When it comes to convincing simulation of guitar amps, cabinets, speakers and microphones, a few hardware manufacturers seem to be pushing the boundaries. Will the Torpedo blast the competition out of the water?
For guitarists and bassists, recording the intricacies of the very personal sounds we've painstakingly developed is a challenge. The studios where we could record at realistic volumes are largely gone, and we're much more likely to find ourselves in a small project studio or converted bedroom, where the presence of neighbours and compromised acoustics mean that recording at a 'real life' level is likely to be a complete no‑no.
In recent years, we've seen a legion of amp and cabinet simulators, running on computers or proprietary hardware. Most claim to deliver the tone and feel of our guitars or basses, played through a dizzying selection of amps, cabs and mics, and they do so with differing degrees of success — but I've not heard any that quite capture both the sound and feel of the interaction between me, my guitar and a real amp and cab.
In an attempt to provide this elusive feel, a smaller band of manufacturers took an entirely hardware‑based route, and developed emulators designed to be plugged into an amplifier's speaker output, in line with (or even replacing) the speakers. These emulators typically contain a dummy load and associated attenuation that can reduce an amp's output to its speakers to a neighbour‑friendly level or below. Often, using sophisticated EQ filters, they can provide a recording output that emulates the sound of a very limited selection of speaker cabinets. The success or failure of these products depends not only on the accuracy of their speaker emulations, but also on the reactivity designed into the dummy load, which defines the way that the emulator mimics the frequency‑dependent changes in the load impedance that a real loudspeaker presents to an amplifier. The more accurate this load-impedance replication, the better the amp's output stage can generate a facsimile of the negative feedback that it gets from a real speaker, thus ensuring that the output valve distortion and 'feel' of the amp are as close as possible to that which the player is used to from his or her usual speaker cab.
Two Notes Audio Engineering straddle both camps with their Torpedo VB101, a combination of dummy load and digital emulation, designed to replace or to run in parallel with an existing speaker cabinet. This device features convolution algorithms that are said to replicate accurately the sound that an amplifier would make in a studio through a selection of guitar and bass loudspeaker cabs and a range of mics and mic placements.
The Torpedo VB101 is a stylish, black‑bodied 2U rackmount device, whose silver and light grey front panel's most prominent visual features are the two blues of the large dot‑matrix screen and two‑digit numerical display, and the purple of the Two Notes logos.
The front panel is divided into five sections. On the far left you'll find the holes of the ventilation grille that cools the dummy load. The next division contains the input‑gain rotary encoder, with its associated nine‑segment LED ladder and overload indicator. The centre section is where the action is, carrying the two displays plus the large Navi(gation) rotary encoder and the switches necessary to set up the VB101, navigate its menus, adjust its parameters and load, recall, modify and save its presets. The output section follows, carrying the output‑gain rotary encoder, with an LED ladder and overload indicator identical to that of the Input section, together with a stereo headphone socket and its accompanying volume control. The last section features a power on/off switch.
The rear panel contains all the digital and analogue input and output connectors. Starting over on the extreme right you'll find the cooling fan. Next to it, the S/PDIF and AES-EBU digital sockets surmount the amp‑input and speaker‑output jacks and the XLR of the line‑level analogue input. Roughly central, the USB, MIDI In, MIDI Out/Thru, word clock in and the optional Ethernet connectors sit above the balanced XLRs of the mono Pre Simulation and stereo Post Simulation line‑level analogue outputs. Power is fed in via a fused IEC mains socket.
Despite the complexity, getting started is pretty simple: at the most basic level, you can plug your amp's speaker lead into the amp input on the VB101 (make sure you're using the 8Ω tap if it's a valve amp) and plug in a pair of headphones. Once you've powered up the VB101, a relay activates the dummy load — at which point you can safely turn on your amp.
Input and output gain can be set semi‑automatically. The VB101 boots up with the test preset 00 and with the SGA (Safe Gain Adjustment) function active. The SGA is Two Notes' proprietary automatic gain‑reduction function, designed to quickly correct any saturation of the input or output section that would result in digital overload. SGA can be switched off entirely or activated for input, output or both. As well as setting initial levels, the SGA will also deal with any aberrant level increases that occur from the player getting excited or from a preset recall, since input and output gains can be stored individually for each preset when global input/output gain settings are not being used.
To use the SGA on power-up, you need to turn off the output gain, headphone output volume, and input gains of any analogue or digital devices attached to the VB101. Then you set your amplifier controls, just as you would normally with a speaker attached, set the VB101's input gain to the maximum value for that function and play normally. You'll see the 'Gain Reduction' message on the main display, which disappears once the Torpedo has set the input gain to avoid clipping. Then you can set the maximum value for the output gain and play again until the same message disappears. That's the Torpedo gains set for your amp, so turn up the headphone volume and/or the input gains of any connected devices, select a preset using the Navi rotary encoder, press the OK key and you're good to go.
