Why don't new synths have 'that vintage sound?'

Discussions about anything analog, digital, MIDI, synth technology, techniques, theories and more.
User avatar
moremagic
Active Member
Active Member
Posts: 314
Joined: Mon Dec 28, 2009 9:29 am
Real name: A.M.
Gear: electribe er-1, sp-303, sp-404sx, s612, nuo 2, sl-1200 mk ii, beatstep, eurorack, tb-03, vx-99
Location: nc

Re: Why don't new synths have 'that vintage sound?'

Post by moremagic » Fri Jan 23, 2015 10:25 pm

upwind or downwind? :-p

User avatar
Stab Frenzy
Moderator
Moderator
Posts: 9723
Joined: Tue Jun 06, 2006 5:41 pm
Gear: Eurorack, RYTM, Ultranova, many FX
Location: monster island*
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by Stab Frenzy » Sat Jan 24, 2015 12:47 am

sneakthief wrote:HideawayStudio: you forgot to mention the importance of the power supply design, especially when it comes to ripple and sag. Recapping a psu may only affect a portion of the sound that specific psu's impart.
It depends, if there are caps in the signal path and an identical replacement is no longer available it can subtly change the tone of an instrument as well.

User avatar
calaverasgrande
Expert Member
Expert Member
Posts: 1038
Joined: Thu Jul 07, 2011 5:29 pm
Gear: MG1, MP201, MF101, MF102, Taurus 3, SH09, KPR-77, Streichfett, Dark Energy, X0Xb0x, Dronelab, Synsonics Drums, Machinedrum, Modular.
Band: N.S.V.
Location: New York
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by calaverasgrande » Sat Jan 24, 2015 1:08 am

Stab Frenzy wrote:
sneakthief wrote:HideawayStudio: you forgot to mention the importance of the power supply design, especially when it comes to ripple and sag. Recapping a psu may only affect a portion of the sound that specific psu's impart.
It depends, if there are caps in the signal path and an identical replacement is no longer available it can subtly change the tone of an instrument as well.
I was pretty ambivalent about the affect that power supplies have on gear. Volts is volts right?
Well my amplifier tech kind of gave me a lesson on tube amp power supplies.
According to him the main difference between Mesa Boogie amps and many other very similar amps that are based on Fender circuits is that they use multiple smaller caps in place of two large caps to achieve the same amount of rectified DC filtering. This means that Mesa's have very stiff power supplies which makes them almost sound solidstate compared to Marshall, Traynor, Orange etc.
In other words, on a Marshall you are deep cycling two large caps with the rectified AC. On a Mesa you are at most half cycling about a dozen smaller caps for the same amount of DC.
There are also other types of caps. A lot of folks swear the only difference between a 70's Musicman bass preamp and the current one is tantalum caps on the old version. I've only owned the newer ones of those so I can't testify.
Whatsoever thy hand findeth to do, do it with all thy might; for there is no work, nor device, nor knowledge, nor wisdom, in the grave

User avatar
Stab Frenzy
Moderator
Moderator
Posts: 9723
Joined: Tue Jun 06, 2006 5:41 pm
Gear: Eurorack, RYTM, Ultranova, many FX
Location: monster island*
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by Stab Frenzy » Sat Jan 24, 2015 2:22 am

The power supply is hugely important, the rest of the circuit is basically just there to modulate the power supply. :idea:

The company that I used to work for doesn't use solid state voltage regulators in their power supplies for two reasons; they're usually the first thing to die in a circuit and the owner/designer didn't like the way they made his circuits sound. I don't know if it was more of a difference with the high voltage valve circuits we were making than it would be with solid state, but the results were very good regardless.

User avatar
madmarkmagee
Active Member
Active Member
Posts: 416
Joined: Tue Feb 15, 2011 8:18 am
Gear: King Korg
ESQ-1
Reface DX
Microbrute
Shruthi-1
Pittsburgh System-1
Juno 106 (needs work)
Location: Sydney

Re: Why don't new synths have 'that vintage sound?'

