Proton Granular Synth!

This forum is for discussing Rack Extensions. Devs are all welcome to show off their goods.
User avatar
etyrnal
Posts: 316
Joined: 24 Jan 2016
Contact:

02 Feb 2017

I was mostly thinking of this as a discussion, and a thought experiment, and clarification. It seems like we are on the same wavelength give or take.

Sent from my XT1575 using Tapatalk

WongoTheSane
Moderator
Posts: 1851
Joined: 14 Sep 2015
Location: Paris, France

02 Feb 2017

selig wrote:
WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
etyrnal wrote:Very often though, samples are already normalized. In other words they've been adjusted so that the peak sample in the wave file is at 0 dB. So leveling is not exactly what we're talking about here. We're talking about adding gain to the entire sample while at the same time limiting.
I think you skipped one step. I was answering this:
buddard wrote:I like the idea of being able to squash the signal by boosting the quiet parts of the sample, like an extreme one-knob version of Selig Leveler! It turns the sample source into more of a "texture" to work with, basically ignoring the original dynamics. :puf_smile:
...I'm talking leveling and you're describing automatic gain control. Those are two different processes. What you describe is certainly easy to implement, but it's not leveling and probably won't give the results you expect. My argument was that implementing a Leveler (with just one knob) in Proton would be probably more difficult than a gain control (it would either require normalizing the sample first or scanning it for the peak, which isn't necessary for compression or gain control), and that using a proper Leveler provides more control at the cost of a few seconds, which I don't consider as "the long long way around".
A single knob (two, max) would be easy to do a quick "leveling" effect, two knobs would cover everything. If you assume peaks of 0 dBFS, then you don't need a Target knob. If you assume you want to level the entire dynamic range of most samples, you don't need the Curve knob. That leaves the Recovery (which could also be preset), and primarily the Blend knob. Easy to do something very workable IMO.


Sent from my iPad using Tapatalk
Weird, I would have said: remove Recovery (set to fast by preset), remove Blend (set to wet), keep Curve (it's how I use it usually, except Recovery: default position for normal sound processing, and fast for effect samples, wavetables, etc). Obviously only in the context of Proton, I'm not suggesting that for the device itself :D
I typically set Curve after Target, choosing how 'far' into the dynamic range I want to lift. Blend is the #1 control for me, the one I use to set the final amount of the effect. In other words, the depth of the effect. I'm constantly fine tuning Blend and to a lesser degree, Recovery. I typically set and forget Target/Curve, since they are set according to the dynamics of the source (and wouldn't change unless the source changed).

Blend controls how much you increase the low level signals, from no change to 100% leveling - isn't that what we're talking about here?


Sent from my iPad using Tapatalk
Oh, but you mean when working on normal vocals, right? I was only talking about the cases where you need to remove dynamics as much as possible (and not really care whether it's realistic or not as the sound is going to be mangled anyway). In this context, wouldn't setting Wet to 100% and using Curve to determine how much of the lower sounds will be lifted (acting more or less like a threshold: everything above is maxed out, everything below is silent) make sense?

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

02 Feb 2017

WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
etyrnal wrote:Very often though, samples are already normalized. In other words they've been adjusted so that the peak sample in the wave file is at 0 dB. So leveling is not exactly what we're talking about here. We're talking about adding gain to the entire sample while at the same time limiting.
I think you skipped one step. I was answering this:
buddard wrote:I like the idea of being able to squash the signal by boosting the quiet parts of the sample, like an extreme one-knob version of Selig Leveler! It turns the sample source into more of a "texture" to work with, basically ignoring the original dynamics. :puf_smile:
...I'm talking leveling and you're describing automatic gain control. Those are two different processes. What you describe is certainly easy to implement, but it's not leveling and probably won't give the results you expect. My argument was that implementing a Leveler (with just one knob) in Proton would be probably more difficult than a gain control (it would either require normalizing the sample first or scanning it for the peak, which isn't necessary for compression or gain control), and that using a proper Leveler provides more control at the cost of a few seconds, which I don't consider as "the long long way around".
A single knob (two, max) would be easy to do a quick "leveling" effect, two knobs would cover everything. If you assume peaks of 0 dBFS, then you don't need a Target knob. If you assume you want to level the entire dynamic range of most samples, you don't need the Curve knob. That leaves the Recovery (which could also be preset), and primarily the Blend knob. Easy to do something very workable IMO.


