Hello
I do have something to ask... Every now and then someone asks for a VST emulation of some vintage synth that is not avaiable from the big software synth companies. Inevitably, someone mentions Reaktor ensembles and Synthedit versions. We all now, that those synths can only imitate the sound architecture (that is, how many OSC, LFOs, etc and the way they are connected) but they are not really emulating they synth, with all its quirks and defects thatn turned into features after 30 years.
Now I have the question: How much is the sound of a synth defined by its architecture? Are the sounds of the MiniMog defined only by its circuitry? Or maybe the architecture forced the musicians/programmers into certain sounds that will now be forever associated with it?
Will anyone notice in a mix?
Got me thinking...
Emulations - Architecture vs. real emulation
-
- KVRist
- 180 posts since 19 Apr, 2004 from Espoo, Finland
Often it is a case of both. Minimoog in particular has quite a characteristic sound, but certainly a lot of it are the limitations of the synth too (ADS envelopes, lowest note priority etc).vmartell wrote: Now I have the question: How much is the sound of a synth defined by its architecture? Are the sounds of the MiniMog defined only by its circuitry? Or maybe the architecture forced the musicians/programmers into certain sounds that will now be forever associated with it?
In later synths the circuitry is often more close to ideal and its effect is less (many 80s polysynths).
If you want to know more about emulating actual circuitry, you might want to check my paper on emulating the Moog filter: http://dafx04.na.infn.it/WebProc/Proc/P_061.pdf
-
- Banned
- 705 posts since 29 Jan, 2004
@vmartell:
Well.
You really did a good anylysis of the current situation. Many people (unfortunately) believe, that all you have to do is the simulation of the sound architecture and the signal flow. Some of them even did never really taste the real synthesizer, merely "googled" the written specification anyhow...
That's actually a misere in my opinion.
Even if you do that with standardized tool sets (like SE or Reaktor), you wouldn't be able to simulate the piece of analog stuff approximately.
All they then probably did, was building another synth, which merely has the sound architecture of the original. Sounding and behaving completely different.
But more: often it is NOT EVEN possible to simulate that sound architecture correctly or exactly with those tools cause of the limitatoions or wrong implementation of the details.
And finally (that's in fact the only thing you can be sure of) they want to make us believe, that the printed "logo" or the "faked name" or the "simulated interface" on the so created synthesizers guarant us any "authenticity"...
You can sell an apple as a banana to a men, who never did see a banana or an apple before - that is often the calcula of some suppliers (or they even don't know it any better themselfes)...
I did not even listen to ONE authentic simulation created with SE or Reaktor or any other of those musician LEGO systems. (That also probably was never intented to do with those tools finally ...).
Reaktor is not even able to simulate the digital reference VA hardware synthesizer, the Virus for some reasons. But it has it's strongth in other things of course!
Sound architecture and the special behaviour of the single components are essentially to a synthesizer - digital or analog - that makes no difference.
The only way would be to perform a minutious in-depth (highly professional) analysis of the object of cloning to finally come to an approximative result of simulation (assuming there is enuff knowledgement about the entire subject finally - and last but not least - the ORIGINAL instrument is also physically available for examination).
That *is* actually a "kings dicipline", the upper ligue of DSP plugin programming. And that requires obviously much more flexibility and power in the used programming language than any sound lego system is currently (or ever) able to offer.
.
Well.
You really did a good anylysis of the current situation. Many people (unfortunately) believe, that all you have to do is the simulation of the sound architecture and the signal flow. Some of them even did never really taste the real synthesizer, merely "googled" the written specification anyhow...
That's actually a misere in my opinion.
Even if you do that with standardized tool sets (like SE or Reaktor), you wouldn't be able to simulate the piece of analog stuff approximately.
All they then probably did, was building another synth, which merely has the sound architecture of the original. Sounding and behaving completely different.
But more: often it is NOT EVEN possible to simulate that sound architecture correctly or exactly with those tools cause of the limitatoions or wrong implementation of the details.
And finally (that's in fact the only thing you can be sure of) they want to make us believe, that the printed "logo" or the "faked name" or the "simulated interface" on the so created synthesizers guarant us any "authenticity"...
You can sell an apple as a banana to a men, who never did see a banana or an apple before - that is often the calcula of some suppliers (or they even don't know it any better themselfes)...
I did not even listen to ONE authentic simulation created with SE or Reaktor or any other of those musician LEGO systems. (That also probably was never intented to do with those tools finally ...).