To get the best out of it, though, the Torpedo needs a bit of setting up. The user interface, in addition to the headphone volume pot and the three rotary encoders mentioned above, carries 10 switches which, in conjunction with the Navi encoder, give you screen‑based interaction with the menus governing Setup, Program Save and Compare, Speaker/Mic selection, Mic positioning, simulation Bypass, Post-simulation FX and its Bypass — plus the usual OK and Escape functions.
Entering the Setup menu allows you to configure the VB101. The Audio submenu lets you select the input source, SGA mode (input gain, output gain, or both), whether input gain should follow preset settings, and Output Mode, which allows the simulation to be panned both left and right, or paired with the dry signal.
The last two items in the Audio menu are player‑related. In the case of the VB101, the 'Normal' latency is less than five milliseconds, a delay which is equivalent to you being approximately five feet away from your speaker. This can be reduced to a 'Low' setting of below 3ms if required, to help deal with other latencies elsewhere in the system, but this reduction comes at the expense of losing the Overload parameter in the loudspeaker simulation, which may or may not be a reasonable trade‑off for you. Finally, Load Compensation activates a filter that's designed to correct for the changes in an amplifier's dynamics when connected to a non‑reactive dummy load, in order to restore the 'feel' of the amp to the player.
The Sync menu settings determine the VB101's clock source and, where applicable, the sampling rate of its outputs, from 44.1kHz to 96kHz, while a Device ID setting lets you give different IDs to multiple Torpedos connected via USB to the same computer... I should be so lucky!
The remaining functions and their menus govern the sounds. The Program switch lets you save and, using the Navi encoder and the OK key, name the presets that you create. The Compare switch controls Two Notes' neat implementation of the 'common compare' function. In this implementation, the function's set of temporary parameters is re‑initialised each time the Torpedo is restarted, so that it always loads the start-up preset. You can independently modify either the preset and its 'compare clone', or both. Selecting another preset doesn't overwrite the compare clipboard, so you can quickly compare a modified preset with an existing one or, by loading another preset into the clipboard, you can instantly compare two existing presets. You can load a different preset into the compare clipboard simply by going into Compare, recalling the preset and again holding the Compare key for a couple of seconds. The only slight issue is that recalling or loading a preset changes the name of both sides of the Compare function to that preset but, other than that, it's a very usable way of doing things.
As you'd expect, the Spkr/Mic switch lets you choose which emulated guitar or bass speaker cab and mic you want to use. There are eight mics and (currently) 30 cabinets, giving a possible total of 240 combinations, as well as a huge number of possible variants. Two Notes have pre‑loaded 36 presets to get you started, and you can add your own up to a maximum of 100. If you need more, the Torpedo Remote software allows you to store an unlimited number of presets on your computer.
If your favourite cab/mic setup doesn't exist, fear not! Not only are Two Notes frequently uploading new cabinet simulations for you to download, but the downloadable Torpedo Capture software enables you to record the sound of your own cab and mic in your room and use the results (saved as a TUR file) to create your own set of impulse responses. The Capture software will also allow you to load (and convert to TUR format) third‑party impulse responses in WAV format, or indeed to utilise any audio file in that format to create weird and wonderful impulse responses if you're into extremely experimental sonic textures.
In the Miking menu, your inner sound engineer can position the virtual mic relative to the virtual speaker. You have control over the distance from the speaker (0‑3m), and the mic's position relative to the 'Center' axis of the speaker. These two controls interact, so that the maximum off‑axis position is the edge of the speaker at 0m distance, and is 1m at maximum distance. You can also Position the mic at the front or back of the cab. Perhaps the most unusual control parameter is Variphi, where you control the distance between two separate simulated micropones. As you vary the distance, you vary the phase relationship between them. If you've used a cocked wah pedal to set the frequency content of your sound, you'll get the idea, the difference being that you've got 3m to play with, so there's a wider variety of periodic phase relationships.
The last two functions in the Miking menu are Overload, which simulates the distortion and compression that you get from overdriving a loudspeaker, and the dry/wet mix between the unprocessed sound and the selected simulation.
The final control switch brings up the Post FX menu, which includes a five‑band, non‑parametric EQ, an exciter, a compressor and a spatial enhancer, which creates a psycho‑acoustic, pseudo‑stereo effect that seems to work best over headphones.
