Where the Best Analog Emulations Succeed & Where They Don't

For computer based music makers. Discussions about plug-ins and stand alone computer synth gear.
Naive Teen Idol
Supporting Member!
Supporting Member!
Posts: 196
Joined: Sun Apr 27, 2008 5:14 am
Gear: Roland Jupiter 6, Studio Electronics MIDIMini, DSI Evolver, Linn LM-2 Linndrum, Roland TR-77, Roland GR-300, Novation Remote SL 25

Where the Best Analog Emulations Succeed & Where They Don't

Post by Naive Teen Idol » Sun Jan 25, 2009 12:53 am

Watching William H's Sonic State review of the Prophet '08 last night, I noticed him (at about 7:45) going into some detail discussing how virtual analogs often fail when it comes to emulating self-oscillating filters ("Real analog filters have a certain juice to them"). It was an interesting point -- and fairly stated, as he said "most" emulations (not all) and actually provided an example of something he felt was difficult to replicate on VA's.

It got me thinking: what DO the best VA's still fail to replicate? And what do they get right today that they might not have five years ago? I know there's a lot of talk about pitch stability...but to be honest, that strikes me as a bit of a phony argument, as it's something the most rudimentary Reaktor programmer could address with a random LFO and a touch of amplitude. I got creamed (perhaps fairly) on the "Show Me the Difference Between a Moog and an ARP" thread for suggesting that I might understand the nature of a synth via a faithful, well-done emulation. But this review made me think a good, informed debate on this subject here could be instructive to a lot of folks.

User avatar
Yoozer
Expert Member
Expert Member
Posts: 1390
Joined: Sun Oct 09, 2005 1:31 pm

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by Yoozer » Sun Jan 25, 2009 1:46 am

Where they don't:
- phase stability, if you could call it that. I wish I still had the link to the video where a Moog 901 was hooked up to a scope and you saw the center of the waveform move back and forth like crazy.
- subtle distortion/drive. Pretty much all distortion is post-filter, post-VCA to be used as an insert effect and it sounds metallic and tinny.
- crazy modulation. Yes, the ubiquitous bass, lead and filter tweaking, I believe that. Go check back with me when Mini-style feedback works as it should, and when the ringmodulation + FM + sync at the same time sounds like it should.
- marketing. Stop blabbering about emulating down to the circuit board level; all you have to do is convince the most fundamentalist analog snob with golden ears that what they hear is actually a Mini.
- interfaces. What works in the physical world does not translate well to screen, so stop trying to meticulously render something that people in the know will see as fake anyway - and give your users the choice in interface - clear, readable, and using the mouse and keyboard to their fullest extent.
"Part of an instrument is what it can do, and part of it is what you do to it" - Suzanne Ciani, 197x.

User avatar
Carey M
Active Member
Active Member
Posts: 635
Joined: Mon Mar 13, 2006 2:08 pm
Gear: Akai Miniak, Fender Chroma Polaris, Moog Sub Phatty, Roland SH-101, Roland Jupiter-4, Sequential Circuits Prophet-600, Komplete 8, Reaper, iPad, beer
Band: Nightsatan
Location: Turku, Finland, EU
Contact:

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by Carey M » Sun Jan 25, 2009 1:59 am

What Yoozer said. And add oscillator hard sync to the list. All analogue emulations sound disappointing in this area, in my opinion, especially in the higher octaves.
Naive Teen Idol wrote:informed debate on this subject here could be instructive to a lot of folks.
:lol: What? :lol:

- CM

User avatar
cornutt
Moderator
Moderator
Posts: 2117
Joined: Sun Dec 02, 2007 6:03 am
Gear: 6th
Location: Rocket City USA
Contact:

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by cornutt » Sun Jan 25, 2009 2:55 am