Post by madmarkmagee » Sat Jan 24, 2015 10:58 am

shaft9000 wrote:There is none more vintage than.... grandpa farting in the wind
:lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol:

nvbrkr
Senior Member
Senior Member
Posts: 812
Joined: Fri May 09, 2008 8:14 pm
Gear: An electric piano and analog synths.
Location: Helsinki, Finland

Re: Why don't new synths have 'that vintage sound?'

Post by nvbrkr » Sat Jan 24, 2015 5:57 pm

GuyaGuy wrote:
VCO8 wrote:
More importantly, what creates that character?
Lack of treble and sub-par components distorting mostly.

And the way oscillators react to control voltage as opposed to quantized MIDI.
I'd listen to this man.

User avatar
Bitexion
Synth Explorer
Synth Explorer
Posts: 4230
Joined: Fri Apr 15, 2005 7:43 pm
Gear: Alesis Andromeda A6
Roland D-50
Creamware Minimax
Yamaha DX7s
Analogue Systems modular
Ensoniq SQ-80
Waldorf Blofeld
Location: Drammen, Norway

Re: Why don't new synths have 'that vintage sound?'

Post by Bitexion » Sat Jan 24, 2015 6:07 pm

Well here's a hint. They're not vintage yet.

User avatar
madtheory
Supporting Member!
Supporting Member!
Posts: 5155
Joined: Mon May 01, 2006 12:45 pm
Real name: Tomas Mulcahy
Gear: Flangebeast Mk1, Plonkotron, Morovdis Arpeggiator, Maplin My First EQ, Jeff Wayne Thunderchild rack, Thermostat, Buck Owens' Moog.
Band: Minim
Location: Cork, Ireland
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by madtheory » Sat Jan 24, 2015 7:40 pm

I think it's hilarious how folk who haven't built much, let alone designed something from scratch, are telling hideaway studio that he forgot about PSUs, then post a load of unsubstantiated mythology about valves. For two reasons- there are no valve based synths apart from a couple of very rare oddities, and second because this is a guy who has repaired and restored some of the best sounding synths ever made, and is trusted by people with very good ears and world class musical skill. His points are all accurate and well made, learn from them.

So fine, make your point about PSUs but you need to back it up with facts, not hearsay and over simplified explanations.

BTW the Minimoog oscillator has a 741 op amp at its heart. Which is an IC. Which proves you can't generalise about something as complex as a synth circuit- and that includes generalising about oscillator scaling, CV and MIDI.

The recapping thing is greatly exaggerated. Most folk can't do a blind before and after AB comparison. If you're doing it right you just replace the broken and unreliable ones to bring the machine back to as specified by the service notes, and any known issues in the field.

User avatar
pflosi
Synth Explorer
Synth Explorer
Posts: 3620
Joined: Tue Nov 13, 2007 6:14 pm
Gear: more than 150 characters...
Location: zürich
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by pflosi » Sat Jan 24, 2015 8:34 pm

madtheory wrote:I think it's hilarious how folk who haven't built much, let alone designed something from scratch, are telling hideaway studio that he forgot about PSUs, then post a load of unsubstantiated mythology about valves. For two reasons- there are no valve based synths apart from a couple of very rare oddities, and second because this is a guy who has repaired and restored some of the best sounding synths ever made, and is trusted by people with very good ears and world class musical skill. His points are all accurate and well made, learn from them.
Even funnier when you consider that the man has actually restored one of the very rarest and oddest of those valve based synths. :thumbleft: I'm still amazed by that work =D>

User avatar
madtheory
Supporting Member!
Supporting Member!
Posts: 5155
Joined: Mon May 01, 2006 12:45 pm
Real name: Tomas Mulcahy
Gear: Flangebeast Mk1, Plonkotron, Morovdis Arpeggiator, Maplin My First EQ, Jeff Wayne Thunderchild rack, Thermostat, Buck Owens' Moog.
Band: Minim
Location: Cork, Ireland
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by madtheory » Sat Jan 24, 2015 8:50 pm

LOL! Yup those are defo in the "very rare oddity" category!

