In the '90s, guitar amps had a strong desire to "make many voices in one box."
Mesa/Boogie's TriAxis and Marshall's 6100 are separate answers to that desire. The aim is nearly the same: hold many voices, from clean to lead, in one box, all in tubes.
But the forms that emerged were opposite. The TriAxis a rack, the 6100 a head.
What I want to say here isn't the surface story of "a rack saves space, a head is easy to carry." The difference of form reaches deeper — into how the power stage is handled and where the voices are made.
Let me line them up head-on, on three axes.
Axis 1: is the power stage "chooseable" or "fixed"?
The biggest difference is here.
The rack TriAxis is a front-stage (preamp) device only. The power amp that finally pushes the sound out is prepared separately. Whether you build a 395 Simul-Class or a 2:Ninety changes the sound considerably. So in a rack, the back stage is open, and you choose and fill it yourself. The sound isn't completed inside the box alone.
The head 6100 is integrated from front stage to output stage. The power stage is part of this head and can't be swapped. Each channel is made assuming it rings on this output stage. The sound is completed inside the box.
This isn't a difference of packaging. It's a difference of "who decides the final push of the sound." The rack leaves it to the user; the head has the designer decide and close it.
Axis 2: are the voices made "at the front stage" or "as a whole amp"?
The next difference is how the many voices are made.
The TriAxis holds voices as front-stage modes. It abstracts the circuits of successive Marks, prepares several voices at the preamp stage, and recalls them by preset. Making the voices apart happens mainly at the front stage. The back stage is their common exit.
The 6100 has a design where its three channels each drive the power stage independently. Clean, crunch, and lead aren't thinly branched sharing a front stage — they're built as separate circuits as far as possible, reaching the output stage. Each voice is close to "one amp."
Put another way, the TriAxis rings a front-stage dictionary of voices on a common back stage. The 6100 packs several near-complete amps' worth into one box. Even the same "many voices," the way the voices are made faces the opposite direction.
Axis 3: is the distance to the player "assemble and recall" or "turn on the spot"?
Last is the relationship with the user.
A rack makes a relationship of assembling and recalling the sound. Make settings in advance, memorize by number, call with a floor controller. Assemble into one system together with effects and other gear. Strong in reproducibility and integration. Sound-making is close to the sense of designing gear.
A head makes a relationship of turning the sound on the spot. The panel before you is within reach; turn a knob and it changes at once. Step and the channel switches. Carry, connect, ring right away. Strong in responsiveness and immediacy. Sound-making is close to the sense of playing an instrument.
The design difference seen in axes 1 and 2 becomes, directly, a difference of usability. The rack, choosing the back stage and assembling voices at the front stage, becomes a "design and recall" tool; the head, fixing the back stage and packing voices into one box, becomes a "turn and ring right away" tool.
Neither is superior
Whichever of the three axes you take, no ranking forms.
The rack suits those who want to choose even the back stage and recall a crafted sound accurately. The head suits those who, in exchange for fixed insides, want to carry and ring right away and tweak directly at hand.
That both existed side by side in the '90s is because what was demanded wasn't one thing. Both the studio's reproducibility and the stage's responsiveness are needed. Not superiority — a difference of premise.
The meaning of recording the two forms at CSL
The differences so far come out directly as "differences of sound."
The sound of a rack where you can choose the back stage, and the sound of a head where the back stage is fixed. A voice made at the front stage, and a voice where a whole amp's circuit rings. These don't convey by comparing spec sheets. Ring them, line them up, and only then is the difference known.
CSL records those two forms as forms. For a rack, which back stage it was paired with; for a head, which channel. Keep them by condition and you can compare "the rack sound" and "the head sound" on the same footing.
Even the same multi-feature, rack and head become different sounds. There's meaning in keeping that difference as sound.
In short
The TriAxis and 6100 realized the same aim of "many voices in one box" in opposite forms.
The rack leaves the back stage open to be chosen and assembles voices at the front stage. The head fixes the back stage and packs near-whole-amp circuits into several channels. As a result, both the sense of sound-making and the sound that comes out change.
Rack or head? It wasn't a story of saving space but of where, and by whom, the sound is completed. CSL keeps those two forms, lined up as sound.