Two of the most "characterful" or distinctive synths in my studio are the Korg 700 and the Roland TB303 with Devilfish mod.
Over the years I've noticed when recording these beasties (especially when set to fairly gnarly or aggressive sounds) that the recorded waveform shows a significant amount of DC offset about the zero line.
Anyway, the other day I took brief possession of a Korg M3 while the owner is between studios. It has the Radius card in it. I've been playing around with it and have been really impressed with the sound. I recorded some parts from it, and the most "analogue" sounding part (a moog-ish lead synth sound from the Radius card) displayed that same DC offset in it's waveform - first time I can remember seeing it from a digital synth!

It's not exactly a scientific double-blind trial, but I wonder if the DC offset could be a characteristic of these "vintage" sounds?

Any thoughts?

--------------------
Adam Inglis
A Disco Ate My BABY!

DC offset or asymmetrical waveform?

The 700 is a pretty damn good synth that really has a lot of it's own character and often over looked, after ages trying to get one I finally just brought a 700s

I remember ages ago checking out a Yamaha AN-200 (AN-1X board inside) on a oscilloscope and the waveforms were quite impressive and didn't seem very static at all, that's all I remember though as I did this ages ago..the AN-1X is a great sounding analogue modeller too

Did I mention I got a 700s? lol

My Moog Taurus pedals always show a waveform with considerably more wavform weight displayed under the mid line than above it. I have no idea what this means, but I understand that it's not anything to be concerned about.

In general I agree that it is probably just such imperfections that help to attract us their sound, however that might manifest itself.

There is one artist I work for that actually likes hiss, so prefers his acoustic guitar passed through an amp to impart its noise and restricted frequency range.

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

I doubt DC offset plays much of a part. In general, it's a bad thing..
Capacitors are one of the first things to fail in older kit, and amongst other duties, they are used to filter out DC. IE, dying caps can easily = more DC leakage.
DC is a bad thing to have in the audio circuit.. It stresses and eventually kills other components. Amplifiers & speakers can't deal with it at all, so it's good to be rid of it.

Some waveforms are asymmetrical anyway. In fact the 303 saw wave is. Kinda looks a bit like DC offset, but it isn't.

I expect books have been written about why analogue sounds the way it does! Personally I would put it down to subtle inaccuracies that analogue circuits are prone to, causing things like slight tuning drift.
And also, the fact many analogue synths have free running oscillators. IE, it's always running, but you only hear it when you open a gate. This matters because it means the waveform is unlikely to start at the same point of it's cycle each time you play it. It is somewhat random, unlike digital which usually starts at the point every single time.

Im sure there are 1000 other likely explanations too. But that would be delving deep into anorak and weak lemon drink territory!

--------------------
Paul

DC offset will tend to introduce even harmonic distortion, if it is enough to drive any stage of the signal chain into non-linearity. Even because it is asymmetric - symmetric distortion gives you odd harmonics.

Even harmonic distortion is considered in some circles to "sound good". I have met one person who liked lots of it - he had cobbled together a "wonderfully hifi stereo" that was full of it.

It does not get past the first series capacitor or transformer though, although in the case of a transformer, that counts as a "stage" for the purposes of introducing distortion.

Not sure what the residual effects on an audio transformer are. It is not something I would want to do without being sure it was not going to permanently magnetise the core.

Quote damoore:

---

if it is enough to drive any stage of the signal chain into non-linearity.

---

This is pretty much what a chum of mine (he wot designed the VCS3) explained to me once.

The old gear with their crude, often expediently and penny pinching cobbled together designs/components, are (apparently) often/sometimes full of non-linearities which can manifest themselves in 'pleasant' oscillator detune, 'pleasant' filter and amplifier overdrive, 'pleasant' resonant feedback in the filter, 'pleasant' envelope shapes and 'pleasant', irregular LFO waveforms,' pleasant' ring mod, etc..

There may well be some subtle 'magic' in it all but whether it's worth frothing over and paying silly prices for is another matter especially in this day and age when the VAST majority of people listen to downloaded MP3s on earbuds or PC speakers or have it pumping out of a club PA with 300% thumping distortion!!!

--------------------
Website / Music Lab Machines / Blog

Not just synth but analogue stuff in general.. It's the non-linearity and randomness of the machine that usually seems to give them "that" sound and also one of the hardest things to emulate

Quote Korff:

---

DC offset or asymmetrical waveform?

---

Asymmetry for sure, but offset in so far as the bulk of the waveform energy is to one side of zero.
Why I noticed it particularly is it can get you into trouble with clipping, when trying to record at optimum levels. The meters in some DAWs must only "look" at one side of the waveform, so you don't realise you're going over until you hear the playback/check the waveform on the screen.

--------------------
Adam Inglis
A Disco Ate My BABY!

Quote Adam Inglis:

---

Asymmetry for sure, but offset in so far as the bulk of the waveform energy is to one side of zero.

---

Virtually all audio and musical equipment designed for the last 70 yeras or more has had some provision to ensure that there is no DC offset at the output. If you have managed to record something with a DC offset then either the source, or the recording chain, is faulty. It does happen, but it's not right and not intended.

As has been said, an internal DC offset, whether by design or fault, can result in an increased level of even harmonic distortion, and most people find even significant levels quite benign and pleasant.

Quote:

---

Why I noticed it particularly is it can get you into trouble with clipping, when trying to record at optimum levels.

---

Ah yes... but then recording with a negligible headroom margin isn't 'optimum' -- partly for this very reason!

Quote:

---

The meters in some DAWs must only "look" at one side of the waveform, so you don't realise you're going over until you hear the playback/check the waveform on the screen.

---

Different digital meter designs vary a lot... which is another reason for recording with an appropriate headroom margin. Only an oversampling true peak digital meter can deliver totally accurate level indications.

Hugh

--------------------
Technical Editor, Sound On Sound

Quote Hugh Robjohns:

---

If you have managed to record something with a DC offset then either the source, or the recording chain, is faulty. It does happen, but it's not right and not intended.

As has been said, an internal DC offset, whether by design or fault, can result in an increased level of even harmonic distortion, and most people find even significant levels quite benign and pleasant.

---

Thanks Hugh. But how can one hear an "internal" DC offset if it doesn't appear at the output? Or are you saying that the harmonic distortion it causes makes it to the output, but without the offset?

Quote Hugh Robjohns:

---

Ah yes... but then recording with a negligible headroom margin isn't 'optimum' -- partly for this very reason!

---

Well, "negligible" was defined in hindsight in these cases!

--------------------
Adam Inglis
A Disco Ate My BABY!

OK re-reading the above posts, I think my use of the term DC offset must be incorrect - what I'm seeing must be the result of extreme waveform asymmetry.
I'll keep an eye out for examples to illustrate...

--------------------
Adam Inglis
A Disco Ate My BABY!

Quote Adam Inglis:

---

But how can one hear an "internal" DC offset if it doesn't appear at the output? Or are you saying that the harmonic distortion it causes makes it to the output, but without the offset?

---

The latter.

hugh

--------------------
Technical Editor, Sound On Sound