GUITAR AMPLIFIER SIMULATION This is a discussion of some of my ideas for computer based simulation of guitar amplifiers. Focusing more on the goals than the "hows" of such a technology, I wish simply to present my views of what I'd personally like to see developed. I should warn you: my imagination is bigger than both my intellect and my knowledge of electronics and computers. If you're able to keep this in mind, and tolerate the ideas of a novice, then I think you may find at least a couple of things here worthwhile. Or at least worth thinking about. The time frame is very long term. The tone is almost hopelessly optimistic/futuristic. Of course, if you're dead set against the idea of a computer program trying to emulate a tube amp, then you probably ought to be reading another web page. I'm just offering some possibilities, not conversion. Personally, I hope that the real thing will always be with us. I know that there's a certain organic magic to valve amps that will always elude programmers. Having said that, however, I see no reason not to try to get as close as we can. I don't see this as an end sum game, where the two technologies are locked in either/or competition. I hope my perspective on this becomes clearer as you read on. Main Goals of Guitar Amp Simulation: 1. Stunningly Realistic Sound/Feel. 2. Accurate Simulation of the Parameters/Components. 3. Ease of Use 4. Open Platform for Programmers. Goal One is the "duh" (obvious) goal. And maddeningly elusive. We all know great guitar amps, especially tube amps, make our guitars sound and feel like they're alive-like we're more alive. How much of this can be captured in the digital domain is open to debate, but if we're gonna try, let's see just how close we can get. Goal Two is tied in with the first one. A program that doesn't simulate the "sag" of a tube rectifier will have a hard time emulating a Vox AC30 or a Fender Bassman. Other elements of a tube amp-the circuit, the voltages used, the types of tube employed, the physical placement of components, lead dress, the weave of the transformers, the kinds of capacitors and resistors used- all these and more have profound effects on the sound and feel of the amplifier. And these are things guitarists may want to experiment with, free from worrying about getting electrocuted for their mistakes. Goal Three: I can hear it now: "Well if you have all the design parameters as part of the program, only engineers are going to be able to use it!" Not necessarily. Just because the program itself juggles all these things doesn't mean users have to. There could be various editing levels, only the very deepest of which would allow users to delve into the real engineering, the real complexity. Easier editing modes could hide a great deal of this mess, leaving it up to the computer [really, the programmer] how to supply B+ voltage, coupling circuits, and heater current for that new 12AX7 gain stage you just added. If just clicking on "Add gain stage" is too hard, then one could have even easier levels, where one could, say mate the crunch channel of Marshall's 30th Anniversary amp to the power amp section of a Vox AC30. (Or connect all 3 channels of the Brit's front end to the Vox's power amp). And if THAT's too hard for you, just use the box as a library of amps. Turn the unit on, select an amp, plug in your guitar, adjust the settings to your liking, and wail away. Not so different from using real amps, save for the first two steps' order. Goal Four: Just as the first two goals are intertwined, I think these last two are as well. Instead of having a program that only one company makes applications for, I'd much prefer to see an open platform where third parties could develop various programs for the unit; such as whatever editing levels or software they wanted to write. Or highly interactive tutorial programs that could take novices and show them how to do their own mods. The base would be the real engineering level-making most of the same decisions real world tube amp builders make-component choice and placement, chassis material, voltages, circuit topology, etc. But one could build all sorts of software that could interface with this foundation. BENEFITS of AMP SIMULATION: 1. More guitarists get into working on tube amps 2. Zoom box with hundreds/thousands of amps 3. Test new amp ideas quickly, easily. One of the main benefits I see for this technology is that many more guitarists could get into working on, or at least understanding guitar amps. Virtual guitar amps break down several substantial barriers that I think have kept lots of guitarists in the dark about tube amps. First, when working on real tube amp, there's a very real danger of destroying your amp or your irreplaceable self. If this isn't a barrier to experimentation, I don't know what is. Imagine if that, in order to play music, one had to know music theory COLD. One mistake, one wrong chord, and you're a goner! Think of all the great musicians who've written such amazing music over the years, so many without the benefit of formal knowledge of harmony and rhythm. John Lennon even used "pandiatonic clusters" and "Aeolian cadences" without having a clue to what they were! Back