Trace the history of the guitar amp from the design side and you notice one value being polished again and again.
The interaction of picking and distortion.
The large input born the instant a string is struck — how distortion receives it, crushes it, returns it. From the Fender Bassman to Marshall's master-volume amps, Mesa/Boogie's cascading gain, the Soldano SLO-100's gain distribution, the Diezel VH4's separation — the lineages differ but the destination is close: making the attack feel good.
What people call "playing feel" mostly lives right here.
So what happens if a player judges that entire value unnecessary?
Read Allan Holdsworth's gear choices from there, and every one of them makes sense.
Starting point: he wanted to be a saxophonist
Holdsworth said, all his life, that he never really wanted to play guitar. As a child he longed for the saxophone, and he openly named horn players — John Coltrane, Cannonball Adderley, Charlie Parker, later Michael Brecker — as his influences.
This isn't just an anecdote. It's the root of a performance philosophy.
The guitar is, structurally, very close to a percussion instrument. The volume peaks the instant you strike, then decays. In principle, the player can touch the sound only at the moment of attack. Vibrato and bends add something afterward, but that's bending a decaying sound — not re-launching the sound itself.
A wind instrument is different. As long as there's breath, you can reshape a sounding note after the fact — louder or softer, brighter or darker.
What Holdsworth worked at his whole life was closing that gap.
His legato is famous not because it's fast. The direction of causality matters: he arrived at that technique to erase the distinction between a picked note and an unpicked one. Erase it, and a phrase stops being a sequence of strike-points. It becomes a single line.
In other words, he was trying to remove the strike-point from the guitar.
Once you're here, the gear decisions make themselves.
The decision to step down from Marshall
In the '70s Holdsworth used Marshall — a 50W head and 4x12. Soft Machine, Tony Williams Lifetime, Gong, U.K., the Bill Bruford band. Early on, a Vox AC30, a Hiwatt, and the British Burman ran through too. His sound in that era was, broadly, British tube amps.
But he stepped down from it. After that, the amps he made central were the Norlin-era Lab Series L5, the British Hartley-Thompson, and the American Pearce G1.
All three are 100% transistor, solid-state amps.
By the common sense of the time, that was a strange choice. The late '70s to early '80s was exactly when transistor amps were being fixed as "cheap substitutes." Those after high gain went to boutique tube amps; only those after clean stayed with solid-state machines like the Roland JC-120.
Holdsworth chose solid-state for a reason that fits neither camp.
The Lab Series L5, a solid-state amp from under the Norlin umbrella (the same capital as Gibson), has a very clean, uncolored, even clinical sound. Holdsworth used it mainly for chords, inserting a Dynacord digital delay between two of them for depth. The Hartley-Thompson, also transistor, was a touch warmer and became his lead center in the early '80s.
And the Pearce G1 clinches that this choice was no accident.
This all-solid-state amp, released in 1982 by Dan Pearce, came as the G1a combo, the G1b powered rack head, and the G1c preamp-only unit. It splits a clean channel from an overdrive channel and includes compression/limiting and multi-band filtering.
The key point: Holdsworth was both an early endorser and a voice in the design. The later spec change known as the "1.5 mod" — adding gain and low end to the lead channel — is said to have been at his request.
So he wasn't settling for an off-the-shelf transistor amp. Taking solid-state as the premise, he had it shaped toward his own demands.
That's an entirely different motive from the era's usual "use transistors as a cheap substitute."
Line up the properties a tube amp has, and you can see what he was avoiding.
Sag. The rectifier can't supply enough current, so the voltage drops the instant you dig in; the sound swells a beat late.
Power-stage compression. The harder the input, the more the peak is crushed, shrinking the gap between strong and weak.
Output-transformer saturation. The low end's rise is dulled and pushed back.
These are all temporal distortions that happen around the attack. And in rock guitar amps, that very distortion has been prized as the source of pleasure. The design philosophies of Soldano and Diezel I wrote about were, at bottom, ways to pull out this temporal distortion deeply without muddying it.
For Holdsworth, these were, at the very least, things he "couldn't have happening on their own."
For someone who wants to erase the strike-point, a device that emphasizes the strike-point is unnecessary.
Solid-state responds honestly to the input. Hit hard and it's hard, hit soft and it's soft, with no delay. Less distortion also means the player's control of time passes through intact. He used solid-state not as a "substitute for tubes" but for a property tubes don't have.
That said, if we end the story at "he abandoned tubes," his later gear becomes inexplicable. We'll get to that.
Incidentally, Holdsworth's modified Hartley-Thompson is also known for the solo Eddie Van Halen recorded on Michael Jackson's "Beat It." A player who's a symbol of high gain recorded a signature solo on an amp born from the opposite philosophy. A design's intent and its use don't always match.
The role of distortion is reversed
Here we need to unbundle the word "distortion" once.
In rock, distortion plays two broad roles.
One is tonal. Harmonics increase, the sound thickens, it turns aggressive.
