About all those amp-sims...

[Mark Hammer]

A variety of folks including the ROG folks, and more recently the VERY prolific MartyMart, have been busily churning out JFET emulations, inserting JFETs where tubes were in the original design.  These have been the product of great diligence, and made a lot of people VERY happy, so I'm not going to take anything away from them.

But....... (you knew there had to be one, eh?  )

Here's the thing.  Such simulators essentially mimic the input stage of the various amps they aim for, and often do so with the very same power supply parameters.  If you switch over from the pedal-guys channel to the amp-o-phile's channel and tune in, what you hear is a great deal of comment about the importance of the output stage and other things.  It may be the way such and such a cab and speaker complement give an amp a distinctive tone (would a classic 59 Bassman be the same amp with a 4 x 12 Celestion stack instead of 4 x P10Rs?, would a JTM45 still be one with a single 15" JBL?).  It may be the unique properties that the output transformer provides (think about the strong link between Hi-Watt amps and Partridge transformers).  It may be the differences in plate voltage used on the very same output tubes between one year issue and another.  And so on.

Now, it is the case with just about any subfield of audio that there are aficionados who make a huge deal out of little things that don't really matter unless you have "golden ears".  But the widespread assumptions about post-preamp things that make a big difference are too well entrenched, even among the level-headed and impartial, and too well-reflected in commercial products whose post-preamp properties are only understood long after the product itself acquires a very favourable reputation among users.  Indeed, it may be that a great many classic and desirable amps share a large number of things in common about the input stage, but are easily differentiated by things that happen after the input/preamp stage, which are well beyond what one can easily capture in a simple discrete circuit emulation.  The most obvious example for me would be the post output-transformer feedback presence circuit.

In which case, my question is:  "Have we exhausted the possibilities of name-brand preamp emulations that might be truly different and idiosyncratic?"

I'm not asking this because I think the folks who have so nobly produced these things so far were wasting their time, or because I thought that none of the amp characteristics could be captured by focussing on preamp design.  Rather, it may well be a waste of anyone's time to pursue additional sims.  It would be nice for everyone to be able to own a blackface Super Reverb and brownface Deluxe Reverb, but quite frankly their different qualities are not to be found in their respective preamps.  Same goes for a whole slew of other amps that everyone would like to own.

[Arn C.]

So, is there a way that the power amp sections of these amps could be simulated in the similar fashion as the preamp sections?  Maybe even less the transformers.  Maybe a circuit could replace the transformer or maybe a smaller transformer could be used like the little ones from  Mouser.
Also if one was made, logically the output needs to be no more than what these preamp circuits put out.

Hmmm.......

Peace!
Arn C.

[Quackzed]

there have got to be some scaled down/small transformer power amp emulators out there! There's just gotta!! 

..nice flat chip amp power section fed by "power amp emulator" fed by jfet preamp ROG sims??!!

what? no? no!!!???
That would be cool...i'm gonna go look around!!

I think it must be a cool feeling to "jfet-ize" your dream amp tho... probably why theres such an interest in the ROG idea of replacing tubes with jfets, there are lots of tube amp schematics out there to "jfet-ize"
and you feel your in uncharted territory with a "new" one...

plus they sound really good !
but yeah... it starts to get redundant... especially where alot of different amps use the old rca/fender input section with some tweaks...still
"Let there be songs, to fill the air!"
and get to work on a power amp emulator!!! 

[WGTP]

One of the things about output sections that is most obvious, even to me, is that the push-pull designs cancel the 2nd order harmonic which is not happening with the CS stages in the pre-amp, or the output stages of some of the amps like Fender Champs or some of the boutique models.  I think I said that right.   

[PenPen]

Interesting this should come up. I was thinking the same a while back. I am also guilty of trying to sim an amp, but to be honest simulating the preamp is sometimes more of an attempt to create a pedal with SOME of the character of an amp. For example, the one jfet Bassman sim I whipped up a while back was aimed at making a dark boost and coloration pedal, not as much for simulating the amp itself. I have been wanting to put together a 30w SS amp, modelled after a Bassman. I plan to use Marty's Bassboy for the preamp, but I'm still researching the power stage. I DON'T want to go the easy route and just slap in a lm3886, I want to replicate the 6L6 tubes as closely as possible. Still haven't found a SS power part that has the same relation to a 6L6 as a jfet has to a 12AX7. It won't sound the same, most likely, but it will be fun to attempt at least. And it should have at least SOME of the dark character of the Bassman, it won't nail it exactly, but you should be able to tell they are related.