from the analogy, one can fry thousands of simulated circuits without being in any danger whatsoever. You're just learning and creating-with complete and total safety. Second, kind of the flip side to the first point, is that because you're in no danger, you don't know have to know squat before opening up a virtual amp. You can just go in, poke around and change things, without years of reading old, moldy engineering texts from the 30's and 40's. Of course, the more one knows going in, the more likely you are to get circuits that sound the way you want them to. But it's not a prerequisite. Not having to have at least several months of study of complex material means we've just nuked another obstacle to potential amp tinkerers. Third, there's no need for costly parts, tools, workbench, and workspace. Apart from the cost of the amp simulator, the overhead is almost nonexistent. (At least until you decide you're ready to work on real circuits.) So, we've eliminated another important barrier. Fourth, related to the second, is that having many different editing levels and tutorial programs can greatly speed up the learning process. Most electric guitarists today know that an amp has both a preamp section and a power amp section. That's all they would need to know about circuitry to "mix and match2 preamps and power amps. Take a preamp from list A, drag it over to your choice of power amp from list B, and whoola! You've just designed a new guitar amp! When they're ready to learn about gain stages, they can hit the "add gain stage2 or "delete gain stage" icon, and bang! they've just designed (okay, modified) yet another new amp or preamp! The different user levels and tutorials can spoon feed just the amount of knowledge the guitarist is read for. Users could get pretty deep into tweaking circuitry without even knowing Ohm's Law, never mind all the equations in the Radiotron Designer's Handbook. " and 4: You don't need to know much of anything to start, and once you do start, learning's gonna be a lot easier than the way you're grandfather did it. There could even be programs written that would help guitarists spot and trouble shoot mistakes. So that when something they've made sounds totally horrendous, or blows up the virtual amp, the program could suggest some reasons why, and how to avoid such problems in the future. Even if it's not a mistake as such, the program could help. "I want my amp to sound { brighter, darker, warmer, less trebley, more trebley, less distorted, more distorted, smoother, rougher,..} and a program could offer helpful hints, perhaps with some sample "text" problems, and how the program "fixed" them. These would be circuits the guitarist could play through, so he could hear as well as see the changes. EDITING LEVELS: Here are some ideas for what different editing levels could do. I'm sure you could come up with lots better' ones. 1. EARLY USER LEVEL. Perhaps level one. Where you'd just plug in, turn the unit on, pick an amp from the library, twist the knobs to you liking, and play away. The key here is "twist the knobs". No dang scrolling! I'm envisioning a multi-knobbed interface, that could be as big as a normal amp's control section. Perhaps different companies could make different knob controllers, some with chicken heads, some with oven dials, some with Marshall like knobs, some with body orifices to stick the guitar cord into. Now, when you selected a "Fullerton Deluxe-AB763" amp, you'd get the first knob to be "Volume2, then next one "Treble2, then "Bass2, "Volume" (for the vibrato channel), "Treble2, "Bass2, "Speed2, and finally "Intensity". (Hmmm, sounds like "I'm your density".) Any additional "pots" on the main controller would remain off. Well, I guess you could add a bias control. Or a control to vary the B+ voltage on the first gain stage's plate. Or vary the value of the coupling cap on the second gain stage and the value of the first stage's cathode resistor bypass cap in some predetermined ratio or math based function. But let's not get in that deep yet. Next, when you switched to a Custom Hiwatt 100, the controls would run "Normal Volume2, "Brilliant Volume2, "Bass2, "Treble2, "Middle2, "Presence2, and "Master Volume". For your solo, you chose one of several Marshall "Plexi" #1959 100 Watt heads, the controls being in the same order as on the genuine article. [Unless, of course, you decide to alter this stock arrangement.] The controller might have as many as twenty or more "pots". Those not designated for use in a specific amp would remain off or unlighted. Those in use might have a liquid crystal display telling you what they were. Maybe you could do like Demeter and color code the things-one color for treble, another for mids, yet another for bass. Hmmm, possibly a fourth color for that mid shift control you added. I suppose you could have your guitar actually plug into the controller on its way to the virtual amp preamp. That way, you could have a four input Marshall have all four inputs. The control would "sense" which input you were using, and send your signal to the corresponding point in the circuit. Another way would be to have one input, with a "dial" switch that would let you chose between different inputs. Or bridge them. Or whatever. Other jacks might become active when you add an effects loop to your virtual amp. This would be useful for using "real2 effects boxes along with your unreal amps. One needn't be restricted by typical locations of effects loops-they could come just after the first gain stage, between the simulated power amp and simulated speakers, or wherever else you'd like them to appear. (No, the preamp wouldn't feed the effects units a hundred watts of real power if one chose that latter scenario.) But the basic deal is that when you choose an amp, you see the same controls-or at least order of them- as you would on the original amp. This, of course, being subject to user modification. 