Guitar amp Vs PA?

[darrin_h2000]

I suspect that in either case, 100% of guitarists in the audience would say the performing guitarist's tone sucked

I would say that they would know 9/10 that it was done by modeler...but as far as them honestly saying it sucked I wouldnt go that far...people who dont know how to tone in thier amps might sound sucky...but most modelers take away that possibility by having the correct setting in a patch.

 

[dintymoore]

For me I want to hear guitar on stage coming through speakers, only... no horns. That's the sound of the instrument.

But in the end, you're carving the air in the room so there are probably cases where just using the pa is alright... I guess. But you'll never see an Elvis impersonator as good as Elvis, and an amp will always be the real deal.

I don't like the philosophy of doing the least you can get away with to get paid, and if people want to go through the pa to get "their sound" that's cool but it usually seems more just laziness.

 

[mshilarious]

You can't play a modeler, it's simply a signal generator following an algorithm.

It's not a signal generator, not at least in the manner of say a synthesizer. Take a look at the page I linked--those functions are way too simple to accurately model even a single tube, but they building blocks. So if you have, for example:

f(x) = ax^3 + bx^2 + cx + d

That is entirely a function of x, the input signal, and a few constants (a,b,c,d; think of them as "mix" controls). The squared term yields second-order distortion (preamp tube); the cubed terms odd-order (push-pull output into transformer), and the regular ol' x is the mix of the original signal (d would be intended to cancel offset).

As I said, too simple for a proper model, just a building block.

But it's not the same as a synthesizer at all, which is a time-based function: a MIDI signal triggers a tone generator:

f(x,t) = sin (x,t)

which has to feed an ADSR envelope. Much more complicated, and yet musicians are able to play synthesizers, aren't they?

This gets back to my other thread: nobody that is playing a tube amp that possesses all of the time variant and invariant nonlinearities is playing anything like the natural sound of their instrument. Why not plug your guitar into a very clean, very high input impedance FET DI into the board and PA? Doesn't that sound good to you? Does that react in all of the nonlinear ways you ascribe to a tube amp? Of course it doesn't. And actually there is nothing special about the FET; I could give you a 12AX7 DI that behaved in the same manner. But that's exactly the way I play bass (although I just use a transformer). And actually, I record guitar that way too, just listening to the natural sound of the instrument. Later on while mixing I'll mung it good with various plugs, but I don't monitor with them.

Thus I think we need to dispense with the idea that it's impossible to play through any type of modeler because they are synthetic. The modern guitar tube amplifier is synthetic too; it's an analog additive synthesizer triggered by an input signal. Look at something like the Mesa rectos and you see all kinds of features purposely intended to alter that synthesis. It's more akin to a theremin or a Hammond organ than an acoustic guitar.

 

[darrin_h2000]

Well they could make one using tubes...but the modeler would put off too much heat and the amount of the tubes it would take to do the switching just one of those chips in a pod does, it would have to be the size of a football staduim.

But in the vain of the Seymore Duncan convertable which did use tube preamp modules for its modeling its still pretty big...I wish I could find one of those on craigslist...it think that is the best possible scenario.

 

[Lt. Bob]

Well they could make one using tubes...but the modeler would put off too much heat and the amount of the tubes it would take to do the switching just one of those chips in a pod does, it would have to be the size of a football staduim.

But in the vain of the Seymore Duncan convertable which did use tube preamp modules for its modeling its still pretty big...I wish I could find one of those on craigslist...it think that is the best possible scenario.

yeah, they did sound very good when they worked.
But I knew several guys back in the day that used them and they were pretty prone to problems.

Nice amps though if you get one that works all the time.

 

[ggunn]

Right, here goes... I'm just plain fed up of dragging my valve guitar amp everywhere with me so I was wondering (because it'll be a long while before I can check for myself) if any of you had ever considered ditching your guitar amps and running the following set up?:

guitar > effects > Line 6 POD > PA system

Every considered it?

Would I still be able to get a good guitar sound / tone going from the POD to our PA (Mackie SRM450's)?

pros / cons?

It sounded like such a good idea (not hauling 100 lbs of gear) that I had to give it a shot. After one gig I went back to my amp. I think it would have been OK had the other guitarist in the band been doing the same thing, but he was playing through a Fender Twin in the crunch zone. The two did not mix well at all; no matter what I did with distortion, overdrive, amp/cab simulation, etc., my tone sounded thin and artificial next to his. I was disappointed.

 

[miroslav]

I dont know where you shop...but the POD X3 is $500...

What waste of money...that's already half-way toward a decent tube amp!!!

Look...I didn't say ALL modelers cost $100...but man, there's a lot of cheap crap out there being used in quantity by folks who really DO believe that they have *real tube amp tones* coming out of those little gadgets.

 

[miroslav]

Yeah...you can call it "analog additive synthesis" if it works for you, but no matter how you cut it, there IS a difference when you simply have *code* that is always read/executed in the same manner, every time you trigger it.
Tube amps don't behave that way, they interact with the player and guitar/pickups, and with his touch, in a way that a modeler never can, no matter how many variations of code you provide with it.
For me…it's all just too close to MIDI and sequencing...where it’s basically presets (algorithms) being triggered, and nothing more.