[mojotron]

I agree that the total sound of an amp does not come from just the preamp, but I'm not sure if chasing after the more esoteric aspects of an amps sound would be that beneficial. After all, you have to plug everything into something that we have to infer is implicitly "less than what you wanted" anyway. For me, the preamp circuit is generally the most approachable and arguably one of the most significant contributors to an amp's characteristic sound. Although, by matching a speaker-sim with the preamp circuit... into a bypassed VAMP (or it's more famous cousin) and you might have a bit more to play with...

I think where all this work is going is a great place. If we look down the road a year from now (maybe a lot sooner), I think that we will see more things turn up like "I figured out how to design in that Fender Twin jingly sound" - which I think we have sort of gotten past actually... Or "I took this sound that I got from the Dr Boogie and and added it to this JTM-sounding circuit... doesn't sound like either, but...". But, I see this work moving more in the path of new design from the insights gained, rather than refining a simulation that would have no way of really being totally accurate in a variety of environments.

All of this work has been very inspiring for me, but I really don't want it to go in the direction of perfecting a sim, rather finding/developing new ideas about how certain characteristic sound are formed - then roll that into a design - possibly with some other ideas where the result is greater then the sum of the parts.

[WGTP]

I also wonder if replacing the tonestack/filters in the sims, with a 6 band EQ wouldn't allow for lots of sims from the same circuit.   One of the things I have gotten from the sims, is the importance of EQ to the distortion (nothing new here) 

[Mark Hammer]

A well-reasoned and reasonable reply.

Yeah, I don't want to stymie anyone's enthusiasm and creativity.  I guess there just comes a point in any endeavour where you have to say "Maybe we've pretty much exhausted the possibilities for X at the moment.  Let's take a step back, breath deep, and try and take in the big picture.  What is it that my current approach will NOT allow me to do in a fully-realized way?"

Of course the elephant in the tent is that no matter how outstanding any amp-sim is, eventually you, um, have to play it through an amp.   If it is the case that amps impose their own signature (and I enter that, not because I doubt it, but in a rhetorical sense, like a lawyer would), then the JFET-based amp-sim would be expected to lose its own signature as it passes through the true amp.  That doesn't mean that it doesn't produce something extremely pleasing to the ear and musically valid, but if FET-based facsimiles is what you want, expect the authenticity to be a little wanting.

And yes, the real goal is not to simply try to replicate the design of an amp in solid-state form, but rather to be able to produce the distinctive qualities of that amp; the qualities that producea certain feel that eventually begets a certain playing style.  I think it is a pretty safe bet that folks play differently out of an AC30 than they play using a Dual Recto or Polytone, and that they do so because of the loop between what hits the ears and how it encourages hand movement and note choice to eke out those sounds a little more.

[GFR]

Some ideas could be borrowed from this:

[Arn C.]

I just got finished talking to a guitar playing friend of mine who owns a Vox Valvetronix Guitar Amp.
He said that 12ax7a's are used for the power amp section(to simulate the power amp section) then this is applied to the real power amp(which is non tube).  I will assume that the real power amp is clean, merely boosting the amp sims.  May be some food for thought!

Peace!
Arn C.

[Melanhead]

Hmmm ... I was thinking of cloning my modded JCM800 ... I added an extra tube stage and did quite a bit of tweaking ... It would definitely be different ... I just have to dig up the schema as I did it about 7 years ago! ...

Mark, I know what you mean about the poweramp section though ... It definitely adds to the flavor of the amp. I do a lot of recording with my 22 watt boogie and run it hot to get more powertube distortion than preamp distortion ... very sweet and hard to emulate.

[squidsquad]

 I think a lot of *magic* come from power tube sag...which is why we all like to turn our amps up.
An emulation of THAT...would be benificial to those who go direct into a computer to record without bothering others....and still get the sound.  Perhaps adding tailored compression to a sim would achieve it.

[Quackzed]

ampulator...I ran across that too , and theres a peavey (a few?) amp with a solid state "tube emulation" type power amp section(at least thats how it's advertized! ) i think the classic or heritage vtx.. think the vtx series??! - means VALVE something something...
crate has a similar line...
          i agree that the idea ain't to make a ss amp that sounds exactly like tube amp x... what would the point be?
but rather to find the underlying priciples and designes and circuit fragments to be able to make something new. I mean if you get unexpected results while outside "the book" of amp design who knows what results are waiting for you...
... I feel jfet amp sims are kind of like the super massive black holes of effects box design.It's kinda where the action is for some(newest discovery), it's a new approach and naturally lots of people want to "study"/ get into it . it's when your hungry to do something "outside" the box that providence will slap some meat on yo plate!
gotta get me some jfets man... dig.
let's throw some wacked ideas at the wall and see what sticks...
where is the next effects/audio breakthrough? laser strings? plasma cones? projection stereo? microtubes?
i really do love the mad scientist factor of effects building...