2. Early Intermediate User Levels: Before I delve into these, let me explain that I don't view this at all as a purely vertical "easy-hard" "simple-complex" world. I see lots of room for the horizontal, which I think would make itself manifest if we had an open platform where independent programmers could come up with their own user levels. A: COMBINATION LEVELS: In broad terms, what I see is lists from which guitarists could take one thing from list A and add it to another from a list B. Or C. Here are some possible combinations: 1. List A: Preamps List B: Power Amps 2. List A: Preamps List B: Power Amps/Speaker 3. List A: Preamps List B: Power Amps List C: speakers 4. List A: Amps List B: speaker List C: cabinet 5. List A: preamps List B: phase splitter List C: Push-Pull Power /inverter Amps This can get pretty involved, as one adds more of these larger building blocks. I think 1. is the most obvious to use. It lets you choose what preamp you want and mate it with whatever power amp you want. I'm not just talking stand alone units, either; I'm talking about taking the preamp and power amp sections out of combo and head amps. [I'm assuming that the phase inverter/splitter (if one is used) is part of the power amp.] I'd like to see a list of several hundred, or even a few thousand guitar preamps and amps. Many would be modified variations on stock designs. Some examples of the "mix and match" approach? Sure. Bogner 3FISH" preamp into a VHT power amp. Mesa Mark IV front end into the power amp of a #1959 100 Watt Marshall. Vox AC30 Top-Boost preamp into a tweed Champ power amp. Ampeg V-4 into a new type of guitar power amp using the legendary 300B power triodes. ADA MP-1 serving as the front end for a 50 Watt Marshall modified with THD's Yellow Jackets to have two EL84's in Class A, cathode bias. 5F6-A Fender Bassman preamp into a Matchless DC30 power amp. (I guess I might have the program nuke the presence control on this one, since it seems to be more a part of the Bassman's power amp than its preamp. Plus, I don't think Matchless amps use negative feedback, which is required for this style of presence control.) Nothing stops us from feeding one preamp into another preamp. Or making some kind of line out of a head amp, and running it into another preamp-power amp combination. Yeah, like Guitar Player magazine did with a THD Hot Plate, run an old Marshall wide open into an inductor based attenuator, then use it as an overdrive "pedal" for a Fender Twin. Or, how about being able to have simulated stomp boxes to slam into the amps' inputs? Let's continue making lists: 6. List A: Combo amp List B: mic List C: mic placement 7. List A: Head amp List B: speaker type List C: Cabinet type You can make up your own lists, using parameters like these: mic type, mic position, room architecture and materials (how about those broken beer bottles you place just below the mic to get additional reflections?), speaker, and cabinet type. Of course, any of these you could break down into their own sub-parameters. You could choose from a list of a hundred different speakers. Or maybe "build" you own, based on such variables as cone material, magnet type, size of components, shape of components, etc. Then, of course, you may chose to get deep into cabinet construction. Material type, thickness, cabinet size, front or back loaded speakers, open back or closed back,..... How about some nice old tube mics to record with? Oh, you say you want an old trusty SM-57? No prob. Where would you like it? What position? Possibly two mics to get some cool phase cancellation happening in the virtual room. What? You want one mic moving in an arc around the cabinet? And the other one spinning horizontally 3.56 times a minute? Sure. Finally, let's just add a signal from a good line out of the power amp. At this point, I think it's clear using lists to deal with variables has its drawbacks. There are better ways to incorporate these things into our models and user interfaces. I'll let you design them. I'm having enough problems as it is dealing with all the parameters. B: Basic Preamp Topology Mods: On the vertical scale, one could argue this is a bit father down (or up) the axis of complexity, as compared with our previous mix'n'match approach. Here, there are icons for "Add Gain Stage2, "Delete Gain Stage2, "Add Cathode Follower2, "Delete Cathode Follower". Or at least an edit menu, with icons of generic gain stages and cathode followers. Let's deal with gain