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
  #241
Lives for gear
 
xanax's Avatar
I don't know how deep DIVA goes in SPICE level.. but another example is Cytomic which uses SPICE on The Drop (MS-20 filter emu) here is what they have to say about their circuit modelling:

Cytomic uses the same circuit modelling techniques that are used in professional circuit simulation packages but optimised for real time use. The structure including all feedback loops and useful non-linearities of the original circuit has been preserved. The algorithm uses actual values for resistors, capacitors, diodes, transistors and other circuit components. The voltages and currents are solved in real time and match as closely as possible to the analog circuit in both the time and frequency domains.

How is going into such detail stupid & lazy?.. and i doubt these guys would go in so deep if it didn't result in something more accurate..
Frankly your 4 step model approach sounds kind of primitive next to such advanced emulation study..
Old 28th July 2014
  #242
Lives for gear
 
antto's Avatar
i'm speaking in general
you don't have to know much about DSP to pick up spice and a .gif of the circuit diagram of an effect pedal, and throw it into SPICE
then you import your wav file and run DC simulation with "export to wav" sort of (you can do this in LTspice, i've done it)
but you see, this doesn't really help you, plus it eats loads of CPU and thus you wait, it usually can't be ran in realtime
so, i call this stupid, because you'll get your .wav at the end, but you won't learn anything, nor will you be able to use this in your DAW

i know Andy (Cytomic), in fact, he was the one who spent quite some time trying to teach me how to deal with the TPT maths
Andy is a slightly different story (as far as i know him)
he insists on using the actual names of the variables and equations as when dealing with analog circuit analysis
us DSP coders tend to put names like "x" for input and "y" for output
he really comes from the direction of circuit solving (so me speaking with him was hard)
he also builds the circuits in his lab, scopes them, and so on

so he really does very deep emulations of actual circuits (filters, etc..) and i bet most of you (including me) don't have the ears to appreciate the accuracy
i'm pretty sure he knows all the maths in order to write the same code that SPICE would use to generate all the detailed analysis in a simulation
see, this is what i meant with lazy - you'll be lazy if you don't optimize all that, because it's uber detailed
let me highlight the first sentence:
"Cytomic uses the same circuit modelling techniques that are used in professional circuit simulation packages but optimised for real time use."
see, it's optimized, as i said, loads of the complex things actually simplify (identical behaviour and everything, just lest math operations and also less CPU usage than SPICE)
spice is supposed to be able to simulate arbitrary circuits, so you get a penalty for that, you can't really optimize an "unknown" circuit too much
while, in emulating a known circuit - first of all, the circuit is known, and "fixed" (that means, the components won't magically change their order, neither will new components suddenly appear in the middle)
so you can optimize most things
a simple example is 10 resistors in series, SPICE will simulate all of them, but they are really equal to one big resistor at the end
you'd be lazy if you don't optimize that sort of thing
Old 28th July 2014
  #243
Lives for gear
 
kirkelein's Avatar
 

Quote:
Originally Posted by Analog Prophet View Post
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.
Sorry to disappoint but I still love it to bits. It's just the chorus that sounds a tiny bit muffled compared to the other one but even with that it's still highly useable. It'll be the last synth I ever part with
Old 28th July 2014
  #244
Deleted User
Guest
Quote:
Originally Posted by kirkelein View Post
Sorry to disappoint but I still love it to bits. It's just the chorus that sounds a tiny bit muffled compared to the other one but even with that it's still highly useable. It'll be the last synth I ever part with
Actually the Junos sounds best without that terrible chorus
Old 28th July 2014
  #245
Lives for gear
 
antto's Avatar
Urs and Andy aren't stupid nor lazy, quite the opposite, both of them are doing hard work for those algorithms to run in realtime and be as accurate as possible

i said stupid and lazy because some people (from the discussions about ACB) were really thinking that the white coats would have this magical uber fast CPU where they just throw a "circuit" and it then runs a realtime circuit simulation
no, that would be both stupid and lazy and it just doesn't work that way
the white coats did the same thing i am talking about here - analyze the circuit, stare at it, write code, test it, tweak it, rewrite it, test it again, and again, and so on
and they even did some things a little bit sloppy

so essentially, it's the good old art of making a model and deciding which things can be optimized, approximated, and so on, yet keeping the "important" aspects still there
and this varies from dev to dev
as you can see where there are different emulations of the same synth from different devs - each has slightly different things accentuated
and the quality varies too