Sent from my iPad using Tapatalk
Weird, I would have said: remove Recovery (set to fast by preset), remove Blend (set to wet), keep Curve (it's how I use it usually, except Recovery: default position for normal sound processing, and fast for effect samples, wavetables, etc). Obviously only in the context of Proton, I'm not suggesting that for the device itself :D
I typically set Curve after Target, choosing how 'far' into the dynamic range I want to lift. Blend is the #1 control for me, the one I use to set the final amount of the effect. In other words, the depth of the effect. I'm constantly fine tuning Blend and to a lesser degree, Recovery. I typically set and forget Target/Curve, since they are set according to the dynamics of the source (and wouldn't change unless the source changed).

Blend controls how much you increase the low level signals, from no change to 100% leveling - isn't that what we're talking about here?


Sent from my iPad using Tapatalk
Oh, but you mean when working on normal vocals, right? I was only talking about the cases where you need to remove dynamics as much as possible (and not really care whether it's realistic or not as the sound is going to be mangled anyway). In this context, wouldn't setting Wet to 100% and using Curve to determine how much of the lower sounds will be lifted (acting more or less like a threshold: everything above is maxed out, everything below is silent) make sense?
I'm just thinking of how to adapt the concept of the Leveler to become a dynamics eliminator. In that case I'd want a depth control to determine how much to bring up the low level signals. The Curve control only determines the RANGE that you lift - it's the Blend that determines how FAR you lift that range. Small Curve settings would not make sense in this application even though they DO make sense with vocals. This is because with vocals you do NOT want to lift the LOWEST signals; the breaths, the background signals, etc.

But with a sample, you basically want to lift the entire signal - the only question is how much lift? Think of a piano decaying to silence - you want to make the piano basically not decay at all, right? That requires a large Curve setting so that the entire decay can be lifted if desired. From there it's up to the Blend control to determine how MUCH you lift the signals. Hopefully I'm making sense here…


Sent from my iPad using Tapatalk
Selig Audio, LLC

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

02 Feb 2017

etyrnal wrote:
selig wrote:
etyrnal wrote:
selig wrote:
etyrnal wrote:
selig wrote:
etyrnal wrote:
selig wrote:
etyrnal wrote:
selig wrote:[quote="WongoTheSane"][quote="etyrnal"]Very often though, samples are already normalized. In other words they've been adjusted so that the peak sample in the wave file is at 0 dB. So leveling is not exactly what we're talking about here. We're talking about adding gain to the entire sample while at the same time limiting.
I think you skipped one step. I was answering this:
buddard wrote:I like the idea of being able to squash the signal by boosting the quiet parts of the sample, like an extreme one-knob version of Selig Leveler! It turns the sample source into more of a "texture" to work with, basically ignoring the original dynamics. :puf_smile:
...I'm talking leveling and you're describing automatic gain control. Those are two different processes. What you describe is certainly easy to implement, but it's not leveling and probably won't give the results you expect. My argument was that implementing a Leveler (with just one knob) in Proton would be probably more difficult than a gain control (it would either require normalizing the sample first or scanning it for the peak, which isn't necessary for compression or gain control), and that using a proper Leveler provides more control at the cost of a few seconds, which I don't consider as "the long long way around".
A single knob (two, max) would be easy to do a quick "leveling" effect, two knobs would cover everything. If you assume peaks of 0 dBFS, then you don't need a Target knob. If you assume you want to level the entire dynamic range of most samples, you don't need the Curve knob. That leaves the Recovery (which could also be preset), and primarily the Blend knob. Easy to do something very workable IMO.


Sent from my iPad using Tapatalk
Recovery would be everything above 50% or like 75%. It can all be intelligently done with a single knob. If you want to do you could even have the last couple of percent be going into slight saturation.

Sent from my XT1575 using Tapatalk
Recovery is a "time" constant, so percent make no sense to me in that regard. Saturation doesn't make sense either, at least not while leveling.

Again, the idea was to remove the dynamics from the original audio if desired, not to compress/limit/saturate it.


Sent from my iPad using Tapatalk
Musical qualities are ratios... Treating things like percents and ratios gives musical results...

Sent from my XT1575 using Tapatalk
Not following…If so, then what ratio or percent is 200 ms Recovery Time?


