There has been a lot of odd claims made on this thread over the months.
I dont buy the "all the solid state harshness comes from the first few milliseconds of the attack overloading the input" for one moment. That harshness would last only as long as the input was clipping; and the same harshness would be observed in any amp topology if you were clipping the input. That the amps might respond differently depending on topology is clearly true, but if we are talking that initial transient, then once that has passed then the harshness caused by it is gone. (Yes this ignores more subtle things like bias shifting. Again this would work itself out pretty quickly in any competently designed amplifier.)
Railing against the harshness of a (presumably, distortion or overdrive) pedal output plugged into a hi-fi, or a FRFR cabinet is silly. That's not how pedals are designed to be used. The guitar amp / speaker combination has an extremely sculpted frequency response that smooths things so it is tonally acceptable to the player. The harshness of an overdrive / distortion pedal is partly down to the circuit deliberately cutting lows to prevent them farting out on bass notes; and the fact that clipping generates higher harmonics.
Claiming that a delay or reverb smooths out harshness is silly. The effected signal might be tonally shaped, but the dry signal is not.
I dont think modelling amps have any place in this discussion (other than to observe that they are solid-state). Their working premise is to feed the guitar signal to some DSP that emulates an existing amplifier; perhaps with Impulse responses for cabinet emulation for DI / FRFR cabinets,; etc. So any harshness there must be down to the emulation.
Printer2 is broadly right, in my opinion. EQ can go some way to "fixing" things. But if the amp itself is generating the overdrive / distortion that the player is aiming for, then its a bit more nuanced: the (usually fixed) low / high frequency shaping (voicing) that goes on through all the overdrive stages of the amplifier then become a lot more important: its harder to EQ out stuff from an unpleasantly voiced amplifier.
Cheers, and regards,
Ant
I dont buy the "all the solid state harshness comes from the first few milliseconds of the attack overloading the input" for one moment. That harshness would last only as long as the input was clipping; and the same harshness would be observed in any amp topology if you were clipping the input. That the amps might respond differently depending on topology is clearly true, but if we are talking that initial transient, then once that has passed then the harshness caused by it is gone. (Yes this ignores more subtle things like bias shifting. Again this would work itself out pretty quickly in any competently designed amplifier.)
Railing against the harshness of a (presumably, distortion or overdrive) pedal output plugged into a hi-fi, or a FRFR cabinet is silly. That's not how pedals are designed to be used. The guitar amp / speaker combination has an extremely sculpted frequency response that smooths things so it is tonally acceptable to the player. The harshness of an overdrive / distortion pedal is partly down to the circuit deliberately cutting lows to prevent them farting out on bass notes; and the fact that clipping generates higher harmonics.
Claiming that a delay or reverb smooths out harshness is silly. The effected signal might be tonally shaped, but the dry signal is not.
I dont think modelling amps have any place in this discussion (other than to observe that they are solid-state). Their working premise is to feed the guitar signal to some DSP that emulates an existing amplifier; perhaps with Impulse responses for cabinet emulation for DI / FRFR cabinets,; etc. So any harshness there must be down to the emulation.
Printer2 is broadly right, in my opinion. EQ can go some way to "fixing" things. But if the amp itself is generating the overdrive / distortion that the player is aiming for, then its a bit more nuanced: the (usually fixed) low / high frequency shaping (voicing) that goes on through all the overdrive stages of the amplifier then become a lot more important: its harder to EQ out stuff from an unpleasantly voiced amplifier.
Cheers, and regards,
Ant
I am finishing up another JFet based boost pedal for my friend who plays bass, using it as a front end to "warm up" his Ampeg SCR-DI preamp pedal.
The boost can be run "clean-ish" or can select between two levels of diode limiting in the local NFB (Source to Gate). I tried this out with both guitar and bass, and to me it sounds really nice. The diode limiting gives a bit of grit but not in a harsh sense, and the NFB is adjustable so you can dial the gain into a sweet spot that allows you to dig in for more grit/crunch.
I am planning my next build around the same idea, but with a bass/treble tone control stage, and maybe add a couple more features if they work out.. so more of a full preamp than simple boost.
As Ant mentioned
The boost can be run "clean-ish" or can select between two levels of diode limiting in the local NFB (Source to Gate). I tried this out with both guitar and bass, and to me it sounds really nice. The diode limiting gives a bit of grit but not in a harsh sense, and the NFB is adjustable so you can dial the gain into a sweet spot that allows you to dig in for more grit/crunch.
I am planning my next build around the same idea, but with a bass/treble tone control stage, and maybe add a couple more features if they work out.. so more of a full preamp than simple boost.
As Ant mentioned
That is a really important point.. probably why I prefer mild to moderate overdrive in a tube amp compared to super high gain amps, there can be some "unpleasant" (to me) raspy artifacts in high gain amp distortion that once you attempt to EQ it out on the amp tone controls or at the mixing desk, pretty much kills the ''good sounding stuff'' along with it. Cheers!
Very much this.
I'm not aware of any commercially available amplifiers that allow any user adjustment of those internal voicing bits that in large part define an amplifier's tone, especially those that are generating distortion. That would be interesting, although more complicated to navigate than a FMV tone stack. Yeah, most of us in this forum can break out the soldering iron and adjust stuff ourselves to taste, but most guitar amp users wont have this luxury.
Case in point: I really want to like Mesa Boogie amps, as many players do. But I've never plugged into one yet where I can get on with the gain channel. They're clearly not short of gain, but it always seems in the wrong places tonally to me. Yes, that's a "me" thing.....
And I'm not going to spend big money to buy one to see if I can re-voice it to my liking!
Cheers, and regards,
Ant.
Had the same happen to me with plugging into a modern Marshall DSL amp at the store a few years back, and at a jam where I tried playing through a relatively newer model Orange amp. The overdrive/distortion voicing on those amps seem to be geared more towards the thrash metal crowd or whatever the new kids are playing these days (!)
I think a modelling amp with myriads of presets may be an excellent tool in the hands of a professional musicians who is capable to fine tune all the sound paramaters to his requirements. Being and old school guitar player who played for a lifetime through DIY stuff, I like to keep things as simple as possible. And sometimes I fiddle a little bit with voicing, keeps me going and taste varies over the years. Decades ago I checked asymmetric (H2) distortion vs symmetric (H3) distortion, Ge-/Si-Diode clippers and OTAs as well as peak clipping. Teemuk Kytälla wrote a very nice book about all this. All in all I think all is said meanwhile - but not by everyone yet.
Multieffect/modelling is great for coverbands.
First time I tried the Boss amp, that came with my Les Paul, I went from being Eric Clapton to become Angus Young just by one click on a knob😊
Cheers!
First time I tried the Boss amp, that came with my Les Paul, I went from being Eric Clapton to become Angus Young just by one click on a knob😊
Cheers!
You brought up something I did not think of. If the reverb or delay smooths out the signal it must do it in feed-forward mode. The harsh initial attack is over well before the reverb or delay signal makes its way to the mixer. If you are using hifi speaker the roll off due to guitar speakers hitting the back of your legs might not apply though.
One would think. But sometimes the note (or notes)'s frequencies are riding upon this big, low frequency (relatively DC for a "few milliseconds") wave, which serves to shift the bias of the stage that's clipping. So perhaps this is what's being heard and complained about. Solution is have a brick wall high pass filter at, say, 79Hz.
But then someone else will make a post complaining they cant get that swooshie effect when they really pound the strings, which they somehow wove musically into their playing.
I acknowledge that bias shift during clipping can be an issue. Hell, overdriven guitar tube amps have been prone to this for years. But I'd expect that any bias shift due to the transient initial attack will also be temporary; I'd expect that any competently designed amp recovers from this bias shift fairly quickly after the clipping from this transient has passed.
A brick-wall filter would be relatively easy to try. Has anyone ever done it in a DIY or commercial design? Personally, I'm not sure that it would help much; I think that if this bias shift is a problem, there are better ways to overcome it by preamp design itself.
Cheers, and regards,
Ant
Could such brickwall filter be made without changing the tone of the guitar? (Hereby meaning the sound or hmm..)
Cheers!
Cheers!
So the number "79" is supposed to let 80Hz and above pass through, supposing all meaningful tonal characteristic of any guitar is above the low E string in frequency. Dont know what that would do to the drop D folks; I'm sure a better value could be chosen with the same desired effect.
I've built a few incompetently designed ones, where the problem was too much low frequency getting on through the gain stages, sometimes even blanking out chunks of sound in discernible timeframes. Perhaps I just wasnt a talented enough player to incorporate that as an effect.
Made my Flamma pedal do that on one setting, using a Kala acoustic nylon string with its piezo "Shadow" preamp. Whoever did that one must've let a lot of low frequency stuff on through, cause it knocks the Flamma's setting right off its rocker, of course recovering moments later. Same setting with my electric guitar as input doesnt do it and cant be made to - whatever big "DC" component coming from the nylon guitar just isnt there.
Had an idea one time to make everything a function - more or less - of the input envelope, a signal such as that which controls the frequency of an auto wah (Think Jerry Garcia on "Estimated Prophet"). B+ power supply voltages, bias voltages collapsing or increasing as a function of the amplifier input signal RMS value. With so many people in the churn thinking about this stuff, of course that's been already thought of - and patented.
When I was in college, I made a combo envelop filter with 2nd order filter having variable Q - think bouncy-wah. These days it would be tap-tempo auto-wah, of course riding on your playing. So tap-tempo amplifier, with all the internal voltages wiggling around in sync with your playing along to a beat. They do that some for sure already; this would be exaggerated wobbling B+ collapse or stage bias. How would that sound? Tonal tremelo initiated by your playing? I have no idea... Someone will do it in a matter of time.
Im sorry. I meant wont the high-pass filter affect the guitars tone above 80hz? Will it be a passive filter on the guitar/first pedal/preamp? Can such a filter be made transparent?(Apart from it low-cutting😊)
Cheers
Cheers
If you investigate the signal characteristic of open low e-string - be it with an oscilloscope or a spectrum analyzer - you will discover that the fundamental harmonic H1 is relatively small, compared to H2 and H3. So, with the fundamental totally removed you will not observe much difference. The ear construes the fundamental according to the difference of harmonic overtones, and thats it. A low frequencey cutoff of 150Hz does not do any harm to the guitar sound, but saves output power by reducing rumble frequencies in the sub region.
It should not.
And not mess with the bias points of clipping circuits as the guitar's sound rides up and down upon those rumble frequencies. So we have some numbers to start with - DSP filters, just enter the value and listen.
This can be perceived as a "good thing" or a "bad thing", depending on the player of course and also what type guitar and amp setup they are using. It's often referred to as the "mid hump" or "bump". EQ'ing at the distorter stage is an important factor, maintaining clarity for the upper end while avoiding muddiness and fuzzy-ness on the lower strings. You can shift that cutoff frequency downwards from 700 Hz and it makes a very noticeable difference in the overall effect. A few variants of the Tube Screamer do this, or have a variable pot in the RC to shift the cutoff and bass gain (Voice control on Zen drive). Another approach is to incorporate two filter networks with different frequency cutoffs and slopes. IMO EQ at the distorter is at least as important to the overall sound as the distorter itself.
DIY, no expense too great (compared to a commercial effort) you could have an array of bandpass filters, each followed by the same "distorter" circuit. Then mix each output level like a graphic EQ.
Better yet, hex pickup with each string having it's own bandpass filter in the appropriate range for the string, then each with its own distorter. I'd bet that would yield some clean chord based distortion sound.
Hex pickups have been around for a long time; I'm curious why they havent ever "caught on", except as a synth controller. Too many wires in the connector is all I can think of. Shoot, if my playing was "worth it" I'd have six amplifiers on stage; one setup for each string. Imagine how fat - even Tom Scholz didnt do that.
Has been done (google cycfi pickups) and it does not sound like one single guitar anymore, actually it sounds like 6 guitars (or keyboards) playing synchronous single-note lead lines all mixed together. It doesn't sound distorted at all and it doesn't sound like chords being played. All the beauty -- the gritty IMD products -- of the conventional guitar distortion is lost.
So that's it. I remember some early guitar synth had what they called "hex fuzz". Guess it just didnt work, else everybody would be using it 40 years later...
There are also iterations of that idea, which divide the signal to several passbands and then distort each pass-band individually (e.g. Traynor TS series, Quadrafuzz).
At first it sounds like a solid idea; eliminating much of the IMD products. But in practice it creates IMD at fixed frequencies and starts to sound artificial and "synthesized".
At first it sounds like a solid idea; eliminating much of the IMD products. But in practice it creates IMD at fixed frequencies and starts to sound artificial and "synthesized".