet-6? I find it wild that you can have 5 rings of synths, each dedicated to one method or another. Its more about the evolution of Everything.

 

Never had the economy for the programmer back then, but after a while the muscle memory gets the hang of things...

[CHarrell]

47 minutes ago, Tusker said:

Maybe the issue is not tech but culture?

 

These synths are decades old and people complain about their sounds being dated and stale, yet we have instruments that are centuries old and people are still finding fresh and modern sounding ways to use them. 

[ProfD]

Synths, samplers and ROMplers were instrumental (no pun intended) in a lot of music produced from the 1970s through early 2000s.

 

Synth programming and sound design were a huge part of the creative process in music production. A few sounds became iconic. 

 

For the better part of 3 decades and counting, KBs have contained a reasonable facsimile of every musically useful sound needed to play, compose and produce music.

 

Nowadays, synth programming and sound design seem to have been primarily relegated to hobbyists and music composers.

 

Performing musicians can dial up pre-fabricated sounds and get on with the business of playing songs. No need to program anything. Tweak...maybe.😎  

[Iconoclast]

On 6/27/2024 at 7:10 AM, ABECK said:

...have we long since topped out on what can be done with Subtractive and FM synthesis?...

Technology wise, absolutely not. The real question is what can our tiny human brains and hands manipulate in a meaningful way. 

FREQUENTLY the simplest instruments are my most useful.  I think music designers ESPECIALLY IN THE KEYBOARD SPACE are not worried enough about user interface. They are more worried about their feature list than what actual features get used in the product.

High quality VST's are way more flexible and sonically near indistinguishable from analog instruments. But having the knobby interface is inspiring and once you know the instrument, is quick to work with.

I think that recently there's been a proliferation of poly AT and modulation strips, and I think that's a great direction; improving the expressiveness of the instruments interface.

Someone else posted earlier about the current predictability of the signal chain. i.e. Osc-Filter-Amp, with 3 or 4 segment envelope(s).  It would be interesting to see if someone could break that paradigm and still have a giggable, easily programmable platform.

When it comes to FM...it seems to me the ENTIRE problem is the interface. Most FM VST's have a more understandable interface and are much easier to edit than anything built into hardware, and in the case of FM, I think it's soundwise, 100% indistinguishable from the original.

[D. Gauss]

When i told the keyboard player in my roots-rock band that need not bring his synth to gigs, lest he be ridiculed mercilessly by the rest of the band, it was the best subtractive synthesis result ever.

[YashN]

On 6/27/2024 at 9:43 PM, J.F.N. said:

The Roland D50 was imho an interesting evolution of the usage of subtractive synthesis, I have spent a lot of time hanging over that two row display, and quirky page navigation...

 

https://www.muzines.co.uk/articles/l-a-synthesis-what-is-it/1975

Very interesting article. Thanks for sharing. I had a D-10 as my first real synth.

 

L.A. synthesis is part of the broader Sample + synthesis category. In this line of synths, it exploits the perception of how sounds merge into one when played close enough, specifically here, with the Attack being a complex waveform that isn't easily synthesized by classic Subtractive Synthesis but that is easily reproduced by playing a Sample with some limited control over it. The tail end of the overall sound can then be done with the subtractive part, a simpler sound but with more control.

 

The other side of the underlying perceptive principle has some Spatial effects.

 

The D-50 Joystick makes me think this is really also a Vector Control synth, since we do have 4 'Partials' in a sound unit but I'm not sure as I don't see people talk of it as a Vector + L.A./S+S Synth.

 

That fundamental structure of 4 'Partials' in S+S (Partial here really a more complex sub-unit of sound than just a sine in Roland terms), has been kept in many subsequent Roland synths, an architecture that is rather easily reproduced in a Kurzweil V.A.S.T. synth. My Roland D-10 has been in storage for a long time.

 

The D-10 & D-50 also had a small set of fixed Algorithms where you could choose structures that have a Ring Modulator or not between either a PCM Sample and the Synth part, or two Synths, and different routings, etc...

 

[J.F.N.]

37 minutes ago, YashN said:

Very interesting article. Thanks for sharing. I had a D-10 as my first real synth.

 

L.A. synthesis is part of the broader Sample + synthesis category. In this line of synths, it exploits the perception of how sounds merge into one when played close enough, specifically here, with the Attack being a complex waveform that isn't easily synthesized by classic Subtractive Synthesis but that is easily reproduced by playing a Sample with some limited control over it. The tail end of the overall sound can then be done with the subtractive part, a simpler sound but with more control.