Quote:
Frankly your 4 step model approach sounds kind of primitive next to such advanced emulation study..
it's not a 4-step model, i was listing which things you actually care about in the context of being able to use a theorethical filter. "use" not "analyse"
so you just care about those 4 things more or less, you don't care about the individual capacitors, or the biasing, or all the little atoms when you're using an analog filter, do you? you just input some signal, tweak the knobs, and you hear the magestic result
well, that's all.
to emulate this, whether you'll use some "equations" or full spice simulation, or hire scientists to do the maths by hand with paper and "B2" pencils, or you'll simulate the circuit at atomic level on a super computer - it doesn't matter, as long as you can get the same output given the same input and conditions - that means "if it sounds and responds the same"
all i was saying that, if you're realistic, you have to optimize it, and you may also have to make some compromises in order to get it to run in realtime at a low enough cpu usage
"3rd gen" VA is really about doing less compromises, and also using proper algos where it makes a difference

EDIT: and excuse me for the off topic
Old 29th July 2014
  #246
Lives for gear
 

Quote:
Originally Posted by Analog Prophet View Post
Regarding newer analog synths my Minitaur has some of that older sound that I like.
A lot of ppl feel this way about the minitaur (myself included). Which begs the question, if moog was able to recreate the vintage taurus oscillators so well with the minitaur (and for such an affordable price) why can't they do the same with the model d?

I think everyone would love to be able to get a clone of the model d oscillators in a $600 module format like the minitaur.
Old 29th July 2014
  #247
Lives for gear
 
Eigenwert's Avatar
Quote:
Originally Posted by D_Davis View Post
It's weird to me how so many people here seem to think that it is the instrument that provides the soul, organic nature, liveliness, inspiration, or dynamic sound, when, in fact, it is the musician.

There is no such thing as a bad instrument, only a bad musician.

A good musician can make great music using anything as a tool.

Some of the best music I ever made was made with a broken 4-track, a broken guitar, a $5 Radioshack microphone, and some shoes tumbling in a dryer and pots and pans for percussion.
You are a pure hero. The Chuck Norris of Gearslutz. However, to mere mortals there exists such a thing as a quality difference between tools. Those ultra cheap screwdrivers tend to break very easy. Of course not, if Chuck Norris uses them. A Minimoog will still make beautiful noises if someone who doesn't understand synthesis tries out what the functions do. Of course, if you were auditioning a Yamaha DJ-X it would sound just as good.

However, we are ordinary Gearslutz and - hey - this is Gearslutz! "We are the super musicians who still are stellar when using the ****tiest gear available" is another forum.
Old 29th July 2014
  #248
Lives for gear
 
robot gigante's Avatar
Quote:
Originally Posted by Eigenwert View Post
You are a pure hero. The Chuck Norris of Gearslutz. However, to mere mortals there exists such a thing as a quality difference between tools. Those ultra cheap screwdrivers tend to break very easy. Of course not, if Chuck Norris uses them. A Minimoog will still make beautiful noises if someone who doesn't understand synthesis tries out what the functions do. Of course, if you were auditioning a Yamaha DJ-X it would sound just as good.

However, we are ordinary Gearslutz and - hey - this is Gearslutz! "We are the super musicians who still are stellar when using the ****tiest gear available" is another forum.
The unholy offspring of Chuck Norris and MacGyver to be precise, but close enough.

Anyway, to be blunt, legendary vintage gear is legendary for a reason. I don't care if it's a U47, a Neve 1073, a JMI AC30 Top Boost, or a Minimoog. All attempts to say it's just subjective, just nostalgia, or whatever are just bull s h i t. Their track record is proven, and it's awesome, and if you can't hear why, maybe get into knitting, weight-lifting, Dungeons and Dragons, stamp collecting, whatever - the world is yours friends.

With modern gear, maaybe some of it will reach that status. Most of it won't.
Old 29th July 2014
  #249
Gear Head
 

Quote:
Originally Posted by antto View Post
all you really need from that filter is:
- input signal
- output signal
- cutoff frequency
- resonance amount
you dont mention any single poles here or any other details that add to the sound..

but sure, when you model the details the filter is not a simpletone one.
Old 29th July 2014
  #250
Lives for gear
 
antto's Avatar
Quote:
Originally Posted by betterbox
Quote:
Originally Posted by antto
all you really need from that filter is:
- input signal
- output signal
- cutoff frequency
- resonance amount
you dont mention any single poles here or any other details that add to the sound..

