[pinnahertz]

 
In earlier posts, he showed how synths are also not inherently limited in dynamic range, just like they are not limited in producing ultrasonics.

Not at all. He showed a case of a software synth generating a signal deliberately chosen to produce ultrasonics, but would never by chosen to make music. He did not address all "synths", not even close. It's one specific case, and while it may occur in other software synths as well, it's not representative of "synths" at all, certainly not being used to produce musical signals.

Inherently, synths are only limited by the creative minds used to write their software, and the processing power of the hardware they run on, both of which are continually evolving.

If only it were that simple. The real limitation in producing sound in a synth is a combination of the usability of the tools, understanding of the technology, and a very strong artistic sense. The strong need of both technical aptitude and artistic aptitude occurs strongly in synth programming. That combination is rare in individuals, they generally favor one aptitude strongly over the other, but is the real limit.

Please don't mention human audibility (that's a given).  

Good you understand that.

 

[pinnahertz]

  I mentioned that in practice, it doesn't mean that producing music at so high sampling rate is possible.
 
For some reason, when I played that square wave my CPU load reached ~50% or so and this is only one synth doing one very basic thing. It caused my computer to lag, and I heard clicks and pops every sec or so. Of course I could still render it but I couldn't play it in real time (without the mentioned problems).

So your example wasn't even playable. That scales what you have "proven" down just a bit.

Granted, producers will have better hardware than I but producing music involves a lot of processing. I don't think it's possible to have a combination of hardware and software that let's you to do the whole producing in 192kHz. At least not yet.

Might it not be that you were asking your specific software synth to do something it was never intended to do? I have a lowly laptop with several different DAW apps, each of which can run tracks and tracks of software synths running patches rich and complex. No delays, and I have yet to find the track limit, played live without rendering. The composite mix is far more demanding demanding than generating a simple square wave.

They aren't (always) inherently limited but practically, they always are, to some extent.  

If you want to verify, download a function generator app and try your squarewave with that. I'll be it works just fine and doesn't tax your CPU.
 
Sounding more and more like what you've proven is that your software synth can't actually produce that signal in a playable form.

 

[StanD]

  I mentioned that in practice, it doesn't mean that producing music at so high sampling rate is possible.
 
For some reason, when I played that square wave my CPU load reached ~50% or so and this is only one synth doing one very basic thing. It caused my computer to lag, and I heard clicks and pops every sec or so. Of course I could still render it but I couldn't play it in real time (without the mentioned problems). Granted, producers will have better hardware than I but producing music involves a lot of processing. I don't think it's possible to have a combination of hardware and software that let's you to do the whole producing in 192kHz. At least not yet. They aren't (always) inherently limited but practically, they always are, to some extent.

Intel will be glad to have you buy a new system with the latest and greatest CPU. After that Synth developers will be glad to consume all available CPU cycles for some new doodad and so the viscous cycle continues. By the way which CPU/Clock are you running? Did you check to see what else was going on your PC while you were running the synth software? You have enough RAM to avoid paging?

 

[VNandor]

  So your example wasn't even playable. That scales what you have "proven" down just a bit.

 
 
 
Let me explain that. First of all, I could render the project without any problems and I could play it back without any problems. (Just as I've already said that in my prevoious post). What I couldn't do well is to play the synth real time inside the daw when I applied a high amount of oversampling under the "draft" tab.


I wouldn't even need to use that by the way, and only enable it for rendering. However if a producer wanted to hear how the rendered file is going to sound then the oversampling under the draft tab must be the same as under the render tab. The plugin's help  point out to that if you don't believe me.

I just activated it out of curiousity to see what happens. Then I concluded that, practically it cannot be used that way (using high amount of  real time oversampling) as it quickly overloads my CPU so ultimately it doesn't help and makes the sound even more different than the rendered file would be.
As for why I activated it for rendering, the spectrogram I showed wouldn't look nearly as convincing without oversampling.
I'm not sure if you were misinterpreting what I was saying on purpose or you really just didn't understand it.
 
 
 
 
 

  Might it not be that you were asking your specific software synth to do something it was never intended to do

Well I asked the software to do 64x oversampling real time. It's probably there because in some cases it has its uses. I've already noted I didn't "have to" use it, I only used it to see how my computer reacts. What I had to use is the oversampling for rendering.
 
 

  Sounding more and more like what you've proven is that your software synth can't actually produce that signal in a playable form.

Just to make sure you get what I'm saying, I repeat again the .wav file could be played back perfectly. The problems I mentioned were only present when I tried to play the synth inside the daw.

 

[pinnahertz]

   
 
 
Let me explain that. First of all, I could render the project without any problems and I could play it back without any problems. (Just as I've already said that in my prevoious post). What I couldn't do well is to play the synth real time inside the daw when I applied a high amount of oversampling under the "draft" tab.
 
<snip!>
 
Just to make sure you get what I'm saying, I repeat again the .wav file could be played back perfectly. The problems I mentioned were only present when I tried to play the synth inside the daw.

You created an unplayable synth patch you could only use after rendering to prove that "synths produce ultrasonics".  
 
Got it.

 

[spruce music]

   
 
 
Let me explain that. First of all, I could render the project without any problems and I could play it back without any problems. (Just as I've already said that in my prevoious post). What I couldn't do well is to play the synth real time inside the daw when I applied a high amount of oversampling under the "draft" tab.


I wouldn't even need to use that by the way, and only enable it for rendering. However if a producer wanted to hear how the rendered file is going to sound then the oversampling under the draft tab must be the same as under the render tab. The plugin's help  point out to that if you don't beleive me.

