The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Synths for sale  Search Gearslutz Go Advanced
Why does older analog synths sounds better than newer? Keyboard Synthesizers
Old 28th July 2014
  #211
Quote:
Originally Posted by Sharp11 View Post
I provided you a link, did you read it?

Yes, you created a test that was relatively bias free, except for the fact that it told listeners there would be a difference and pointed them to where those differences would be - then came a slew of CB-related posts to try and discredit the result.

Your CB comes in the form of believing those who disagree with your particular theory have "little experience" without any evidence to support this claim (and at the same time apparently being unwilling or unable to understand the statistical significance of your own test result), then you go off into anecdotal land to further try and confirm your bias, a sure example of logical fallacy.
Re the link yes of course I read it (despite being well aware of what it is) the question is how did you read and interpret it?

Yes I quite understand how you think Im now confirming the bias of my apparent 'belief'. May I ask how do you know that belief is a result of confirmation bias? You dont - unless you have ammassed a broad spectrum of evidence showing my lack of objectivity? Creating an argument for something - does not automatically mean confirmation bias at work.

In terms of the results of my first test - and my apparent difficulty in understanding the statistical significance (which I actually didnt) - we will see. Again - your own bias has mislabelled me as not understanding the statistical significance - when in fact I did and commented on this. You confuse a need to further investigate a hypothesis - with a misunderstanding of statistics. I guess that suits your argument of me being biased. Isnt that what confirmation bias is?

Again - the difference is - Im actually investigating in the real world.
Old 28th July 2014
  #212
Registered User
Quote:
Originally Posted by archfrenemy View Post
It is the imperfections and flaws of those older synths that give them a unique and beautiful sound. The tend to be much more pitchy, steppy, slurred, harsh, mellow, grainy, distorted, washy, raw and drifty... That is why I love them. Flawed sounds are complex, musical and inspiring.
I've been reading these sorts of comments a lot lately. But all of those adjectives, really? I can buy a vintage synth on ebay and hear all of those flawed, complex, musical, and inspiring things?

I'm sorry, but with all due respect, I'm a bit skeptical about how pronounced these things really are, and to what extent they're in the ears of the beholder. And I know that I am not alone.

Other than a bit of oscillator drift (which Urs Heckmann has modeled nicely into his Diva synth), I'm not sure about these other "flaws" or the "musical and inspiring" mojo that these machines are supposed to have.

If, as a player, you feel these things while sitting in front of a vintage machine and playing it, great, but as to how many of these characteristics are actually audible for your listeners to hear, well, I remain skeptical.
Old 28th July 2014
  #213
Lives for gear
 
kirkelein's Avatar
 

I don't view them as flawed at all. It's all electrical components behaving in a non-linear fashion, which is always good in anything pertaining to beauty. Trees are beautiful and cliffs and crashing waves. Bird songs and human faces; pelages and plumages. It's all complex, non-linear, fractal stuff. Now it may have been an accident that the old components had more of this, but it's certainly something they were aware of when designing these instruments, and I'd wager were quite happy about too. Today the components used are more geared toward math and less toward music, so you generally get more sterile results from your circuitry, unless you want to spend big bucks on NOS parts, discrete components, or build your own chip foundry.

Higher tolerance for example, is supposedly good for music, but bad for computers. Vintage synths are flawed as computers, but since they're not actually computers but indeed musical instruments, they're not flawed at all (in that regard).
Old 28th July 2014
  #214
Lives for gear
Quote:
Originally Posted by kirkelein View Post
I don't view them as flawed at all. It's all electrical components behaving in a non-linear fashion, which is always good in anything pertaining to beauty. Trees are beautiful and cliffs and crashing waves. Bird songs and human faces; pelages and plumages. It's all complex, non-linear, fractal stuff. Now it may have been an accident that the old components had more of this, but it's certainly something they were aware of when designing these instruments, and I'd wager were quite happy about too. Today the components used are more geared toward math and less toward music, so you generally get more sterile results from your circuitry, unless you want to spend big bucks on NOS parts, discrete components, or build your own chip foundry.