 

The other side of the underlying perceptive principle has some Spatial effects.

 

The D-50 Joystick makes me think this is really also a Vector Control synth, since we do have 4 'Partials' in a sound unit but I'm not sure as I don't see people talk of it as a Vector + L.A./S+S Synth.

 

That fundamental structure of 4 'Partials' in S+S (Partial here really a more complex sub-unit of sound than just a sine in Roland terms), has been kept in many subsequent Roland synths, an architecture that is rather easily reproduced in a Kurzweil V.A.S.T. synth. My Roland D-10 has been in storage for a long time.

 

The D-10 & D-50 also had a small set of fixed Algorithms where you could choose structures that have a Ring Modulator or not between either a PCM Sample and the Synth part, or two Synths, and different routings, etc...

 

 

Yes, you could say the D50 makes kind of a vector synth with the joystick indeed.

 

Another thing I always liked was the synth parts, ie. the "oscillators", making the D50 extremely flexible with the combo of PCM attack samples and the oscillators, reminding of the concept in the legendary instruction video with Steve Porcaro (where he's high as a kite) of how to do exactly the same thing, before D50, with an EMU sampler and an Oberheim Expander. Which hindsight made the D50 a really inexpensive synth even when it was first introduced, in comparison.. hehehe!

 

 

 

[YashN]

13 minutes ago, J.F.N. said:

 

1. Another thing I always liked was the synth parts, ie. the "oscillators", making the D50 extremely flexible with the combo of PCM attack samples and the oscillators,

 

 

2. reminding of the concept in the legendary instruction video with Steve Porcaro (where he's high as a kite) of how to do exactly the same thing, before D50, with an EMU sampler and an Oberheim Expander.

 

 

1. It's not just an "Oscillator" though, it's a full-fledged digital version of a traditional analogue subtractive signal path with controls: after the oscillator, there's a 'TVF' (Time-Variant Filter) as the filter and a a TVA (Time-Variant Amplifier) as the Amplifier, hence a "synth" and hence "Sample + Synthesis"

 

2. Yes, a creative way of doing a similar thing, except you'd need 4 'Partials'  & the Joystick to do the same thing as the D-50, and you would also need all the structures with or without Ring Mod and various routings if you want to duplicate the whole structure of the synth. Quite an expensive way too, but if you have all the modules and various synths and samplers, and a nice X-Y controller, probably the most flexible-while-expensive way of all. But at least with just two things, you can already explore the '2 Sounds merging into one' perceptive effect. For the spatial effects, you probably need all 4 'Partials'.

 

Incidentally, I've been taken up doing some unconventional things again with my Kurzweil, the kind of thing that if you search Youtube for it, you won't find a thing about it on Kurzweil. Otherwise, I have a Joystick module in preparation - it will need some time soldering and planning support and placement and cable length. That will open up yet another massive door for Control/Synthesis here. I could also just use an X-Y controller on the screen or something like that, but I want the Physical Control.

[J.F.N.]

3 minutes ago, YashN said:

 

1. It's not just an "Oscillator" though, it's a full-fledged digital version of a traditional analogue subtractive signal path with controls: after the oscillator, there's a 'TVF' (Time-Variant Filter) as the filter and a a TVA (Time-Variant Amplifier) as the Amplifier, hence a "synth" and hence "Sample + Synthesis"

 

2. Yes, a creative way of doing a similar thing, except you'd need 4 'Partials'  & the Joystick to do the same thing as the D-50, and you would also need all the structures with or without Ring Mod and various routings if you want to duplicate the whole structure of the synth. Quite an expensive way too, but if you have all the modules and various synths and samplers, and a nice X-Y controller, probably the most flexible-while-expensive way of all. But at least with just two things, you can already explore the '2 Sounds merging into one' perceptive effect. For the spatial effects, you probably need all 4 'Partials'.

 

Incidentally, I've been taken up doing some unconventional things again with my Kurzweil, the kind of thing that if you search Youtube for it, you won't find a thing about it on Kurzweil. Otherwise, I have a Joystick module in preparation - it will need some time soldering and planning support and placement and cable length. That will open up yet another massive door for Control/Synthesis here. I could also just use an X-Y controller on the screen or something like that, but I want the Physical Control.

 

Yes, I am very familiar with this machine, what I mean is "oscillator" as in the sound generator, as opposed to the sample, of the partial when in Synthesis mode.

 

Good luck with your controller project, ambitious!!

