SQ80 emulating VSTi available

[tomg]

full analog simulation rather than just an emulation, might become feasible in a couple of years.

Ok

[olepro]

Like 1-2-Many said.
Rainers Ensoniq site and all his research on the synth, yes even buying up the last SQ/ESQ1 spare parts before they were distroid and make them available to us is fantastic.
We owe you a big thank you for keeping this great synth alive.
And now Sigfried who's done the SQ8L vsti he must on the thank you list too.
You are so cool both of you !

I'm looking VERY forward to se the finished SQ8X.
I will buy it no matter the price (allmost)
Now you know that Sigi.....

[r_buchty]

Transistors switch at near light speed. This switching forms the leading edge of oscillator saw and square waveforms. If the waveform isn't correct it won't matter what follows.

Getting a bit esoteric here, aren't you?

There is no such thing as "the" transistor. That analog stuff just has too many tolerances, i.e. none is alike. And especially with music synthesis this is a *bad* thing as you get all kind of nastinesses which the designers then tend to avoid e.g. by matching individual transistors, or buying expensive laser-trimmed stuff.

If things would be as dramatic as you describe, there could never be "the" typical Moog filter sound, "the" typical SSM2040 filter sound, "the" typical WASP filter sound, etc.

Thankfully, those characteristics are derived from the "macroscopic" effect, e.g. how the individual components are arranged.

Going digital will introduce other challenges, but getting the sound of a transistor switching at near light speed (which it does inside your CPU as well, btw., and there probably *much* faster than your average audio transistor which won't be any good in the GHz area anyway...) isn't.

Don't get me wrong. I'm a VST fanboy but there's theory and then there's reality.

Agreed

[r_buchty]

You mean Antii from Smartelectronix who coded the Taurus 1?

That's the one.

A lot of people complained about my list that it's not accourate and I had to defend myself with several examples (for example Nubi which is a Hammond Emulation, but seing as every hammond sounds different right from the start - then you can't nail it with an emulation, same with Moogs).

The second problem is that people tend to have somewhat, erm, distorted memories about how certain machines sound.

Do you remember the contest Peter Gorges did a couple of years ago in the German Keyboards magazine? There he used a collection of various synths to mimic stereotypes of other synths, asking the people to detect which was the original and which was fake.

Noone got all right. Sure, there were some easy ones everyone spotted right away. But the tough ones could only be detected in a direct a/b comparison and would never be audible in a mix.

*Und? Könnt ihr zwei vielleicht schon mehr über den SQ8X sagen? *

Das überlasse ich ganz Sigi... Ich war ja nur die "architektonische Wissensbasis"

[meeks]

Was out all weekend..is the link dead? Or is it the workplace filter?

[tomg]

Transistors switch at near light speed. This switching forms the leading edge of oscillator saw and square waveforms. If the waveform isn't correct it won't matter what follows.

Getting a bit esoteric here, aren't you?

There is no such thing as "the" transistor. That analog stuff just has too many tolerances, i.e. none is alike. And especially with music synthesis this is a *bad* thing as you get all kind of nastinesses which the designers then tend to avoid e.g. by matching individual transistors, or buying expensive laser-trimmed stuff.

If things would be as dramatic as you describe, there could never be "the" typical Moog filter sound, "the" typical SSM2040 filter sound, "the" typical WASP filter sound, etc.

Thankfully, those characteristics are derived from the "macroscopic" effect, e.g. how the individual components are arranged.

Going digital will introduce other challenges, but getting the sound of a transistor switching at near light speed (which it does inside your CPU as well, btw., and there probably *much* faster than your average audio transistor which won't be any good in the GHz area anyway...) isn't.

Don't get me wrong. I'm a VST fanboy but there's theory and then there's reality.

Agreed

That's all very academic. Not to drag this out because we do agree but how fast, how many cpu cycles, at what sample rate, opposed to applied forward voltage?

You are aware that sample rate only allows state changes at a fixed frequency. This means that you have to wait for the next sample to see if something is turned on no matter how fast the machine is. In order to do full analog simulation an almost infinate sample rate would have to be used. I've the feeling it's going to take quite a few more years to get it right.

In the mean time I'm thrilled that this stuff works as well as it does. Huge thank you to you and Siegfried for the SQ-8L. It's one of the best instruments so far and may be the best VSTi emulator ever.

BTW I've wanted real time spice from the very beginning.

[mistertoast]

If you want to emulate the subatomic level, we have ages to go. :->

If you want to get a VSTi that operates within the range of the tolerances exhibited in physical specimens of a synth sampled at CD rates, it's more a modeling and programmer problem than a MHz problem.

[mistertoast]

>>You are aware that sample rate only allows state changes at a fixed frequency.

But todays computers are fast enough, with good programming, to calculate many intermediate samples. Enough to push the quality above CD and the audio equipment the music is played on, as well as most human ears.

The trick is how to model what goes into those samples. It's really hard. As others have pointed out, the more variability in the original instrument, the easier it is to get close to the real thing.

[tomg]

If you want to emulate the subatomic level, we have ages to go. :->

If you want to get a VSTi that operates within the range of the tolerances exhibited in physical specimens of a synth sampled at CD rates, it's more a modeling and programmer problem than a MHz problem.

It's always a matter of what's good enough for most of us, me included. Historically the best musicians have gone the extra mile to get extraordinary so I think we will eventually get there, just no time real soon.

[mistertoast]

True. I think the question really goes to the heart of what an instrument is. There's such a range of guitars, pianos, etc. But we want the leading edge of a filter sweep from an 80s synth to sound just so.

Kinda thinks our music is relying more on gimmicks than song composition.

[c_huelsbeck]

The yousendit.com download limit on my comparison mp3 was reached, so I re-uploaded:

http://download.yousendit.com/B4F2F32E5571DBBA

[mistertoast]

I love that comparison file. I can hear differences, but if you mixed them up, I'm not sure I could identify them all correctly.

[tomg]

>>You are aware that sample rate only allows state changes at a fixed frequency.

But todays computers are fast enough, with good programming, to calculate many intermediate samples. Enough to push the quality above CD and the audio equipment the music is played on, as well as most human ears.

The trick is how to model what goes into those samples. It's really hard. As others have pointed out, the more variability in the original instrument, the easier it is to get close to the real thing.

It's nothing that time and money won't fix.... eventually.

[mistertoast]

>>It's nothing that time and money won't fix.... eventually.

Yes, and I bet if there were more money in this industry, the time would come sooner.

[mistertoast]

Speaking of money, I sometimes get blue about how developers in this biz are always strapped for cash. But then I read about the dev teams for classic hardware synths, and the stories are just as bad. Hardware or software, it's always been a lousy business.