Higher tolerance for example, is supposedly good for music, but bad for computers. Vintage synths are flawed as computers, but since they're not actually computers but indeed musical instruments, they're not flawed at all (in that regard).
You are correct, but there are efforts to keep things more musical. You can build Mutable Instruments Shruthi-1 by yourself, use a filter design of your liking and cheap parts from your local Radio Shack. That will be the start of your journey to analog sound design.
Old 28th July 2014
  #215
Lives for gear
 
kirkelein's Avatar
 

Quote:
Originally Posted by coffee View Post
You are correct, but there are efforts to keep things more musical. You can build Mutable Instruments Shruthi-1 by yourself, use a filter design of your liking and cheap parts from your local Radio Shack. That will be the start of your journey to analog sound design.
I'm totally in favour of DIY stuff but I wish there were more poly options out there. I'd probably have bought an Ambika already if it just had more pots on it, but maybe some day. I would like to add some wavetables to my toolbox.
Old 28th July 2014
  #216
Registered User
So I've got an app for my iPhone that makes my MP3s sound like they're being played on an old record player. Very convincing.

I've got another app that makes my movies look like they were shot on 8mm film, and several $0.99 apps that can make my photos look like Polaroids from the 70's or truly vintage pics from the early 1900s.

Where's the Vintage Synth Mojo App? That will take a recording of my Moog LP and give it all that mojo that a Model D has? That will take a Jupiter-80 and make it sound like a Jupiter 8? Or whatever you believe Diva is "missing" from an OB8, where's the app or VST that will "fix" that?

And I'm not talking about "well you can take such and such EQ and set it to this and get this filter VST and set it to that and and then boost the thingamajiggy using plugin number 5,241..."

I'm talking about a bona fide, "make your modern synth sound like a vintage synth" app. Or VST.
Old 28th July 2014
  #217
Lives for gear
 
D_Davis's Avatar
 

I use plug ins to add wow and flutter and tape his to my recordings.
Old 28th July 2014
  #218
Lives for gear
 
archfrenemy's Avatar
 

Quote:
Originally Posted by keybdwizrd View Post
I've been reading these sorts of comments a lot lately. But all of those adjectives, really? I can buy a vintage synth on ebay and hear all of those flawed, complex, musical, and inspiring things?

I'm sorry, but with all due respect, I'm a bit skeptical about how pronounced these things really are, and to what extent they're in the ears of the beholder. And I know that I am not alone.

Other than a bit of oscillator drift (which Urs Heckmann has modeled nicely into his Diva synth), I'm not sure about these other "flaws" or the "musical and inspiring" mojo that these machines are supposed to have.

If, as a player, you feel these things while sitting in front of a vintage machine and playing it, great, but as to how many of these characteristics are actually audible for your listeners to hear, well, I remain skeptical.
My vintage synths of choice are older digital controlled analog synths. They are different animals due to slow processors, simple or bad programming, digital artifacts, hackability and can be pretty amazing with the right Sysex / midi controllers. My current vintage synths are the SCI Sixtrak, SCI Prophet 600 and JX-8P. Now if you are trying to do a saxophone solo on one, then you will never hear any of those adjectives from my earlier post... However, when you push those filters they tend to step and break up into harmonics instead of have the standard smooth filter close. That is caused mostly by the slower processor having the stepped response to internal waveforms and parameter changes. I know because I tried the the new processor upgrade on the Prophet 600 and it lost all of that beautiful lofi filter distortion. I went back to the old processor immediately. We are not talking about a subtle difference in sound. That sound is unique, musical and very hard to recreate via effects.

Also, my use of the word "flawed" was not intended to be negative. I tend to love music and performances for their flaws. I am bored to death by the bulk of today's mainstream perfectly edited, processed, pitch corrected and over EQed polished crap. 5 minutes or less on song writing / meaning (other than a possible discussion on how to best pander to a stereotypical portion of the population) and then days of intense turd polishing. Just deliver some authenticity, flaws and all.
Old 28th July 2014
  #219
Lives for gear
 
xanax's Avatar
Quote:
Originally Posted by keybdwizrd View Post
Where's the Vintage Synth Mojo App? That will take a recording of my Moog LP and give it all that mojo that a Model D has? That will take a Jupiter-80 and make it sound like a Jupiter 8? Or whatever you believe Diva is "missing" from an OB8, where's the app or VST that will "fix" that?