 

 

[YashN]

13 minutes ago, J.F.N. said:

 

Yes, I am very familiar with this machine, what I mean is "oscillator" as in the sound generator, as opposed to the sample, of the partial when in Synthesis mode.

 

Good luck with your controller project, ambitious!!

 

 

 

Yes, that's correct, or simply a traditional subtractive synth but implemented in digital. With some of the ALGs, you can just use it as a mix of two subtractive synths too or mix of two PCM (for a 'Tone' of which there are two: Upper & Lower), and Samples can also be looped, the Ring Mod providing additional harmonics. I had a lot of fun with the D-10, although I was bummed that I couldn't get the D-20 because it had the sequencer. Well, I hacked the D-10 Rhythm Pattern Sequencer by changing the envelope mode of non-percussive sounds that I laid inside the Drum patch and composed around 2-3 albums worth of songs with just that. That internal Pattern-chaining sequencer is my preferred way of creating songs since then.

 

Thanks, but the controller project is rather simple: thankfully I am recuperating the Joystick from a discarded Yamaha SY-22 so do not have to start from scratch. It's still a very DIY affair for the rest, like cabling, casing, placement, maybe some metal work that I am trying to avoid. I am thinking of wood too to contrast with the stalk metal of the K2500XS. I'll probably post some pics in the Repair/Mods section and some sounds when I'm done - it's going to be one heck of a powerful Vector Control Synth for sure: 4 choices among all the engines possible with V.A.S.T., all of which I built myself. That's around 20^4 combinations... for the basic implementation. There's way more after that too after I tackle everything again in some under-utilised configurations. This is going to be fun.

[J.F.N.]

24 minutes ago, YashN said:

 

Yes, that's correct, or simply a traditional subtractive synth but implemented in digital. With some of the ALGs, you can just use it as a mix of two subtractive synths too or mix of two PCM (for a 'Tone' of which there are two: Upper & Lower), and Samples can also be looped, the Ring Mod providing additional harmonics. I had a lot of fun with the D-10, although I was bummed that I couldn't get the D-20 because it had the sequencer. Well, I hacked the D-10 Rhythm Pattern Sequencer by changing the envelope mode of non-percussive sounds that I laid inside the Drum patch and composed around 2-3 albums worth of songs with just that. That internal Pattern-chaining sequencer is my preferred way of creating songs since then.

 

Thanks, but the controller project is rather simple: thankfully I am recuperating the Joystick from a discarded Yamaha SY-22 so do not have to start from scratch. It's still a very DIY affair for the rest, like cabling, casing, placement, maybe some metal work that I am trying to avoid. I am thinking of wood too to contrast with the stalk metal of the K2500XS. I'll probably post some pics in the Repair/Mods section and some sounds when I'm done - it's going to be one heck of a powerful Vector Control Synth for sure: 4 choices among all the engines possible with V.A.S.T., all of which I built myself. That's around 20^4 combinations... for the basic implementation. There's way more after that too after I tackle everything again in some under-utilised configurations. This is going to be fun.

 

Here's interesting "inside information" about the sound generated when a partial is in "S" (Synthesis) mode (from a post on the GS forum):

 

"It is Roland’s “dirty secret” that the square/pulse is generated in exactly the same way as Casiios CZ Phase Distortion Synths. You start with a cosine, as you increase the phase distortion, the shape tends towards a square.

 

The Saw is generated by multiplying (aka ring mod) that square with another cosine, which flips the second half of the cosine the other way up (pos/neg), so what you get is the first half of a cosine repeated, which is approximately a Saw at double the frequency of the cosine.

 

The resonance is done with windowed synced sines, again just like the Casio CZs.

 

Phase Distortion and Windowed Synced Sines were patented by Casio, so Roland were being bloody cheeky using them. Also, Phase Distortion is a Non-Linear function, meaning it produces frequencies that did not exist in the source waveform.

 

But Roland claims whatever the D50 was doing was a Linear function. Hence Linear Arithmetic Synthesis.

 

It’s nonsense, it’s all a smokescreen so that they were not sued by Casio and Yamaha (Yamaha had won a case that stated any synthesis that used changing the phase readout of a Sine or Cosine was covered by the FM patent they had with Stanford University)."

[YashN]

I remember that post. Quite interesting.

 

The Linear Arithmetic Synthesis part isn't about the methods of producing the oscillator waveform, I think it's about the simple mixing of the various branches of the Sample + Synthesis architecture, and the mixing of the 'Partials'.

 

Probably incorporates the Ring Mod that is post-Oscillator as well.