[Sir H C]

As an amp junkie I have to chime in.  The output transformer is the hardest part to emulate.  The real ones saturate, filter, and do other weird things at they are hammered past their limits by many old amps.  It is a noticable difference when you change the OPT in some of the reissue amps.  The iron does make a difference on the sound.  Trying to get this right, or even close, is not trivial and would be a lot of effort. 

When you look at many of the classic amps of the past, they are nearly identical in layout but have very different sounds based on the components inside (Fender tweed Bassman -> Marshall JTM-45).

[Mark Hammer]

I think a lot of *magic* come from power tube sag...which is why we all like to turn our amps up.
An emulation of THAT...would be benificial to those who go direct into a computer to record without bothering others....and still get the sound.  Perhaps adding tailored compression to a sim would achieve it.

Yes and no.  Remember that compression simply addresses the level of the signal.  Sag, because it is power-supply based, changes how the amp handles the signal.  That's one of the reasons why tube-bias tremolo sounds different than opto-isolator or other non-bias types of tremolo.

Such sag is, I suspect, quite dynamic in nature, which means that things like simple fixed "dead battery" circuits probably won't cut it.  Doesn't mean that sag can't be mimicked or passably faked (i.e., close enough for R & R) in solid-state form, just that it will probably end up being a little more complex or just different than we might have thought.  I guess what one needs to get started on it is some type of detailed analysis of transients produced under "sag" and "no sag" conditions.

As an amp junkie I have to chime in.  The output transformer is the hardest part to emulate.  The real ones saturate, filter, and do other weird things at they are hammered past their limits by many old amps.  It is a noticable difference when you change the OPT in some of the reissue amps.  The iron does make a difference on the sound.  Trying to get this right, or even close, is not trivial and would be a lot of effort. 

When you look at many of the classic amps of the past, they are nearly identical in layout but have very different sounds based on the components inside (Fender tweed Bassman -> Marshall JTM-45).

I think we're on the same page here.  All of which leads me to suggest that our collective attention might benefit us all if it was redirected to understanding more about what's going on "after the phase splitter".

[GFR]

I just got finished talking to a guitar playing friend of mine who owns a Vox Valvetronix Guitar Amp.
He said that 12ax7a's are used for the power amp section(to simulate the power amp section) then this is applied to the real power amp(which is non tube).  I will assume that the real power amp is clean, merely boosting the amp sims.  May be some food for thought!

Peace!
Arn C.

There´s a Valvetronix patent.
Output signal converter for ampliciation or attenuation of tube amplifier output while maintaining output properties
6,175,271
6,229,387

Nice readings.

[Sir H C]

I think a lot of *magic* come from power tube sag...which is why we all like to turn our amps up.
An emulation of THAT...would be benificial to those who go direct into a computer to record without bothering others....and still get the sound.  Perhaps adding tailored compression to a sim would achieve it.

Yes and no.  Remember that compression simply addresses the level of the signal.  Sag, because it is power-supply based, changes how the amp handles the signal.  That's one of the reasons why tube-bias tremolo sounds different than opto-isolator or other non-bias types of tremolo.

Such sag is, I suspect, quite dynamic in nature, which means that things like simple fixed "dead battery" circuits probably won't cut it.  Doesn't mean that sag can't be mimicked or passably faked (i.e., close enough for R & R) in solid-state form, just that it will probably end up being a little more complex or just different than we might have thought.  I guess what one needs to get started on it is some type of detailed analysis of transients produced under "sag" and "no sag" conditions.

As an amp junkie I have to chime in.  The output transformer is the hardest part to emulate.  The real ones saturate, filter, and do other weird things at they are hammered past their limits by many old amps.  It is a noticable difference when you change the OPT in some of the reissue amps.  The iron does make a difference on the sound.  Trying to get this right, or even close, is not trivial and would be a lot of effort. 

When you look at many of the classic amps of the past, they are nearly identical in layout but have very different sounds based on the components inside (Fender tweed Bassman -> Marshall JTM-45).

I think we're on the same page here.  All of which leads me to suggest that our collective attention might benefit us all if it was redirected to understanding more about what's going on "after the phase splitter".

Oh and phase splitters affect sound too.  If the devices are not matched, you lose some of the cancellations talked about elsewhere.

[Skreddy]

We're currently brainstorming a transformer-saturation exercise in this thread...
http://www.elixant.com/~stompbox/smfforum/index.php?topic=42872.0