Sent from my iPad using Tapatalk
Why would recovery time be needed? The calc is based on an individual sample point. For every individual sample point you are basically calculating the difference between the samples current level and 0db. This calculation I'm talking about his pure math. Visually picture a sample waveform in your mind. Now take that image as a JPEG and move it into an image editor in your mind. Now tell the image editor that you want to resize the Image Grab the bottom of the image and move it up towards the top. Eventually if you keep raising it you will get a flatline. All other values in the middle will be calculated as ratio of distance from the top. Of course eventually everything will be a flatline if you keep moving it upwards. Obviously some of the calculations would have to be done using natural functions, but again it's still just a simple mathematical calculation. In the example of resizing a JPEG image of a waveform, which would basically be the same thing for making calculations on an audio waveform except now you have to make a calculations in the negative and positive. So for example if you wanted to raise the lowest level in the sample by 50% of the distance between that samples current level and 0 decibels, then you calculate how much gain is required to increase that samples level by 50% of the distance between itself and 0db. For samples that are louder you do the exact same thing you calculate the difference between that current sample level and zero DB in decibels, and then you raise bye the percentage of that many decibels. Once again if using a natural function makes more sense then of course we would have to do that. But the concept Still Remains. It is literally a simple mathematical calculation. It's a ratio. Call it a fraction if you want.

Sent from my XT1575 using Tapatalk
That's not how any dynamics devices work that I'm aware of, FWIW.


Sent from my iPad using Tapatalk
right, but in Dynamic processors sometimes the producer or the musician wants to achieve certain musical effect based on playing back the sample as is. But we're talking about granular synthesis where we are potentially going to be throwing away those beautiful Dynamics, because we are trying to create something entirely different than what we started with. And if a person doesn't like how much Dynamics it's losing then don't twist the knob up for that particular patch.

Sent from my XT1575 using Tapatalk
That's exactly what I'm saying, not sure where the confusion is coming from here.


Sent from my iPad using Tapatalk[/quote]
And for example let's say that a person has a drum Loop loaded into a granular synthesizer, and they have the length and emitter frequency set so that you're actually hearing full drum hits, then just run the output of the granular synth through a leveler or a compressor limiter or whatever it is that you're trying to do. But anytime you're down dealing with grains it would make no sense to me to go to all the trouble of putting some high class high-precision compression between the sample and the emitter.

Sent from my XT1575 using Tapatalk[/quote]
OK then, this feature is not something you would use. But I bet if it sounds really cool to do this, you would change your mind on this… ;)


Sent from my iPad using Tapatalk
Selig Audio, LLC

WongoTheSane
Moderator
Posts: 1851
Joined: 14 Sep 2015
Location: Paris, France

02 Feb 2017

selig wrote:
WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
etyrnal wrote:Very often though, samples are already normalized. In other words they've been adjusted so that the peak sample in the wave file is at 0 dB. So leveling is not exactly what we're talking about here. We're talking about adding gain to the entire sample while at the same time limiting.
I think you skipped one step. I was answering this:
buddard wrote:I like the idea of being able to squash the signal by boosting the quiet parts of the sample, like an extreme one-knob version of Selig Leveler! It turns the sample source into more of a "texture" to work with, basically ignoring the original dynamics. :puf_smile:
...I'm talking leveling and you're describing automatic gain control. Those are two different processes. What you describe is certainly easy to implement, but it's not leveling and probably won't give the results you expect. My argument was that implementing a Leveler (with just one knob) in Proton would be probably more difficult than a gain control (it would either require normalizing the sample first or scanning it for the peak, which isn't necessary for compression or gain control), and that using a proper Leveler provides more control at the cost of a few seconds, which I don't consider as "the long long way around".
A single knob (two, max) would be easy to do a quick "leveling" effect, two knobs would cover everything. If you assume peaks of 0 dBFS, then you don't need a Target knob. If you assume you want to level the entire dynamic range of most samples, you don't need the Curve knob. That leaves the Recovery (which could also be preset), and primarily the Blend knob. Easy to do something very workable IMO.


Sent from my iPad using Tapatalk
Weird, I would have said: remove Recovery (set to fast by preset), remove Blend (set to wet), keep Curve (it's how I use it usually, except Recovery: default position for normal sound processing, and fast for effect samples, wavetables, etc). Obviously only in the context of Proton, I'm not suggesting that for the device itself :D
I typically set Curve after Target, choosing how 'far' into the dynamic range I want to lift. Blend is the #1 control for me, the one I use to set the final amount of the effect. In other words, the depth of the effect. I'm constantly fine tuning Blend and to a lesser degree, Recovery. I typically set and forget Target/Curve, since they are set according to the dynamics of the source (and wouldn't change unless the source changed).