[J.F.N.]

15 minutes ago, YashN said:

I remember that post. Quite interesting.

 

The Linear Arithmetic Synthesis part isn't about the methods of producing the oscillator waveform, I think it's about the simple mixing of the various branches of the Sample + Synthesis architecture, and the mixing of the 'Partials'.

 

Probably incorporates the Ring Mod that is post-Oscillator as well.

 

"Linear Arithmetic synthesis, or LA synthesis, is a term that was invented by the Roland Corporation when they released their ground-breaking D-50 synthesizer in 1987.

 

It was the first keyboard to combine a synthesizer with sampled waveforms that could be modified.

 

Linear arithmetic makes one think of simple math, such as addition, and indeed, the primary way that LA synthesis created sounds was by adding sampled waveforms to digitally-generated waveforms."

[AROIOS]

3 hours ago, J.F.N. said:

... the legendary instruction video with Steve Porcaro (where he's high as a kite) ...

The brass pitch envelope trick Steve demoed at 33:40 applies equally well to Oriental plucking sounds like Chinese Zheng and Japanese Koto, on which the Attack often starts sharp but quickly return to normal.

High or not, that nerdy passion/energy Steve demonstrated is infectious. For a second, I had the urge to look up an MC-500 on Reverb.com 😃

[J.F.N.]

1 hour ago, AROIOS said:

The brass pitch envelope trick Steve demoed at 33:40 applies equally well to Oriental plucking sounds like Chinese Zheng and Japanese Koto, on which the Attack often starts sharp but quickly return to normal.

High or not, that nerdy passion/energy Steve demonstrated is infectious. For a second, I had the urge to look up an MC-500 on Reverb.com 😃

 

Totally!!

 

When I found this clip, the full version on YouTube (my keyboard buddy back in the days had the VHS and we've watched it bunch of times then), I got seriously inspired by the way he's using different types of sequencers for different purposes and then dumps the sequences in-between them, and I ended up hooking up all gadgets I have to fiddle around with them in the same fashion.. (check my signature...) I didn't remember that part, and for sure didn't have the abundance of gadgets back then either...

 

 

Beware of the MC series sequencers, they suffer from a timing bug when synced up via midi in loop mode, making them useless for anything else than strictly linear stuff if they're to be in sync with something else...

 

I had an MC80, and it seems all MC series are infected 300, 500, 50, 80, and the MkII version of them too, obviously the same code base, so I ended up selling mine after finding out the hard way.

 

Did a ridiculously great deal in total though, when I bought mine, used, it had the expansion card with sounds in it, terrible sounding so I ripped it out and sold it on Reverb, dude in Florida (I'm in Europe) bought it for close to 3 times what I paid for the sequencer, and then when I sold the sequencer, Reverb too, it was almost unreal what I got for it, Epic!

 

 

[Sundown]

I’ll say a few things …

 

You can still get some great analog-ish pads and comp sounds from a Korg M1 (I own the software version). The acoustic emulations sound dated and cliché, but there is something great about the late 80’s and early 90’s ROMplers when it comes to sitting in a mix. The “width” of the sounds makes them easy to arrange and mix and there’s something about the 12-bit/16-bit content that sounds right in an arrangement.

 

But there are also times that older instruments fall short, and if I try to use my original Roland D-20 in a composition today, excluding a few sounds, it’s going to sound thin or tweezy. And I’m going to spend a lot of time trying to manage resonances with EQ to get it to sit properly. So a lot of it depends on the specific instrument. I wouldn’t mind a vintage D-50 or D-550, but the sluggish response to MIDI data (which sounds like transient smearing) makes me pause.

 

I actually don’t own any hardware instruments newer than my Kurz PC361, unless you count my digital piano. I own a lot of software instruments and software samplers, but being that I’m a studio rat, I haven’t bought any new boards.

 

Part of it is the cost barrier, part of it is the sheer overload of sounds I already have (largely from software), and many of the modern boards are too fat for compositions. They’re tremendously fun to play in the store and they sound great if you’re just playing some riffs, but they’re far too bloated for a dense mix. I would have to strip the sounds down to make them more useable. Patches that are most valuable for real work don’t sound fantastic on the shop floor, and that’s no doubt a challenge with the manufacturers. They have to present the board in a way that gets attention, and these huge stacked bass sounds and swirling pads win-out over more practical sounds.

 

So I think it’s a balance in the end. You can get a great, aggressive lead sound with a TX81Z with some EQ and delay, and you can get a phenomenal EP from the latest offerings. And hopefully when you blend them both you get a great arrangement.