Yoozer wrote: - phase stability, if you could call it that. I wish I still had the link to the video where a Moog 901 was hooked up to a scope and you saw the center of the waveform move back and forth like crazy.
I get your point, but I don't know that the 901 is a good example. It's an ancient design with no temp compensation whatsoever, and notoriously difficult to keep in tune.
- subtle distortion/drive. Pretty much all distortion is post-filter, post-VCA to be used as an insert effect and it sounds metallic and tinny.
An excellent point. A fair amount of what we associate with the sound of analog fiters, particularly on Moogs, is clipping that occurs either at the input to the filter, or in the resonance loop. There's too much of a tendency to think of distortion, or other waveshaping, as an add-on.
- crazy modulation. Yes, the ubiquitous bass, lead and filter tweaking, I believe that. Go check back with me when Mini-style feedback works as it should, and when the ringmodulation + FM + sync at the same time sounds like it should.
That is something that requires some fairly involved processing. Plus, to really do it right, the processing loop has to be running continuously, not just when a note is played. That's something that modular emulations in particular struggle with, the fact that you can create a patch on a real modular that will drone indefinitely. It makes the software eat up a lot of processor cycles because everything is running all the time, which is probably why most VAs don't do it. I think TimeWARP 2600 is one of the few that does (and that was probably because Pearlman insisted :mrgreen: ).
- marketing. Stop blabbering about emulating down to the circuit board level; all you have to do is convince the most fundamentalist analog snob with golden ears that what they hear is actually a Mini.
Good point, but the counter-argument is that a subjectivist will never admit that any emulation can approach the real thing, no matter what. So it becomes a pointless exercise.
- interfaces. What works in the physical world does not translate well to screen, so stop trying to meticulously render something that people in the know will see as fake anyway - and give your users the choice in interface - clear, readable, and using the mouse and keyboard to their fullest extent.
I'm glad someone besides me thinks this. OK, it was cute when the first classic-synth emulations appeared, and you edited patches by manipulating screen images of the hardware controls. But it's a very sub-optimal way of using a screen and a mouse. To pick a good example, I like a lot of the things that Absynth has done. It uses the screen as a screen, and not a pretend panel. Further, operating systems and desktop managers today have user interface standards intended to make it easier for the user to learn multiple applications, which VA programmers subvert when they create non-standard interfaces.
Switches, knobs, buttons, LEDs, LCD screens, monitors, keys, mice, jacks, sockets. Now two joysticks!

MitchK1989
Active Member
Active Member
Posts: 663
Joined: Mon Apr 24, 2006 1:46 am
Location: Nova Scotia

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by MitchK1989 » Sun Jan 25, 2009 5:38 am

I think the problem with most soft synths is in fact the whole lack of drive or inappropriate placement of drive in the signal chain, but there are softsynths that do drive well IMO.

I'd check out the following:

Schwa Olga: Quirky GUI that's a love/hate thing (you can get taped on english labels for GUI params by clicking on the white bowtie on the left) but the central drive knob introduces instabilities and saturation/clipping at multiple points in the signal chain at once that just adds an overall analog grime IMO.

ImpOSCar: The filter definitely drives. Definitely.

D-Cam Synthsquad: Not out yet. The new bundle from fxpansion deals with audio rate analog stuff in one of the three synths, they all seem to do better in the analog emulation domain in many ways compared to most, and amber is pretty interesting (actual emulation of divide down circuitry to create string machine esque tones)

I'm afraid those are the only actual analog modeled synths I've really tried (or in the case of D-Cam, researched/listened to tons of sound examples... On the site there's just usual song examples, but the DSP developer has been explaining lots of things and posting dry comparison examples and things on KVR for the past week or so)... I've never had much interest in the generic nord/virus style VA softsynths... Just stuff that goes for somewhat purist analog style quirky stuff with drive in the signal chain and whatnot and wavetable/fm/additive digital stuff...

Naive Teen Idol
Supporting Member!
Supporting Member!
Posts: 196
Joined: Sun Apr 27, 2008 5:14 am
Gear: Roland Jupiter 6, Studio Electronics MIDIMini, DSI Evolver, Linn LM-2 Linndrum, Roland TR-77, Roland GR-300, Novation Remote SL 25

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by Naive Teen Idol » Sun Jan 25, 2009 6:08 am

MitchK1989 wrote:I think the problem with most soft synths is in fact the whole lack of drive or inappropriate placement of drive in the signal chain, but there are softsynths that do drive well IMO.