And I'm not talking about "well you can take such and such EQ and set it to this and get this filter VST and set it to that and and then boost the thingamajiggy using plugin number 5,241..."

I'm talking about a bona fide, "make your modern synth sound like a vintage synth" app. Or VST.
well that's sort of what latest gen plugs that do component modelling are for such as Diva.. UAD have been doing it for years in the sound processing processing domain going through great lengths to emulate every component in hardware circuits.. Roland's Aira line is probably the latest in vintage hardware emulation, with their controversial 'Analog Circuit Behavior' technology: AIRA — Analog Circuit Behavior - YouTube
Old 28th July 2014
  #220
Lives for gear
 
pinkerton's Avatar
 

To really answer this question you'd have to define "better" in an objective sense, without synonyms. Then you could end this easily.
Old 28th July 2014
  #221
Gear Guru
I've heard many Juno 6's and many of them didn't sound the same. All the Juno 60's I've played sounded the same.

Blahdunno why
Old 28th July 2014
  #222
Lives for gear
 
xanax's Avatar
a lot of same model analogs don't sound the same.. mostly due to internal revisions.. the slightest change of a component can alter the sound.. but also due to age & calibration issues..
Old 28th July 2014
  #223
Lives for gear
 

Quote:
Originally Posted by xanax View Post
well that's sort of what latest gen plugs that do component modelling are for such as Diva..
Does Diva actually do what you think component modelling is?
Old 28th July 2014
  #224
Gear Guru
Quote:
Originally Posted by xanax View Post
a lot of same model analogs don't sound the same.. mostly due to internal revisions.. the slightest change of a component can alter the sound.. but also due to age & calibration issues..
I know ... But I've never found it more drastic than the Juno 6. I'm guessing something goes wrong with the VCA
Old 28th July 2014
  #225
Lives for gear
 
kirkelein's Avatar
 

I was sad to hear my Juno-60 didn't sound as crisp with the chorus on as did my Juno-6 which I sold for the former. Noise was less prevalent, confusingly. But somehow that's a bit charming. Like your cat ruining a lamp because you ignored him. It's bad but it's kind of good.
Old 28th July 2014
  #226
Lives for gear
 
maisonvague's Avatar
 

Quote:
Originally Posted by EDGEK8D View Post
We are a complicated bunch, people.
We are indeed!

I'm glad you agree we'd all go nuts if we tried to be objective at all times.

But wait -- wouldn't that give you more business if we did?

Thanks for sharing your stories!

Quote:
Originally Posted by archfrenemy View Post
That sound is unique, musical and very hard to recreate via effects.
I know what you mean, man. I love wonky old synths.
Old 28th July 2014
  #227
Quote:
Originally Posted by keybdwizrd View Post
I've been reading these sorts of comments a lot lately. But all of those adjectives, really? I can buy a vintage synth on ebay and hear all of those flawed, complex, musical, and inspiring things?

I'm sorry, but with all due respect, I'm a bit skeptical about how pronounced these things really are, and to what extent they're in the ears of the beholder. And I know that I am not alone.

Other than a bit of oscillator drift (which Urs Heckmann has modeled nicely into his Diva synth), I'm not sure about these other "flaws" or the "musical and inspiring" mojo that these machines are supposed to have.

If, as a player, you feel these things while sitting in front of a vintage machine and playing it, great, but as to how many of these characteristics are actually audible for your listeners to hear, well, I remain skeptical.
I can tell you that a Prophet VS has flaws that are clearly audible, and it's a hybrid synth. Those flaws, such as the aliasing, truly help make it wonderful and unique. Listen to the grit on the FilmStrings patch. One of my favorite sounds of all time.
Old 28th July 2014
  #228
Lives for gear
 
xanax's Avatar
Quote:
Originally Posted by SimonInAustralia View Post
Does Diva actually do what you think component modelling is?
your question sounds loaded.. this is what their website has to say:

