VA Vs A

[aciddose]

Yes I was half-way referring to the 106 voice daughter-boards without realizing the 60 doesn't use any of these at all.

The whole assembly though from that era is shoddy at best. The PCB can have significant issues over time as left-over flux paste eats away at pads, solder joints come loose and other issues occur. The circuits themselves aren't protected much if at all, the switches and potentiometers at this point should be almost unbearable to use unless the unit had been locked in an air-tight container and never touched...

I just have trouble thinking of what might make a significant difference other than a more severe failure of the circuit in these if variations can't be blamed on something like the daughter boards.

It may be as simple as needing to trim all the voices?

[ghettosynth]

I just have trouble thinking of what might make a significant difference other than a more severe failure of the circuit in these if variations can't be blamed on something like the daughter boards.

It may be as simple as needing to trim all the voices?

That's the majority of it right there. Beyond that it's more likely that differences have to do with the monitoring/recording signal chain.

That said, people use broken synths all the time and don't know it.

[fluffy_little_something]

I know what those classic synths sound like because I grew up when there was nothing else, yet. I know their sound from countless Fusion, Disco, and other songs and instrumentals. And I have heard the Minimoog in real life, but not that Sub 37 thingy as I said. So don't put words into my mouth

Your words.

Just had a look at the Minimoog panel (I know it was not used in the video), interestingly it has the same three pulse wave options (without width control) as Sylenth.

So, the fact that you aren't familiar enough with the mini to know the waveforms, AND, you think that simple feature alignment with sylenth is "interesting", tells more than you seem to realize.

Indeed, we seem to have a different definition of the verb to emulate. I have long abandoned that nerdy obsession with copying (which is what you seem to understand by emulating) stuff.

Again, you don't understand enough to know what you don't understand. I'm not, and never have, been talking about copying.

No, I have never had anything to do with the Sub 37 and I have not said otherwise. I only know it from videos and pictures, and I never said otherwise.

Indeed, I have never owned a Minimoog, so I am not familiar with the controls. And unlike to some people here it is nothing special to me, by modern standards it is just a primitive machine. I don't even like its sound character. I prefer the Oberheim sound.

Yes, indeed. Sylenth might as well have had 2 or 4 or 5 different pulse options, but it has the same three judging from the controls. Don't know if it's coincidence or not...

I focus on the main things, not on details that don't really matter to me. Certainly Sylenth is not as good as Diva or Xils stuff. But I bet that embedded in a song most people can't tell the difference between a good Sylenth (let alone Diva, Lush, etc.) sound and a hardware sound. They might notice a difference if they heard the very same song with just that one sound replaced, but even then they might not know which is which, not to mention that it would be irrelevant regarding the result, i.e. the music.

Emulation does not imply that the result is the very same as the original. Just think of the Arturia emulations, which differ quite a bit from the originals. Basically it is more about imitating or even just being inspired by.
There are rather few (attempted) strict emulations in the sense of copying all the details of the original.

[fluffy_little_something]

Again, of course Sylenth, Dune, etc. can emulate the sound a Minimoog makes. There is nothing voodoo about the Minimoog I am not saying it is a 100% identical, but why would it have to be?

So, I'm just going to tell you, that you are not using the word "emulate" in the same way that many of us, mostly those that disagree with you, are using the word. To you, it seems to mean "have a similar architecture to the original," i.e., that as long as it has a similar set or superset of controls to the original, and a similar set of features, e.g., both have a 24dB LPF, then it can emulate the original.

And even if they are a good emulation (Monark and Diva) there is still just a difference from the real thing. (Sylenth is not even in the galaxy)

The Moog just has an effortless oomph to the bass and the way the filter thumps when it jumps etc... there is a liveliness there that just puts a smile on the face... and that has nothing to do with being pedantic about perfect emulation.

Most people who have hardware synths also have a relatively powerful PA, combo etc. People who only use software tend to use smaller monitors, me included. I wonder what modern plugins sound like on a good combo On the headphones Sylenth for instance has a decent bass, but who knows what it would sound like on a subwoofer...
I really liked the bass of Arturia's SEM...

[aciddose]

Once you're talking about "sounds like" you've already jumped into the realm of subjectivity and there is no need for discussion as it will very likely be entirely unproductive.