I just activated it out of curiousity to see what happens. Then I concluded that, practically it cannot be used that way (using high amount of  real time oversampling) as it quickly overloads my CPU so ultimately it doesn't help and makes the sound even more different than the rendered file would be.
As for why I activated it for rendering, the spectrogram I showed wouldn't look nearly as convincing without oversampling.
I'm not sure if you were misinterpreting what I was saying on purpose or you really just didn't understand it.
 
 
 
 
 
Well I asked the software to do 64x oversampling real time. It's probably there because in some cases it has its uses. I've already noted I didn't "have to" use it, I only used it to see how my computer reacts. What I had to use is the oversampling for rendering.
 
 
Just to make sure you get what I'm saying, I repeat again the .wav file could be played back perfectly. The problems I mentioned were only present when I tried to play the synth inside the daw.
 
 
 

Well I think you should get an award for persisting beyond all reason until you push an irrelevant point into total irrelevance.  Maybe we could call it irrelevance squared?

 

[VNandor]

  You created an unplayable synth patch you could only use after rendering to prove that "synths produce ultrasonics".  
 
Got it.

It did produce ultrasonics. There rendered file was what I expected. I could play the patch if I didn't use real time oversampling. I think you are doing this on purpose.

 

[VNandor]

 
Well I think you should get an award for persisting beyond all reason until you push an irrelevant point into total irrelevance.  Maybe we could call it irrelevance squared?

Seriously, I was just nitpicking the statements like these.

  Yes, the analog synths could produce ultrasonics. And we live in a time when there is a fascination with analog synths, so they are fairly thick on the ground.  But digital synths and sampling synths seem to be hanging at 16/48, thus very little ultrasonic capability.

 

   
2. Samplers and syths have restricted dynamic ranges and frequency ranges compared to many acoustic instruments or groups of acoustic instruments.

 
Both ones has been adressed since, what I missed from these is something like "practically" or in "any real life applications" or anything else along these lines.
 
And after that the arguing continues because apparently, the spectra of the waveform I created with an actual software that is being used edm producers is not good enough to prove the theoretical limits of the synth is beyond what they seemed to assume. The theoretical limits, I think I should stress it again, not the practical one. But when I made the post, I didn't know they meant practical limits because they didn't mention it.

 

[Yuri Korzunov]

Spectrum of syntesizer at analog output is infinite. Due aliases for digital one though For analog one harmonics distributed to infinity. It is matter of noise floor and spectrum anayzer sensitivity.
 
What can limit dynamic range of synthesizer? Implementation only.

 

[pinnahertz]

  It did produce ultrasonics. There rendered file was what I expected. I could play the patch if I didn't use real time oversampling. I think you are doing this on purpose.

YES, I am.  And here's why:

 
In earlier posts, he showed how synths are also not inherently limited in dynamic range, just like they are not limited in producing ultrasonics.

That's what some people are getting from your posts.  It's one post by one user, but he no doubt reflects many, many more who are not active participants.
 
The conclusion in that post  is completely wrong and misleading.  We already have several posts that seem to have been mislead by your misinformation.  
 
"Synths" is the first problem.  "Synths" were not used.  What we have here is one example of a specific software synth that was coaxed deliberately. and with effort, into producing ultrasonics.  That example doesn't represent "Synths" in general at all, nor their lack of dynamic or frequency limitations, it represents one specific case where the user deliberately forced the result.   
I'll give you a little credit for noting the situation was impractical, but that's not what some people are getting from this. 
 
To prove what Lazy is saying you'd have to look at the output of a huge number of synths, both software and hardware, of all the different types, sample a wide range of patches, look at the resulting spectra, then draw your conclusions.  Or, do what I did, look at the raw technology, the sample rates involved, and the practical application of synths in general.  
 
I have no idea why it's important to cling to the heavily biased theoretical example. The demonstration was good and correctly done, but has no relevance to the real world of "synths" producing ultrasonics.
 
The continual reinforcement of it is, in my opinion, irresponsible, as it advances an idea that influences incorrect conclusions by others.

 

[sonitus mirus]

You guys still at it?
 
Too bad this study didn't include EDM as one of the 10 music genres tested. 

 
Dynamic Range Across Music Genres and the Perception of Dynamic Compression in Hearing-Impaired Listeners
 
Now I finally know where Schlager stands with regards to dynamic range.

 

[StanD]

  You guys still at it?
 
Too bad this study didn't include EDM as one of the 10 music genres tested. 

 
Dynamic Range Across Music Genres and the Perception of Dynamic Compression in Hearing-Impaired Listeners
 
Now I finally know where Schlager stands with regards to dynamic range.

Hearing impaired audiophiles can hear ultrasonics, so they could do the ultrasonic EDM thing.

 

[pinnahertz]

  You guys still at it?
 
Too bad this study didn't include EDM as one of the 10 music genres tested. 

 
Dynamic Range Across Music Genres and the Perception of Dynamic Compression in Hearing-Impaired Listeners
 
Now I finally know where Schlager stands with regards to dynamic range.

This link made it all worth while.  Thanks!

 

[pinnahertz]

 
Hearing impaired audiophiles can hear ultrasonics, so they could do the ultrasonic EDM thing.

Ultrasonic EDM: dogs and bats shuffling and fist-pumping.  Can't wait. 

 

[StanD]

  Ultrasonic EDM: dogs and bats shuffling and fist-pumping.  Can't wait. 

It'll make for an interesting mosh pit.