Blend controls how much you increase the low level signals, from no change to 100% leveling - isn't that what we're talking about here?


Sent from my iPad using Tapatalk
Oh, but you mean when working on normal vocals, right? I was only talking about the cases where you need to remove dynamics as much as possible (and not really care whether it's realistic or not as the sound is going to be mangled anyway). In this context, wouldn't setting Wet to 100% and using Curve to determine how much of the lower sounds will be lifted (acting more or less like a threshold: everything above is maxed out, everything below is silent) make sense?
I'm just thinking of how to adapt the concept of the Leveler to become a dynamics eliminator. In that case I'd want a depth control to determine how much to bring up the low level signals. The Curve control only determines the RANGE that you lift - it's the Blend that determines how FAR you lift that range. Small Curve settings would not make sense in this application even though they DO make sense with vocals. This is because with vocals you do NOT want to lift the LOWEST signals; the breaths, the background signals, etc.

But with a sample, you basically want to lift the entire signal - the only question is how much lift? Think of a piano decaying to silence - you want to make the piano basically not decay at all, right? That requires a large Curve setting so that the entire decay can be lifted if desired. From there it's up to the Blend control to determine how MUCH you lift the signals. Hopefully I'm making sense here…


Sent from my iPad using Tapatalk
Right, I understand. My method does indeed leave silent parts when the signal is below a certain threshold (on purpose when preparing vocals for a vocoder for instance), which isn't desirable if the *whole* sample has to be maxed out. I hope you're prototyping already, because that would be a great addition to Proton (and granulars/wavetablers in general)... :lol:

User avatar
etyrnal
Posts: 316
Joined: 24 Jan 2016
Contact:

03 Feb 2017

selig wrote:
etyrnal wrote:
selig wrote:
etyrnal wrote:
selig wrote:
etyrnal wrote:
selig wrote:
etyrnal wrote:
selig wrote:
etyrnal wrote:[quote="selig"][quote="WongoTheSane"][quote="etyrnal"]Very often though, samples are already normalized. In other words they've been adjusted so that the peak sample in the wave file is at 0 dB. So leveling is not exactly what we're talking about here. We're talking about adding gain to the entire sample while at the same time limiting.
I think you skipped one step. I was answering this:
buddard wrote:I like the idea of being able to squash the signal by boosting the quiet parts of the sample, like an extreme one-knob version of Selig Leveler! It turns the sample source into more of a "texture" to work with, basically ignoring the original dynamics. :puf_smile:
...I'm talking leveling and you're describing automatic gain control. Those are two different processes. What you describe is certainly easy to implement, but it's not leveling and probably won't give the results you expect. My argument was that implementing a Leveler (with just one knob) in Proton would be probably more difficult than a gain control (it would either require normalizing the sample first or scanning it for the peak, which isn't necessary for compression or gain control), and that using a proper Leveler provides more control at the cost of a few seconds, which I don't consider as "the long long way around".
A single knob (two, max) would be easy to do a quick "leveling" effect, two knobs would cover everything. If you assume peaks of 0 dBFS, then you don't need a Target knob. If you assume you want to level the entire dynamic range of most samples, you don't need the Curve knob. That leaves the Recovery (which could also be preset), and primarily the Blend knob. Easy to do something very workable IMO.


Sent from my iPad using Tapatalk
Recovery would be everything above 50% or like 75%. It can all be intelligently done with a single knob. If you want to do you could even have the last couple of percent be going into slight saturation.

Sent from my XT1575 using Tapatalk
Recovery is a "time" constant, so percent make no sense to me in that regard. Saturation doesn't make sense either, at least not while leveling.

Again, the idea was to remove the dynamics from the original audio if desired, not to compress/limit/saturate it.


Sent from my iPad using Tapatalk
Musical qualities are ratios... Treating things like percents and ratios gives musical results...

Sent from my XT1575 using Tapatalk
Not following…If so, then what ratio or percent is 200 ms Recovery Time?


