Algoritm : new FM synth from RS

This forum is for discussing Rack Extensions. Devs are all welcome to show off their goods.
User avatar
challism
Moderator
Posts: 2401
Joined: 17 Jan 2015

Post 18 Feb 2021

deeplink wrote:
16 Feb 2021
I'm interested to see what the back of it would look like... Hoping it's got a handful of CV support.
Here you go!
back.jpg
You do not have the required permissions to view the files attached to this post.
~~~~~~~~~~~~~~~~~~~~~~
ReasonTalk Rules and Guidelines

User avatar
antic604
Posts: 669
Joined: 02 Apr 2020

Post 18 Feb 2021

challism wrote:
18 Feb 2021
deeplink wrote:
16 Feb 2021
I'm interested to see what the back of it would look like... Hoping it's got a handful of CV support.
Here you go!

back.jpg
Yep, that's very plausible... :clap:
Music tech enthusiast.
DAW, VST & hardware hoarder.
My "music": https://soundcloud.com/antic604

User avatar
marcuswitt
Posts: 178
Joined: 17 Jan 2015

Post 18 Feb 2021

challism wrote:
18 Feb 2021
deeplink wrote:
16 Feb 2021
I'm interested to see what the back of it would look like... Hoping it's got a handful of CV support.
Here you go!

back.jpg
Cool, ALGORITM comes with balanced Outputs! Finally! Thanks for that promising image! 😁

Yonatan
Posts: 1304
Joined: 18 Jan 2015

Post 18 Feb 2021

My only question...

Is this "thing" doing any AI?

Yonatan
Posts: 1304
Joined: 18 Jan 2015

Post 18 Feb 2021

All I want is a video where Ryan explains different synth modulations and what makes FM unique...

Ryan is a good educator, that is for sure.

TritoneAddiction
Posts: 2847
Joined: 29 Aug 2015
Location: Sweden

Post 18 Feb 2021

I'll definitely check it out when it's released.
So far FM4 is my favorite FM synth. It's a pretty perfect synth imo. Nice and simple and sounds awesome. Looks good too.

User avatar
Loque
Moderator
Posts: 8653
Joined: 28 Dec 2015

Post 18 Feb 2021

challism wrote:
18 Feb 2021
deeplink wrote:
16 Feb 2021
I'm interested to see what the back of it would look like... Hoping it's got a handful of CV support.
Here you go!

back.jpg
Great! It's one of those new wireless devices 👍
:reason: 11, Win10 64Bit.

AnotherMathias
Posts: 33
Joined: 29 Sep 2020

Post 18 Feb 2021

turn2on wrote:
17 Feb 2021
This is very interesting to listen about classic true FM design with only sine waveforms into operators.
Look, TX81z and his brother DX7 (rev.2) - based on the 8 waveforms! :) why people try to think that only Sine operators is a classic??) look at any other FM synth, only sine is a real limits.
I don't think that the DX7II had anything but sine waveforms. The patches were the same as the first DX7, but with the addition of some stuff like layering and splits, and a newer DAC (which the DX7S also had).
Other than the 4-operator TX81Z (and the mostly similar keyboard version DX11), I think the first DX to use non-sine waveforms was the SY-77.

I'd be in favor of a sine only implementation for Algoritm. When modulating non-sines with other non-sines, the overtones can quickly become a big mess. And with 7-9 operators, you'd really have to have a lot of focus and restraint to avoid it all becoming fizzy, clanging chaos.

Looking at the preview picture though, it seems that the filter might just be another module that you can use in a similar way as an operator. It would be quite possible that there are other modules as well, such as analog-style or wavetable oscillators, or maybe ones playing user samples.
It all could be a bit too much, but I generally trust Reason Studios to keep their devices somewhat manageable, and not go too overboard with complexity.

deeplink
Posts: 226
Joined: 08 Jul 2020
Location: Dubai

Post 18 Feb 2021

AnotherMathias wrote:
18 Feb 2021
turn2on wrote:
17 Feb 2021
This is very interesting to listen about classic true FM design with only sine waveforms into operators.
Look, TX81z and his brother DX7 (rev.2) - based on the 8 waveforms! :) why people try to think that only Sine operators is a classic??) look at any other FM synth, only sine is a real limits.
I don't think that the DX7II had anything but sine waveforms. The patches were the same as the first DX7, but with the addition of some stuff like layering and splits, and a newer DAC (which the DX7S also had).
Other than the 4-operator TX81Z (and the mostly similar keyboard version DX11), I think the first DX to use non-sine waveforms was the SY-77.