Setting up the Torpedo VB101 for first use with your amplifier (with or without cabinet) is as simple an operation as the manual suggests, and tweaking the unit's setup parameters to suit your particular situation is a proverbial breeze. For the purposes of the review, I fed the VB101 from the loudspeaker output of my favourite test valve amplifier, a GE6L6‑loaded THD Univalve. I also ran the Univalve's transformer‑isolated unbalanced line output to the Torpedo's analogue input. The post‑simulation stereo and pre‑simulation mono analogue outputs went to my venerable Tascam SX1 digital mixer/recorder, which feeds two Tannoy active monitors. This setup allowed me to record and to directly compare the pre- and post-simulation outputs of the Torpedo and to evaluate its simulations when driven by the Univalve's line output, as well as allowing me to gauge the playing feel of the VB101's reactive load in comparison to that of the real cabinet.
Once I'd got the Torpedo set up to my satisfaction, using its dummy load in reactive mode, I started a quick run through its guitar cab presets, with a '67 Telecaster and a '72 Les Paul Deluxe as my sources, and the Univalve running pretty much clean. Without knowing the room in which Two Notes record their IRs, it's impossible to gauge just how accurate the mic/cab simulations are in absolute terms, but when I swapped mics on a cabinet, the sonic changes reflected those I'd expect to hear if I changed, say, a SM57 for an MD421 or a C451. I also heard the expected change when I swapped the simulations of a Marshall‑style 4x12 for a Deluxe 1x12 while using the same microphone. Taken together, these results would seem to be confirmation that the speaker/microphone simulations can be regarded as being pretty accurate. That the VB101's converters run at 192kHz/24‑bit and that all internal processing is at 96kHz/32‑bit floating-point might conceivably also have something to do with that apparent accuracy.
I was particularly impressed by the sonic potential inherent in the mic-positioning capabilities of the Torpedo. Spending time working on the distance and off‑axis position of the mic in relation to the cabinet is one of those tasks that, out in the real world, really repays the time and effort spent in getting it right. Repositioning a microphone to a virtual accuracy of 30mm in distance and 10mm off‑axis might be regarded as being a bit pernickety, but I've been known to run from control room to studio and back to do something very similar — though, admittedly, not always with such great (and repeatable) precision. It is to the Torpedo's credit that changes of that magnitude (or should that be minitude?) have an audible effect on the sound.
Comparing the pre- and post-simulation outputs showed just how effective the simulations are. The unsimulated output sounded just as characterless as you'd expect, but in a live environment (where you might want to feed a lower‑wattage guitar amp for stage monitor purposes, while simultaneously sending the sound of your main 100‑Watt head flat out through a 4x12 into the PA), this feature would prove invaluable. For recording purposes, this separation enables you to record a stereo simulated track together with an unprocessed mono track that could then be used to drive a different simulation, in order to give you a second sound (or third, fourth, fifth and so on) to mix in. The short time I spent mixing a 4x12 cab with the speaker distortion turned up into a 1x12‑based clean sound served to remind me just how powerful this approach can be.
Listening to the Univalve's fat, Les Paul‑based overdrive sound through the VB101's 4x12 simulations was a real pleasure. The Univalve has a very detailed sound, even when the power stage is being overdriven, and the VB101 kept that subtlety intact. Winding up the loudspeaker distortion parameter gave me the effect of having a high‑wattage Univalve driving speakers into distortion, and I found some really heavy lead sounds lurking in the corners, especially with the 4x12 simulations. I don't have the space to keep a 4x12 in my music room (plus I have neighbours), so the chance to run the Univalve through a succession of virtual 4x12 cabinets, balancing its power-stage distortion with the simulated loudspeaker distortion, was an experience to revel in.
The Post FX section's EQ, exciter, compressor and spatial enhancer all work as you'd expect. Whilst I can't easily think of a recording situation where I'd look to process the output of a digital system before tracking it, these effects will come into their own through a PA or if you're just using headphones to listen with.
Finally, I switched the VB101's input to the line output of the Univalve and tried driving the mic/cab simulations with that source. I felt that this way round I got perhaps a little more detail in the clean and crunch sounds, but that there was nothing much to be gained once I got into real overdrive territory. With this setup I was also able to compare the feel of the VB101's dummy load in reactive mode to that of the Univalve running into a real‑world speaker cabinet. I know exactly how my Univalve feels when driving the THD 2x12 that I normally use it with. Comparing the two, I was a lot happier with the feel of the real cabinet on clean and crunch sounds, but didn't really find an enormous amount to complain about on more overdriven sounds. In the past I have played through hardware dummy-load speaker emulators that gave me a better feel than either the VB101 or the Univalve, but bearing in mind that these emulators cost between 15 and 50 percent of the price of a Torpedo VB101, its dummy load performed well. For me, the most playable VB101 setup I came across was running the Univalve, its volume attenuated slightly by its internal Hotplate, into the THD 2x12 and driving the VB101 with the Univalve's line out — the proverbial best of both worlds.