Sent from my iPad using Tapatalk
Why would recovery time be needed? The calc is based on an individual sample point. For every individual sample point you are basically calculating the difference between the samples current level and 0db. This calculation I'm talking about his pure math. Visually picture a sample waveform in your mind. Now take that image as a JPEG and move it into an image editor in your mind. Now tell the image editor that you want to resize the Image Grab the bottom of the image and move it up towards the top. Eventually if you keep raising it you will get a flatline. All other values in the middle will be calculated as ratio of distance from the top. Of course eventually everything will be a flatline if you keep moving it upwards. Obviously some of the calculations would have to be done using natural functions, but again it's still just a simple mathematical calculation. In the example of resizing a JPEG image of a waveform, which would basically be the same thing for making calculations on an audio waveform except now you have to make a calculations in the negative and positive. So for example if you wanted to raise the lowest level in the sample by 50% of the distance between that samples current level and 0 decibels, then you calculate how much gain is required to increase that samples level by 50% of the distance between itself and 0db. For samples that are louder you do the exact same thing you calculate the difference between that current sample level and zero DB in decibels, and then you raise bye the percentage of that many decibels. Once again if using a natural function makes more sense then of course we would have to do that. But the concept Still Remains. It is literally a simple mathematical calculation. It's a ratio. Call it a fraction if you want.

Sent from my XT1575 using Tapatalk
That's not how any dynamics devices work that I'm aware of, FWIW.


Sent from my iPad using Tapatalk
right, but in Dynamic processors sometimes the producer or the musician wants to achieve certain musical effect based on playing back the sample as is. But we're talking about granular synthesis where we are potentially going to be throwing away those beautiful Dynamics, because we are trying to create something entirely different than what we started with. And if a person doesn't like how much Dynamics it's losing then don't twist the knob up for that particular patch.

Sent from my XT1575 using Tapatalk
That's exactly what I'm saying, not sure where the confusion is coming from here.


Sent from my iPad using Tapatalk[/quote]
And for example let's say that a person has a drum Loop loaded into a granular synthesizer, and they have the length and emitter frequency set so that you're actually hearing full drum hits, then just run the output of the granular synth through a leveler or a compressor limiter or whatever it is that you're trying to do. But anytime you're down dealing with grains it would make no sense to me to go to all the trouble of putting some high class high-precision compression between the sample and the emitter.

Sent from my XT1575 using Tapatalk[/quote]
OK then, this feature is not something you would use. But I bet if it sounds really cool to do this, you would change your mind on this… ;)


Sent from my iPad using Tapatalk[/quote]
Of course i would! LOL! The boost knob, which I'm now thinking should be called "Acceleration", as I've described it, would solve it perfectly fine -- and it wouldn't make the CPU cry.

Sent from my XT1575 using Tapatalk

User avatar
etyrnal
Posts: 316
Joined: 24 Jan 2016
Contact:

03 Feb 2017

WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
selig wrote:
WongoTheSane wrote:
etyrnal wrote:Very often though, samples are already normalized. In other words they've been adjusted so that the peak sample in the wave file is at 0 dB. So leveling is not exactly what we're talking about here. We're talking about adding gain to the entire sample while at the same time limiting.
I think you skipped one step. I was answering this:
buddard wrote:I like the idea of being able to squash the signal by boosting the quiet parts of the sample, like an extreme one-knob version of Selig Leveler! It turns the sample source into more of a "texture" to work with, basically ignoring the original dynamics. :puf_smile:
...I'm talking leveling and you're describing automatic gain control. Those are two different processes. What you describe is certainly easy to implement, but it's not leveling and probably won't give the results you expect. My argument was that implementing a Leveler (with just one knob) in Proton would be probably more difficult than a gain control (it would either require normalizing the sample first or scanning it for the peak, which isn't necessary for compression or gain control), and that using a proper Leveler provides more control at the cost of a few seconds, which I don't consider as "the long long way around".
A single knob (two, max) would be easy to do a quick "leveling" effect, two knobs would cover everything. If you assume peaks of 0 dBFS, then you don't need a Target knob. If you assume you want to level the entire dynamic range of most samples, you don't need the Curve knob. That leaves the Recovery (which could also be preset), and primarily the Blend knob. Easy to do something very workable IMO.


Sent from my iPad using Tapatalk
Weird, I would have said: remove Recovery (set to fast by preset), remove Blend (set to wet), keep Curve (it's how I use it usually, except Recovery: default position for normal sound processing, and fast for effect samples, wavetables, etc). Obviously only in the context of Proton, I'm not suggesting that for the device itself :D
I typically set Curve after Target, choosing how 'far' into the dynamic range I want to lift. Blend is the #1 control for me, the one I use to set the final amount of the effect. In other words, the depth of the effect. I'm constantly fine tuning Blend and to a lesser degree, Recovery. I typically set and forget Target/Curve, since they are set according to the dynamics of the source (and wouldn't change unless the source changed).

