No, unfortunately Pavel is wrong. I listen to the distortion plugins that are available, unlike some other people with strong opinions. I don't use such plugins and I don't recommend that others use them. Nor do I use digital guitar amps. It has nothing to do at all with hum and buzz. I don't know where such ideas come from. No use for that stuff real or simulated.
In terms of modeling distortion, okay, suppose you want to model the first several harmonics and other other distortions produced by a guitar amp when a low E and a low A are struck together. The interval is a perfect 4th, and it will produce some beat notes in the guitar itself that will be included in the signal to be distorted. Also, say you are sampling at 48 kHz. To get the 7th harmonic if you want to go that high, you will need to upsample 7 times, distort, and SRC back to 48kHz in real time. And we don't want to hear any aliasing or artifacts of antialiasing filters. And it should sound like a guitar amp including the great sounding speaker distortion of something like this Alnico - Celestion Blue - Celestion - Guitar, Bass & Pro Audio Speakers which doesn't sound like exactly like any other speaker. You could try it, but I think you will find it isn't all that easy.
And, sure Pavel could add a little second or third harmonic to something and say, "see, its easy!" But aren't we talking about distortion that people actually like? People do like the sound of things like well distorted guitars, not simple digital distortion. One can't prove this or that as Pavel claims unless we are using examples people actually seem to like the sound of, and doing that isn't so easy as Pavel seems to think. In other words, if there is some kind of "sweet distortion," it is more likely than not going to turn out to be subtly complex in some ways, although maybe not in all ways. Not simple stationary distortion.
Last edited:
Guys, you're not going to get terribly far with defining whether "sweetness" can be added via DSP (although if its either linear or harmonic distortion, then, yes, pretty easy to add), if you have no idea what on earth "sweetness" actually means in terms of an effect. Obviously simulating something like overload/clipping behavior from another amplifier/system is (even if it's attempted to be reproduced on an amplifier far away its own clipping) going to be a lot tougher.
And whether one person likes/dislikes said effect that already exist is another question entirely. To each his/her own there.
And whether one person likes/dislikes said effect that already exist is another question entirely. To each his/her own there.
Right. Basing your entire world view on plug ins available to wannabee record producers isn't really a thorough trawl of available methods and solutions. And Pavel is talking about reproduction not recording. very different.
Nelson Pass of course does it the classy way. One of his designs has adjustment so so can select between a bit more 2H or a bit more 3H. No one can agree on the best setting, but all have their preferences.
Nelson Pass of course does it the classy way. One of his designs has adjustment so so can select between a bit more 2H or a bit more 3H. No one can agree on the best setting, but all have their preferences.
This is not what Mark is talking about, there are no claims here about sampling classic mics, i.e. original Neumann's with unobtainium VF14 tubes. I don't don't see any discussion of modeling the large signal/distortion characteristics of SS vs. tube, in fact the video states the better small diaphragm mics are not that well emulated which is simply physics.
Richard: I'll take your blur and raise you a legible multitone output from an AP courtesy of TomChr (which I have posted before). Note this is just a 40W LM3886 based hybrid, but it's still lower distortion than the average pre-amp. MiniDSP is many times worse.
Attachments
Last edited:
Well, in the case of this mysterious "tuning" of the reproduction that started the whole late trend, this is a circumstance where the end user wants to inject themselves into the "recording" process, so there's a healthy bit of overlap.
Hard to argue with NP's ethic -- tune to one's liking.
Hard to argue with NP's ethic -- tune to one's liking.
That is great performance...... Now try this.... apply multitone to the source (via DSP?) and output to a preamp and thru power amp all together and what does it then look like at the output. ??
THx-Richard
Perhaps it isn't the distortion, but the output transformers response to distortion that gives a tube warmness? Warmness seems often related to tiny amounts of saturation which is related to noise. Where as just noise seems a little harder to correlate, but usually is acting on other factors in some way.
The sweet, nearly wet, sound is still a bit mysterious. Maybe that is why some people will pull up in their Rolls Royce to an audio show and go drop as much as a house on Tidal gear.
The sweet, nearly wet, sound is still a bit mysterious. Maybe that is why some people will pull up in their Rolls Royce to an audio show and go drop as much as a house on Tidal gear.
I don't understand the question, first of all the crest factor of the multitone is a potentially large factor in the levels possible, second the source material would totally obscure the noise floor and I can't see any information from such an experiment.
You could take source material and add deep holes in it and see the infill, Dean Jensen did this ages ago but now with DSP you could do much better. I have some examples on Jan's site, one could easily take any piece of music and create holes <-120dB down at any intervals you want. Personally I would take a whole minute of music sampled at 96k and do it with a 5.7M point FFT.
Last edited:
He is quite knowledgeable, and his presentation is excellent.. That said..
page 14: 4.8 uH is NOT the inductance of 10 feet of wire. It is the inductance of the LOOP formed by the wire and the return current path of the measuring device. The inductance of a 10 foot length of wire is 150 nH. The meter reports that PLUS the inductance of the entire loop path.
Page 20: On the east coast, we use single bushing power transformers. On the west coast, they use two. So about 5% of the return current is via the earth, on tother side, it is different. (note the picture on page 22 is that of a single bushing transformer.)
Page 24. He show an isolated secondary and not an autotransformer, and he doesn't include the pole earthing conductor. The statement that none of the load current flows through the earth is incorrect.
Page 32. There is not a magnetic "null" between the hot and neutral conductors in romex. I explained this to him years ago. Totally incorrect.
I almost attacked him for page 44, but realized it's a Rane and not a Numark..
Page 60 to 62. He totally ignores the induced signals within the chassis caused by the ground loop current flowing through the chassis and induction to the low level internal signal lines. I've explained this to him as well.
Page 90. A coax can become susceptible to external magnetic field when it is bent, as that breaks the symmetry of the coax and the centroids are no longer common. Also, a high gradient magnetic field can couple to a coax, such as if a coax is tangent to an E core transformer's edge.
Page 91. bad magfield pic again. this produces a magfield that falls off as 1/spacing of hot to neutral..
page 92. Finally, that is accurate.
page 94. Current along the axis of a cylindrical conductor on the outside produces ZERO internal magnetic field. The core wire of a coaxial cable CANNNOT see any magnetic field caused by current on the shield.
Page 98. Shield current caused magnetic field cannot induce voltage on the core wire by magnetic induction.
I stopped at page 120...
I must note that the date of this presentation is roughly the same timeframe as when I detailed the problems. So, I suspect this was probably done before I corrected him on points.
John
Sounds unrealistic. The inductance of a 1m of a common single wire is 1uH. 1uH/m is a very good approximation to any single wire.
Well, it is rather the highish output impedance which gives tube amps the name of being warm. This underdamps speakers that are designed for low Z transistor amps.
Designed for?
Over dampening is almost homogenise with speakers in the way that old speakers under dampening was. Yes you CAN lower distortion with it, but you can also kill other other qualities. There certainly seems to be a medium. But I suppose some speakers are literally so over dampened it takes a low Z amp to force them with big power.
Quite, time to remind people about the Carver Challenge again
Yes, accuracy in the classic sense is not important. They will never get it.
How do you measure a 1 meter length of wire without creating a current loop? Many people confuse this issue.
A cylinder of current as within a solid conductor at dc will have magnetic energy stored within the wire at 15 nH per foot. This remains at that level until skin effect begins at some frequency, where the inductance will begin to drop. In the extreme, at very high frequency, all the current will be on the conductor surface. In that case, there will be NO internal energy and no internal inductance.
When one measures any length of wire for it's inductance, one can only do so by closing a loop. It is the inductance of that loop that the meter tells you, not the external inductance of the wire.
So the statement that a ten meter length of wire has 4 plus uH of inductance is just nonsense.
John
I own plugin version of this $8000 harmonic generator for mastering studios. If someone is interested in what plugin can do today, try this. UAD version is $300, and it is the best distortion plugin I have ever used.
Vertigo Sound VSM-3 | UAD Audio Plugins | Universal Audio
It does what they claim, but the effect is different from the distortion added at the power amp stage, in my opinion, . I guess speaker and amp interaction is the key.
Vertigo Sound VSM-3 | UAD Audio Plugins | Universal Audio
It does what they claim, but the effect is different from the distortion added at the power amp stage, in my opinion, . I guess speaker and amp interaction is the key.
- Status
- Not open for further replies.