I'd check out the following:
I'd also add Reason 4's Thor, which has a 9-mode shaper and also a ladder filter that allows you to place the drive before or after the self-oscillation feedback loop. It def. adds an element of "grit" you don't find in a lot of other softsynths.

Good answers!

User avatar
Yoozer
Expert Member
Expert Member
Posts: 1390
Joined: Sun Oct 09, 2005 1:31 pm

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by Yoozer » Sun Jan 25, 2009 2:38 pm

MitchK1989 wrote:Schwa Olga
I find that it's complete overkill in the distortion / drive department - pretty much impossible to coax anything warm and smooth out of it.
Naive Teen Idol wrote:I'd also add Reason 4's Thor, which has a 9-mode shaper
Yeah, you might want to compare that to the Virus B/C/TI shapers ;)
cornutt wrote:I get your point, but I don't know that the 901 is a good example. It's an ancient design with no temp compensation whatsoever, and notoriously difficult to keep in tune.
And it's exactly that what I want to see mimicked. It's easy enough to generate a perfect sawtooth with a band-limited impulse train; it's the imperfection that is the hard part.
That is something that requires some fairly involved processing.
If they claim that they've mimicked the circuit in software, it shouldn't. Let someone do 3 years of research to come up with something good; even if you have only a single voice of polyphony bringing a Core i7 to its knees, at least you'll have the correct code.
Good point, but the counter-argument is that a subjectivist will never admit that any emulation can approach the real thing, no matter what. So it becomes a pointless exercise.
Actually, it's not pointless; you just have to make 'm eat their own words and show them that they're full of c**p ;).

Premise 1: we record digitally in almost all cases
Premise 2: we have awesome A/D conversion available
Premise 3: recording a (legendary analog synth) with an expensive interface does not make the subjectivist frown, unless he's fond of tape, but chances are they can't be bothered with all the work.
Premise 4: there does not have to be a difference between a perfect emulation's series of bits stored to a harddisk and a series of bits that came from an awesome A/D converter which the original (fat, warm, creamy, juicy, more steak adjectives here) (legendary analog synth) was fed into; after all, we end up with a finite resolution, finite count of bits.
Premise 5: our ears are not perfect and are easily tricked; our sonic memory is not perfect.
Premise 6: I contend that a resolution of 24 bits and 192khz can be good enough. Like your eyes eventually stop at discerning detail at a distance, your ears stop (at a very well-defined point) in discerning frequencies.
Premise 7: our tactile memory is not perfect; e.g. our fingers won't care if there's a small difference in Newtons required to turn a knob on the original and to do the same on a rebuilt original.

Scenario: Construct the emulation. Put it in a case that resembles that of a vintage synthesizer so that you can't hear or see that there's different hardware inside. Connect high-resolution controls to said emulation. Let said subjectivist play on it and judge.

What's left then is similar to the DeBeers argument against synthetic diamonds; if the girl's not carrying the carbon equivalent of a dead African on her finger, it doesn't count, whine whine, sob sob.
I'm glad someone besides me thinks this.
I asked Arturia and they basically told me that the original interface already costed that much time and money to make that a clear-cut alternative wasn't possible, cost-wise. Furthermore, there's the marketing aspect of having a cute 3d render.
To pick a good example, I like a lot of the things that Absynth has done.

NI Massive and FAW Circle do an even better job in some cases :).
Further, operating systems and desktop managers today have user interface standards intended to make it easier for the user to learn multiple applications, which VA programmers subvert when they create non-standard interfaces.
Yes, but I don't think desktop interface standards should be forced to apply to synthesizers. First; desktop interfaces are supposed to be dull and looking like eachother. This is not a quality you want to see in a musical instrument. Second; the actual controls on a desktop are too limited and not aesthetically pleasing. Sliders in Windows are for quantized values, rotary knobs are not used at all. Using color to signal active/inactive is replaced by a checkbox. Tabs appear on top of a window, not somewhere in between.
"Part of an instrument is what it can do, and part of it is what you do to it" - Suzanne Ciani, 197x.