stages first. A sizable chunk of electric guitardome knows that more gain stages means more gain, and thus, more overdrive. And I think these same folks are intuitive enough to work the equation in reverse. While they may be baffled looking at an actual schematic, they could probably figure these out: Fullerton Preamp: IN---Gain Stage---EQ---Volume---Gain Stage--- 70's California Lead Preamp IN---Gain Stage---EQ---Volume---Gain Stage---Gain Stage---Master-- 90's Shred Preamp IN--Gain----Volume---Gain---Gain---Gain---EQ---Master----- Stage Stage Stage Stage I left out the more than likely cathode follower between the Shred Channel's fourth gain stage and the EQ. It'd probably be there in the virtual circuit; but, no need to give the guitarist more info than he or she needs. So, if they wanted to add a gain stage to the Fullerton preamp, they could just put the arrow from their computer at the point in the circuit they'd like the stage to appear, click to "reserve" this space, then click on "Add Gain Stage". And lo and behold, a new virtual gain stage would appear where there had been none before. Or perhaps, one could click on and drag a virtual gain stage to the point in the circuit it was to be added. They might also be able to click on and drag the EQ circuit around. Or the volume/gain control. Or preexisting gain stages. For example, one could move the second stage of a Marshall 100 Watt lead head from the 1970's before the "stock" first gain stage of the high gain channel. They're not the same exact gain stage. One has a 10K ohm cathode resistor with no bypass cap, and the other has a 2.7 K Ohm cathode resistor with a .68 uf bypass cap. Users don't even need to know this, they just need to be able to move stuff around, almost like they do on a word processor. Adding, deleting, copying and pasting gain stages. I like the sound of that. When guitarists were ready, they could go up a slight level of difficulty and make the cathode followers the program had been "hiding" appear. A brief explanation could appear for those unfamiliar with what these circuits are, and how they are most commonly used in guitar amps. Then, it's back to adding, deleting, copying and pasting. And, of course, listening. Hmmm, maybe that's how the lead channel on Marshall's 30th Anniversary amp was developed....'sure has a lot of cathode followers. For all these alterations to circuit topology, the program would have to make automatic changes in the coupling circuitry and the power supply circuitry needed to power the new gain stages and cathode followers. It might add a resistor and a filter cap or two and connect this to an existing B+ supply. It might add a totally separate power supply. It might just assume the new tube is on Venus, so it'll work without filament current. This is why designing the software for these "easy to use" editing levels could prove to be quite difficult. Users, as they became more sophisticated, could always alter these "default" settings/arrangements. Similar to how web browsers have a mode to view the html, there could be a mode to view the actual schematic it all its glory. Gain Stage Library: Just as there might be a library of different preamps, power amps, combos, heads, etc. there could be a library of different gain stages. For example: Brit Stage: 100K, 2.7K. .68 uf Modded Brit Stage: 220K, 2.2K, .47 uf Super Modded Brit Stage: 470K, 1.5K, 1 uf Fullerton Stage: 100K, 1.5K, 25uf Neutered Stage: 100K, 3.OK Bright Stage: 220K, 10K, .1uf One could do the same for cathode followers, but these don't seem to vary as much in guitar amps. C. Advanced Levels: I suppose these are where we start dealing with the actual schematic. And eventually a virtual 3D chassis where we can place parts, move wires around, punch holes. Users should be able to change resistor and capacitor values quickly and easily. One should be able to substitute in different types of capacitors and resistors...and tubes, transformers, chokes, diodes, etc. It should be easy to add bypass caps around resistors. There might be a library of power supplies, coupling circuits, input circuits, various effects loops, reverb and EQ circuits, and whatever else one would like to have. For some of these, simply click on and drag them to wherever you'd like in your circuit. In general, drawing schematics should be as painless as programmers can make possible. It might be cool if you could make any resistor or capacitor variable. Not so much a virtual "pot" on a virtual chassis, though of course, that'd be needed as well. I'm just talking about having the ability to change component values in a continuous function while doing design work. If later, I want to make some of these actual pots, I can do that. And since were adding this capability, let's make it as controllable as continuous MIDI messages are with today's effects units. So, one could change several component values at a time, with one turn of a knob. Some may go up while others go down. Some may go up, then stop or go down as others reach certain predetermined values. I'm not going to get into this all that much, as I'm sure those who really are professionals in this field know what would be needed at the true