Same amp head, and change the cabinet you connect and the sound changes greatly. Everyone knows this by experience. Without buying a new head, just swapping the cab makes it ring like a different amp entirely.
But then, of the final sound, how much is decided by the head and how much by the cab? Sort this out and your sound-making — and your way of recording — stops wavering. This piece takes apart the division of labor between head and cab, and connects from there to "what should be fixed when recording."
What the head makes, what the cab decides
Roughly, split the roles.
The head side makes the skeleton of the sound. The quality of distortion, the gain structure, compression, and playing feel. How it distorts at the front stage and returns at the power stage — much of the sound's "character" is decided here. When you play the same phrase, the rawness, tightness, and grain depend strongly on the head's design.
The cabinet side decides how that signal finally becomes air. The cab is the last analog gate that turns an electrical signal into vibration of air. Physics is packed in densely here.
The physics of speaker and box
Look a little more closely at the elements that decide the cab's sound.
- Speaker size: 8", 10", 12". The larger the size, the easier the low end and the different the throw. That 12" became standard for guitar amps has a compromise of range and manageability behind it.
- Cone and magnet: the weight and shape of the paper cone, the magnet type (alnico or ceramic) change compression, how the highs come out, how it breaks up. Even the same 12", a different model is a different thing — some tight and modern, some collapsing early into sweet distortion.
- Box structure: open-back or closed-back. Open lets the lows wrap around, spreading front and back. Closed tightens the lows and throws them straight forward. Even the same speaker, change the box and the low-end quantity and directivity change.
- Box material and volume: plywood type, thickness, internal volume, bracing. The box itself resonates and adds color.
These combine to decide the band balance and the sound's "rise and throw." Even the same head, a different cab reaches the ear as an entirely different sound.
Asking "how many percent" is actually hard
Honestly, cutting "how many percent is the cab" as a number is hard.
The head makes the center of the distortion. But how that distortion finally sounds is decided by its figure after passing the cab and mic. Even the same distortion, depending on the cab it passes, goes thick or thin, raw or smooth. Even if the head makes "wild distortion," if a tight cab pulls it together, what reaches the ear is a composed sound. Conversely, a mild head's sound can be pushed and widened by a raw cab.
So head and cab are close to multiplication, not addition. Neither alone decides the sound; both change each other's result. So rather than "how many percent," it's more accurate to think "both are independent variables."
With two variables, you can't compare
Here a recording problem arises.
You want to know the difference between heads, but change the cab along with it and you can't tell which caused the change. Record head A on a tight cab and head B on a raw cab, and you can't separate whether it's the A–B difference or the cab difference. Conversely, you want to know the cab difference, but move the head settings too and again you can't pin down the cause.
Move two variables at once and the comparison itself fails. Like a science experiment: to know one thing, you have to fix everything else.
So CSL fixes a reference cab
That CSL fixes one reference cabinet is for this reason.
Fix the cab and the difference between heads remains as the difference between heads. A reference cab doesn't mean "the best cab." It's a common yardstick for comparing amps on the same footing. If the yardstick's scale changed per model, you wouldn't have measured anything. So whichever model we record, we pass it through the same reference cab.
Mic position is the same idea. Mic positions are infinite, and a small move changes the sound. But CSL narrows to the representative points Center and Edge. Not to be exhaustive, but because it prioritizes being able to compare with conditions aligned. Prepare every position and options increase, but cross-model comparison becomes impossible. Fixing is a choice for comparability.
Combo amps are a slightly different story
That said, combo amps are a different case.
A combo is designed as one sound, with a speaker and box chosen specifically for that amp. Head and cab aren't separate; they make "that model's sound" as an integrated whole from the start. In this case, swap to a reference cab and the sound that model originally rang is lost.
So for a combo, CSL records with the combo's actual speaker in addition to the reference cab. The reference cab as a yardstick for comparing with other models. The actual speaker to keep that one amp's raw figure. The purposes differ, so we record both. Fixing for comparison, and the model's true voice — the two are made to coexist.
In short
Sound is decided by the multiplication of head and cab. The head makes the skeleton of the distortion; how it becomes air is decided by the cab — speaker size and material, open or closed box. So rather than cutting "how many percent," it's better to think both are independent variables.
With two variables, you can't compare. So CSL fixes the reference cab and mic position, and keeps the head difference as the head difference. Only for combos, it also records the actual speaker's sound. Fixing isn't unfreedom — it's a promise for discussing sound on the same footing.