I'd be in favor of a sine only implementation for Algoritm. When modulating non-sines with other non-sines, the overtones can quickly become a big mess. And with 7-9 operators, you'd really have to have a lot of focus and restraint to avoid it all becoming fizzy, clanging chaos.

Looking at the preview picture though, it seems that the filter might just be another module that you can use in a similar way as an operator. It would be quite possible that there are other modules as well, such as analog-style or wavetable oscillators, or maybe ones playing user samples.
It all could be a bit too much, but I generally trust Reason Studios to keep their devices somewhat manageable, and not go too overboard with complexity.
*Another wild Mathias appears

User avatar
turn2on
RE Developer
Posts: 325
Joined: 13 Mar 2015

Post 18 Feb 2021

Yea, my typo about 8 waveforms of TX81z) Its Dx11 brother of course.
Anyway, FM synthesis world of only sine - is limited. And this is words of Yamaha, Roland and many more companies.. marketing? Not at all.
From one side "only sine" OPs is enough..
If you look at the 1984 Yamaha TX816, there 8 modules! 1 module TF1 = 1 DX7. So TX816 = 48 Sine-wave operators.. and this is in 1984! :clap:
2021 - time for something modern. No needs to say "sine-ops" is a classic.
Powerful classic of FM possibilities is a FS1R with spectral waveforms.
Something modern FM - placed into Yamaha Montage / ModX... their FM-X.. If you try to dive into them.. (I know...."workstation today???"), you can find really wild power of the FM synthesis. So its much more interesting, and create acoustic emulated instruments, that Dx7 cant.

Using not only sine- waveforms, is just a way to have more harmonics from single OP.

So, what today Yamaha said about FM waveforms:
The original DX7-type FM was built on simple Sine wave Operators interacting. Each of the DX7's six Operators output just a Sine wave. Since, by nature's design, Sine waves contain no harmonics, this meant it was necessary to use at least two Operators to make a complex wave shape or you had to FEEDBACK the Operator on itself to generate complexity.

In the new FM-X engine, you get a variety of what are called “Spectral Forms” or wave shapes. There are two different Sawtooth / Pulse or may be Resonant waveform types. All the harmonics in the harmonic series are generated. For those familiar with analog synths – this gives us the Sawtooth family of waves.
When Spectral Form = Sine you'll initially have no harmonics while these other six Form types are rich in harmonic content right away.


DX7 is a perfect classical sine-fm machine based on the FM theory. One thing, if modern synth translate DX7 patches (load properties for sine-waveform into operators and translate all settings), and another if this is a really modern Operators that have some logic into waveforms.
As example using wavetables, easy way to have complex solutions for single operator. Only sine waveforms never can sound like other waveforms under FM. Yes, sine forms oriented to the acoustic results. But Yamaha and many other developers find various waveforms for complex results that help to create acoustic sounds that not possible to DX7.

I believe, Algorithm not only "sine-based". 7-9 operators is not something special today. And internal filters too.
Now we don't know anything about his possibilities.

So, we add to FM wavetables. Somebody add user-samples, somebody anything else. And this is also the way to use waveforms, sine-waveforms... So, we already include tons of sine waveform types to device. Just work with sines over and over. Creating overtones, or use pure sine tone without any overtones at all.
Or back to complex odd Pulse harmonics, Sawtooth..

FS1R use some variations of the waveforms.. and was been a king of the FM in sound possibilities.
OpSix also make additions to the FM: 21 waveforms (classic waveforms, bit-reduced sine waves, additive waveforms, multiple noise, and more).

I'm just try to say, good to have sine-machine that can translate to sine-OPs DX7 patches today. In the way of PX7.
But also interesting to have more modern additions.

User avatar
lowtom
Posts: 90
Joined: 29 Sep 2017

Post 19 Feb 2021

Classic or not, I really don't care. Let's wait and see what RS cooked in their lab.
DX7 did not even do proper FM back then. It was not capable of doing so. It was phase modulation synthesis really. If someone want "true" FM, DX7 is not it.