but sure, when you model the details the filter is not a simpletone one.
what poles? what details? what sound?
you're obviously thinking about some specific filter on some specific synth.
if you followed the conversation, i said those 4 things to SimonInAustralia, who was commenting on a quote from the u-he site related to the "ZDF" filters

in short, i wasn't talking about any specific filter, nor any specific synth
in fact, it doesn't have to be exactly those 4 things either, some filters don't have resonance, others have more than one output (SVFs), others may have two cutoff parameters, who knows

also, when i play with the analog synths i have - even tho i'm a sort of "nerd" - i don't really think about the individual poles, nor the powersupply, nor the little electrons
no, i just tweak the knobs and hear the sound, that's all
if the stuff that is inside that synth somehow changed from electronic components to bio-electric cells, or digital VA code/equations, or programmable alien bacteria (or whatever else) and it produces the same output in the same conditions - would i notice? would it matter?

i really don't think it matters what kind of "stuff" is behind the sound, as long as it sounds as intended.
Old 29th July 2014
  #251
Lives for gear
 
antto's Avatar
there are no "virtual transistors" in neither "TPT" nor "ZDF"
those two things are more about preserving the filter topology, and removing the unit sample delay in the feedback path
with those two things alone you can get a linear filter (a filter which doesn't have saturation)
so at that point - that's just elegant equations
for actual saturation - you have to add it on top <somehow>
Urs explains that he does it iteratively, and that's the usual way, which is used for such tasks
he also explains that with complex signals, or extreme conditions - it will take more iterations (thus more cpu usage)

TPT/ZDF also doesn't mean that there is any crosstalk, nor a virtual powersupply in the sense that audioconsult wants
i can't speak about Urs's code, as i haven't seen it, but i can speak about my own, if someone asks
but i feel this thread is not the right place to talk about these things
Old 29th July 2014
  #252
Gear Guru
 
Karloff70's Avatar
 

Quote:
Originally Posted by antto View Post
so he really does very deep emulations of actual circuits (filters, etc..) and i bet most of you (including me) don't have the ears to appreciate the accuracy
And I'll bet more people have the senses to tell the difference than you think. Including you.

I am so, so sick of this dumbing down of 'we can't tell the difference anyway'.
Old 29th July 2014
  #253
Lives for gear
 
antto's Avatar
well, for me, it was relatively recently, when i was still on the same side as the mortal users (few years ago)
as a mortal user who has no idea about the things in-depth, you imagine those differences (when speaking of analog vs virtual analog)
but once you go to the other side, and start modeling an actual analog synth, and start comparing your model - there sure are differences, but NOT where you imagined they were

i'm also lazy, and stupid
normally, on todays VAs you'll see bandlimited oscillators (or at least very decent antialiasing) ... there are all sorts of techniques, from mip-mapped tables to BLEP/BLIP/BLIT/whatnot
but in my "emulation" - i just generate a stupid simple sawtooth, that aliases with all of the possible digital glory, at 4x oversampling, and feed it to the "VCF"
so, yes, i decided to let the oscillator alias, because i didn't hear it as a problem, you could call me a sloppy dev, i am
other devs are probably sloppy in other places, some more, others less
it comes down to the result tho
if Diva sounds great - do you really care whether Urs was sloppy somewhere?

again, i'm speaking in general, i hope i don't offend anyone

specifically in Andy's case, i said that i can't appreciate the accuracy, because i have zero experience with those specific filters he's emulating (as well as the synths they come from)
i've heard demos from The Drop, and i've seen videos from the developement process in his lab
but i feel exactly like a mortal user again, because i have no clue where to listen for the differences
only when you start modeling a specific synth or circuit - you get an idea which are the weakest points in the model, and then the iterative process of rewriting, re-testing, rewriting... till the differences get smaller and smaller
and at that point, only you notice, mortal users don't really know where to look, they just hear the whole thing at the output

if you know how many months i've been running the same static waveforms and adjusting the code, hours, days..
that is because the model didn't have the proper equations, and there comes all the ugly approximations and cheats, you adjust one thing till the difference gets smaller but as you do this, the "cheat" bites you in the a$$ and another difference increases elsewhere, so when you can't get all the cheats to behave - you erase it and try something else, again

anyway
i think it's not too fair to compare or expect Diva to sound and behave exactly like some specific synth, especially if Diva doesn't have the modules from that synth
it's okay to compare whether it sounds analog, but that is already known - it does, if it has no digital artefacts, and if so many people got it wrong in the blind test
people should just use it as it was intended, i am sure it can do things beyond what analogs do, afaik it had models of digital oscillators too (jp8000 iirc)
Old 29th July 2014
  #254
Lives for gear
 
xanax's Avatar
I also find it a bit ironic that you out of all people claim "most of you (including me) don't have the ears to appreciate the accuracy" after posting at least a hundred posts in the AIRA thread over the "flawed" TB-3 filter envelope algo.. anyways this is getting way off-topic, perhaps better to take this discussion to the DIVA thread eh?
Old 29th July 2014
  #255
Lives for gear
 
antto's Avatar
because while i may have spent 4 years studying the 303, i have not much clue about most other existing synths (and their filters) in such depth, plus i'm in a noisy room, listening on cheap HIFI speakers ;P~
sorry for the offtopic
mods can erase my posts, i'll stfu now
Old 29th July 2014
  #256
Gear Head
 

