My wife and I have always been drawn to blues music. We particularly enjoy
Muddy Waters, John Lee Hooker, B.B. King, Howlin’ Wolf, Albert King, Freddie
King, Buddy Guy, T-Bone Walker and John Mayer.

There’s something about that music that I’ve always loved. It doesn’t need a
thousand notes. A guitar, an amplifier and a few well-placed notes can suddenly
turn into a conversation.

B.B. King is perhaps the ultimate example. He could make a single sustained
note, a bend and that unmistakable vibrato say more than many guitarists manage
in an entire solo.

I wanted to see how close I could get to that world using what I already have:
my guitar, an iRig and my Linux workstation.

The goal was simple. Plug in the guitar, play through virtual amplifiers and
effects, record when I want to, and keep the latency low enough that it actually
feels like playing through an amplifier.

What started with plugging an iRig into a USB port turned into an exploration of
ALSA, PipeWire, JACK, REAPER, Neural Amp Modeler, cabinet impulse responses and
TONE3000.

Blues as a conversation

One of the things I find fascinating about blues is that I don’t really think of
it as playing notes. I think of it more as communicating.

A blues phrase can sound like a question. The next phrase answers it. A bend can
sound uncertain, angry or pleading. Vibrato changes the character of a note
after you’ve already played it.

And then there is silence.

A pause can sometimes say more than another five notes. Keith Richards has often
talked about the importance of space and silence in music, the idea that the
notes you don’t play are part of the music too. That philosophy fits the blues
perfectly.

Silence creates anticipation. It gives the previous note somewhere to go. It
gives the listener a moment to absorb what was just said before the guitar
speaks again.

In that sense, a rest isn’t an absence of music. It’s part of the phrase.

The same thing happens when we communicate with each other. The literal message
is only part of what we hear. Tonality, timing and the pauses between words can
sometimes communicate more than the words themselves. The same sentence can
sound loving, angry, uncertain or sarcastic simply because of how it is
delivered.

At least in my humble opinion, blues works in much the same way. The notes are
the words, but the bend, vibrato, attack, dynamics and silence give those words
their meaning.

Perhaps that’s why relatively simple blues can be so difficult to play
convincingly. You can learn the vocabulary, the pentatonic scales, the blue
notes and the familiar licks, relatively quickly. Learning how to say something
with them is another matter.

If blues is a conversation, silence is the moment when you listen.

Major and minor: the vocabulary

The minor and major pentatonic scales provide a wonderfully simple vocabulary
for having that conversation.

The minor pentatonic gives you much of the tension, grit and melancholy we
immediately associate with blues:

minor pentatonic: tension, grit, question

The major pentatonic can give the same phrase a sweeter, warmer and more
resolved character:

major pentatonic: warmth, optimism, answer

The really interesting territory is often between the two.

B.B. King was a master of moving between major and minor colours within the same
musical conversation. Add bends, blue notes, vibrato, timing and, importantly,
silence, and five little notes suddenly become a surprisingly large vocabulary.

That’s how I’m trying to approach learning blues guitar. Instead of asking:

Which scale should I play?

I try to ask:

What am I trying to say, and which note helps me say it?

The amplifier and effects can change the voice. They can’t decide what the voice
has to say.

Rhythm: where you put the note matters too

There is another part of this idea that fascinates me. Rhythm isn’t only about
playing at the correct tempo. It’s also about where you place yourself inside
that tempo.

Jamaican music is a wonderful example. Reggae and Rastafari-influenced music can
create an extraordinarily relaxed feeling by leaning into the space around the
beat. Musicians can play slightly behind the pulse rather than attacking every
note mechanically on top of it.

I like to think of it as playing just a little behind the heartbeat.

That doesn’t mean playing slower than the band or playing out of time. The pulse
remains clear, but the note is allowed to arrive with a tiny amount of patience.

That tiny difference changes everything.

Ahead of the beat: urgency, tension, pushing
On the beat:       precision, stability
Behind the beat:   space, weight, relaxation

This isn’t literally a rule that Rastafari musicians play at some measurable
speed below a human heartbeat. It’s a way of describing a particular rhythmic
feeling. The groove can feel deep and relaxed precisely because nobody is
desperately trying to arrive early.

Blues has plenty of this too.

A guitarist can play exactly the same note over exactly the same chord and make
it feel completely different simply by moving it slightly in time. Come in early
and the phrase can feel impatient. Hit it squarely and it feels decisive. Let
the rhythm section arrive first and place the note fractionally behind it, and
suddenly the phrase can feel heavy, relaxed or almost conversational.

That brings me back to communication.

When we speak, we don’t deliver every syllable at mathematically equal
intervals. We hesitate. We accelerate. We leave a pause before an important
word. Sometimes we deliberately make someone wait a fraction longer for the
answer.

Music does exactly the same thing.

Perhaps the vocabulary isn’t just notes, bends, vibrato and silence. Timing
belongs there too.

The note tells you what was said.

The timing can tell you how it was meant.

Getting the guitar into Linux

My guitar connects to an IK Multimedia iRig USB interface.

Ubuntu recognized it immediately:

lsusb

The iRig appeared as:

1963:00c2 IK Multimedia iRig USB

ALSA could also see the recording interface:

arecord -l

which showed:

card 4: USB [iRig USB], device 0: USB Audio [USB Audio]

Before touching REAPER, I tested the iRig directly with ALSA. Once I could hear
a clean recording of the guitar, I knew the guitar, iRig and Linux USB audio
stack were working.