[aciddose]

What I mean by this is that for example it doesn't matter which monitors/headphones/system I use, the timbre of whatever I play back remains the same. I don't notice any difference at all unless there is some problem such as distortion or similar.

Of course some systems will have more emphasis on high or low frequency, my hearing seems to adjust for this fairly well much like your eyes adjust depending upon the level of light. In a candle-lit room you can still see that an orange is orange and an apple is red (assuming they are of course...) and then looking at the same objects in very intense light I'm sure you'd observe the same colors, although you might notice smaller elements of the texture of each or observe more about the over-all shape due to the shadows they cast.

Still though, these things don't actually change. When we're talking about timbre differences between instruments, for myself at least I'm referring to something more akin to the difference between red and orange. Not blue and red, but red and orange, similar although a notable and distinct color. I would hope we're not focusing on minor details of the texture which would normally not be noticed - don't let yourself be mislead to believe that these are the only elements which differ. No, the difference is often more like an orange vs. a banana, then even someone blind can tell the difference by odor alone.

An analogy in subtractive synthesis might be the difference between a 12db and 24db low-pass filter. The same filter, not two different ones. So, only the slope of the filter is different while everything else is identical. This timbre difference would likely not be noticed in a large number of cases, but in some it is critical. I find a 12db low-pass is essential for some brass sounds (for example) while a 24db low-pass is essential for some flute or bass sounds.

The timbre differences I refer to are on the same scale at minimum and often far greater.

[pdxindy]

But I would need the settings from the Moog because I don't feel like experimenting for an hour.

With the Moog, you need a minute or two to make a good sound... You don't need an hour of experimenting to find a sweet spot. The Moog is like one big sweet spot and you turn a couple dials and then start playing. Couple quick adjustments and you have another beautiful sound.

These sorts of sounds are not complicated. If you have to experiment for an hour on your softsynth, that means you are trying to make up for a fundamental deficiency.

[aciddose]

Emulate:

1. obsolete; Ambitious or envious rivalry
2. Ambition or endeavor to equal or excel others (as in achievement)
3. Imitation

These are a good set of definitions to allow one to grasp the complete intended meaning of the word.

I use the word taking into account a few things:

1. If you have a rival you wish to compete with and an ambition to equal or exceed this rival, you would consider anything less an "unsuccessful" or "attempted" emulation and anything more a "successful" emulation.
2. Generally the goal of emulation is not to entirely replace a thing, but to equal it to the point that using both would be redundant.
3. In all examples of emulation there are cases in which both the original and the emulator are best suited to a particular task. For example in the case of synthesizers, I would never make an attempt to use a software synthesizer to replace analog hardware when I want to work in electronics or make modifications. I would never use an analog hardware monophonic synthesizer to quickly play a polyphonic part.

So, in these cases the majority of emulations of a particular synthesizer will fail when it comes to modifying the electronics themselves or using them with other electronics.

The majority will be of great benefit when I want to very quickly and with minimal effort listen to what a polyphonic version of a preset played back on a monophonic synthesizer would sound like.

[ghettosynth]

Once you're talking about "sounds like" you've already jumped into the realm of subjectivity and there is no need for discussion as it will very likely be entirely unproductive.

Exactly, a DX7 Moog Bass "sounded enough like" a Moog to convince quite a few people to sell their analogs.

[fmr]

One useful thing I can say about this is that although yes, there are minor differences due to component variations, they will never be on the scale you seem to be referring to.

The juno60 is famous for having very poorly constructed, unreliable packages for the filter circuits and other similar issues. I'm sure if you're an owner you know all about this. Small differences in moisture in the air in combination with faulty packages could lead to a wide variety of difficult to predict effects on the filter performance.

Am I the only one seeing a contradiction in here? For one, I never said anything about the scale of the diferences, nor that they are timbral (actually, some of them are related with the envelopes) and secondly, you start to say that the diferences are minimum and end saying that even small diferences in moisture in the air in combinantion with faulty packages could lead to a wide variety of difficult to predict effects on the filter performance.