User avatar
sneakthief
Newbie
Newbie
Posts: 78
Joined: Fri Jun 02, 2006 9:43 am
Location: Berlin
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by sneakthief » Sat Jan 24, 2015 11:03 pm

I have tons of respect for HideawayStudio. No question. For the sake of covering all angles in this thread, it seemed important however to bring up PSU design.

There's a lot of things that can happens as a PSU ages. As I mentioned, PSU voltage ripple & sag play a big role in sound and can affect frequency response (especially in the low end), transient response, noise, cross-talk/signal bleed & tuning stability. Needless to say, it's pretty obvious that any voltage instabilities can affect the VCO CV.

And to quote Mike Peake regarding some of the oldest synth PSU, namely the Moog Modular 910: "Some will understand that AC ripple and an unregulated -10V line and germanium power transistors, that 821 diode, etc. might be part of the beloved total."

Here's one random example: I was repairing a CS70m that started to warble as you played more notes - and the LFO would freak out. Holding down 6 notes caused massive distortion.

Scoping the PSU lines revealed a lot: the CS70m runs on +/-15V for the audio (typical), 5V for the massive 2-board CPU and +10V / -5V for the logic. Those last 2 rails would sag as you hit more keys and you could see the ripple increase all across the board. Furthermore, the logic voltages were being regulated down from the main +/-15V rails, so all the voltages were impacted. Hence the warbling & distortion. This is an obvious example of what happens when a PSU can't deliver the goods.

The first logical step was a full PSU recap on this 35-year old 30kg behemoth. Unfortunately that only helped a small amount, so either the transformer and regulators weren't able to deliver the same amount of current as before AND/OR the entire board was using more power as all its hundreds of caps aged.

When a client doesn't have the resources to pay for a full recap, the dilemma is whether to replace the whole PSU or supplement it. Not wanting to impact the sound, I decided to simply disconnect the +10/-5V rails and use a small Meanwell +/-15V switching PSU that could safely fit inside the case. Regulators and filtering brought those voltages down to the proper levels.

Everything worked perfectly now and the biggest change was that the low end completely opened up, thanks to the +/-15V rails being freed from the logic circuits.

I faced this kind of problem before when working on an MKS-80 that couldn't always power MPG-80, depending on the building that it was plugged in. I assume it had to do with the AC quality. Anyhow, the quick and dirty solution was to connect the MPG-80 to a 9V wallwart and not drag down the MKS-80's transformer. Musicians in Berlin often don't have a lot of money to pay for full synth restorations, so such improvisations are occasionally called for.

Once again, this had an effect on the low end frequency response of the MKS-80.

One final example: I chose to replace the PSU of an Arp Omni, simply because it was faster and more reliable than working on one that's known to catch fire. The owner and I sat down with it after the transplant (and ripping out all the horrible old tantalum capacitors) and did some comparisons between before and after recordings. I didn't expect much of a difference considering it's mainly a stringer, but the bass synth had a wider frequency response. Go figure.

Anyhow, that's my 2 cents.
madtheory wrote:I think it's hilarious how folk who haven't built much, let alone designed something from scratch, are telling hideaway studio that he forgot about PSUs,
Over the years I've repaired a wide range of synths - stuff like P5's (Rev 2 & 3), Synthex, OB-Xa, Xpander, CS70m, JP8, TVS, etc. These were my patients just in March:

http://sneak-thief.com/Synth_Hospital.J ... spital.JPG[/img]

FWIW I've built a ton of junk over the years... synths, drum machines, eq's, compressors, mixers, fx - you name it:

Image
sneak-thief: raw electrofunk!
http://sneak-thief.com

User avatar
madtheory
Supporting Member!
Supporting Member!
Posts: 5155
Joined: Mon May 01, 2006 12:45 pm
Real name: Tomas Mulcahy
Gear: Flangebeast Mk1, Plonkotron, Morovdis Arpeggiator, Maplin My First EQ, Jeff Wayne Thunderchild rack, Thermostat, Buck Owens' Moog.
Band: Minim
Location: Cork, Ireland
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by madtheory » Sun Jan 25, 2015 11:33 am

I really don't think that's the point here. You've given examples of PSUs that were not functioning to spec, whereas hideaway pointed out how the older stuff has noise and instability from several sources, even when in spec. In comparison a PSU is pretty straightforward it either works or it doesn't. So- it's a bit harder nowadays to have that "slop" as Dave Smith calls it because IC specs are so much better now. If anything makes the "vintage" sound, it's that. Minimoog osc being a great example because it uses the 741 which most budding EEs have struggled with.

User avatar
HideawayStudio
Supporting Member!
Supporting Member!
Posts: 1387
Joined: Fri Jan 09, 2009 10:41 pm
Real name: Dan Wilson
Gear: 163 tubes in a large wooden box!
Band: Shortwave
Location: UK
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by HideawayStudio » Sun Jan 25, 2015 12:03 pm

No worries Sneak - I know you far more than dabble with this stuff...

And many thanks for the kind words of support but easy guys.. lets not turn this into a slagging match...

There have been good points raised all around - PSU interactions included...

In fact that one is interesting and I've certainly experienced issues along these lines.

However - I first avoided this subject because in fact we've been able to design super smooth PSUs for many decades now and in fact the ripple/regulation specs on discrete PSUs in the Minimoog are well on a par with modern integrated linear PSUs. The difference is that its so very much easier to design them now than it was 35+ years ago. A well designed regulated supply will continue to run quite happily even with surprisingly tired capacitors as their very nature is to counteract ripple/load changes so really, in some ways, IMHO its much more about the quality of the original design in the first place.

However - where this can go horribly wrong is where the cabling, connectors and solder joints distributing the PSU rails are aging - this recently turned out to be the root cause of some pretty crazy tuning instability on what used to be Joy Division's ARP Quadra - the crimped receptacles in the PSU connectors to various pcbs were making such poor contact due to years of tarnishing that the slightest knock or sudden change in load drove everything crazy and eventually the CV DAC failed due to a resulting ground lift! The slightest increase in resistance in some of these rails outside of the voltage sensing loop in a voltage controlled synth is a problem for sure.

Perhaps the most interesting case is in the beast itself - It turns out a chunk of Novachord 346's eerie nature is due to a fair degree of mains hum "poisoning" the signals in the dividers. The ripple is fairly low but with 72 monostable tube dividers which are much more sensitive to jitter than later bistable circuits (the Novachord predates the bistable!) they interact in a harmonically related manner which renders an almost grainy/screechy character. The complex 7 rail regulated PSU was completely rebuilt but its worth remembering that there is less than 20uF of capacitance across the !entire! PSU rail distribution to over 144 tubes!! This is because large value caps simply didn't exist in 1938 and this makes the active noise cancelling PSU in the Novachord even more remarkable. Unfortunately I have long suspected it is not working quite as efficiently as it should due to running at 50Hz and the addition of safety grounding (Novachord wasn't earthed with a 75 year old "death cap" to chassis!!).

I've experienced very similar interactions with EMI from transformers too eg. in the Korg Sigma which, just like the Novachord, can be heard most markedly in the higher registers.

I do agree largely with Madtheory's stance on recapping. I think recapping has to be viewed with pragmatism. Speaking as a design engineer, one thing that is often overlooked/misunderstood is that in 80% of cases the value of caps in an electronic design are based rule of thumb/part availability/cost. In the vast majority of both analog and digital designs from the 80's and earlier it really doesn't matter if you're using 100N or 220N HF decoupling with 4u7,10,22 or even 100uF bulk capacitance at each supply decouple. In fact you will find some manufacturers stick to one set of values and another to their own. My point is that it often therefore would require a very significant change in capacitance before any tangible effects are perceived. The same is often the case for DC blocking in audio paths - a manufacture may choose to use 470N, 1uF, 2u2, 10u etc - and although there is some science behind this, again, you will find that more often than not its in fact cost/availability that is driving their choice - so again, in many cases don't expect even a 30% change in capacitance to have that much effect.