That made troubleshooting everything else considerably easier.

PipeWire, JACK and REAPER

My setup has one complication. The input and output are different devices.

The guitar enters through the iRig, while my speakers are connected to my
monitor over the RTX 4090’s HDMI output.

The audio path therefore looks like this:

Guitar
  |
iRig USB
  |
REAPER
  |
NVIDIA HDMI
  |
Speakers

Ubuntu was already running PipeWire:

pactl info | grep "Server Name"

which reported:

Server Name: PulseAudio (on PipeWire 1.0.5)

I installed PipeWire’s JACK compatibility:

sudo apt install pipewire-jack

and started REAPER through it:

pw-jack ./reaper

Inside REAPER I selected:

Options > Preferences > Audio > Device > JACK

PipeWire then handles routing the iRig input through REAPER and sends the result
back out through HDMI.

REAPER doesn’t have to record

This was probably my favourite discovery.

Create a track, select the iRig as its mono input and enable Record Arm and
Record Monitoring.

You don’t actually have to press Record.

REAPER processes the incoming guitar in real time:

Guitar > iRig > REAPER > FX > Speakers

This was one of those small discoveries that completely changed how I thought
about the DAW. It was Alessandro Duca who taught me this little trick.
REAPER wasn’t merely somewhere to record the guitar anymore.

It had become the amplifier.

Sometimes it’s the simplest piece of advice that completely changes how you use
a piece of software.

Adding real amplifier models

REAPER’s effects were enough to prove that the setup worked, but the next step
was to start playing with actual amplifier models.

I experimented with Neural Amp Modeler (NAM) and its native Linux LV2 plugin,
which allows captures of real amplifiers to be loaded into REAPER.

Then I discovered TONE3000, which provides native Linux plugins including LV2,
VST3 and CLAP.

For REAPER I installed the LV2 version under:

~/.lv2/TONE3000.lv2

The interface needs WebKitGTK on Ubuntu:

sudo apt install libwebkit2gtk-4.1-0

After restarting REAPER, I could browse and load amplifier models directly from
the FX chain.

Naturally, I went looking for B.B. King.

Chasing B.B. King’s sound

B.B. King’s famous later setup was refreshingly simple:

Lucille
  |
Lab Series L5
  |
2x12 cabinet

I found a B.B. King Lab Series L5 preamp capture.

Because it’s a preamp capture rather than the complete amplifier and speaker
combination, I added a cabinet IR afterward.

My version looks like this:

Epiphone Les Paul
  |
Neck humbucker
  |
iRig
  |
REAPER
  |
Lab Series L5 preamp
  |
L5 2x12 cabinet IR
  |
Reverb
  |
Speakers

For the cabinet capture I started with the Shure microphone version.

My Les Paul obviously isn’t Lucille, but the neck humbucker through a warm,
clean L5 model gets into very interesting territory remarkably quickly.

The rest is in the fingers.

Unfortunately.

Record clean, listen through the amplifier

There’s another useful consequence of this setup.

REAPER can record the clean guitar signal while I listen to the processed
version:

                    Recorded clean guitar
                           |
Guitar > iRig > REAPER > L5 > Cab > FX > Speakers

The performance and the amplifier aren’t permanently tied together.

After recording, I can replace the L5 with something completely different. I can
change the cabinet, microphone, reverb or EQ while leaving the original
performance untouched.

In practice, I can concentrate on playing first and worry about the exact
amplifier sound later.

Automating effects

Once the guitar is inside REAPER, effects don’t have to behave exactly like
physical pedals.

A wah pedal is a good example. I can record the guitar first and then automate
the wah movement afterward.

The same applies to volume, pan, reverb, delay, gain, EQ and many other plugin
parameters.

The guitar itself is recorded in mono, but that doesn’t mean the entire mix has
to remain mono.

I can record a second performance and pan the two takes:

Take 1                                      Left
Take 2                                     Right

The small differences between the two performances create the stereo width.

Another approach is to keep the guitar in the centre and use stereo delay and
reverb to create the space around it.

I particularly like that approach for blues. The guitar remains focused and
intimate while the room around it becomes larger.

The final setup

What I ended up with is essentially this:

Guitar
  |
iRig
  |
PipeWire / JACK
  |
REAPER
  |
Amp model
  |
Cabinet IR
  |
Effects
  |
HDMI
  |
Speakers

What started as an experiment to get my guitar working properly under Linux
ended up giving me an entire virtual guitar rig.

But perhaps my favourite discovery from the whole experiment has nothing to do
with Linux.

I started this because I wanted the technology to help me get closer to the
sound of the blues musicians my wife and I love.

That led to PipeWire, JACK, REAPER, NAM, TONE3000, amp captures, cabinet impulse
responses, microphones, automation and signal routing.

After building all of that, I ended up back at the same lesson I started with.

The technology can give me B.B. King’s amplifier.

It can approximate his cabinet.

It can give me compression, reverb and EQ.

It can even move the wah pedal for me.

But it cannot decide when to bend a note.

It certainly cannot decide when not to play one.

And it cannot decide whether that note should land exactly on the beat or linger
just behind it.

If blues is communication, the equipment gives us the voice.

We still have to figure out what we want to say.

I still have lots to learn to be able to play well, but for now it’s food for
the soul.