engineering levels. TUBE TYPES to CONSTRUCTION OF NEW TUBES: As the technology progressed, I think it'd be great if we could choose among not only different tube types, but variations on specific tube types that have been manufactured over the years. Different 12AX7's from Telefunken, Mullard, RCA, Tung-Sol, Sovtek, and other factories. Same with other preamp tone bottles, and power tubes such as EL34's, 6CA7's, KT77's; 6L6's, 5881's, KT66's, EL37's; EL84's, 6BQ5's; 6V6's, 6AQ5's; KT88's; 6550's; KT90's; 7027A's, and so on. When changing from a sturdy old Mullard EL34 to a more modern equivalent, the program might suggest modifications if the circuit you were using was likely to fry the new, more temperamental tubes. At even more advanced stages, wouldn't it be nice if we could vary the construction of the tubes themselves? All the physical, chemical, electrical, and other properties-and how these interact with one another!-so that we could come up with our own versions of these tubes. We could have some "ideal" tubes that had no microphonics, if we liked. Possibly even create totally new tubes that have never actually been built in the real world. Man, talk about customization! Many guitarists may not realize, however, there already exist dozens of tube types beyond our old guitar amp stanby's. Let's add these valves to our tube library. Tubes like the famous 300B power triode, 5691 dual triodes, EF86 and 5879 preamp pentodes (Hello Vox! Hello Gibson!), and whatever else we see perusing Glass Audio, Angela Instruments' catalogue, RCA tube manual's, and the like. For some of these tubes, it'd be nice if they came with several sample circuits we could start off with, rather than always making our own from scratch. For the power tubes, wouldn't it be cool for each of them to have their own mini libraries of single ended, parallel single ended, push-pull Class A and push pull Class AB1 schematics? One needn't even know what these valves were to use them. We could have objects, er, amps already pre-designed in our tube amp library that incorporated these odd-ball tubes into their designs. I've always wondered what a power amp with 300B's would sound like with an electric guitar. Would it sound good for clean stuff? How about when pushed into overdrive? At least we guitarists would finally have access to true triodes in our output sections, instead of the 3triode" option on so many pentode equipped amps today. Additional Features: MORPHING I'd love for there to be several types of "morphing" from one amp to another. I'm not so sure how this would be done. I can imagine taking a given preamp with say, all 5.0 K Ohm cathode resistors, and morphing it to one to a series of new values (say 2.7K for the first stage, 10K for the next one, and 1.5K for the third.) But changing a Fender Deluxe into a Soldano SLO100 poses more serious challenges. But possibly more opportunities as well. I think this can be done. It'd be very cool to be able to morph between different amps with a pedal, so we could control the "sweet" spots in between them as we played. I'd like it if the pedal sweep took us from a clean Twin style amp to a dimed Hiwatt, and finally to a 50 Watt Marshall driven by a Klon Centaur. And that's just for a pedal that moves up and down. If we had some side to side action happening as well, we could throw in a several additional amps. 3 to an axis yields 9 amps, and that's before we're talking about the in-between "morphed" positions. Man, I feel like Craig Anderton! (Sorry Craig, I actually mean that as a compliment.) SIGNAL TAPPING: [No, this isn't the latest resurrected shred technique. It's for circuits, not frets.] It'd be great to be able to "tap" the signal at various points in a circuit to see what it sounds like along the way. After the first gain stage's coupling cap, after the volume pot, after the interstage attenuating 470K resistor with the 470 pf bypass cap. After that new low pass filter you put in. Just before the tone stack. You get the idea. AUTOMATIC ADJUSTMENTS FOR RIG/ROOM (Semi-)Automatic Adjustments for the rigs you play through and the rooms you play in. Maybe you could somehow communicate to your preamp the kind of power amp, speaker and enclosure you were playing through. And possibly aspects of the room you were playing in. The unit would make automatic adjustments, which you could fine tune, or just "cancel" or ignore. Would it be possible for the simulator to put out some kind of "test tone2, which would be fed back to it with a certain mic? The program would then analyze the results, and optimize the post virtual amp EQ settings for (what it considered) optimal results. With Artificial Intelligence (if it ever gets off the ground), your preamp might "learn" your preferences over time. Effects of Having Good Amp Simulation: A lot of the impact of this on the guitar playing world would depend greatly on the four goals I outlined earlier. The first one is by far the most critical. All the damn parameters in the world don't mean a thing if great tones, or at least some good ones, aren't there. We've all had experiences with products that had a gazillion knobs, but not one good sound. That's exactly what I DON'T want to see happen with this, though I stongly suspect it's what the first few generations of these devices may be. If we can be patient, I think better days lie ahead. For die hard Fender amp enthusiasts, I don't think these units will ever satisfy them. That's cool, though, as 1. I set goal one (GREAT TONE) as something to strive for, not necessarily attain, and 2. I do not see this technology as inherently threatening to real tube amps. The two could easily compliment one another. Before I get into that, let's look at some of what we've got now in the way of tube amp emulation. 