The other is temporal. Distortion crushes loud sounds and lifts soft ones. As a result, volume variation shrinks, decay slows, and notes sustain longer.
Mainstream high-gain amps have been designed for both at once — with the tonal side as the lead. The texture of the distortion is the product value; words like "grain" and "separation" are used to evaluate it.
Holdsworth's usage leans almost entirely toward the temporal side.
What he wanted was a state where volume variation is small, decay is slow, and the sound keeps ringing at an even thickness no matter where you take it in a phrase. To approach a wind instrument's sustain, you need that. So he needed gain, but it had no need to be aggressive at all.
His later setup backs this up. When he used two Yamaha DG-80 112s, one was clean and the other had its gain pulled back to a very shallow crunch, with the master volume opened wide to compensate.
That's almost the reverse of a normal high-gain approach. Not distorting to gain volume, but gaining volume without distorting. Distortion is added only enough to sustain the note.
The very value placed on the amount of distortion is different.
He didn't refuse tubes. He refused things moving on their own
Here we need material to keep the story from getting too simple.
Holdsworth returned to tube amps.
In the late '80s, Mesa/Boogie enters his rack — a Mark III head, a .50 Caliber, a Quad Preamp, a Simul-Class 295 power amp. From the '90s he even used a Dual Rectifier. All gear a solid-state purist would never pick.
So the summary "he's the guy who abandoned tubes" is inaccurate.
What was going on, then? The answer is in the device attached to the output of those Mesas.
The Juice Extractor. A signature device Holdsworth conceived in the mid-'80s and Rocktron later productized.
Inside, it's a reactive load box. It receives the amp's output in place of a speaker and converts it to line level. Not an attenuator like a power soak, but a replacement for the speaker itself. He sent the signal through it into a solid-state power amp such as a Carvin.
Holdsworth's stated aim was clear. Taking only the preamp output loses a large part of a tube amp's sound, because power-stage saturation and the preamp/power interaction drop out. He wanted to keep that, and record at conversation volume.
In other words, he wanted the power stage's nonlinearity.
This is the important part. What he refused was not tubes. It was the conditions that inevitably come with a tube amp.
That the power stage only works at high volume. That the day's room sound gets mixed in. That there's no guarantee of making the same sound again tomorrow.
These are elements that disturb, from the outside, the "time" he was trying to design.
So he cut the power stage off from volume. After the Rocktron version was discontinued because misuse caused amp failures, he kept building his own, calling it "The Harness" — even making two versions, one for amps with a master volume and one without.
Here we need one more piece to avoid jumping to conclusions.
He did not abandon the speaker. On the contrary, he was obsessive about the speaker and the mic to a degree quite abnormal for a guitarist.
The Coffin — a box for fixing things in place
Holdsworth built his own speaker enclosure — a box he called "The Coffin."
Inside a fully sealed box, he enclosed the speaker and the mic together. The mic was a Neumann U87. The baffle slid in and out so he could swap speakers as needed — 8", 10", 12", 15", Celestion, Jensen, JBL. The mic stand could move, but once the sound was set, it was fixed in that position and not moved.
He built three sizes: a portable small one, a slightly bigger one, and a very large one. The largest box was the best, he said. The reason is plain: the bigger the box, the lower its internal resonant frequency, dropping below the guitar's audible range — and then it's easy to EQ out.
His motive, in his own words, is clear too. Guitarists waste enormous time in the studio moving the cab and mic around chasing the same sound. That's a waste.
With this box, he could load it into a car and take it. Plug the amp into one end, the console into the other. As long as the mic stays in the same position, the sound is the same.
This is not a "device that removes the speaker." The opposite. It's a device for fixing in place the speaker and mic — the layers that misbehave most.
How serious he was about that fixing is shown by the fate of a 2002 Tokyo recording. During the making of the live album All Night Wrong at the Roppongi Pit Inn, staff wary of theft removed all the mics set up on the first day before the second. It could have been rebuilt. But Holdsworth refused, and said, in effect, "forget it."
If it can't be returned to the same position, it isn't the same conditions. For him, mic position wasn't a finishing step you adjust later — it was on the side of the preconditions that make a performance hold together.
Line these up and you see that the Juice Extractor and The Coffin are products of the same idea. One removes volume, the other the room and mic position, from the variables.
He wasn't reducing layers. He was fixing, one after another, the variables he didn't want to move.
And the one thing he deliberately left free after all that is the time of the sound.
This is not a mere recording trick. It's a question of how much of the amp you place under your own control.
His relationship with Rocktron doesn't end there. In the '90s Rocktron released the ProGap preamp loaded with artist presets, and among the preset providers, alongside Steve Lukather, is Holdsworth's name. His rack held an Intellifex too. Tube, transistor, digital, and homemade — there's no trace at all of caring about the pedigree of the means.
It's not that he didn't care. He cared about something else.
The control layer was outside the amp
One more thing defined his sound decisively: the spatial effects.