Thus I think we need to dispense with the idea that it's impossible to play through any type of modeler because they are synthetic. The modern guitar tube amplifier is synthetic too; it's an analog additive synthesizer triggered by an input signal. Look at something like the Mesa rectos and you see all kinds of features purposely intended to alter that synthesis. It's more akin to a theremin or a Hammond organ than an acoustic guitar.

I really dig you positive/proactive perspectives on modeling and by all means, people who can accept using them (for whatever)....should. Modelers can be another tool, just like you have real pianos and electric pianos…etc…but it's just not ever going to happen in a way that will wipe out tube amps or the guitarist's love of them...so there will always be this division of views.
That’s the only point I’m making and view I’m against…where some people are convinced that modelers are actually a direct replacement for the tube amp they are modeling!!!
I for one will never use a modeler to track or play through live. If I was in a band where I had little say, and someone told me I couldn't play through my amp and could only model through the PA…
…...well, then I just wouldn't play that gig.

It’s about much more than just the sounds coming out of the output....

 

[darrin_h2000]

For me…it's all just too close to MIDI and sequencing...where it’s basically presets (algorithms) being triggered, and nothing more.

So you are hating on MIDI too...
I draw the line at hiring an orchestra everytime I need one

 

[miroslav]

No...didn't say I hate MIDI.
Back in the early-to-mid 90s...I was ensconced with MIDI!!!

I just simply grew tired of the whole sequencing thing, and got away from all the "synth" sounds and went back to more traditional instruments...
…though I'll still throw in a string or horn sample into the mix when needed.

Do you do a lot of full, orchestral recordings...or do you play guitar driven Rock/Pop?

 

[darrin_h2000]

Do you do a lot of full, orchestral recordings...or do you play guitar driven Rock/Pop?

Actually I am working with the KC Philharmonic members that do the Ballet work...but Ive done plenty of rock records.

 

[cavedog101]

I have been keeping my eyes open for just that...remember the seymore Duncan Convertable?...it had 4 swappable channels and several modules to model the sounds of different amps.

You're kidding....Right?

I've owned four of them. Not EVER a MODEL of anything other than a two channel ALL_TUBE head that you could swap different voiced preamp tubes around to create different attack, decay, types of gain-staging. Oh, and the amp had to be OFF or you'd burn yer fingers and probably get a taste of the 600 volts to plate that was available.

NOT a modeling amp......NOT four channels....and the 'modules' were only part of the preamp section and were simply individual tubes doing the same job that ALL preamp tube sections do.

And before you try and tell me about the owners manual (which I have right here in my desk) giving EXAMPLES of how the different combinations of the preamp modules can be used to EMULATE the sounds of then popular amps, please understand that side by side with several of the popular amps to be emulated, it was still a different beast and a beast of its own BTW. I know this simply by using it with these amps in pairs for years.

As for modeling.

I liked Christie Brinkley.

As for guitar players named King, all the rockers ripped on Freddie a lot more than all the others.

 

[mshilarious]

Yeah...you can call it "analog additive synthesis" if it works for you, but no matter how you cut it, there IS a difference when you simply have *code* that is always read/executed in the same manner, every time you trigger it.

Tube amps don't behave that way, they interact with the player and guitar/pickups, and with his touch, in a way that a modeler never can, no matter how many variations of code you provide with it.

I don't have a problem with that opinion, in fact I'd be willing to stipulate that no current modeler can proper emulate a tube amp.

What I can't accept is that no accurate mathematical model of a tube amp could ever be created. It might take a huge amount of DSP, but it kinda flies in the face of physics to say that it can't be done. A lot of textbook physics describes ideal rather than real-world devices, but there are also mathematical explanations for those nonlinearities. It's just up to the rigor of the model to properly and completely consider the behavior and interaction of every component. What level of effort is required to do that accurately? It is likely that a particular modeler has done that for 600 devices? Probably not.

Code doesn't have to execute in the same manner every time you execute it; it's possible to introduce as much randomness as you desire. Although randomness is super-easy to implement in an analog circuit too. I'll give you a tube amp that has a sample-and-hold circuit that determines the amount of preamp gain based upon the instantaneous noise voltage of a zener diode at startup. Do you want such an amp? I would doubt it, but that can be implemented in an analog or digital circuit.

An algorithm can interact with your guitar in the same manner as an analog amplifier--that is to say, it can't. Until you start talking about acoustic feedback, the guitar-->algorithm/amplifer path is a one way street. And there doesn't have to be any reason that an algorithm cannot feed a speaker and therefore end up with the same acoustic interaction. That's obviously tricky with a PA since feedback through the monitor system is probably not tolerable. If the guitarist's technique requires feedback, that would disqualify running direct through the PA, I should think.

Anyway, back to algorithms as a musical technique. They can do all kinds of things that analog circuits can't do very easily. It doesn't have to be that exotic, either. Let's say you want an overdriven tone that is all fourth and fifth overtone, with no second or third and nothing higher. That's a very difficult task for an analog amp, but quite easy for an algorithm. That is the sort of creative technique I'd like to see guitarists consider rather than trying and possibly failing to simply emulate an amplifier.