1. Roland's COSM technology: The products that comes closest in spirit (though not yet in sound) to what I'm envisioning come from Roland: the VG-8 and GP-100, both of which employ Roland's COSM (Composite Object Sound Modeling). I've tried them both, and was impressed, particularly when I considered this important fact: this is the FIRST generation of such products. Even the first generation of transistor amps/preamps/overdrive boxes which TRIED to emulate tube tone weren't as good. [I'm not talking about the first transistor guitar amps of the 1960's, but later versions which touted their ability to emulate the sound of tubes.] Art Thompson and his fellow reviewers at Guitar Player, who have a definite preference for older tube gear, had very good things to say about the VG-8, including its amp emulations. From conversations with Art, and from having read most of his output at Guitar Player, I can tell he's normally isn't a huge fan of the latest whiz-bang digital effecter. I'm not taking what he says as gospel, (Shit, I disagree with his opinions quite often) but I think he's a good, public reference point. Having used these COSM based products myself, I think we've got a LONG way to go. A very, very long way. I really was NOT pleased with most of the overdrive tones in the VG-8. A friend of mine, and a great guitarist, bought a VG-8, and the only way we could get the distortion to sound even tolerable was to roll back the highs to about 8 o'clock. This seems to be the secret behind why the "CLAPTONE" setting sounded as good ("not bad") as it did. I had better luck with the GP-100. Perhaps it was the rig I was running it through. Speakers are certainly important to getting an amps "voice" to come through. (Yeah, I know, they ARE part of the circuit/voice for real tube amps.). Still, for all of the fun I had with the GP-100, it was lacking in ways I can't quite describe. I guess it's the "I know it when I hear/feel it" test we all put musical instruments through. Despite the shortcomings of Roland's version of Composite Object Sound Modeling, I would very much like to see research on it continue-whether it's Roland flying solo, or preferably with more companies and universities joining the fray. The parallel is with the "physical modeling" synthesis, first made popular by Yamaha. Much of the research took place in universities, most notably Stanford, whom I heard sold Yamaha some of the rights to their work. Regardless of how it came to be, and keeping in mind physical modeling's present limitations, it seems to be doing well. Reviewers are quite pleased with how natural and expressive physical modeling synths are compared to other forms of synthesis. While acknowledging its faults, the keyboard community appears to be interested in further development and refinement of this new form of synthesis. Guitar players are generally a more technophobic bunch, more likely to be skeptical of new technologies. This can be a healthy thing. I don't expect electric guitarists to embrace all kinds of emulation; I just hope they can be patient with it. Let it grow a little more before damning it. Non-Computer Based Tube Amp Emulation 2. Tech 21 I'd never recommend one to a blues freak, but this company's offerings are a gas for experimenting in the studio on the cheap. While they're okay live, given a choice, I'd opt for a Mesa V-Twin or Matchless Hot Box for (true) footpedal tube colors. I find Tech 21's products to be light years ahead of the transistor amps of yore when it comes to getting tube-ish sounding tones out of a non-tube box. Plus, I find them to have a wider palette of tonal colors. I think these Sansamp boxes miss much of the dynamic FEEL of valves, and I still haven't been able to get them to feedback as nicely as a real tube amp. But we are talking tube amp emulation, and I think they're at least tolerable for some applications. 3. Eric Pritchard's Deja-Vu amps For those unfamiliar with Eric, he's an amp builder from West Virginia who claims he's solved the tube puzzle with his "tube emulators". I've only read one review of his products, and that was in Vintage Guitar, whose reviewer claimed he liked Eric's "Amp 11". At one point, Mr. Pritchard was aiming for the high end market, but now he seems to have shifted course, and wants to hit the mid-priced amp segment. I've included him here, even though I've never tried one of his amps, (and it seems almost no one else has) because he at least SOUNDS like he knows his shit. And he doesn't seem to have an attitude about it. He certainly believes in what he's doing, but he doesn't have the nasty edge, that say, that ROSK guy came across as having when he was "feuding" with Guitar Player magazine. The other reason I've included Deja-Vu Audio, is that I'd like to hear from those who have tried his amps. And maybe indirectly pass on a note that if his 12AX7 emulators are as good as he claims, why not build some rough equivalents to some of the various tube preamps and footpedals we have today? It'd be nice to compare them side by side, at prices that wouldn't make us mortgage our houses. I sure want to hear the total amp packages (pre and power) he makes, but I imagine they'd be pricey enough to make getting distribution more problematic. 