Blend controls how much you increase the low level signals, from no change to 100% leveling - isn't that what we're talking about here?


Sent from my iPad using Tapatalk
Oh, but you mean when working on normal vocals, right? I was only talking about the cases where you need to remove dynamics as much as possible (and not really care whether it's realistic or not as the sound is going to be mangled anyway). In this context, wouldn't setting Wet to 100% and using Curve to determine how much of the lower sounds will be lifted (acting more or less like a threshold: everything above is maxed out, everything below is silent) make sense?
I'm just thinking of how to adapt the concept of the Leveler to become a dynamics eliminator. In that case I'd want a depth control to determine how much to bring up the low level signals. The Curve control only determines the RANGE that you lift - it's the Blend that determines how FAR you lift that range. Small Curve settings would not make sense in this application even though they DO make sense with vocals. This is because with vocals you do NOT want to lift the LOWEST signals; the breaths, the background signals, etc.

But with a sample, you basically want to lift the entire signal - the only question is how much lift? Think of a piano decaying to silence - you want to make the piano basically not decay at all, right? That requires a large Curve setting so that the entire decay can be lifted if desired. From there it's up to the Blend control to determine how MUCH you lift the signals. Hopefully I'm making sense here…


Sent from my iPad using Tapatalk
Right, I understand. My method does indeed leave silent parts when the signal is below a certain threshold (on purpose when preparing vocals for a vocoder for instance), which isn't desirable if the *whole* sample has to be maxed out. I hope you're prototyping already, because that would be a great addition to Proton (and granulars/wavetablers in general)... [emoji38]
my creative genius always seeks the most miraculous developments... Lol

Sent from my XT1575 using Tapatalk

dana
Posts: 335
Joined: 29 Apr 2015
Contact:

03 Feb 2017

etyrnal2 wrote:
dana wrote:Its a bit of a cpu killer when using playing multiple long release notes (or rapid-fire retriggering)

Apart from that, its really amazing!

the thing is... when you're writing, a patch all by itself, and there's no other sounds going on, and you're making it super lush and smooth, you're not considering how much of that will even be perceptible when the rest of the music is playing with it...

meaning, you might be wasting all the extra super high density, because once the rest of the music is going with it, some of that is going to be a huge waste... so, when the rest of the music is playing, you can reduce the "Life" until you get a satisfying result that still gives the listener the 'feel' or the 'impression' of the effect... i.e. you might have something cranked up so 'amazing' that 30% of the extra "quality" isn't even going to be detected when the rest of the reverb tails are washing all over the place etc...

perception is king in the end
To everybody moaning that i'm complaining about the cpu usage, when you don't need extra.. well, that track i wrote with 3 instances required each to be bounced, its a top of the line macbook pro 2015 i7 quad core 16Gb ram.. Its not exactly a sloucher. Each of the channels is hardly doing very much.

When will reason people stop trying to justify deficiences in software? If we didn't mention these things, reason wouldn't be where it is today.

I think there is a definite room for improvement in the cpu usage.

User avatar
etyrnal
Posts: 316
Joined: 24 Jan 2016
Contact:

03 Feb 2017

WongoTheSane wrote:
etyrnal wrote:You're kidding, right?

So, how do you You do this with samples in the factory sound bank etc.
1) Drag the sample from the FSB to the sequencer.
2) Drop a Leveler in the Insert FX slot.
3) Press P to play the sample in a loop.
4) Set Target on Leveler to whatever the Peak Hold is.
5) Push Curve to +24db or whatever you like. 100% wet gives best results.
6) Bounce in place.
6a) (optional) Normalize.
7) Bounce result to New Samples.

Takes about 10 seconds + duration of the sample.
etyrnal wrote:Were all smart enough to realize that we could take the long long way around. But why?

It's like saying, "why make Proton, when you could just NN-19 mumble mumble Sample Start mumble mumble automate mumble mumble patch mumble mumble lfo adsr hack patch?"

Why is it that when you suggest a new feature that would offer quite useful Improvement, there's always somebody who's got a suggest that you could do it all from scratch and manually? It's like an argument against Improvement and evolution. I just don't get that.
Calm down, I was just describing a workaround that I use everyday (with Expanse these days).
You for got step 0... BUY a new "Leveler" Re

So, Your solution is to buy another Rack Extension?