That said - more often now I am seeing signs of extreme aging in caps from the very era we, for rather too long, maybe took for granted - ie. the 80's and what with the number of furry green legged leaky/gooey caps I've encountered recently I think we are going to see a lot more bizarre behavior in our cherished synths in coming years.

But - despite all this - I am still firmly of the impression that a big chunk of what we attribute to the "vintage" sound in synths is down to op-amps (or their discrete counterparts) and there is no doubting the original 741 was a pretty grim choice for audio use. Furthermore, I entirely agree with Tom's response to the SMT vs PTH thing but what you must remember is that in an SMT IC, by its very nature (being modern), is likely to be carrying the latest silicon revision of even some of the venerable op amp designs and thus really don't expect an equally new PTH version of the same device to sound any different.

User avatar
sneakthief
Newbie
Newbie
Posts: 78
Joined: Fri Jun 02, 2006 9:43 am
Location: Berlin
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by sneakthief » Sun Jan 25, 2015 12:29 pm

madtheory wrote:In comparison a PSU is pretty straightforward it either works or it doesn't.
Having built dozens of synth PSUs using different topologies, I have to disagree.

First of all, every PSU has inherent ripple, which quite as I mentioned often affects tuning stability and transient/frequency response. Furthermore, 70's synths often had PSUs that were inefficient, prone to voltage/current ripple & voltage sag especially in hotter environments - straight out of the factory.

These bad brand-new PSUs had a myriad design flaws: for their intended usage they were typically under-spec'd in terms of cooling capabilities, filtering and current delivery. One thing to keep in mind here is that voltage regulators usually become less efficient as they heat up and also as the load increases, so too does ripple.

Other PSU factors that affect the sound include subtle and not so-subtle mains hum from poor grounding, improper AC input filtering, power switches that affected tuning as a result of oxidation (SH-101), excess heat which interferes with VCO/VCF stability, poor ventilation, etc.

I would argue that older synths PSUs also don't cope as well with mains voltage rises and drops. (BTW, here's a great thread which covers many technical aspects of power supply design and measurement:
http://electro-music.com/forum/topic-51694.html )

Not to mention fatal errors like missing crowbar over-voltage protection (ARP PSUs), tantalums that short-circuit (many synths), sending too high voltages to CMOS chips (OB-X, OB-Xa), capacitors with too high ESR (which affects ripple current) or too low temperature ratings (which affects capacitor longevity and also ripple current).

Older voltage regulators, transformers, power transistors, capacitors and bridge rectifiers often had poor performance characteristics. I've seen many boards with blackened PCB's around the rectifiers! I could also link a dozen service notifications for synth PSUs that required changes shortly after their release. This still happens! The mid-90's Allen & Heath GL-2000 console had the worst PSU I've ever seen in such a desk and was subsequently completely redesigned within a couple of years.

Bottom line: the quality of such components has increased incredibly over the past 40 years. Designers also now know how to build better PSUs, instead of relying on crazy Rube Goldberg-esque regulation schemes.
sneak-thief: raw electrofunk!
http://sneak-thief.com

User avatar
sneakthief
Newbie
Newbie
Posts: 78
Joined: Fri Jun 02, 2006 9:43 am
Location: Berlin
Contact:

Re: Why don't new synths have 'that vintage sound?'

Post by sneakthief » Sun Jan 25, 2015 12:36 pm

PS. It's true - building good PSUs has been possible for decades! All I'm saying is that in practice, things got more... fuzzy ;)

My take is that compromises in PSU design account for maybe 10% of the so-called vintage sound.
sneak-thief: raw electrofunk!
http://sneak-thief.com

Post Reply