:reason: :refill: :re:

User avatar
RoryM0
Posts: 165
Joined: 21 Jun 2017

Post 19 Feb 2021

If Algoritm is capable of freely routing at least 6 operators in a way that matches all 32 algorithms of PX7 then I'd really like to imagine that it will be able to open PX7 patches. That would be a serious boon.

DecafDreams
Posts: 159
Joined: 07 Oct 2020

Post 19 Feb 2021

RoryM0 wrote:
19 Feb 2021
If Algoritm is capable of freely routing at least 6 operators in a way that matches all 32 algorithms of PX7 then I'd really like to imagine that it will be able to open PX7 patches. That would be a serious boon.
I hadn't even thought about this.

I wonder if they'll offer a tool to convert .SYX into Algoritm like they did for PX7: https://www.reasonstudios.com/dx7-px7-converter/ (still available!)

I'd love it if they also offer a discount to PX7 owners for us to 'upgrade' to Algoritm...

Maybe they won't do any of this and this is their take on FM and nothing to do with imitating the DX7 at all...

User avatar
Loque
Moderator
Posts: 8653
Joined: 28 Dec 2015

Post 19 Feb 2021

DecafDreams wrote:
19 Feb 2021
RoryM0 wrote:
19 Feb 2021
If Algoritm is capable of freely routing at least 6 operators in a way that matches all 32 algorithms of PX7 then I'd really like to imagine that it will be able to open PX7 patches. That would be a serious boon.
I hadn't even thought about this.

I wonder if they'll offer a tool to convert .SYX into Algoritm like they did for PX7: https://www.reasonstudios.com/dx7-px7-converter/ (still available!)

I'd love it if they also offer a discount to PX7 owners for us to 'upgrade' to Algoritm...

Maybe they won't do any of this and this is their take on FM and nothing to do with imitating the DX7 at all...
AFAIU, the envelopes are different. So, open or converting might be possible to a certain degree. Interesting idea for sure.
:reason: 11, Win10 64Bit.

User avatar
RoryM0
Posts: 165
Joined: 21 Jun 2017

Post 19 Feb 2021

Loque wrote:
19 Feb 2021
AFAIU, the envelopes are different. So, open or converting might be possible to a certain degree. Interesting idea for sure.
Good point, I hadn't considered the envelope settings, or literally any of the other settings which might not translate and are obviously really important for the overall sound of a PX7 patch :?

User avatar
EnochLight
Moderator
Posts: 6829
Joined: 17 Jan 2015

Post 19 Feb 2021

Personally I have zero interest in actual DX7 (or even PX7) compatibility. I’m looking forward to something new and unique with that Propellerhead spin.
Win 10 | Reason 11 Suite |  Studio 1 4.6 | i7 3770k @ 3.5 Ghz | 16 GB RAM | Reactor 1TB SSD | RME Babyface Pro | Panorama P-4 | Akai MPC Live 2 & Force | Roland System 8, TR-8 7x7 Expansion, MX1, JX8P, TB3 | Ensoniq SQ1 Plus | Korg Minilogue XD

User avatar
dusan.cani
Posts: 243
Joined: 16 Oct 2018
Location: Slovakia

Post 19 Feb 2021

EnochLight wrote:
19 Feb 2021
Personally I have zero interest in actual DX7 (or even PX7) compatibility. I’m looking forward to something new and unique with that Propellerhead spin.
:thumbs_up:

User avatar
selig
RE Developer
Posts: 9054
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

Post 19 Feb 2021

AnotherMathias wrote:
18 Feb 2021
I'd be in favor of a sine only implementation for Algoritm. When modulating non-sines with other non-sines, the overtones can quickly become a big mess. And with 7-9 operators, you'd really have to have a lot of focus and restraint to avoid it all becoming fizzy, clanging chaos.
Depends on the algorithm - on the DX7 there are only three algos that have all OPs modulating other OPs (16, 17, 18). In most cases there are 2-3 modulations at most between OPs, using multiple "stacks to create essentially "layers" for building sounds. Otherwise there are only 7 of 32 that have more than 2-3 OPs modulating each other (1, 2, 7, 8, 9, 14, 15).
Selig Audio, LLC

User avatar
challism
Moderator
Posts: 2401
Joined: 17 Jan 2015

Post 19 Feb 2021

EnochLight wrote:
19 Feb 2021
Personally I have zero interest in actual DX7 (or even PX7) compatibility. I’m looking forward to something new and unique with that Propellerhead spin.
100%

And I appreciate your using the Prop name there.
DX7 has been done to death at this point. Bring on the future, Propellerhead style!
~~~~~~~~~~~~~~~~~~~~~~
ReasonTalk Rules and Guidelines