 

[ggunn]

An algorithm can interact with your guitar in the same manner as an analog amplifier--that is to say, it can't. Until you start talking about acoustic feedback, the guitar-->algorithm/amplifer path is a one way street. And there doesn't have to be any reason that an algorithm cannot feed a speaker and therefore end up with the same acoustic interaction. That's obviously tricky with a PA since feedback through the monitor system is probably not tolerable. If the guitarist's technique requires feedback, that would disqualify running direct through the PA, I should think.

Given a sufficient amount of code and processing power, any deterministic system can theoretically be modeled; I'd think that feedback could be modeled as well. It could be cued by holding a sustained note longer than a certain length of time. DISCLAIMER: I'm an EE but I'm not a bitbanger, so I could be wrong.

 

[Lt. Bob]

Let's say you want an overdriven tone that is all fourth and fifth overtone, with no second or third and nothing higher. That's a very difficult task for an analog amp, but quite easy for an algorithm. That is the sort of creative technique I'd like to see guitarists consider rather than trying and possibly failing to simply emulate an amplifier.

that's actually quite an interesting example and idea.
That WOULD be cool but wouldn't that be getting into synth territory?

 

[mshilarious]

that's actually quite an interesting example and idea.
That WOULD be cool but wouldn't that be getting into synth territory?

Kinda, but so does an octave pedal. Phasers and flangers are time-variant systems, hard to call those anything but synthesizers, and with a ring modulator. Or a talkbox . . .

The overall point is that if this kind of DSP had been available to Hendrix and his peers, they likely wouldn't have decried the lack of perfect emulation of the amps they were using (again, they would still have had to use them anyway), instead they would think, "Hey this is a neat toy; how can I use this?"

 

[Lt. Bob]

Kinda, but so does an octave pedal. Phasers and flangers are time-variant systems, hard to call those anything but synthesizers, and with a ring modulator. Or a talkbox . . .

The overall point is that if this kind of DSP had been available to Hendrix and his peers, they likely wouldn't have decried the lack of perfect emulation of the amps they were using (again, they would still have had to use them anyway), instead they would think, "Hey this is a neat toy; how can I use this?"

Absolutely ..... they'd have used it in some amzing way.

 

[darrin_h2000]

You're kidding....Right?

I've owned four of them. Not EVER a MODEL of anything other than a two channel ALL_TUBE head that you could swap different voiced preamp tubes around to create different attack, decay, types of gain-staging. Oh, and the amp had to be OFF or you'd burn yer fingers and probably get a taste of the 600 volts to plate that was available.

NOT a modeling amp......NOT four channels....and the 'modules' were only part of the preamp section and were simply individual tubes doing the same job that ALL preamp tube sections do.

And before you try and tell me about the owners manual (which I have right here in my desk) giving EXAMPLES of how the different combinations of the preamp modules can be used to EMULATE the sounds of then popular amps, please understand that side by side with several of the popular amps to be emulated, it was still a different beast and a beast of its own BTW. I know this simply by using it with these amps in pairs for years.

To be fair I really want the head itself, that held 4 modules...that was the first pic that came up...it still had real tube preamps to switch from one or another...like the tech21 head with tubes...doing it with circuts and not DSP.

 

[darrin_h2000]

Kinda, but so does an octave pedal. Phasers and flangers are time-variant systems, hard to call those anything but synthesizers, and with a ring modulator. Or a talkbox . . .

The overall point is that if this kind of DSP had been available to Hendrix and his peers, they likely wouldn't have decried the lack of perfect emulation of the amps they were using (again, they would still have had to use them anyway), instead they would think, "Hey this is a neat toy; how can I use this?"

All my footswitches are DSP...my envelope filter...my pitchshifter...I know that Jimi would have loved those.

And he would answer in this thread..."Shame on all of you for keeping my brother Darrin down with your backward thinking."

 

[cavedog101]

To be fair I really want the head itself, that held 4 modules...that was the first pic that came up...it still had real tube preamps to switch from one or another...like the tech21 head with tubes...doing it with circuts and not DSP.

You really dont know a thing about these amps do you. Weren't you the person who recently posted something to the tune of 'people posting reviews of equipment they've never used'...

So what is it you "want to be fair" about?? And where did you hear that the amp heads held "four" modules?

Do you really understand the concept of the amp? Do you know what the modules do? Are you assuming that theres some sort of magic switcher that allows you to change the modules like you would change programs in a modeler?

The actual concept is more of a Lego set than a continuously switchable modeler. There ARE some aspects of flexibility to this that allows you to shape the amp as far as a general sound and its feel and response. There's lots of ways to drive the amp from each one of its TWO channels. It has a true variable power transformer that allows adjustments of the actual rms output. It has your standard volume, treble, middle, bass, reverb controls on each channel.......but its not really a modeler in the strictest sense of the definition.

You should try and check your research more. Because sometimes, it appears you know nothing of what you speak.

Just a friendly suggestion.