But why even buy a Leveler, when you could build one out of a long chain of cleverly combined/patched Reason 2.5 built-in devices?

:idea: "if You fight for your limitations, you get to keep them." ~ author unknown
Last edited by etyrnal on 03 Feb 2017, edited 1 time in total.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

03 Feb 2017

etyrnal wrote:
WongoTheSane wrote:
etyrnal wrote:You're kidding, right?

So, how do you You do this with samples in the factory sound bank etc.
1) Drag the sample from the FSB to the sequencer.
2) Drop a Leveler in the Insert FX slot.
3) Press P to play the sample in a loop.
4) Set Target on Leveler to whatever the Peak Hold is.
5) Push Curve to +24db or whatever you like. 100% wet gives best results.
6) Bounce in place.
6a) (optional) Normalize.
7) Bounce result to New Samples.

Takes about 10 seconds + duration of the sample.
etyrnal wrote:Were all smart enough to realize that we could take the long long way around. But why?

It's like saying, "why make Proton, when you could just NN-19 mumble mumble Sample Start mumble mumble automate mumble mumble patch mumble mumble lfo adsr hack patch?"

Why is it that when you suggest a new feature that would offer quite useful Improvement, there's always somebody who's got a suggest that you could do it all from scratch and manually? It's like an argument against Improvement and evolution. I just don't get that.
Calm down, I was just describing a workaround that I use everyday (with Expanse these days).
You for got setp 0... BUY a new "Leveler" Re

So, Your solution is to buy another Rack Extension?

But why even buy a Leveler, when you could build one out of a long chain of cleverly combined/patched Reason 2.5 built-in devices?

:idea: "if You fight for your limitations, you get to keep them." ~ author unknown
I don't think that's what was being said at all. Someone was trying to be helpful by suggesting another way to get the desired results, that's all.

If this thread is to be helpful to anyone, I think it best to stay on topic…how 'bout that Proton?!? ;)


Sent from my iPad using Tapatalk
Selig Audio, LLC

Peter

03 Feb 2017

selig wrote:If this thread is to be helpful to anyone, I think it best to stay on topic…how 'bout that Proton?!? ;)
I dunno about y'all but I'm about to be using Proton in every project. I'm learning to appreciate Mode 1 more - instant techno atmosphere! Run something through Mode 1, add the kHs Resonator to find a sweet note to accentuate according to the composition and modulate wetness with something like Ammo. There's my tip of the day. :thumbs_up:

Q: Hey Peter what do you use to make music?

A: Proton.

:D

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

03 Feb 2017

Peter wrote:
selig wrote:If this thread is to be helpful to anyone, I think it best to stay on topic…how 'bout that Proton?!? ;)
I dunno about y'all but I'm about to be using Proton in every project. I'm learning to appreciate Mode 1 more - instant techno atmosphere! Run something through Mode 1, add the kHs Resonator to find a sweet note to accentuate according to the composition and modulate wetness with something like Ammo. There's my tip of the day. :thumbs_up:

Q: Hey Peter what do you use to make music?

A: Proton.

:D
Another tip - use the Comb (resonating) filter (with keyboard tracking) in Proton to add a pitch element to non-pitched samples in Mode 1. Tune the filter to 261.6 Hz (or an octave relative to that) to track accurately. :)


Sent from my iPad using Tapatalk
Selig Audio, LLC

User avatar
KirkMarkarian
Posts: 292
Joined: 13 Dec 2015
Location: Tucson, AZ
Contact:

03 Feb 2017

Can Proton/Fritz make stuff like this:

User avatar
etyrnal
Posts: 316
Joined: 24 Jan 2016
Contact:

03 Feb 2017

KirkMarkarian wrote:Can Proton/Fritz make stuff like this:
Yes.

Sent from my XT1575 using Tapatalk


User avatar
Exowildebeest
Posts: 1553
Joined: 16 Jan 2015

03 Feb 2017

KirkMarkarian wrote:Can Proton/Fritz make stuff like this:
Ikeda :thumbs_up:

User avatar
KirkMarkarian
Posts: 292
Joined: 13 Dec 2015
Location: Tucson, AZ
Contact:

03 Feb 2017

Well, I delved in deeper to Proton - it can aid in creating the sound I am looking for - but for making those types of click/pop patterns, I had to get the Propulsion sequencer. I can design the synth sounds by hand, but the means of using Proton or Fritz to actually create those types of click/pop patterns doesn't seem to exist, either in option 1 or 2. I am unable to slow down the pulses to just make repeatable clicks or pops.

