Analog - VA shootout?

[coyote]

Inspired by a thread on SSS....

 

There have been B3 shootouts which included a number of clones vs. the real thing. In addition, someone once posted here a mini-Rhodes shootout which did likewise. Has anyone ever published results from an analog-VA shootout?

 

I'd like to see some blind A/B listener tests of well-programmed synths of both VA and analog persuasion. The test ought include some renditions of well-known synth passages AND some newly created items, with each passage programmed & played on both the analog and the VA. In theory, the person who says "analog sounds much better than VA" ought be able to listen to the first couple seconds of the first passage and immediately identify whether it's analog or digital, even before hearing the 'other' rendition of the same passage - and then accurately repeat that assessment with every passage played.

[Tusker]

Coyote:

 

Some time ago one of the guys on midiwall did a little version of this. He had short snippets on a nord lead and a technosaurus microcon for comparison I believe. The guessing was all over the place. The mixed results were then explained by the fact that they were using mp3's and that the snippets were short.

 

I think I can hear the difference between certain synths, given enough of a sound sample. But the difference between analog synths can be pretty big too.

 

Best,

 

Jerry

[eric]

Keyboard magazine did one of these back in 1999 as a cover story. They called it "Analog Heaven." This was in a bit earlier era of VAs, but did include the Nord Lead, Virus, AN1x, Novation, etc. compared to each other and then to a Prophet 5. Interesting comment was that the P5 had a random analog "punch," but was still "vanilla" sounding and most of the editors would have chosen one of the VAs over the P5 for their versatility.

 

There has not been a more updated VA shootout in Keyboard, other than one issue that had the Lead 2, JP8000 and AN1x all pitted against one another.

 

Regards,

Eric

[Bobro]

Originally posted by coyote:

In theory, the person who says "analog sounds much better than VA" ought be able to listen to the first couple seconds of the first passage and immediately identify whether it's analog or digital, even before hearing the 'other' rendition of the same passage - and then accurately repeat that assessment with every passage played.

What a ludicrous theory, sure you didn't mean it for the "humor in music" thread?

 

Analog fans are continually talking about sound in the mix, even to the point of knocking their own instruments as too "fat" in this area, and the feelings of movement, change, drift and "aliveness", which are usually not readily apparent in a few seconds of an isolated track.

 

-Bobro

[coyote]

The guy who posted about it on SSS wasn't listening to the synths "in the mix". Besides, if a track doesn't sit well in the mix you can usually eliminate frequencies via EQ til it does sit well - while a "too thin" synth cannot get beefed up enough via EQ to ever sound good.

 

My A/B test concept has nothing to do with a mix. Once you get to a mix, there are too many variables to accurately judge a single element of that mix. I'm just taking the statement "Virtual analog is cr@p compared to real analog", a sentiment I've heard voiced here many times, and proposing a way to test it (and test just the synths themselves, definitely not in a mix).

 

To be more specific, here are a few things I'd do... Prevent the listeners from seeing what synth is being played, or from comparing notes. Then on each synth, play ten seconds of:

a) a Jan Hammer-type lead

 

then on each (and in a different order than before):

b) the Emerson "Lucky Man" solo

 

then on each (and in a different order than before):

c) the "Tom Sawyer" intro & synth melody

 

etc. Have each listener record whether he thinks a passage is being played on analog or VA as it's being played. (Ideally, the test subjects are all the folks who have made a statement similar to "VA is cr@p".)

 

Originally posted by Bobro:

Originally posted by coyote:

In theory, the person who says "analog sounds much better than VA" ought be able to listen to the first couple seconds of the first passage and immediately identify whether it's analog or digital, even before hearing the 'other' rendition of the same passage - and then accurately repeat that assessment with every passage played.

What a ludicrous theory, sure you didn't mean it for the "humor in music" thread?

 

Analog fans are continually talking about sound in the mix, even to the point of knocking their own instruments as too "fat" in this area, and the feelings of movement, change, drift and "aliveness", which are usually not readily apparent in a few seconds of an isolated track.

 

-Bobro

[alfonso]

Judge from mp3?

...that's a bit our limit...

[Max Ventura]

Oh come on guys, I had TWO guys who really owned Minimoogs believe that a part I played on a JP 8000 was indeed a Mini, and I went as far as declaring it was one of the "R.A. Moog" models from early '71.

For that sake, I also had my guitar teacher (a renowned NY session player back then) swear the tone from my "Marshall 800 was wonderfully chunky" onstage...I was using a Roland JC-120 and a huge effects pedalboard; the Marshall stack was also there, but it belonged to the band next to us.