The Juno-60 have some poorly components? They can be faulty? Maybe so, but what's up? If my unit doesn't sound like your unit, do I have to have mine serviced because yours is the one with the right sound? And what if I "like" the sound of mine, even if the components are faulty, as you say? What if I simply don't bother anymore to have the unit serviced (which would be difficult around here, anyway) as long as it plays?

How many analogues have been built that have some poorly components? I have two sound modules here that have problems due to poorly components and design flaws (yet they sounded good then, and still sound good now).

That's the quirks of analogues. And after all, it seems you are agreeing with me. And if I was going to create a perfect emulation based on my (implied faulty) unit, you would claim that emulation is not perfect, yet it was, except not when compared to your unit, but to mine).

If for nothing else, you just stated a reason why VAs are more convenient, and software even more.

[pdxindy]

Am I the only one seeing a contradiction in here?

It does not seem like a contradiction to me.

The idea is there that analogue is by nature having a significant variation between different synths of the same model. It seems to me he is saying that is not so. That there are some specific instances of poorer construction does not contradict that.

[fmr]

Am I the only one seeing a contradiction in here?

It does not seem like a contradiction to me.

The idea is there that analogue is by nature having a significant variation between different synths of the same model. It seems to me he is saying that is not so. That there are some specific instances of poorer construction does not contradict that.

The key here is what "significant" means. Since this is highly subjective, let's agree to disagree. But for people so picky finding diferences between hardware units and their software emulations, seems like you become much more tolerant when talking about diferences between two units of the same model.

[aciddose]

Significant in this case means measurable, beyond a few cents of tuning difference or a few db of amplitude or that sort of thing.

The majority of component variations will contribute insignificantly to the output signal in the majority of cases.

In an envelope it is possible that 30+ year old cheap quality electrolytic capacitors may have leaked or been subjected to damaging environments (heat, smoke, vibration) in addition to the expected five to ten year lifespan.

In those cases we can not count this as "component variation", this is "component failure".

"Component variation" would be small in-spec variations due to selection tolerances, generally 1%, 5%, 10%, 20%, 30%.

A variation of 5% in envelope time is easily countered by providing 5% less than the minimum desired time (1ms - 50us) and 5% more than the maximum (20s + 1s).

After this point you simply adjust the control by the amount of variation (maximum +/- 5%).

For example, given a fader length of 3", this would be an adjustment of 5/16", or 1/4" + 1/16th".

[aciddose]

Regarding whether software could "reproduce" this effect, although I'd ask why on earth would you want to... This is extremely simple. Any software could simply randomly adjust within some tolerance of preset parameters.

This seems backward and pointless to me however as you can get the same effect by slightly adjusting the parameters you input.

[fmr]

In those cases we can not count this as "component variation", this is "component failure".

"Component variation" would be small in-spec variations due to selection tolerances, generally 1%, 5%, 10%, 20%, 30%.

A variation of 5% in envelope time is easily countered by providing 5% less than the minimum desired time (1ms - 50us) and 5% more than the maximum (20s + 1s).

After this point you simply adjust the control by the amount of variation (maximum +/- 5%).

For example, given a fader length of 3", this would be an adjustment of 5/16", or 1/4" + 1/16th".

So, let's pick your exemple. One of the things I observed was exactly regarding the envelope and its effect on the filter. My Juno-60, admitedly, as never serviced, since it has been hardly used for many years now. Yet it still works OK, and plays, and the faders are behaving as normal as I can remember.

Going to your max tolerance of 30%, this would mean that the envelope has to be "significantly" diffferent on the software than it is on the hardware to give the same results. If this happens when it is all at zero (and since you can't go below that), there will be a noticeable difference. The same applies if the decay os the release are noticeably slower or faster in my unit, and lead to bigger or smaller decay times when near max, and I can't have those in the software because it was modelled after an envelope that behavews quite differently.

Yes, I could change the capacitors in my unit. But what if I like it the way it is? I would want the software to "emulate" that behaviour, because that's the one I want. Yet, it was the the "wrong" behaviour.

I hope I clarified my point now. What one sees as "faulty" can be seen by other as a desirable "feature" and that's where a software emulation cannot help, unless the programmer would allow for "tuning" the components (which is possible, of course, but not sure id desirable).

Anyway, I'm glad there are so many soft emulations. Many of the emulated synths are ones I could never pick anymore, and this allows me to have an idea of what it was.