User avatar
Jagwah
Posts: 1835
Joined: 16 Jan 2015

Post 20 Feb 2021

I'm rather excited about this.

m.arthur
Posts: 96
Joined: 21 Oct 2017

Post 20 Feb 2021

turn2on wrote:
18 Feb 2021


When Spectral Form = Sine you'll initially have no harmonics while these other six Form types are rich in harmonic content right away.


This is so silly. You're quoting some Yamaha ad copy for their "FM-X" engine and then stating "see what Yamaha thinks about FM today!". Well duh, when trying to sell something, they are going to talk up the importance of whatever they are selling.

You're still not stopping to THINK this through, even though it's right there in front of you -- the fancier waveforms are just a shortcut to complex waveforms, that's it. What do 6 sine-only operators modulating each other do? THEY CREATE COMPLEX WAVEFORMS. The only difference here is "do I arrive at a complex waveform immediately, because it's a choice for a given operator" vs. "do I CREATE a complex waveform through frequency modulation between operators" -- that is, the Whole Point and Purpose of FM synthesis as invented by John Chowning.

The "right away" is the key part of that stupid Yamaha ad copy. You can "begin" with a complex waveform. Ok, great. But you keep insisting this somehow results in "more possibilities" and without it a synth is incredibly limited, etc etc. It's just not true. You're thinking (and arguing) with Subtractive brain, and it doesn't apply here. You can make all the harmonically-rich and complex waveforms you can dream of just using Sine waves. It's literally the point of FM synthesis.

User avatar
turn2on
RE Developer
Posts: 325
Joined: 13 Mar 2015

Post 20 Feb 2021

I'm only try to talk about Op possibilities today. Sine operators can produce harmonicas-rich waveforms, yes. But what the price using only sine waveforms? Quantity of the operators! And more longer way than it can be.
What if some simple waveforms for building complex results - is already into operator? This help to build faster the same results, and save quantity of the used operators for other needs. This not change FM principles at all.

My previous talk about limitations - is only about possible creative / utility using of the Operators with various waveforms in faster way.
With only sine OPs - you just need more OPs. For what? If OP have sine and much more waveforms, its a real bonus.

And at the end, you not said, how much active sine operators you can need to use for plan to emulate results, that can be possible from the 4 operators, that not use only sine or may be static simple or complex waveform, but can use much more (think about active multi waveform changes into one Operator). This is what last hardware FM synth do. Like a OpSix with operator width. This help to make faster and creative results.

I'm going from the other side (not how to work with modulations, but how to work with Operators and what it does). About possible activity controls of the selected internal active multi waveforms. This can be FM-helper part, that not kill basic FM idea.

When Sine is a movable - additionally approve morphing by Chebyshev filter (as example)? A result - you have additional control that help with faster results to have more, than traditional static waveform source into operators. You can say - this is just only complex waveforms. But this only said that you look at this only from one side. I talk about non-static complex waves, that in the way of traditional manipulations of Level/Tune changes, can change source wave. And this not at all. This waveform - have own pre-algorithm of the waveform, all what you can dream be possible to add. S&H, Filters, RM, Harmonic shifts, Quantisation, etc) :thumbup:

6-9 operators to build complex waveforms under sine?
Or 4 operators that include the same sine and various simple or complex waveforms (that can also work with morphing control), this is a more flexible modern way to have fast, rich and experimental results, than just using many OPs with sine.

Orientation to the J.Chowing theory as a only one principle way to understand FM at all - why?
Yes, its FM theory and nobody trying to remove basic sine as main part of FM from OPs.
But why not to use something more after 50 years from the FM theory (for faster and richer results?)

Ok, I can use Supersaw that modulate chebyshev filter effect of Sine at realtime plus traditional Level/Ratio changes to make complex results, and all for something more to modulate speak Vowel phrase. (Yes, why not using phrases waveforms? This possible by active multi-waveforms source in OPs. Just non static use of the OPs).
Results? This is a real FM results. Frequency modulation. I don't answer now how much sine-based Operators you really need to reproduce it. :D You already said that 7-9 OPs and only sine enough to have a rich waveform at the end.
Yes, I try to add for FM OPs something more and non-traditional. But hey, look at all hardware synths, they start to going in this way too time ago. And this is not marketing, the is a new possibilities, that hard touched by Sine OPs.
So yes. Bigger OP quantity is better. But if half of this OPs make the same what already do one multiwavform OP.. there no any problems.