The oscillators, filters and envelopes closely model components found in some of the great monophonic and polyphonic synthesizers of yesteryear. Modules can be mixed and matched so you can build hybrids, but what sets DIVA apart is the sheer authenticity of the analogue sound. This comes at the cost of quite a high CPU-hit, but we think it was worth it: Diva is the first native software synth that applies methods from industrial circuit simulators (e.g. PSpice) in realtime. The behaviour of zero-delay-feedback filters when pushed to the limit clearly demonstrates the advantages of this groundbreaking approach. Diva
Old 28th July 2014
  #229
Deleted User
Guest
Quote:
Originally Posted by kirkelein View Post
I was sad to hear my Juno-60 didn't sound as crisp with the chorus on as did my Juno-6 which I sold for the former. Noise was less prevalent, confusingly. But somehow that's a bit charming. Like your cat ruining a lamp because you ignored him. It's bad but it's kind of good.
If you unhappy with your Juno-60 I'm happy to buy it, I'm in Stockholm as well as I see you are. In that case, please send me a private pm.
Old 28th July 2014
  #230
Lives for gear
 

Quote:
Originally Posted by xanax View Post
your question sounds loaded.. this is what their website has to say:

The oscillators, filters and envelopes closely model components found in some of the great monophonic and polyphonic synthesizers of yesteryear. Modules can be mixed and matched so you can build hybrids, but what sets DIVA apart is the sheer authenticity of the analogue sound. This comes at the cost of quite a high CPU-hit, but we think it was worth it: Diva is the first native software synth that applies methods from industrial circuit simulators (e.g. PSpice) in realtime. The behaviour of zero-delay-feedback filters when pushed to the limit clearly demonstrates the advantages of this groundbreaking approach. Diva
Yes, loaded.

Are the components being modelled the actual electronic circuits and individual electronic components (capacitors/resistors/transistors/etc.), in a full realtime circuit simulation of each synth module?

Or are the components in their description the synth modules, and it is modelling the oscillator in full as a 'component', for example?

It uses the zero-delay-feedback filter method from realtime circuit simulators, but is it just that method, just the zero-delay filter method, that it uses, or is it performing a complete realtime electronic circuit/electronic component simulation?


A more comprehensive introduction from the Diva manual seems to have a lot of detail about specifically the zero-delay filter method, and not a lot about complete realtime circuit/individual component simulation, which many seem to believe that is what this sort of thing is doing, and the term 'component modelling' is vague enough for manufacturers to use it a little dishonestly/deceptively in their marketing...

Quote:
the spirit of analogue

Diva captures the spirit of various analogue synthesizers by letting the user select from a variety of alternative modules. The oscillators, filters and envelopes closely model components found in some of the greatest monophonic and polyphonic synthesizers of yesteryear.

But what sets DIVA apart from other emulations is the sheer authenticity of the analogue sound. This comes at the cost of quite a high CPU-hit, but we think it was worth it: Diva is the first native software synth that applies methods from industrial circuit simulators (e.g. PSpice) in realtime. The behaviour of our zero-delay-feedback filters when pushed to the limit clearly demonstrates the advantages of this groundbreaking approach.

about zero delay feedback filters

Most filters in old analogue synthesizers have one or more feedback paths. The input signal passes through a series of controlled resistors and condensers, and the output is fed back into the input (or to another point within the signal path). The classic Moog™ ladder filter has four simple lowpass filter stages and a resonance control that determines how much of the signal is fed back into the input. This feedback loop doesn’t introduce any delay, feedback is practically instantaneous...

Digital models try to reproduce this behaviour by calculating the result of applying four simple lowpass filters to an input sample. Feedback means repeating the current calculation using the results of the previous one.

While real circuitry can process a signal within a few nanoseconds, digital models calculate per sample, adding about a million times as much latency to each iteration. The cumulative effect of feedback latency in digital emulations is very noticeable. Oversampling and higher sample rates help reduce the latency, but conventional digital filters always “smear” at high resonance levels.

Methods of addressing the latency problem have been available for many years, but such implementations either don’t model the complex distortion inherent in real circuitry, or they aren’t suitable for realtime processing. Common to all methods is that they predict output values and use that prediction in the current calculation.

