VA Vs A

[zerocrossing]

yes, a few ....

Understatement of the year. Just search a bit on the internet, there are loads of them. Usually the outcome is that about 60-70% think the soft synth is the analog synth. Sometimes more.

Ok thanks. Good to know.
I recently had a demo of Novation's Bass Station 2 and thought it sounded quite thin and "digital". Go figure.

It depends on what you are doing. A lot of tests are of fairly static easy to reproduce sounds. In that case it is common to sound close enough that any differences are more pedantic than practical.

But start pushing things and modulating things (especially at audio rate), and then it is easier to hear the difference. Analogue does not alias. Digital when pushed does. Analogue by the nature of being analogue also has non-linear behaviors that are cpu intensive to reproduce in digital. This means you get unexpected little sweet spots and lovely tonal variations as you modulate stuff. Also, as you tweak stuff, you never have to worry about digital clipping.

There is something enjoyable about the liveliness of a flexible analogue synth. Once recorded into a song, especially more typical music, people may not hear the difference... but one may still enjoy and find inspiration in that difference in the playing.

^this is the correct answer.

[zerocrossing]

Honestly, I find the a/b comparisons to be the least helpful. I find it much more useful to just spend time with each instrument and kind of pay attention to my feelings about it. It's not scientific, but I usually quickly get a gut feeling about things pretty quickly.

[fmr]

yes, a few ....

Understatement of the year. Just search a bit on the internet, there are loads of them. Usually the outcome is that about 60-70% think the soft synth is the analog synth. Sometimes more.

Ok thanks. Good to know.
I recently had a demo of Novation's Bass Station 2 and thought it sounded quite thin and "digital". Go figure.

It depends on what you are doing. A lot of tests are of fairly static easy to reproduce sounds. In that case it is common to sound close enough that any differences are more pedantic than practical.

But start pushing things and modulating things (especially at audio rate), and then it is easier to hear the difference. Analogue does not alias. Digital when pushed does. Analogue by the nature of being analogue also has non-linear behaviors that are cpu intensive to reproduce in digital. This means you get unexpected little sweet spots and lovely tonal variations as you modulate stuff. Also, as you tweak stuff, you never have to worry about digital clipping.

There is something enjoyable about the liveliness of a flexible analogue synth. Once recorded into a song, especially more typical music, people may not hear the difference... but one may still enjoy and find inspiration in that difference in the playing.

^this is the correct answer.

I'd say this applies mainly (if not only) to modular systems, not regular synths. Besides, in the current state of the art of software synths, you will have a hard time finding one that does aliasing, even when tweaked to the extreme. And analogue has noise too, and other artifacts.

[Chapelle]

---

[thecontrolcentre]

I own several A's and a few VA's ... and I find that even the hardware analog monosynths (I own four) have a difference in their sound, never mind the VA's.

[rod_zero]

I don't think Arturia emulation sound that close enough, Diva, Xils, Lush and other do sound close enough to analog and they have won some A/B test vs analog synths on gearslutz.

At the end of the day is not about "faking it" or holding a piece of hardware. it is about using an instrument that inspires you and you are happy with the sound.

[trimph1]

I own several A's and a few VA's ... and I find that even the hardware analog monosynths (I own four) have a difference in their sound, never mind the VA's.

This.

Then there is the issue of the hand wired analogue synths as well. There is a difference even between one ARP2600 and another. Not a huge difference but it is discernable.

[dcfac73]

How would you scientifically prove it anyway?
A simple A/B audio example blind test wouldn't be sufficient.

If you put up example A from an analog synth and example B from a VA, and people prefer A, you can't conclude from that, that analog sounds better.

If you put up example C from an analog synth and example D from another analog synth, and people prefer C, would you conclude that analog sounds better than analog?

If you put up example E from an analog synth and example F from the same analog synth, and peopler prefer E, would you conclude that this particular analog synth sounds better than itself?

There are rules to conduct "scientific" studies, you know? Like in every other aspect that involves human perception and human sciences, you start by select a test universe that you find is broad enough and representative enough, and then conduct several tests, some of them are just checking tests, etc.

Anyway, if you'd want to compare an hardware synth with a software emulation, you would obviously choose the same patch played on both, like what was done in the past, for example with Arturia Jupiter-8 vs the real Jupiter-8, Tal U-NO-LX vs the Juno-60, or more recently the Matrix-12. Since one is the emulation of the other, this is a quite simple task.

Regarding your (somehow absurd) hypothesis, those would be only significant to test individual tastes and sound preferences, obviously. Would not be the ones chosen in an objective study.

So, to sum it up: You can just compare what is comparable, the rest is apples vs oranges.

Precisely.
I was actually after a test which compared like for like model of synth with like for like patches. I guess I should have asked this earlier.

Thanks for the replies, btw, and thanks for keeping it civil. I was expecting to be ridiculed

[pdxindy]

There are rules to conduct "scientific" studies, you know? Like in every other aspect that involves human perception and human sciences, you start by select a test universe that you find is broad enough and representative enough, and then conduct several tests, some of them are just checking tests, etc.

Anyway, if you'd want to compare an hardware synth with a software emulation, you would obviously choose the same patch played on both, like what was done in the past, for example with Arturia Jupiter-8 vs the real Jupiter-8, Tal U-NO-LX vs the Juno-60, or more recently the Matrix-12. Since one is the emulation of the other, this is a quite simple task.

Regarding your (somehow absurd) hypothesis, those would be only significant to test individual tastes and sound preferences, obviously. Would not be the ones chosen in an objective study.

So, to sum it up: You can just compare what is comparable, the rest is apples vs oranges.