I'm absolutely understand that you stay near to the FM theory. And this is real position of the true FM as it been created 50 years ago. But not try to said that frequency modulation of the other waveforms (exclude sine), is just a complex waveforms at all, Im talk about another. Morphed multi-waveforms with deeper internal OP modulations.
But you can at any moment reseting all this if say "all this is a complex waveforms that possible with X sine operators".

Im not trying to say - subtractive view to the FM is a right. But frequency modulation Is a principal way to produce rich harmonic waveforms. Without this, FM stay forever as FM that was been cerated 50 years ago.

Question only in the way. Did you prefer going 50km to the home around the lake or seat to the boat and go home faster. This is your selection.

You can stay on your point and love only sine-based operators and theory. And this is right. For you if you select this way.
When other musicians can love using any possible basic waveforms (with active morphing from anything to anything simple or complex). How fast you build what you need and how faster another musician build the same thing with operator that include the same Sine and plus many other waveforms (or something like a Chebyshev filter / RingMod / S&H emulation and many more into one operator, that affect to another...). Last musician have much more flexible possibilities. Beacue he can select the way going from Sine or pre-ready solutions.

Real old school "how to build square from the sine".. Why you need to build it using sine waveforms and eat operators when you can already have it as waveform (that can be also not just a waveforms, but like the active multi waveform changeable at the real time + all classic level/ration controls)....

Of course, you can have all sine OPs as a rich pallet of the harmonics to build about you can dream. Rich!!! But absolutely not flexible and easy to emulate what you can do with something like a morphing possibilities of the waveforms into OP.

Musicians is not a Nastradamus to plan very hard end results from the sine waveforms with limited OP quantity.
Why not remove few steps to signal realisation skipping few sine-OPs routings by using one OP with needed waveform right now?

Sine OPs perfect to create something new from sine. But not to emulate what active multi waveform OPs can do without big quantity of OPs.

All this not mean that sine-OP synths better / or complex waveform OPs synths better. This is only question of the flexibility, creativity and the speed of work.
More flexible OPs can be more useful in some scenarios, if waveforms of them not just a static complex waveforms, but much more.

----
Back to the ALGORITHM.
New filter section include at the same time Filter, Filter EG, Keytrack, that can be routed to the OP. Simple compact solution as re-looked interface of EG/Filter/break point. Much more interesting - in Freq Mode (Step/Fade). Waveform breakers, or arpy? Or interesting what is this.

OP7 have at the bottom "OUT". As I understand this can be the main part of the OP output routing. Just select on every OP (or filter) needed output for routing. May be without Algo-table, just select where you need to send OP signal (another OP or main output, or something more, like a filter and other)
Last edited by turn2on on 20 Feb 2021, edited 3 times in total.

User avatar
EnochLight
Moderator
Posts: 6829
Joined: 17 Jan 2015

Post 20 Feb 2021

m.arthur wrote:
20 Feb 2021
You can make all the harmonically-rich and complex waveforms you can dream of just using Sine waves. It's literally the point of FM synthesis.
Me - waiting for Algoritm to release:

Image
Win 10 | Reason 11 Suite |  Studio 1 4.6 | i7 3770k @ 3.5 Ghz | 16 GB RAM | Reactor 1TB SSD | RME Babyface Pro | Panorama P-4 | Akai MPC Live 2 & Force | Roland System 8, TR-8 7x7 Expansion, MX1, JX8P, TB3 | Ensoniq SQ1 Plus | Korg Minilogue XD

User avatar
arnigretar
Posts: 58
Joined: 15 May 2020
Location: Iceland

Post 20 Feb 2021

EnochLight wrote:
20 Feb 2021
m.arthur wrote:
20 Feb 2021
You can make all the harmonically-rich and complex waveforms you can dream of just using Sine waves. It's literally the point of FM synthesis.
Me - waiting for Algoritm to release:

Image
hehe yeah, i agree. always exciting when props have a new synth coming.

ab459
Posts: 271
Joined: 28 Dec 2018
Location: Minsk Belarus

Post 20 Feb 2021

Additionally will be very interesting to test basic things face to face with px7 (who have it, please), with share results.

  • Information
  • Who is online

    Users browsing this forum: CommonCrawl [Bot], Noiserunner and 3 guests