Quote:
Originally Posted by Analog Prophet View Post
Actually the Junos sounds best without that terrible chorus
really? i always pity the poor jupiter owners because they have to live without the chorus.
Old 29th July 2014
  #257
Lives for gear
 
xanax's Avatar
^ agreed i love the BBD chorus of Juno/JX series

@antto: it just felt kinda condescending to assume that others here on GS don't have the ears to appreciate accuracy when a lot of us have also spent years/decades on a particular synth
Old 29th July 2014
  #258
Lives for gear
 

Bought something Vintage and something new.....both have oodles of analog goodness





Love this SH1000 just drips in character. Telemark will be arriving today.
Old 29th July 2014
  #259
Deleted User
Guest
Quote:
Originally Posted by Jesse381 View Post
A lot of ppl feel this way about the minitaur (myself included). Which begs the question, if moog was able to recreate the vintage taurus oscillators so well with the minitaur (and for such an affordable price) why can't they do the same with the model d?

I think everyone would love to be able to get a clone of the model d oscillators in a $600 module format like the minitaur.
I would order one in less then a second, even before discuss the purchase with my wife
Old 29th July 2014
  #260
Deleted User
Guest
Quote:
Originally Posted by betterbox View Post
really? i always pity the poor jupiter owners because they have to live without the chorus.
To clarify myself, I'm not talking of the later cheesy plastic Junos. But don't underestimate a Juno-60 without chorus. The factory presets are, as with many similar synths not something you sell your grandma for, but when program it well it is a bad cat with claws and attitude.
Old 29th July 2014
  #261
Lives for gear
 

There is definately some mojo from the older synths like my Odyssey mk3 and will be interesting to hear how the new Korg Odyssey holds up to it with SMT technology.

You can build mojo synths like the Oakley modular which I feel has mojo in heaps wheras the Doepfer sounds a bit modern to my ears.

The SH1000 I just bought needs recapping but I'm going to leave it as it is due the lovely sound it has. Compared to the 101 it sounds a lot more soulful. The Telemark V2 synth I also just bought has that analog sound that I love (from the youtube demo's) will see how it sounds in the real world when it arrives today.

My old Esq-1 hybrid sounds more authentic than some of todays new breed analog/analog-hybrids.
Old 29th July 2014
  #262
Deleted User
Guest
Quote:
Originally Posted by Captain Proton View Post
The SH1000 I just bought needs recapping but I'm going to leave it as it is due the lovely sound it has. Compared to the 101 it sounds a lot more soulful. The Telemark V2 synth I also just bought has that analog sound that I love (from the youtube demo's) will see how it sounds in the real world when it arrives today.
For me as electronic component novice, but lover and user of (some) older synth: What is recapping and how to know if the synth needs to be recapped?
Old 29th July 2014
  #263
Lives for gear
 
Eigenwert's Avatar
Quote:
Originally Posted by antto View Post
well, for me, it was relatively recently, when i was still on the same side as the mortal users (few years ago)
as a mortal user who has no idea about the things in-depth [...]
So you are suggesting you were not a mortal user anymore. That's your thing.
But suggesting education is not open to the mortal user is another thing.

When I read someone thinks he's not a mortal user and mortal users in general had have no idea about the things in-depth, I'll surely not read the rest of their post.
Old 29th July 2014
  #264
Lives for gear
 
antto's Avatar
Quote:
Originally Posted by Eigenwert
When I read someone thinks he's not a mortal user and mortal users in general had have no idea about the things in-depth, I'll surely not read the rest of their post.
i probably didn't used the right words, but i didn't say that i've stopped being a "mortal user"
you would have noticed that if you read the whole post, but you didn't
and when i was only a "user" and less a "coder" - i was very curious about how sound/synthesis works and all that, but no, i didn't know in-depth

Quote:
Originally Posted by Eigenwert
So you are suggesting you were not a mortal user anymore. That's your thing.
But suggesting education is not open to the mortal user is another thing.
just where have i said anything like that?