Precisely.
I was actually after a test which compared like for like model of synth with like for like patches. I guess I should have asked this earlier.

Thanks for the replies, btw, and thanks for keeping it civil. I was expecting to be ridiculed

The emulation is always at a disadvantage. If someone had a VA and the task was to emulate it with an analogue synth, the analogue synth would also be at a disadvantage.

The original doesn't have to emulate anything, it just has to sound good. The emulation not only has to sound good, but it has to sound the same as the original which is harder.

Since in these comparisons, the VA is always the emulation, it is not a fair comparison of sound quality, because everyone focuses on the capability to copy something exactly which is a different task.

[Chapelle]

---

[dcfac73]

https://www.youtube.com/watch?v=N_n2ctb40B8
Just came across this for the Sem V. Closed my eyes and couldn't tell one from the other.
Is listening to Youtube audio sufficient rather than being in the same room when they're played? Who knows, but they sound close enough ,and , as many of the Youtube comments indicate, the difference probably wouldn't be noticed in a full mix.

Love this guy's video reviews btw.

There would probably be slight differences in the way those 2 authentic Sem's sound compared to each other as well. The Sem V would've been modelled after a different unit from the ones in this comparison video. I think a definitive comparison would be to ab blind test the software with the exact unit it was modelled on. But that would be nit-picking.

Here's another video comparing the Roland System 1 SH101 emulation from 12.15 onwards....

https://www.youtube.com/watch?v=3KAAcTS8GqI

[zerocrossing]

I own several A's and a few VA's ... and I find that even the hardware analog monosynths (I own four) have a difference in their sound, never mind the VA's.

I don't think anyone argues this. If all analog synths sounded the same, why would there be more than one?

However, I feel that there is a set of characteristics that an analog synth will exhibit that are not evident or as evident in virtual analogs. Of course, some get these better than others. I know, I know, "What are the characteristics?" Sadly, in my head I think of them as "warmth" and "organic" sounding, which I know is meaningless outside my own head.

What I can tell you is that I know it when I hear it. The best representation of what I'm talking about comes from my ATC-X QFS. In software, the best I think I've ever heard is the new Roland Sh-2 plug in. Diva and XILS Labs synths have it to varying degrees. My Virus Snow does not have it. Lately, I've been hearing demos of the Modulus .002 that has it in spades (I know it's a hybrid).

[zerocrossing]

Since in these comparisons, the VA is always the emulation, it is not a fair comparison of sound quality, because everyone focuses on the capability to copy something exactly which is a different task.

See, I totally disagree here. I have no burning need for an exact copy of anything and probably wouldn't know if it I heard it. The last classic analog synth I owned was a Juno 106 and it was at a time when we called "classic" " new." So... while I have a general idea of what it sounded like, there's no way I could, or would care to, have an exact representation, especially considering that most consider the Juno 60 to have sounded better.

But I digress. I think that outside a few ultra purists, an exact emulation isn't as important as getting a general good approximation of what we consider to be "analog" qualities. I think we know what these are, right? I mean, enough of us agree about what VAs are the best at sounding analog, right? It's not an accident most of us like Diva. It's not some lemming like urge is making the SH-2 plug in all the rage on Gearslutz right now. (me included even though I have no idea what a real SH-2 sounds like)

[jopy]

I'm a great fan of VA synths; I will never stop using Ace or Zebra. Having used an (analog) Roland JX10 back in the 90's, I really appreciate the depth of possibilities in modern VA units. Outside of huge rack modulars that cost $10k+, no analog synth can even approach the complexities of many VAs.

However, about a year ago I got a Prophet 08, and then a Boomstar 4075 soon after that. I never intended to use them to replace the VAs, but rather, just to use instead of a laptop in a live situation. I've had too many crashes and dropouts to trust a laptop live. My decision to go to analog had nothing to do with a belief that they would necessarily sound better, but was grounded more in practical issues. I could just have easily have gone with a Nord, but the Prophet actually spec'd out a little better for the price given my needs.

The strange thing is, I now almost never use the VA stuff for anything resembling analog synthesis even when home recording. The differences, as have been noted here, are in the act of performance, and the ways the tools respond when pushed and pulled around. This is certainly not to say that one is better than the other, but there is a different feel and it results in different music. I feel driven to play the Boomstar and the Prophet in a way that I never felt driven to play even my favorite VA synths. You can say it's all subjective or in my head, but I am behaving differently and I hear the same thing from nearly everyone I've known who buys analog gear after using ITB for a while.

So, I think the A-B tests are pretty misleading for all the reasons already mentioned earlier in the thread. They all assume you're already at a patch, that you are not manipulating in real time in response to other musicians as part of a performance, and that the sound does not need to be adjusted rapidly because you just figured out that the dynamics of the room are so boomy that you need to pull back on your decay times, or the drummer is louder than you remembered.

[dcfac73]

Just on a bit of a tangent-
Even many well known acts don't use analog on stage anymore (let alone the original synths used in their recordings)....

https://www.youtube.com/watch?v=wNnM2MMaUvo

https://www.youtube.com/watch?v=0Bu1fkgUM7s

https://www.youtube.com/watch?v=s5meDSitsd8

https://www.youtube.com/watch?v=MFUVylPVVMc

One fan managed to take this footage....

https://www.youtube.com/watch?v=a_7URJHwduE

Looks like they use off the shelf midi controllers,ipads, and software. Looks like they may be just playing back pre-recorded audio as well.

This bloke seems to be the exception. Bloody hell, that's almost porn....

https://www.youtube.com/watch?v=7eolvRvvLuU