3. Vudu Tubes. Or something like that. I may get this wrong, but hopefully my little history lesson won't be too far off: Back in the 1970's, when Mesa was making their first Boogie amps, they were experiencing difficulties with overly microphonic preamp tubes. So, for the first couple of gain stages (maybe it was just one), they used something called "Fetrons", which were transistor based, but which operated in high voltage circuits similar to those real 12AX7's ran in. Reports I've come across claim they didn't sound as good at tubes, but that the fetrons did reduce problems with microphonics. Mesa later dropped the Fetrons if favor of using all tubes. I hadn't heard a thing about Fetrons since, until at the 1994 Winter NAMM, I came across the booth for Vudu Tubes. They had what appeared to be a new version of the Fetron, which they claimed worked in the exact same circuits real tubes did, and sounded just like real tubes. Yeah, yeah, the same old story. They had a Peavey Classic 20 and a Fender amp ( it might have been a Super Reverb) equipped with these "Vudu Tubes" in the preamp positions, with regular power tubes. (They hadn't yet developed power tube equivalents.) I only heard them very, very briefly in the Peavey. To their credit, they didn't sound like a really harsh transistor fuzz, but without having another Classic 20 with all tubes to do A/B testing, it was hard to judge them. Especially considering the cacophony of noise that is NAMM! As soon as Terry Buddhing from GP arrived, the amps were promptly turned off. An indication of the confidence these folks had in their product? Well, since then, there's been no new information about these new fetrons. I was hoping the guitar mags would do some serious testing of these products, in a wide range of tube amps. But, as I've said, nothing more ever came out of Kirkwood Radio Labs, which developed these Vudu Tubes. Anyone out there familiar with this company? I'd like to hear from you. There are other tube amp emulators (ART Power Plant, Peavey Transtube, etc.) out there, as well as some dang good overdrive boxes (Ibanez Tube Screamers, Klon Centaur) based on solid state technology. Though we're getting a bit off the focus on real time amp modeling via a computer, I thought it worth including these analog methods of getting tube style tones. Okay, so we've got a ways to go yet. But let's see what some of the positive possibilities might be IF there were to exist VERY GOOD SOUNDING tube amp emulation of the kind I've referred to. I warn you, this is going to sound hopelessly futuristic/optimistic. It's just an exploration, not any kind of prediction, so be cool, okay? ONE: It could be just a "phase" guitarists go through. As young players, they might enjoy hot rodding virtual amps on their Mac's and PC's or COSM preamps. Eventually, they'd mature into real tube amps. But they would have learned a lot about them from having used, and been enticed by, the simulated amps. Yeah, it's possible some guitarists would get stuck in the digital realm. Well, so be it. It'd just be one more tonal color we could have. Kind of like when Shawn Lane talked about how for some songs, synth versions of real instruments are better than the real deal. Taking the amp simulation program to an extreme, I'd like to be able to emulate the solid state tube amp emulators we have now! TWO: Instead of competing with actual tube amps, the library of amps on the computer could serve almost as a form of free advertising manufacturers: "You think the VHT Pittbull sounds good here, man, you oughta hear the real animal!" Since it's very doubtful that a program could ever truly nail every nuance and sensation of a genuine valve amp, I don't see it as a forgone conclusion that manufacturers of amps will view this technology as competition. THREE: Amp emulators could be an easy way for guitarists to demo some aspects of an amp or preamp's sound, without having to annoy salesman by trying out every amp on the premises. They could narrow their choices down a bit, so when they DID get to the store, their visit might prove more efficient and less tiresome for customer and salesman alike. It's one thing to hear a tape or CD of an amp, but to be able to (almost) play it for several months before throwing down plastic?!?! FOUR: As a way to spread the tube gospel. This could be more of a factor as the prices of these emulators came down. Think of what Peavey has done with their Classic series-made it possible for guitarists who couldn't possibly afford a Matchless, Mesa, or Fender amp to have a taste of the