i'm on both sides, a hobbyist user (highly unfamiliar with most synths and music and etc..) and also a hobbyist coder (i write sloppy code, find most maths hard to understand, learn by making mistakes)

if you want to continue on this - let's use PM, just don't bend my words
Old 29th July 2014
  #265
Lives for gear
 
antto's Avatar
now, to say something on topic:
perhaps, circuit design knowlage and experience

when they designed the old analog synths years ago - they were more or less doing something new, not seen before
there was no internet where you could check how to make a good sawtooth oscillator

now, after that many years, and loads of experience in design (as well as research, improvement in knowlage and so on) there are a number of known techniques for achieving common things, now you can google "DIY good sawtooth VCO" and you'll probably find very decent information which you can just use directly

now, whether older synths sound better or not.. that's subjective IMO

EDIT: also, i think in the past, things were more seen as a whole, while today, things tend to be designed in sepparate (and distinct) parts.. "modules" or "sections"
Old 29th July 2014
  #266
Lives for gear
 
kirkelein's Avatar
 

Quote:
Originally Posted by Analog Prophet View Post
For me as electronic component novice, but lover and user of (some) older synth: What is recapping and how to know if the synth needs to be recapped?
It's the changing of capacitors, which help regulate the voltage levels within the unit. It's especially important in the power supply where a failing cap can make a synth impossible to calibrate or tune properly.

If it's an old unit it may well have caps that perform a little worse giving it a more unstable, 'organic' quality, but it isn't safe to assume. Many vintage units would see little to no difference after a recap. I'd wager Rolands in general hold up really well in this regard.

But there's a difference between aged and broken. Most people wouldn't bother changing caps until a unit starts acting up, or you can physically see that a specific cap is busted (the top of the cylinder bulges outward). But I can see how one could see it as beneficial to recap even if a unit's working properly if striving for a clean and stable sound.

But other people will have to speak about their experiences with this as I have never actually done it myself.
Old 29th July 2014
  #267
Lives for gear
 
Eigenwert's Avatar
Apologies to Antto for getting him wrong

However, the fact that sound is subjective should be clear to everyone
Old 29th July 2014
  #268
Lives for gear
 
Sharp11's Avatar
 

Quote:
Originally Posted by antto View Post

when they designed the old analog synths years ago - they were more or less doing something new, not seen before
there was no internet where you could check how to make a good sawtooth oscillator
The origins of the synthesizer date back to the late 1800's - that's a long time ago.

Do yourself a favor and study the history of audio before you begin pontificating - I'd suggest starting with the telephone, there are some real surprises there
Old 29th July 2014
  #269
Lives for gear
 
xanax's Avatar
Quote:
Originally Posted by antto View Post
now, to say something on topic:
perhaps, circuit design knowlage and experience

when they designed the old analog synths years ago - they were more or less doing something new, not seen before
there was no internet where you could check how to make a good sawtooth oscillator

now, after that many years, and loads of experience in design (as well as research, improvement in knowlage and so on) there are a number of known techniques for achieving common things, now you can google "DIY good sawtooth VCO" and you'll probably find very decent information which you can just use directly

now, whether older synths sound better or not.. that's subjective IMO

EDIT: also, i think in the past, things were more seen as a whole, while today, things tend to be designed in sepparate (and distinct) parts.. "modules" or "sections"
i gotta disagree with the above.. go do some research on Bob Moog.. he actually developed the VCO, ADSR etc.. as modules.. synths as you know them are the end result of already tons of development from modular design dating back from the 60's.. and synthesis work from even a lot longer ago (as stated just above) they may not have had the internet.. but they had better: school. Bob Moog, Dave Smith etc all have degrees in computer science, electrical engineering, physics..etc Bob did his early work at Columbia-Princeton Electronic Music Center & Dave presented early midi charts at the Audio Engineering Society.

I feel like today most synths are way too dumbed-down, everything is integrated & not many are really forward thinking or breaking new barriers (on the contrary trying to rehash 30 year old designs).. only in the modular world do i really see exciting things happening.. as for DIY yea it's cool to be able to google stuff & work on it in your basement but we're also one step closer to amateur hour..
Old 29th July 2014
  #270
Lives for gear
 
lovekrafty's Avatar
 

Quote:
Originally Posted by Captain Proton View Post
Bought something Vintage and something new.....both have oodles of analog goodness





Love this SH1000 just drips in character. Telemark will be arriving today.
The sh-1000 is a real charmer, lovely little synth
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