User avatar
KirkMarkarian
Posts: 292
Joined: 13 Dec 2015
Location: Tucson, AZ
Contact:

03 Feb 2017

nooomy wrote:I made some house with proton :)

This is beautiful music!

Marc64
Posts: 741
Joined: 24 Jan 2015
Location: Sweden
Contact:

03 Feb 2017

One thing I think is alittle wierd that on the lfo's I turn clockwise to change speed up and on mod env len I turn counter clockwise.
I tend to turn the wong way on them :)
Maybe alearnign thing but... :)

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

03 Feb 2017

Marc64 wrote:One thing I think is alittle wierd that on the lfo's I turn clockwise to change speed up and on mod env len I turn counter clockwise.
I tend to turn the wong way on them :)
Maybe alearnign thing but... :)
LFO is "rate" (in Hertz), going from low to high. Envelope is "length" (in seconds), going from short to long.


Sent from my iPad using Tapatalk
Selig Audio, LLC

Marc64
Posts: 741
Joined: 24 Jan 2015
Location: Sweden
Contact:

03 Feb 2017

selig wrote:
Marc64 wrote:One thing I think is alittle wierd that on the lfo's I turn clockwise to change speed up and on mod env len I turn counter clockwise.
I tend to turn the wong way on them :)
Maybe alearnign thing but... :)
LFO is "rate" (in Hertz), going from low to high. Envelope is "length" (in seconds), going from short to long.


Sent from my iPad using Tapatalk
yeah I understand that but I get confused alittle when they turn the oposite way for the same rate :)

Skickat från min SM-G935F via Tapatalk

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

03 Feb 2017

Marc64 wrote:
selig wrote:
Marc64 wrote:One thing I think is alittle wierd that on the lfo's I turn clockwise to change speed up and on mod env len I turn counter clockwise.
I tend to turn the wong way on them :)
Maybe alearnign thing but... :)
LFO is "rate" (in Hertz), going from low to high. Envelope is "length" (in seconds), going from short to long.


Sent from my iPad using Tapatalk
yeah I understand that but I get confused alittle when they turn the oposite way for the same rate :)

Skickat från min SM-G935F via Tapatalk
When both are in Sync mode, this is true - probably not the best decision but I believe we are not stuck with it as it would break backwards compatibility. Still, thinking of the LFO as a rate, and the Env as a length helps. Remember, the Env isn't always in "loop" mode, so making a one shot envelope "length scale" control a "rate" control didn't make sense. Sorry for the confusion - would probably do it differently now if I had the chance!
:)


Sent from my iPad using Tapatalk
Selig Audio, LLC

User avatar
nooomy
Posts: 543
Joined: 16 Jan 2015

03 Feb 2017

KirkMarkarian wrote:
nooomy wrote:I made some house with proton :)

This is beautiful music!
Thank u! I really like the proton re!! ;)

User avatar
Catblack
Posts: 1020
Joined: 15 Apr 2016
Contact:

03 Feb 2017

KirkMarkarian wrote:Can Proton/Fritz make stuff like this:
That's funny, I saw his data.tron installation about a month ago. On one hand it was impressive, but I wouldn't call it something I'd want to sonicly emulate. But to each their own. You can probably get this sort of sound from Proton. You can get a lot better ones, too.
If you ain't hip to the rare Housequake, shut up already.

Damn.

AJ_3000
Competition Winner
Posts: 98
Joined: 19 Jan 2015

04 Feb 2017

Ottostrom wrote:Kind of a shame they kept the style of their other REs around the screen part. I know the functionality is the important part, but I also want to feel inspired by the look!
Personally I really like the way the devices look. The grey enamel / Hammerite finish complete with the computer style logo are perfectly suited for such weird and wonderful instruments. These Rack Extensions could have been brewed up in the subterranean laboratory operated by a B-movie-star-style mad scientist called Dr Selig, a genius now in thrall to the sinister 'LoveOne Consulting' corporation whose aim is world domination through mind-control using audio waves: instruments liberated by the forces-of-good in the form of the mighty Propellerheads swooping down from their secret Scandinavian mountain-top base on behalf of liberty, justice and Reason.

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 28 guests