User avatar
Carey M
Active Member
Active Member
Posts: 635
Joined: Mon Mar 13, 2006 2:08 pm
Gear: Akai Miniak, Fender Chroma Polaris, Moog Sub Phatty, Roland SH-101, Roland Jupiter-4, Sequential Circuits Prophet-600, Komplete 8, Reaper, iPad, beer
Band: Nightsatan
Location: Turku, Finland, EU
Contact:

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by Carey M » Sun Jan 25, 2009 4:23 pm

Yoozer wrote:Actually, it's not pointless; you just have to make 'm eat their own words and show them that they're full of c**p ;)
I agree. And as a self-confessed analogz4life-idiot, I'm perfectly ready to admit that I'm full of c**p :D
Yoozer wrote:Premise 4: there does not have to be a difference between a perfect emulation's series of bits stored to a harddisk and a series of bits that came from an awesome A/D converter which the original (fat, warm, creamy, juicy, more steak adjectives here) (legendary analog synth) was fed into; after all, we end up with a finite resolution, finite count of bits.
Sadly this perfect emulation does not exist. At least not yet.

- CM

MitchK1989
Active Member
Active Member
Posts: 663
Joined: Mon Apr 24, 2006 1:46 am
Location: Nova Scotia

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by MitchK1989 » Sun Jan 25, 2009 9:19 pm

or maybe no one has stuck enough wood paneling and moog labels on it yet.

User avatar
Carey M
Active Member
Active Member
Posts: 635
Joined: Mon Mar 13, 2006 2:08 pm
Gear: Akai Miniak, Fender Chroma Polaris, Moog Sub Phatty, Roland SH-101, Roland Jupiter-4, Sequential Circuits Prophet-600, Komplete 8, Reaper, iPad, beer
Band: Nightsatan
Location: Turku, Finland, EU
Contact:

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by Carey M » Mon Jan 26, 2009 7:30 am

MitchK1989 wrote:or maybe no one has stuck enough wood paneling and moog labels on it yet.
I hate to use the roll-eyes emoticon...

:roll:

But hey, I'm open to suggestions... What current analogue emulations sound 100% like the real thing? I'd like to try them out. I prefer self-contained synths, but I'm even willing to try out a few VSTis.

- CM "none of my synths have wood paneling"

MitchK1989
Active Member
Active Member
Posts: 663
Joined: Mon Apr 24, 2006 1:46 am
Location: Nova Scotia

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by MitchK1989 » Tue Jan 27, 2009 1:03 am

Carey M wrote:
MitchK1989 wrote:or maybe no one has stuck enough wood paneling and moog labels on it yet.
I hate to use the roll-eyes emoticon...

:roll:

But hey, I'm open to suggestions... What current analogue emulations sound 100% like the real thing? I'd like to try them out. I prefer self-contained synths, but I'm even willing to try out a few VSTis.

- CM "none of my synths have wood paneling"
Oh, I'm not claiming they've reached 100% yet, but they're more than good enough to make kickass music with that up until recently more or less required analog gear (in 1996 making BoC style music entirely in the box was out of the question, now I reckon you could pull it off quite well if you were good at what you did, had a couple of really nice VA synths, and some tape emu plugins... Probably VTAPE, since it's the only one I've found that actually models wow and flutter properly instead of only modeling tape saturation and compression) so assuming making music is the end goal, the current crop of analog emulations is EXTREMELY good. I know olga is too much drive for some people, but that's partially the point. It's not supposed to sound like a moog or an oberheim. It's supposed to sound like an old half broken imported russian synth that never was full of grime and covered in stains.

I didn't mean to imply that subjectivity with regards to interface and appearance was 100% of the reason people prefer real analog to virtual... I think it's probably a 50/50 split though.