real thing. I think all those Peavey's out there, while not being bad amps, serve to whet the appetite for high-end tube guitar amps. (And keep selling tubes, which helps keep tube factories in business.) I wouldn't be surprised if something similar happened with these virtual amp machines. (No, they certainly aren't real tube amps, but their effect on the market could be similar.) FIVE: Education. I've already alluded to this. Guitarists, even if the sims didn't sound all that great, by being able to work on inherently safe circuitry could learn a great deal more than they otherwise would. They might move on to building, or at least modifying their own tube amps. They'd be more comfortable talking with those who did work on their amps, and it'd be more likely they'd get what they wanted out of amp mods. SIX: Communication (communion?) with manufacturers. If guitarists could get inside commercial amps innards on their computers, they might come up with good modifications, or discover safety problems that need to be rectified. This could possibly lead to companies catering to customization of guitar amps-send us your favorite schematic you've come up with, and we'll build it for you. Dan Torres already seems to be heading in that direction, and this is WITHOUT the benefit of good simulation. SEVEN: Have techs just mod the virtual amps. If a person gigs with their Amp Sim Preamp, why not have some programs customized by amp techs? The guitarist wouldn't even need to know what was done to the circuit. He or she would just have to groove on the ear candy. EIGHT: Trade schematics/amps over the Internet. I believe people are already trading their own customized presets for multi-effects units on the net. Imagine trading "amps"! NINE: Use the Internet as a way for small manufacturers to become discovered. It would greatly reduce some of the problems of distribution, at least for those aspects of an amp that can be accurately simulated by a program. People could "try" the amps on the Internet, then buy the physical product if they liked it. Of course, such simulations, if done poorly, might misrepresent the amplifier. TEN: Little "Zoom" boxes that could "house" thousands of pedals, preamps, amps, and speakers. It'd sure be nice to have an insanely versatile, light weight, great sounding box to bring to quick jams, or even gigs. In addition to whatever controls you had on the "amps", you might bring a post virtual amp EQ to adjust for playing through different speakers in different rooms. This could be as simple as a little Boss Graphic EQ. It might save tweaking each preset amp individually. Or at least reduce the amount of fiddling necessary. ELEVEN: As a way to test circuits, possibly reduce product development time. I hesitated to include this, as I thought it painfully obvious, but here it is. TWELVE: As a way to build circuits that normally wouldn't be practical or safe. Wanna hook up six 12AX7's (12 stages) in parallel? Go right ahead! Wanna destroy hundreds of Marshall Plexi heads by using a Variac? Go right ahead! Want to hook up each gain stage to it's own power supply, no problem. An aside: This makes me think of (part of) an idea I had for footpedal tube boxes, each with just one gain stage (or gain stage plus cathode follower). Maybe some boxes with just coupling circuits and/or passive EQ networks. There'd be different types of boxes-some with high plate voltage, some with low, some with large plate resistors, some with low, some with large value cathode resistor bypass caps... Each stage on its own wouldn't be enough to produce overdrive. (If that's what you were after) But you could hook the different individually housed and powered gain stages together in any order you wished. So, you'd be building your own preamp, without dicking around with dangerous amp innards. (Yeah, I'm very familiar with Duncan's Convertible amps, thank you.) Obviously, one could never nail the total sensation of standing in front of an actual tube amp and playing. (Unless the amp were off-Duh!) However, artists such as Neil Young mic their amps and run them through a P.A. It might be reasonable to work towards simulating the feel of standing in front of a P.A. speaker fed by a signal from a mic'ed guitar amp. Phew! Done. I don't know if any of the ideas I've put forth might prove useful, but I just felt like throwing them out there. Before offering criticism, please reread the many caveats I've put in this. Mainly, my very limited background in computers and electronics, my over-active imagination, the hopelessly/ridiculously optimistic "technology-can-solve-anything" tone I took, and my less than perfect proofreading. Also keep in mind, this is my page, not your page, nor a USENET discussion group. So be cool. Thanks. I'm sure thar are lots of reasons why "X will never be possible because of Y". The low sampling rate of digital technology compared to the infinity of information offered by analog might be one thing I'll hear about. Thanks for reading getting this far. I hope you found at least a little something worth your while in all of this. Keenan monadyne@user1.Channel1.com Minor editing and formatting by jorman@aol.com http://users.aol.com/jorman/