Diva vs Analogue - a real world test

[zerocrossing]

Anyway, that's why I feel I don't have the bias that a lot of people talk about. I own software because I like what it is, and I own hardware because of what it is and I think there is a difference. You'll hit me, but I think the biggest differences (still even after this test) are in analog synths, and in those the biggest differences seem to be where things get a bit overdriven, like in filter overdrive, or even analog distortion. If someone can show me a synth that sounds like a DSI using it's feedback, I'd love to hear about it. Monark sounds great, but it's very simple. I like the way things sound when they're breaking.

Any reference recording of what you're talking about, like the mentioned DSI feedback?

I don''t have anything handy but I believe it's gone over in this video:

It seems the video has just a little, but I think Monark's feedback is capable of pretty similar stuff (btw you have two different type choices there), especially if combined with filter resonance.

Yeah, it's similar, but the Prophet 12 is like a space shuttle compared to a Model D. Not that you'd want a space shuttle for all your transportation needs, but sometimes you do. Is there a softsynth with feedback that's as good as Monark's that's got the modulation options, oscillator morphing, FM and AM, waveshaping and effects of a Prophet 12. I don't think so. Monark doesn't even respond to aftertouch. The 12 has a 61 key keyboard with velocity and aftertouch, two pressure sensitive touch strips, assignable mod and pitch wheels and an almost 1:1 feature per knob interface.

If you see my comments around here and GS, you'll see that I'm a firm believer that if you're looking for an all digital synth you're best off ITB unless you can afford a Solaris... and even then you might be better off ITB. I guess if Diva can fool you into thinking you're listening to an OB-8, then ______ could fool you into thinking you're listening to a Solaris.

Anyway, I'm still of the mind that ITB is super great for many things. I recently compared Oddity 2 to the ARP filter on my ATC-X (which I love) and the software was scarily similar. Monark can sound great. (Both those instruments don't even respond to aftertouch. ) But, when you're talking about things with feedback, filter overdrive and analog distortion circuits, software can be less convincing or tax your CPU to death. If you notice the "real world test" didn't feature anything like that. I found this same difference in guitar amp simulators. In your basic clean sounds, they can be totally convincing, but when you really push things, they don't fare as well as the AxeFX or Kemper hardware devices. Don't get me wrong, I still use them on software synths and drums.

In the end, I still stick to my original stance, which is that working ITB is awesome, but you're missing out if you don't have at least one analog around... and I'm not talking a dusty old Matrix 1000 or the like, but something that really lets you push it. You can pick up a MoPho for $250 or so on eBay. It can really scream in a way that no software I've ever heard can.

[david.beholder]

I'm always amazed at how obvious sarcasm isn't obvious to everyone.

So do I especially when it's coming with farty clownade.

Speaking on topic, have you ever played OB8?

[aciddose]

Maybe the OP is a genius. He fooled 60% of us guinea pigs.

The test is equivalent to a coin toss, so you're only 10% off.

Fully within the error bounds expected given the sample size.

[aciddose]

If someone wants to really test this, set the envelope to zero.

Zero sustain, whatever.

Check the output signal of the analog synthesizer (preferably a true analog like an MS-20) vs. the output of a plug-in.

You should find that the analog synthesizer produces a signal, likely with crossover distortion and other properties associated with electronic circuits.

The plug-in however will likely produce awkward silence.

This is not a test of the subjective interpretation of two equivalent objects. This is an objective measurement.

[aciddose]

https://soundcloud.com/aciddose-1/crossover-distortion

VCA CV input = connected to ground. Equivalent in most cases to amplitude envelope sustain = 0.

Gain = +60db.

These effects such as crossover distortion are present in a full-scale signal. They add harmonic content at ~(-60db) which is entirely audible especially in cases such as low frequency (50hz) filter self-oscillation.

Again, measure the two signals from the analog subtractive synthesizer as well as the software subtractive synthesizer.

Try to pass a pure sine signal through the signal path in order to measure with the greatest possible accuracy. (Cutoff = maximum, amplitude = unity.)

The software will not show this harmonic content at the same 1/n magnitude unless it is designed to emulate crossover distortion.

[aciddose]

https://en.wikipedia.org/wiki/Crossover_distortion

[braj]

This guy pulled up beside me today blasting this crazy new hip-hop track, I asked him if that synth bass was real analog or Diva, he couldn't understand what I was saying, so just flipped me the bird and drove off. I guess he didn't care.

[aciddose]

Your post strikes me as incredibly intelligent and wise.

[braj]

Is that sarcasm I smell? anyhow, if you have to take it into a lab to measure it, the difference doesn't matter. If one way or the other helps you actually make or play the music easier or better, then it seems the intelligent, wise thing to do is to use that, use what works depending on circumstance. Want to carry a studio in your backpack? Well, a Jupiter 8 is the wrong damn thing. Have a Jupiter 8 in the studio that sounds and feels boss to use? You would be dumb not to use it.

[Frantz]

I think I can live without cross over distortion being emulated.

[aciddose]

Do you need to take 8-bit audio samples "into the lab" for them to matter?

You're doing your masters in 8-bit now, right?

-60db = 10-bit.

All this is nothing more than a straw-man.

You're just changing the subject, creating a distraction.

What has any of this to do with the point I made?

My point is that in an objective comparison plug-ins and electronic circuits are typically nowhere near alike.

Another point is that in subjective comparisons the results are typically equivalent to a coin-toss.

Why/because I'll tell you why: The reason is that you don't have the ability to hear the difference and assess it subjectively in a reliable way even when it is relatively significant. Relatively significant as in the difference between 16-bit and 8-bit.

[aciddose]

Now, the intelligent part rather than just repeating things I've already said countless times.

This is the reason I gave the example I gave. One produces silence, the other produces a signal you can hear. The difference is plain-as-day, like the brilliant flash of light from an atomic bomb where even the blind smell the smoke from their singed eyeballs.

Denial that this effect (silence vs. signal) is significant is not possible. You must concede that in this situation there is absolutely no comparison.

The difference is as large as possible.

[braj]

It doesn't seem you get the point. The better option will remain relative, subjective, and different for everyone. Some people's music (via workflow, economy, portability, acess) may benefit more with analog or with VA, it all depends on circumstance, on context. I'm not gonna go into atomic bombs here though, LOL.

[Hank the Knife]

Maybe the OP is a genius. He fooled 60% of us guinea pigs.

The test is equivalent to a coin toss, so you're only 10% off.

Fully within the error bounds expected given the sample size.

Well, if I needed three months to make Diva patches sound better than its related OB patches alone, I really would have no time left for audio production, I assume. Besides: tout est relatif.

[aciddose]

This has nothing to do with how something "sounds" or "workflow" or any other subjective nonsense.

It is an objective, concrete proof that there is no objective, concrete comparison between real analog electronics and supposed software "models".

It is simple. Denying it will not make it any less true.

I would never bring up subjective issues because ...



... I am not nearly so intelligent as some of you seem to be.