The Torpedo VB101 dummy load and loudspeaker and microphone simulator is a piece of high‑quality professional equipment that rewards the time spent fine‑tuning its simulations to your requirements. I spent a fair amount of time working on the combination of its Deluxe 1x12 loudspeaker and close‑miked MD421 microphone, and got that pairing sounding very close to the result from my own Deluxe's 1x12 speaker with a close‑miked MD421. As always with simulations replaying through studio monitors, you can't get exactly the same sound and playing experience as being in the room next to an amp, since you haven't got its speaker flapping your trousers round your legs — but in the context of a track, I doubt that a listener would ever notice the difference.
What the VB101 does give you is access to a wide range of cabinets and microphones that you can use to tailor the sound of an amplified guitar or bass pretty precisely to the needs of the track or live situation that you find yourself in. You'll also find some interesting combinations: for example, I'd never thought of the Univalve as a bass amp, but it worked well on my five‑string put through a variety of the VB101's bass cabs. The ability to create your own IRs from real mics and cabs, and also to derive them from any WAV or AIFF audio file, is a major plus in my book. Other significant bonuses in this area are that Two Notes are continuing to add cabinets to the list month on month, and that they have also made available a public beta of Torpedo Exchange, a part of the Two Notes web site that the VB101 user community can use to exchange their self‑created TUR files.
Splendid though it is, the VB101 does face a lot of competition in the recording arena. In addition to the welter of floor and rackmounted preamp/amp/mic/speaker simulators that are already well established in the market at significantly lower price points, there is an increasing number of software programs and plug‑ins that do the same job as the VB101 for a lot less capital outlay. For live use, where I needed sounds from various cabs, I'd much rather take a real amp and a Torpedo on stage than any of the alternatives — which means that for my needs the Torpedo VB101 is virtually perfect.
As a high‑quality piece of kit designed for skilled, professional users, the Torpedo VB101 carries a pretty professional price tag. If you're a guitarist with a selection of great-sounding valve (and even solid‑state) amps, plenty of gigs and recording sessions, then the Torpedo VB101 should be at the top of your audition and shopping lists, because it will take you a long way towards getting the sounds you want, both on stage and in the studio. Mere mortals like me will lust after it, but we'll find its cost difficult — although certainly not impossible — to justify. I'll be very sorry to send it back... for now.
Two software programs (both Mac‑ and PC‑compatible) for the Torpedo can be downloaded from Two Notes' web site. Torpedo Capture allows you to generate your own impulse responses for the unit, while Torpedo Remote is parameter‑editing software that enables you to do anything that can be done on the Torpedo's front panel. This includes modifying, saving and transferring presets both on the VB101 and on the computer, as well as allowing you to manage, move and upload to the VB101 new Two Notes TSR loudspeaker data sets and the user‑generated TUR impulse responses measured or converted through the Torpedo Capture software. The Torpedo Remote is also the means by which firmware updates are loaded into the VB101.
The Torpedo Remote software was easy and intuitive to use, and did everything that was asked of it. The user interface needs a bit of a tweak here and there, as it doesn't quite follow the usual file‑movement paradigms, but other than that minor quibble, it does exactly what it needs to do.
Although I wasn't able to run Torpedo Capture in anger as part of this review — as I don't have the mono 250W into 8Ω transistor amplifier recommended by Two Notes as necessary to drive the loudspeaker during the measurement process — I can confirm that the software is extremely easy to use. This is pretty essential given that (according to the Two Notes web site) the instructions are currently available only in French on the Two Notes forum, and my grasp of that magnificent language is waning fast! However, a quick look round the forum turned up the English pages and an English capture tutorial.
To make a capture you'll need a computer with a soundcard or an audio interface capable of running at 24bit/96kHz, the loudspeaker cabinet and microphone that you want to capture, a mic preamp, a 250W into 8Ω transistor mono power amp with a flat frequency response, and a room where you can make a fair bit of noise.
Once you've plugged the mic into the preamp, preamp out to soundcard in, soundcard out to power amp in, power amp out into loudspeaker, you put the Capture software into test mode, turn the volume of the test signal through the speaker up to your normal playing level, and adjust the input level on the computer so that the bargraph meter on the Capture software reads ‑12 to ‑14dB. After that it's simply a matter of clicking on the Measure button, choosing where to store your TUR file, and waiting while the software sweeps through the frequencies from low to high and calculates the impulse response. After that, you'll have your very own mic/cab simulation ready to be loaded into your VB101.
There are a couple of things to bear in mind with a TUR‑based simulation. The bad news is that you can't move the mic position 'inside' the VB101, so you have to get that right before you start your capture, but the good news is that the sound of the preamp is also captured as part of the process, so you could record impulse responses from your cab using a selection of esoteric mic preamps and vintage microphones. The possibilities that this opens up are legion, even before you start adding IRs from third parties, and WAV files.