That peculiar spread when you play a chord. What many remember as "Holdsworth's tone" is, in truth, often not the amp's tone but this making of space.
He didn't use delay as a device for adding reverberation.
The lineup he used was broad: starting with the Dynacord delay between two '70s Lab Series L5s, then in the '80s an A/DA STD-1 stereo delay, a Yamaha E-1010 analog delay, a Lexicon PCM60, a TC Electronic Spatial Expander, and a Rocktron Intellifex. Different generations, different methods — but what they do is consistent.
Its culmination is Yamaha's UD-Stomp. Born from his relationship with Yamaha's Ryuichiro "Rick" Kuroki, it started from Holdsworth's own idea of an "ultimate delay machine." Inside are eight delay lines, each with independent feedback and panning. Even after it was discontinued in 2004, his fans keep hunting for one.
Eight independent delay lines is excessive as spatial decoration. But think of it as a tool for placing one note at eight positions in time, and the story changes completely.
With delay, he was designing the temporal thickness of a single note. Not reverberation — closer to voices.
The same applies to why he moved to Yamaha's DG series. He met the DG-1000 preamp, later moved to the DG-80 combo, and in his final years arrived at a setup pairing a Hughes & Kettner ZenTera or Switchblade 100 with a Yamaha DG130 power amp. He also used the MagicStomp as a successor to the UD-Stomp.
British tube amps. The transistor machines of Norlin, Hartley-Thompson, and Pearce. Mesa's tube racks plus a homemade load box. The digital of Yamaha and Hughes & Kettner.
In forty years, he lapped the methods about four times.
It's not that he had no attachment to method. He had no reason for attachment to method.
What he wanted to secure was the evenness of the attack, the sustain, and a space to place it in. If he got that, it didn't matter whether the inside was glass, silicon, or DSP. Using a digital modeling amp not to borrow a famous amp's sound but to stably reproduce his own settings — this attitude departs sharply from the average guitarist of the time.
The state of having the control layer outside the amp rather than inside it, he had essentially completed by the '80s.
The right to touch a sound after it's made
And where his philosophy is most exposed is the SynthAxe.
From 1986's Atavachron on, Holdsworth began using this guitar-shaped MIDI controller in earnest. A strange instrument said to have had only around a hundred units made, he's known as one of its few serious users.
What matters is the breath controller attached to it.
Designed by Nyle Steiner, inventor of the EWI, this input device changes volume and tone by breath strength. Holdsworth used it to control the attack and decay of the SynthAxe's sound, and sometimes the timbre itself.
In other words, he obtained a means to reshape a sound already ringing, after the fact.
After striking, make it louder. Softer. Add vibrato. Increase harmonics. All done post-attack.
He said, in effect, "using breath, I feel I'm playing a wind instrument — which is what I think I should have done." With the guitar, a percussion-like instrument, the thing he'd spent nearly thirty years taking the long way around to approach was now directly within reach.
By this point, his gear history should look like a single line.
Stepping down from Marshall, having Pearce reworked, treating distortion as a tool for sustain, cutting the power stage off from volume, fixing the speaker and mic in a box, having eight delay lines built, arriving at the breath controller — all of it does the same thing.
Widen the region the player can touch within the time a sound is ringing.
That's all.
Those who chase tone, and those who chase time
There's almost no record of Holdsworth saying he was satisfied with his sound. The opposite, if anything — he kept voicing quite harsh judgments of his own playing and tone.
Meanwhile, guitarists worldwide have tried to reproduce "Holdsworth's tone" — which amp to use, which chorus and delay to chain.
But copying what he did as a list of model numbers is probably the farthest road. That sound wasn't made by the Hartley-Thompson, nor by the UD-Stomp. Still less does matching the model numbers without reproducing the conditions he fixed — that drops the very part he cared about most. Erase the strike-point, even out the sustain, touch the sound while it rings: one policy running through every gear choice is what produced it.
The philosophy came first, and the gear was decided after.
I think this is a useful viewpoint when surveying the whole history of the guitar amp. Many great amps are designed as answers to a certain performance philosophy. High-gain amps polished the pleasure of the attack because there was a musical demand for it. Holdsworth chose the opposite only because he had a different demand.
Neither is right. But choose gear without knowing which side of the demand you're on, and things won't mesh.
Deepen the distortion and the feel doesn't improve. Add spatial effects and it doesn't feel wider. Often, in such moments, the problem isn't the quality of the gear but that whether you want tone or time hasn't been sorted out.
Do you want to stand the attack up, or erase it? Use the decay, or extend it? Place a note as a point, or draw it as a line?
Holdsworth's gear history is still worth reading because he demonstrated that, with a lifetime's consistency.
Tone is something you choose.
Time is something you design.
Audio
- Allan Holdsworth,
Metal Fatigue(from the album of the same name / ℗1985 Allan Holdsworth), from the official "Allan Holdsworth - Topic" channel. Note: the streaming version is a remaster.