I wish someone would make a real analog with plugin integration like the virus TI... Or even an interfaceless box that acted as a DSP expansion card (like UAD or powercore) but was actually a bunch of analog oscillators and filters that can be used to run various synth and filter effect programs (the envelopes and lfos and such could be digital)... But I wouldn't fool myself into thinking it would make my music any better. It would just be a kickass fun thing to have.

User avatar
Carey M
Active Member
Active Member
Posts: 635
Joined: Mon Mar 13, 2006 2:08 pm
Gear: Akai Miniak, Fender Chroma Polaris, Moog Sub Phatty, Roland SH-101, Roland Jupiter-4, Sequential Circuits Prophet-600, Komplete 8, Reaper, iPad, beer
Band: Nightsatan
Location: Turku, Finland, EU
Contact:

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by Carey M » Tue Jan 27, 2009 7:19 am

MitchK1989 wrote:Oh, I'm not claiming they've reached 100% yet, but they're more than good enough to make kickass music
Even I'm not stupid enough to disagree with this :)
MitchK1989 wrote:I wish someone would make a real analog with plugin integration like the virus TI...
Well, you can't get USB-streaming audio yet, but reKon Audio makes a VST editor for Waldorf Pulse (a rather unexciting analogue for my tastes) MFB-Poly Lite (??) and soon one for a the Moog Little Phatty. You basically get total integration, sans USB-audio and FX-routing ;)

- CM

User avatar
shaft9000
Supporting Member!
Supporting Member!
Posts: 2042
Joined: Tue Nov 14, 2006 3:13 am
Real name: Dave
Gear: Whips chains waxes oils dildos DMT TNT the LHC, and a black rubber duckie
Band: moneymoneymoney
Location: VanNuys, CA USA
Contact:

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by shaft9000 » Tue Jan 27, 2009 9:20 am

MitchK1989 wrote:
I didn't mean to imply that subjectivity with regards to interface and appearance was 100% of the reason people prefer real analog to virtual... I think it's probably a 50/50 split though.

#-o

it's the SOUND, dude. Nothing more - The Sound. the split is 0/100.
it's not a complicated subject, that can somewhow be alleviated by 'modelling circuit behaviors' using digital code.

It's about the reality of the materials in the fuggin' instrument and that's it.

The VA's can do more and do it easier for less money. With all that variety in VA-land why would anyone need an analog synth?? :lol:
2600.solus.modcan a.eurorack.cs60.JP8.Juno6.A6.sunsyn.volcakeys.jd990.tb303.x0xb0x.revolution.
999.m1am1.RY30.svc350.memotron

shaft9000.muffwiggler.com <- singles & mixtape
shaft9000.bandcamp.com <- spacemusic album
youtube.com/shaft9000 <- various synth demos and studies

danbroad
Junior Member
Junior Member
Posts: 163
Joined: Mon Jan 19, 2009 8:26 am
Gear: TSC4 P12 NI9

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by danbroad » Tue Jan 27, 2009 9:54 am

Surely the perfect instrument is one which makes you want to play it, which inspires, which you have fun using. If that's a virtual emulation, a MicroKorg or a Minimoog, so be it.

I play for my hobby, not my job. I can't remember the last time VA or VSTi aliasing, phase issues or inaccurate filter cutoff characteristics affected the fun I had playing and composing. I can, however, remember becoming rather bored with the overall tonal variety I could coax out of my Moog, and selling it.
Less words more noise

MitchK1989
Active Member
Active Member
Posts: 663
Joined: Mon Apr 24, 2006 1:46 am
Location: Nova Scotia

Re: Where the Best Analog Emulations Succeed & Where They Don't

Post by MitchK1989 » Tue Jan 27, 2009 1:12 pm

http://www.vellocet.com/SynthSquadAudio ... gi_res.wav

Check that out and see if you can tell which one of the two sweeps is analog.

One is strobe (a softsynth by fxpansion that's part of the upcoming d-cam synth squad pack) and one is an SH-101.

Post Reply