[Peake]

Quiet fellas, Bob Moog still has some of those awful Voyagers in stock and you don't want to spill the beans about them not being better than VA. Have some respect.

[Dave Bryce]

Originally posted by peake@pacificnet.net:

Quiet fellas, Bob Moog still has some of those awful Voyagers in stock and you don't want to spill the beans about them not being better than VA. Have some respect.

 

dB

[coyote]

I ain't saying anything against real analog. Who knows, maybe it does sound much much better! If so, there's nothing for its promoters to fear; the test could even be a great way for a vendor to promote his wares: "Four out of five listeners yadayadayada...."

[RABid]

What I would find interesting is not a blind test that tries to trick the user, but an upfront collection of recordings of various synchs. Something that would let us compare a Lucky Man lead across synchs along with many other patches. From my experience, Nords and Novations and other brands have a distinct sound but you cannot always compare them side by side. Instead of hiding those traits, make them obvious and turn it into a learning experience.

 

Robert

[Tusker]

Originally posted by Rabid:

What I would find interesting is not a blind test that tries to trick the user, but an upfront collection of recordings of various synchs. Something that would let us compare a Lucky Man lead across synchs along with many other patches. From my experience, Nords and Novations and other brands have a distinct sound but you cannot always compare them side by side. Instead of hiding those traits, make them obvious and turn it into a learning experience.

Id think we have the resources to do that in this community. Id love to have a Keyboard Corner compilation of a bach two part invention (or a jan hammer solo, or lucky man lead) on various synths .just for color. Wed have to discuss standardization versus optimization.

 

Id be willing to contribute a wave file or two.

[realtrance]

Any kind of A/B testing is nigh impossible unless it's done under a very controlled situation. Even very small differences in volume can cause perception of A being "better" than B.

 

The Analogue Heaven shootout, while a fun subjective romp (by people who should know! ), was not -- and didn't seem intended as -- any kind of A/B test.

 

My own feeling once and for all on this is simple: there have been times when I've used my A6 to play my Roland SH-32 (heresy! I know!) and forgotten briefly which one I was plugged into (no, not because of drugs). On the other hand, there are distinct sound things you can do with each kind of synth you can't really replicate on the other.

 

There's a narrow swathe of situations where the possibility of any kind of A/B comparison might be possible, but the uselessness of doing a comparison at that point is that that narrow swathe of situations probably isn't sufficient for anyone to use as a way of determining whether they should go VA or analogue.

 

It's ultimately a subjective decision, and I suggest we keep it that way. Meanwhile, it's always fun to do A/B tests, even if the results are difficult at best to really interpret.

 

rt

[clusterchord]

I think it depends a lot on the type of sound you use for comparison.

 

If you play me couple of slow chords on a standard polysynth sound in mid-hi to hi register while tweaking LPF and/or resonance thruout their range, its almost impossible not to differentiate VA from RA. Same goes for extreme bass register

with some juicy filter funk env going on.

 

anyway, it seems to me there are two issues getting mixed up here; first one is: can we recognize between the two?, and second one is: which one is BETTER?

 

Personally there is no BETTER for me, each excells in its own way, and is used for different things. (Presuming were talking top class RA and VA here).

 

So, to repeat the mantra said here over mil. times; to choose, go in the store, play, figure out what you really like and need, buy it.

[coyote]

If we cannot tell the diff between the two, how can we possibly begin to say which one is better?

 

Those who say "don't look at it as a virtual analog, look at it as a digital subtractive synth" are correct, but they miss the point. These things are designed and marketed as valid replacements for analog synths (in sound as well as function & appearance). It is certainly valid to expect them to sound and perform like analog synths.

 

"Making the differences obvious" instead of 'tricking' people is no test of the hypothesis at all. The statement was "VA sounds like cr@p compared to real analog". The only way to test that statement is a blind test - blind testing is no 'trick', it merely eliminates the prejudice. And someone pointed out that a diff in volume might prejudice the listener. True. Therefore, use a volume meter to verify initial volumes... each synth likely will react slightly differently as a passage is played. That is exactly what we're testing for; it's that kind of difference that may make one sound more 'analog' than another.

 

Of course, such a test will never squelch the argument. But it would be fun anyway, and might get some folks to open their ears - or perhaps rethink their prejudices

[realtrance]

I think it's pretty clear that VA does not "sound like crap" -- that's just some teenager with raging hormones on the line here.

 

So the only questions left to answer are, what are you looking for, how do you find it, and what do you do with it once you've got it?

 

rt