Of course we can’t see into the future either, but our routines are fast enough to deliver at least a few voices in realtime. We calculate the filter with a few test samples and look at the deviation between the prediction and the result, then we use that deviation to calculate a better prediction. The goal, of course, is to close the gap as quickly as possible!

Diva applies a classic trial-and-error principle but includes a rather intelligent way (or so we like to think) of learning from mistakes. As a rule of thumb, Diva’s filters only need to be calculated once or twice in succession. However, it can take up to 15 cycles if e.g. the resonance is very high and the input is very complex e.g. includes noise.
Maybe the developer has gone into more detail elsewhere about exactly what is being modelled, and what the level/depth of modelling is, but there doesn't seem to be very much specific detail in the marketing and manual about it.
Old 28th July 2014
  #231
Lives for gear
 

i think it has to do with thermal noise being pleasing to the ear.
Old 28th July 2014
  #232
Lives for gear
 

Quote:
Originally Posted by Analog Prophet View Post
Yes, Elements sounds great, and older synths are definitely not just nostalgia for all, in same way as a good grand piano or a gong from, say 1979 is nostalgia. Well, with time comes memory, that's true, and in that case is everything past nostalgia, such a book should be nostalgia just because it was written some decades ago and have a special vibe that appeal some. I don't call that just nostalgia, but simply 'a good book'. Good instrument are good instruments regardless if the are from 2014 or from the early days of synths and I simply consider them as 'good instruments' if they are good instruments and appeal to me.
I think this gets rather close to an important factor: With old synths we basically just remember a "best of" collection. The bad to average synths are pretty much forgotten.

Add to this, that due to having a limited amount of parameters and a lot of time spent by a lot of users on each of the "classic" synths it is very well known what strengths those synths have. You just pick the "right" synth for a specific sound. You will not do pads on a 303...

Another major difference between many older synths and many newer synths is the design approach. A lot of older synths (e.g. the Minimoog) where actually not really great engineering, but more like dirty hacks, that where tweaked till they sounded right. Classic example is component selection "by ear". Once you get hundreds of parameters with wide parameter ranges that approach does not work any longer. You basically have to engineer your synths by simulation/small component tolerances/well defined measurements of components.

The "old" approach is always slower, more expensive and leads to a limited feature set. It might be worth the effort for some specific sounds, on the other hand there are also many new synths with huge sweetspots AND huge flexibility...
Old 28th July 2014
  #233
Lives for gear
 
antto's Avatar
SimonInAustralia: you don't have to fully model the analog components with "SPICE" level of detail, in order to emulate an analog filter
a SPICE model is overkill, because it gives you way too many details which you actually don't need at the end
all you really need from that filter is:
- input signal
- output signal
- cutoff frequency
- resonance amount

so, if Diva did spice-level component simulation - that would be awesom and also plain waste of cycles given that you only care about those 4 things i listed
you can get more or less the same things and same behaviour if you do some more work, and figure out how the circuit simplifies into the equations needed to get those 4 things only
and that's most likely what is done in Diva

Urs has been posting in-depth information at KVR if you're interested, you could search it
Old 28th July 2014
  #234
Deleted User
Guest
Well, I didn't like all older analog synths just because they more or less are considered vintage:

Never liked the Korg Poly 800 a no personality at all. The only thing it (it was not mine, but my friend's) was good to was to have as decoration at the rear of the vehicle that transposed us at our marriage.

When the 303 was new I didn't like it - sounded weirded. But times changes. I can't say I am a 303 lover, but still, it has something many other modern synths lacking.

Never liked Prophet VS (I know, it's not analog), but I liked my Prophet 2000 (P2000 and VS was like a duo when they were released) with nice and warm analog VCF.

I never liked the Juno-106 (had first one, sold and later gave it a new trial and bought another... and sold) - could never replace my former Juno-60. Simply to plastic sound.