Reaktor is not even able to simulate the digital reference VA hardware synthesizer, the Virus for some reasons. But it has it's strongth in other things of course!
Sound architecture and the special behaviour of the single components are essentially to a synthesizer - digital or analog - that makes no difference.
The only way would be to perform a minutious in-depth (highly professional) analysis of the object of cloning to finally come to an approximative result of simulation (assuming there is enuff knowledgement about the entire subject finally - and last but not least - the ORIGINAL instrument is also physically available for examination).
That *is* actually a "kings dicipline", the upper ligue of DSP plugin programming. And that requires obviously much more flexibility and power in the used programming language than any sound lego system is currently (or ever) able to offer.
.
-
- KVRian
- 1398 posts since 9 Dec, 2002
Agreed. None of the Reaktor ensembles so far have been close enough to be accepted by yours truly. While something like Junatik is a usable, good sounding synth in itself, a Juno-106 replacement it isn't.
Some emulations are getting close to the targets... but they could still be more accurate (here's hoping that one day we'll have a good replacement for 303/Devilfish amongst other things
)
Regards,
JMH
Some emulations are getting close to the targets... but they could still be more accurate (here's hoping that one day we'll have a good replacement for 303/Devilfish amongst other things
)Regards,
JMH
Now available with added Inherently Suspect Justification!
-
- Banned
- 705 posts since 29 Jan, 2004
Who promised to make a clone (or replacement) of a 303 or DevilFish?jmh wrote:Agreed. None of the Reaktor ensembles so far have been close enough to be accepted by yours truly. While something like Junatik is a usable, good sounding synth in itself, a Juno-106 replacement it isn't.
Some emulations are getting close to the targets... but they could still be more accurate (here's hoping that one day we'll have a good replacement for 303/Devilfish amongst other things
Regards,
JMH
.
-
- KVRAF
- 4822 posts since 14 Mar, 2002 from Somewhere else, on principle
"Real emulation"?
-
- KVRian
- 1398 posts since 9 Dec, 2002
Certain developers have... they've delivered their takes, but nothing so far is "good enough" for my taste. One commercial plugin is more than good enough for my use though, altough it could of course be improved.pustekuchen wrote:Who promised to make a clone (or replacement) of a 303 or DevilFish?
Sadly I cannot disclose any information regarding a possible Devilfish emulation. All I can tell is that the Devilfish I've been thinking of buying has raised the interest of some parties...
Regards,
JMH
Now available with added Inherently Suspect Justification!
-
- KVRer
- 9 posts since 2 May, 2005 from Richmond, Virginia
I dont get why people are so stuck on emulations. I am working on a emulation of the Mircokorg on Reaktor 5. I know it will not sound just like it for one simple reason. The oscillators and filters will sound different. Just like u can have a Korg trition or a yamaha Motiff. Even tho its basicly the same idea they sound different. However the way I see it I rather build a better microkorg then the samething. I do have another question? microkorg have 71 waveforms. I want to sample different wave forms and use them as my oscillators. How can u set the presets to pull up different sounds?
Checkout my site at http://www.soundclick.com/biggrime
-
- KVRian
- 1398 posts since 9 Dec, 2002
The obsession for emulations in my case is simple, getting software equivalents for expensive (damn collectors and fetishists!) for gear I'd love to use.
I don't mind at all if a developer releases a synth that's based on "303 features", as long as the developer doesn't claim it's an authentic sounding emulation. If he does, it'll be compared to the real thing and harsh critique will be given if it fails to deliver what's promised.
As for your Korg issue... vanilla patch with no filter, no modulations etc, one oscillator, full Sustain on the amplitude ADSR envelope. Sample that way, each waveform at a time. That'll go a long way. Depending on the target sampler/synth you might have some limitations regarding the sample length and thus frequency, but I wouldn't worry about that as most sampler VSTis don't care about such issues
Regards,
JMH
I don't mind at all if a developer releases a synth that's based on "303 features", as long as the developer doesn't claim it's an authentic sounding emulation. If he does, it'll be compared to the real thing and harsh critique will be given if it fails to deliver what's promised.
As for your Korg issue... vanilla patch with no filter, no modulations etc, one oscillator, full Sustain on the amplitude ADSR envelope. Sample that way, each waveform at a time. That'll go a long way. Depending on the target sampler/synth you might have some limitations regarding the sample length and thus frequency, but I wouldn't worry about that as most sampler VSTis don't care about such issues
Regards,
JMH
Now available with added Inherently Suspect Justification!
-
- Banned
- 705 posts since 29 Jan, 2004
Aah. The Phoscyon thingy probably ...jmh wrote:Certain developers have... they've delivered their takes, but nothing so far is "good enough" for my taste. One commercial plugin is more than good enough for my use though, altough it could of course be improved.pustekuchen wrote:Who promised to make a clone (or replacement) of a 303 or DevilFish?
Sadly I cannot disclose any information regarding a possible Devilfish emulation. All I can tell is that the Devilfish I've been thinking of buying has raised the interest of some parties...
Regards,
JMH
.
-
- KVRAF
- 5703 posts since 6 Dec, 2003 from Mission Control
If I can program a factory patch and get something close to the sound of the factory patch. In my book that's good enough for software..
Check out "96 Tears" on my Free P-600
http://www.ele4music.com/vst/p-600.zip
90% of these patches sound close enough in a mix to pass for the hardware.
I have never expected software to sound like hardware because it's not hardware. I think it sounds pretty good but I know better than to expect too much.. The bottom line is. If you can get the sounds you want with software great you don't have to buy the hardware... If you are missing that extra something that you can only get with hardware well start saving your money 'cause there is only one way to get the sound of hardware and that's hardware. That is not going to change anytime soon (like before you die)unless we get speed of light computers sooner than that.
Regards
Check out "96 Tears" on my Free P-600
http://www.ele4music.com/vst/p-600.zip
90% of these patches sound close enough in a mix to pass for the hardware.
I have never expected software to sound like hardware because it's not hardware. I think it sounds pretty good but I know better than to expect too much.. The bottom line is. If you can get the sounds you want with software great you don't have to buy the hardware... If you are missing that extra something that you can only get with hardware well start saving your money 'cause there is only one way to get the sound of hardware and that's hardware. That is not going to change anytime soon (like before you die)unless we get speed of light computers sooner than that.
Regards
-
- KVRist
- 193 posts since 18 Jul, 2001
Hi antti,antti wrote:If you want to know more about emulating actual circuitry, you might want to check my paper on emulating the Moog filter: http://dafx04.na.infn.it/WebProc/Proc/P_061.pdf
are you planning on a complete Mini emulation of your own? I really like your ASynth and think a fully blown antti Mini would be worth it. What do you think?
kybernaut
-
- Banned
- 705 posts since 29 Jan, 2004
I just downloaded it.tomg wrote:If I can program a factory patch and get something close to the sound of the factory patch. In my book that's good enough for software..
Check out "96 Tears" on my Free P-600
http://www.ele4music.com/vst/p-600.zip
90% of these patches sound close enough in a mix to pass for the hardware.
I have never expected software to sound like hardware because it's not hardware. I think it sounds pretty good but I know better than to expect too much.. The bottom line is. If you can get the sounds you want with software great you don't have to buy the hardware... If you are missing that extra something that you can only get with hardware well start saving your money 'cause there is only one way to get the sound of hardware and that's hardware. That is not going to change anytime soon (like before you die)unless we get speed of light computers sooner than that.
Regards
With the first try of some patches I can actually say you: It don't even sound aproximately like a hardware synth nor *any* emulation. (It just sounds like any other average SE synth - and that is not because I "hate" SE or something. It is simply a fact. There are also no special extenal modules used, as far I can see - which would be at least necessary to emulate something, right? ...)
(That entire SE-Hardware-Emulation VSTi promborium on that website is by the way an entire and really bad fake to the clueless people!)
But I did also some deeper analysis by listening intensively.
What happens, if you play low notes for example (generally)?
There is a total lost of spectral content for some reasons. Can't you hear that? Very typical for wavetable interpolation with missing sampled keyrange.
All in all it sounds static, week, and livingless. Even purely digital. Pressure and spectral shine of oscillators and any analog filter feeling is completely missing here.
How do you actually come to the conclusion, that it sounds like hardware?
If it sounds like hardware, so then probably like a cheep ROMpler from the 80's (which has got a low quality sample set because of some memory limitation).
Sorry, if I completely demotivated you. But that is for me a total disappointment. And it even shows, that I am not that wrong with my thesis above, which is by the way based on analysis of equipment.
But you gave it away for free. That's honorable!
But please don't tell the people, who have never listened to a real analog hardware synthesizer, that you made a simulation of such. Because you didn't.
That's actually the point we discussed here. The nice screen cannot compensate the total lack of sound quality.
Sorry again.
.
Last edited by blümchen on Wed Jul 06, 2005 8:49 am, edited 1 time in total.