 

Where do I think it’s headed next? I think AI will have a huge impact on articulations and note expression. The “machine gun” effect of playing repeated notes on an older instrument will be non-existent on boards in a few years. The realism that will come from small variations in tone will really take some ROMplers into new territory. As long as I feel like I’m still the player, I’m cool with it… Great sounding boards are great sounding boards.

 

Todd

[AROIOS]

15 hours ago, J.F.N. said:

...I got seriously inspired by the way he's using different types of sequencers for different purposes and then dumps the sequences in-between them...

 

...Beware of the MC series sequencers, they suffer from a timing bug when synced up via midi in loop mode...

 

...Did a ridiculously great deal in total though...

 

 

That tape must have inspired a lot of musicians, especially in the Pop Jazz world, where tons of albums sounded like homework turned in for Steve's class. Russ Freeman's Nocturnal Playground immediately comes to mind. It featured the same Linn9000 drums, TX816/DX7 EPs, Minimoog bass, and Oberheim brass as Steve demoed. To be fair, that was THE setup for Pop Jazz until D-50 and M1 came along, so tons of albums inevitably had that mid-80's sound.

Thanks for the heads up on the MC machines. I'm too spoiled by software sequencers at this point but there's just something oddly satisfying hitting big physical buttons and turning real knobs. 😃

And congrats on more than recouping your cost on those purchases!

[CEB]

Is this an FM thread on why FM is better today than thirty years ago ... because it is.

 

PS:  Well ... I guess I'm overstating some.  the TX816 and DX1 etc.  showed what we could do.

[AROIOS]

16 hours ago, Sundown said:

...You can still get some great analog-ish pads and comp sounds from a Korg M1 (I own the software version)...there is something great about the late 80’s and early 90’s ROMplers when it comes to sitting in a mix. The “width” of the sounds makes them easy to arrange and mix...

Agree 100%. IIRC, it was another post from you that led me to the discovery of this lovely M1 synth brass patch I modified and affectionately named "Katana Runner". 😃

 

 

[Sundown]

42 minutes ago, AROIOS said:

Agree 100%. IIRC, it was another post from you that led me to the discovery of this lovely M1 synth brass patch I modified and affectionately named "Katana Runner". 😃

 

 

 

 

Great sound!

 

I think the M1 in particular stands out when it comes to soft sawtooth, brassy pads and comp sounds. I bought the Triton/Triton Extreme pair last Christmas and there are some good patches in there too, but for 1988 the M1 still does synthetic sounds really well. I almost always kill the reverb and use a modern plugin (e.g. Lex 480L or RMX16) to cover the ambience.

 

Todd

[ProfD]

1 hour ago, AROIOS said:

...this lovely M1 synth brass patch I modified and affectionately named "Katana Runner". 😃

Another excellent example showing that it's all about a how a KB sound is utilized.😎

 

[Tusker]

On 6/29/2024 at 12:33 PM, CHarrell said:

 

These synths are decades old and people complain about their sounds being dated and stale, yet we have instruments that are centuries old and people are still finding fresh and modern sounding ways to use them. 

 

Yes. Acoustic instruments and the sampled (rompler) instruments are different.

 

Acoustic instruments do have a range of expression with just their one "sound" which is broader than a multi-sample. This is why pianists will typically perform on an acoustic rather than a digital piano if you give them a choice. This is why, in an age of beautifully sampled instruments, there is a thriving market for physical modeling instruments. And analog synths. And FM. Sounds that are less static.

 

It's evidence that the sample is not the whole. The snapshot is not the movie. Sampled instruments are a fantastic background to support a singer or a lead instrument. They often contain a huge color palette spanning human culture. That is their superpower. Their timbral range is massive. Yet there are expressive limitations. When was the last time any of us experienced a concert headlined by a multi-sampled synthesizer?

 

There is a world of difference between acoustic and sampled instruments. Each has it's strengths.

[AROIOS]

23 hours ago, Sundown said:

...I almost always kill the reverb and use a modern plugin (e.g. Lex 480L or RMX16) to cover the ambience...

Same here. As was the case on many early Romplers, reverb was not M1's forte.

[ProfD]

18 minutes ago, AROIOS said:

Same here. As was the case on many early Romplers, reverb was not M1's forte.

Most onboard reverb is to sounds the equivalent of ketchup/catsup as a condiment.😁

 

Killing the effects is usually the 1st thing I do when walking up to a KB and auditioning sounds. 