I had mixed feelings for Poly-61


Regarding newer analog synths my Minitaur has some of that older sound that I like.
Old 28th July 2014
  #235
Lives for gear
 
antto's Avatar
well, i've also played with TPT/"ZDF" filters, i've put together a 4-pole ladder, the code looks kinda elegant, and there are no virtual transistors in it, nor resistors
there are a few clever equations and coefficients in them
if you could see the actual code that is executed during a SPICE simulation of that same filter - your head will explode
Old 28th July 2014
  #236
Quote:
Originally Posted by antto View Post
well, i've also played with TPT/"ZDF" filters, i've put together a 4-pole ladder, the code looks kinda elegant, and there are no virtual transistors in it, nor resistors
there are a few clever equations and coefficients in them
if you could see the actual code that is executed during a SPICE simulation of that same filter - your head will explode
I think Urs chooses specific components only with this intensive modelling simulation...interesting what you say about SPICE...however - simple elegant 4 pole code - - it might look elegant as code - but does it capture the characteristics of an analogue filter?
Old 28th July 2014
  #237
Lives for gear
 
antto's Avatar
define "the characteristics of an analogue filter" please
Old 28th July 2014
  #238
Gear Head
 

Quote:
Originally Posted by antto View Post
SimonInAustralia: you don't have to fully model the analog components with "SPICE" level of detail, in order to emulate an analog filter
a SPICE model is overkill, because it gives you way too many details which you actually don't need at the end
all you really need from that filter is:
- input signal
- output signal
- cutoff frequency
- resonance amount
That sounds very much like the simpletone logic that was behind the first generation vintage synth emulation plug ins. Im am sure that a 3rd generation emulation as diva has a more complex structure. What is about the differences of the single filter poles for example? In your simplified model there is no room for such details. And saturation fx..dc offests..crosstalk..powerlimiting fx ..and what else might play a role in the different sonic signatures filters have.

You basically claim that every 24 db filter sounds the same..But synths are not that simple.
Old 28th July 2014
  #239
Lives for gear
 
antto's Avatar
what i really meant is, that SPICE's purpose is to give you lots of details, similar to having a virtual circuit where you can probe and scope at any point, see power consumption and just test if your circuit works before wasting time ordering components and printing PCBs, it's a tool
so SPICE is detailed, yet it doesn't need to simulate individual electrons to be that accurate

you don't need to be able to probe the "signal" at any point on a VA synth, that's why you don't need all the hardcore details
not only does this cost tons of CPU usage, but it's just stupid and lazy
even if you actually used SPICE to simulate your filter - you would really only input a signal, provide the cutoff and resonance parameters, and then get the output
so you will discard most of the intermediate information that is part of the simulation
now, if you're practical, you would only care to get your algorithms to reproduce the output signal, for which you would use the equations that spice uses, but simplify them as much as possible (because most things are really simple) given that you don't need to generate all the details that spice generates besides the output signal
so, you would end up with some tight equations that reproduce the output equivalently to the spice simulation, and it will not eat that much CPU

so just like SPICE doesn't have to simulate the individual electrons in order to be accurate enough - Diva (or another VA/emulation) doesn't have to simulate all components with "SPICE" level of detail in order to produce the authentic sound and behaviour at the end
Old 28th July 2014
  #240
Lives for gear
 
antto's Avatar
Quote:
Originally Posted by betterbox
That sounds very much like the simpletone logic that was behind the first generation vintage synth emulation plug ins. Im am sure that a 3rd generation emulation as diva has a more complex structure. What is about the differences of the single filter poles for example? In your simplified model there is no room for such details.
what about the single filter poles in my "model"?
you know, code is just code, it does only what you tell it to
so if there is code which "adds details" around the poles - there will be

Quote:
And saturation fx..dc offests..crosstalk..powerlimiting fx ..and what else might play a role in the different sonic signatures filters have.
oh, okay, see, i don't have a moog or a moog filter (or any similar analog 4-pole ladder filter) so i wasn't trying to emulate anything specific
i was only trying out TPT/ZDF trying to learn it and see how it works/sounds/behaves
do you realize that you don't have to have a whole synthesizer model (with a virtual power supply circuit) in order to test a TPT/ZDF filter algorithm?

Quote:
You basically claim that every 24 db filter sounds the same..But synths are not that simple.
when did i said that? wtf
i don't even remember the last time i've written/said "24dB filter"
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump