Another Amp In A Box

[bluebunny]

Back in the say, I used to gig a Zoom 9000 DIed straight into the PA - no backline.  I don't think Zoom ever mentioned "AIAB", but at every gig I played with that setup, people would seek me out to say how great the guitar sounded (along with "where's your amp??"). 

Anyway, back to your original programming... 

[Vivek]

post-distortion response and magnitude of response introduced by the amp-speaker combination.

I doubt that Roland made a Amp and Speaker sim, and forgot to mention that it is not meant for use in front of guitar amp

and not meant for Effects return of Guitar Amp

But is actually meant for use direct into mixer.

They surely know they dont want a Speaker Sim going to an actual guitar speaker, leading to "Double Speak"

[iainpunk]

After reading ElectricDruid's analysis of the MT-2, I think it tries to be more of an "amp in a box" (type, not "malicious marketing" designation) than most realize.  Check out the post-distortion EQ:  that gentle high-end rolloff and then the slight peaks at 100Hz and 4kHz -- can you say "cabinet simulator"?

i don't know if i'd call it cabinet simulation or just "BROOOTAL SCOOOP'' sound that was popular in metal around the time the pedal came out.

the magic sauce of the MT-2 is a guitar with lipstick pickups and an old fender champ, this actually sounds quite good

cheers, Iain

[ThermionicScott]

If the post dist EQ on MT2 was meant to be Cab Sim, they would have made it switchable, to avoid 2 cabs in the chain (mt2 with cab sim - effects return- power amp-cab)

Unless you meant it was meant to be used direct into a PA system, not effects loop.

You're taking me way too literally/specifically there.  All I meant was that perhaps the EQ was designed to ape the sound of a closed-back speaker cabinet, not that it was intended as a complete replacement for an amp's speaker(s).  Remember that most buyers of this pedal were just going to plug it into their practice amp that wasn't going to be turned up loud enough to distort or cause a speaker cabinet to "thump" like the real thing, anyway.

I've noticed a tendency around here:  because we spend a lot of time dissecting and categorizing circuits, terms like "soft clipping", "overdrive", "amp in a box", "cabinet simulator", etc have been associated with very specific implementations and uses.  But the outside world treats these ideas more broadly... 

[11-90-an]

I've noticed a tendency around here:  because we spend a lot of time dissecting and categorizing circuits, terms like "soft clipping", "overdrive", "amp in a box", "cabinet simulator", etc have been associated with very specific implementations and uses.  But the outside world treats these ideas more broadly... 

To be fair, those terms mean what we *generally* mean around here, and it's the outside world mixing and jumbling all the terms and due to this, people(and us!) sometimes get confused and sometimes get deceived by big companies' marketing...

[POTL]

If we are talking about the old Marshall turned on at full volume, we mean the entire amplifier circuit, preamp + phase inverter + amplifier, I don't think it is worth considering a certain cabinet, since they can be changed.
I meant the imitation of the preamp only.
It is clear that no transistor circuit can completely replace the lamp, but we can get closer to one degree or another.

I did a lot of comparisons of different circuits and different transistors, tried circuits based on Zvex BOR, Mu Amp like Bsiab, just replaced triodes with JFETs, like in ROG projects, tried old circuits with op amp in the spirit of JHS Angry Charlie.
All methods have their pros and cons.
The MOSFETs seemed to me the warmest, most dynamic, but they were very noisy and I tried Zvex BOR based circuits that have a feedback resistor that, thanks to the Miller effect, reduces the high frequencies (too much for my taste). While there are AMZ mosfet booster based options like Catalinbread RAH that don't have feedback resistors, they might sound brighter, but they all sound muffled and noisy in reviews.
Op Amps are potentially the quietest, most flexible (we can simulate different values ​​of the anode resistors, unlike JFETs), but the overdrive sound itself is very angular, it might be worth spending time analyzing the waveform and playing with diode limitation.
Mu Amps are the most boring, they are not flexible, almost not dynamic and sound too boring.
Single JFETs, dynamic (perhaps slightly worse than MOSFETs), the sound of their overdrive is the most pleasant, but on a clean sound they sound colder than MOSFETs, they cannot simulate different anode resistor values, but they are incredibly quiet, I did an imitation of a Hot Rodded JCM800, which did not make noise at all.

Going back to simulating the whole amp, I think a lot of work can be done and simulated everything.
Considering that most popular amplifiers have a similar layout of a phase inverter (most often a long tail pair) and a classic class B push-pull amplifier, we can use a general template for simulation, focusing only on the preamp.
The physics inverter can be simulated with a single JFET (we will recover the losses after the stack tone), the amplifier can also be simulated with a single JFET or MOSFET (to your liking).This pair of transistors can easily be feedback and add Presence, Resonance and Response.
Well, the cabinet can be easily simulated using an impulse player, expanding the choice of sound.

Perhaps I am mistaken, but if we are talking about powerful amplifiers, such as the Marshall Plexi 100W, then the main emphasis on sound will be in the preamp circuit, the phase inverter and feedback circuit of the amplifier are almost identical in 90% of the amplifiers, only the value of the feedback resistor and the presence / lack of controls.
If we talk about the amplifier, it can be divided into 2 main groups - EL34 and 6L6, which can be used with high and low pass filters.

Yesterday I looked at reviews of effects from Friedman, Diezel and some others that use Op Amps to simulate a lamp, but I realized that the sound is still somewhat square, possibly due to the high speed of op Amps (after all, everyone remembers that they love the same RAT for a slow amplifier).

Overall, I only have 2 ideas left
1) Can a MOSFET be taught to sound as good as a JFET
2) Is it possible to teach JFET to simulate different values ​​of anode resistors.
Thanks to everyone who responded.

[Fancy Lime]

Let me play the devil's advocate for a minute and ask: why? Sure we can try to recreate everything a tube amp does. Including the output transformer, the transformers interaction with the cab impedance e , the phase inverter, ageing power tubes and whatnot. But will all of this make the sound better? Does, e.g., the phase inverter do something pleasant to the sound? Would anyone even hear the difference? Would I? In a band context? I am asking honestly because I have no idea. If yes to all, is it worth the trouble? Wouldn't impulse response modeling give even more accurate results with a fraction of the hassle? I am not trying to discourage anyone from plunging down the rabbit hole. It's just that I have been down that one before and found it not nearly as rewarding as I imagined it would be when I started. I have since come to the conclusion that I like what I like, I try to build stuff that sounds the way I like it, and I don't give a damn if it sound like something else or not. But it took me while of trying to copy tube amp sound to get here.

Cheers,
Andy

[iainpunk]

1) yes, a mosfet with a source capacitor parallel to the resistor could theoretically sound the same as a Jfet without the capacitor.
2) of course you can just change the source resistance to change the equivalent cathode resistance.

i honestly think an opamp with clipping diodes (not in the feedback loop) or a clipping bjt sound more lively and better in general than a tube sounds overdriven. its completely subjective tho.

better yet, instead of clipping, i like the sound of wave-folding even better! much more agressive with a lot less gain, while cleaning up when you play softly! ultimate in dynamics, you dig in, it goes bezerk with overtones, you touch lightly, it dances with overtones while being really clear... and backing off the volume cleans it up. the only down side is that there is not so much sustain as other circuits would have with the same level of THD (total harmonic distortion)

in my experience, most guitar players don't hear or "feel" the diference between the LM308 or a ua741 or a OP07 in a double blind test, we actually did that to settle a debate where i used to do volunteers work. i took my beadboard and all 3 opamps and we just let 3 dudes who claimed to know the difference try them all blind and guess. they all had it wrong and mistaken the op07 or the 741 for the 308.

just out of curiosity, how about designing and tinkering with your own design to get it to sound really really good instead of walking the beaten path of other peoples designs? ever just tried that and gone through with it until it sounded perfect?

andy, impulse response only does the frequency response, not the clipping knee and hysteresis and other non-linearity in the design. i really appreciate the rest of the message you send tho.

cheers, Iain

[Fancy Lime]

@Ian
TBH, I have no idea how they implement the digital models these days. It would be easy to model the entire response, including all dynamic shenanigans, though. Computationally expensive but not difficult. Not something that I would put in a pedal but a plug-in should be able to handle that. That sort of thing must exist, no? What is that called?

Andy

[POTL]

It depends what you mean by your design. I studied the principles of preamp and their key differences in order to understand which part affects the sound in one way or another. The limitation of cold and hotness, the presence and absence of cathode repeaters, the difference between the stacks before and after, and how their range of operation changes depending on the degree of signal distortion. I understand how it all works and what components are responsible for what in a real preamp. Regarding the design of distortion pedals, I think that given the breadth of choice and the limitation of analog components, it is not very easy to come up with something from scratch, but I can make a circuit based on active elements(op amps/ transistors) and get the sound I want (amount of distortion, certain frequencies, etc.) Rather, the question is in choosing an active element around which you can begin to build a circuit. I looked at a few more reviews / comparisons and realized that op amps from manufacturers I know do not sound the way I want when they are directly compared with effects on transistors.

[iainpunk]

getting any active element to sound perfect isn't an exact science, its more of an art. it also depends on what kind of tone your after, but its always possible and doable.

Rather, the question is in choosing an active element around which you can begin to build a circuit

i think that's irrelevant, i am convinced that i can get any type of active device to sound within a small range of each other, but the deliberate choice of elements is that some elements are easier to get certain sounds from than others.

it is not very easy to come up with something from scratch

i strongly disagree, you should have a look through my notebook. its filled with wack and odd ideas. but that's dependent on your personality and way of thinking.

cheers, Iain

[Fancy Lime]

As Ian said, there are many ways to pickle a herring. The big advantage of opamps is that they are very predictable and linear within their intended range of operation. That makes deliberate design from the blueprint up much easier than with components that are less (BJT) or much less (MOSFET) or very much less (JFET) consistent and predictable.

The good news is, you seem know which component in a tube amp has what sonic effect. I think this discussion could get a lot more productive if we discussed each part separately. You tell us what the "principal components of the sound" are that a certain element of a tube amp produces, and we can make much more concrete recommendations how to emulate that.

Cheers,
Andy

[teemuk]

getting any active element to sound perfect isn't an exact science, its more of an art.

I guess the "art" part is within defining what is that subjective characteristics that is defined as "perfect". Everything else in the process follows strict and predictable laws.

It also depends on what kind of tone your after...

Me and everyone else. My perfect tone is not someone else's and vice versa. ...and my perfect tone for noodling blues licks would be very different than my perfect tone for tremolo picking black metal riffs. 
It's all so very subjective.   

[POTL]

As Ian said, there are many ways to pickle a herring. The big advantage of opamps is that they are very predictable and linear within their intended range of operation. That makes deliberate design from the blueprint up much easier than with components that are less (BJT) or much less (MOSFET) or very much less (JFET) consistent and predictable.

The good news is, you seem know which component in a tube amp has what sonic effect. I think this discussion could get a lot more productive if we discussed each part separately. You tell us what the "principal components of the sound" are that a certain element of a tube amp produces, and we can make much more concrete recommendations how to emulate that.

Cheers,
Andy

As for the preamplifier sound framing, if you do not take into account the amplifier, transformer and cabinet, all sound is formed in tube mode (which is obvious). Passive components install voltage dividers and filtering. Filtration applies to both classic RC filters and lattice resistors paired with the internal capacitance of the lamp, as well as to anode and cathode capacitors. Offsetting the tube (hot / cold, etc.) will affect the waveform in the same way as tube amp / follower mode affects not only the sound but also the feel of the game. In fact, filtration can be easily simulated with any type of active ingredient.
It all comes down to work with the waveform, symmetric or asymmetric limiting and the very shape of the change in peaks.
I have done some simulations and have not yet been able to find a suitable solution, with classic diode soft and hard clipping.
I totally agree that op amps have many advantages, they are consistent, affordable, stable, etc.
Over the next few months, the workload at my work will not allow me to test the theory in practice.
Although it would be interesting to get a sound similar to field effect transistors.