 

Manufacturers tend to slather too much ketchup, er, reverb and mustard and relish on everyhting as if it's Frank's Hot Sauce.

 

Maybe I'm thinking too far in advance towards the July 4th holiday.😎

[AROIOS]

35 minutes ago, ProfD said:

Most onboard reverb is to sounds the equivalent of ketchup/catsup as a condiment.😁

 

Killing the effects is usually the 1st thing I do when walking up to a KB and auditioning sounds. 

 

Manufacturers tend to slather too much ketchup, er, reverb and mustard and relish on everyhting as if it's Frank's Hot Sauce.

 

Maybe I'm thinking too far in advance towards the July 4th holiday.😎

Yup, we've come a long way on good sounding onboard verbs.

It's interesting how chorus didn't seem to suffer as much from old tech, especially the Roland ones. It's often such an integral part of the EP patches, replacing it with outboard choruses often alters the sound in a detrimental way. In fact, I often slap a wonderful and free JUNO-60 style chorus on dry synth and EP sounds to fatten them up.

There are also serendipitous finds in older synth EFXs. Some of the "Multi" type effects on Roland's XV and higher SC models worked wonders on guitar and keyboard sounds. It's a shame Roland ditched them on later models.

[YashN]

On 6/29/2024 at 8:59 PM, J.F.N. said:

 

"Linear Arithmetic synthesis, or LA synthesis, is a term that was invented by the Roland Corporation when they released their ground-breaking D-50 synthesizer in 1987.

 

It was the first keyboard to combine a synthesizer with sampled waveforms that could be modified.

 

Linear arithmetic makes one think of simple math, such as addition, and indeed, the primary way that LA synthesis created sounds was by adding sampled waveforms to digitally-generated waveforms."

 

From the horse's mouth:

 

"In a general sense subtractive synthesis and digital sampling are both forms of linear synthesis. LA Synthesis is a hybrid of some of the concepts found in these technologies.
...
These calculations are primarily addition and subtraction, hence the name Linear Arithmetic. On the D-50, sounds are created by combining partials and tones (addition), removing unwanted harmonics (subtraction) and ring modulation (sum and difference)"

- Roland Corporation, 1988
 

Now, turns out the TVF filter isn't a common DSP filter at all but actually involves mixing in another wave and further ring modulation.

 

Hence, more Linear Arithmetic.

 

Since it's not a conventional filter, it doesn't have a conventional filter slope. To my ears, it's close to a something like a conventional 12db/oct and 18db/oct.

 

Clever stuff by Roland!

 

BTW, the little brother, the D-10, has a few more structures than the D-50.

 

Additionally, the joystick control for balance of partials is activated via buttons for pairs of partials (Upper Tone or Lower Tone) and not simultaneously for all 4. That would explain why it isn't a 'Vector' synth.

 

I am not sure I understand the OP's post about Subtractive Synthesis because he somehow removes V.A. which is a big application of subtractive in the digital domain.

 

The title is less ambiguous. The biggest changes along the way have been more options for the Osc Sections, more filter types and slopes, serial and parallel arrangements with or without additional processing in the filter sections, filter banks, Hybrids (Digital Oscs into Analogue VCFs and VCAs), DSP implementations, etc...

[David Emm]

On 7/1/2024 at 2:21 PM, ProfD said:

Most onboard reverb is to sounds the equivalent of ketchup/catsup as a condiment.😁

Killing the effects is usually the 1st thing I do when walking up to a KB and auditioning sounds.

 

I take your point, which is perfectly sensible. The un-effected patches show you what kind of clay you really have.                             

 

I also take the devil's advocate spot, because some host effects are as much a component as the envelopes. Deleting those and running it through the Eventide of the gods may yield less tasty results. Counterintuitive at first glance, but often true. I seem to lower the host effects and then piggyback things through a modern effects chain as often as not. To my ear, that's a best-of-both-worlds benefit. I watch a real-time EQ display so I can nudge the spectrum and panning into the best positions. It embiggens everything. 🤓

[ProfD]

7 hours ago, David Emm said:

The un-effected patches show you what kind of clay you really have. 

Right.

7 hours ago, David Emm said:

I also take the devil's advocate spot, because some host effects are as much a component as the envelopes.

Absolutely. It helps to know if and how the effects shape the sound. 

7 hours ago, David Emm said:

I seem to lower the host effects and then piggyback things through a modern effects chain as often as not. To my ear, that's a best-of-both-worlds benefit.

That's a huge part of real sound design.😎