Guitar rig architectures
A frequent question among guitarists struggling to understand the available technology is "In what order should I connect all of my gear?". The most common answer seems to be "Do whatever sounds good to you", which is correct without being helpful.
It turns out that there are some simplifying principles one can apply. I explain this in terms of "architectures" - or organizing principles - for guitar rigs. Despite the apparently endless proliferation of detailed configurations, every guitar rig can be classified into one of three architectures. I'll describe these first in terms of the signal chain.
- guitar > fx1 > preamp > amp > speaker
- guitar > fx1 > preamp > fx2 > amp > speaker
- guitar > fx1 > preamp > fx2 > amp > iso-box (or load) > fx3 > amp > speaker
Architecture one is the classic "low-gain" architecture, dating back to the genesis of guitar amplification. The distinguishing feature of this architecture is that all effects (if present at all) appear before the preamp. Historically this architecture is associated with the period from the 1950s through the early 1970s. In terms of amplification, think of the classic amps from the "big three": Fender, Marshall and Vox. Effects were limited in scope and diversity. Fuzz, wah, and tape echo were the most common, supplemented somewhat later on by octave fuzzes, compressors, and phasers (including the Univibe).
The Leslie tone cabinet is one additional effect that saw the most use during the period in which architecture one prevailed. This fits into the architecture not in the FX position, but rather as a variation upon the speaker. Most players found that the Leslie's built-in amplifier didn't have enough power to keep up with the sound levels of the day, so they'd upgrade the Leslie's speakers to handle more power and drive them directly from the output of a guitar amplifier.
Architecture one "works" because the amount of distortion introduced after the effects is relatively low and the playing style associated with this architecture tends to "embrace" the sound of distortion following effects.
Architecture two evolved as amplifiers were hot-rodded to deliver more distortion by adding additional tubes to the preamp section. The added gain didn't interact well time-based effects. By this time guitar playing had become more technical; with less space between notes a distorted echo simply doesn't sound good. Other time-based effects became more prevalent during this period with the introduction of bucket-brigade analog delay chips and early digital delays.
Common application of architecture two puts EQ effects (wah, filters) and amplitude effects (treble boosters, compressors, linear boosters) into the FX1 position and time-based effects (echo, chorus, flanger) into the FX2 position.
There's a bit of crossover between architecture one and architecture two. I tend to associate architecture two with the introduction of higher-gain, hot-rodded amps. However, lower-gain amps having built-in reverb are technically members of architecture two since the reverb appears between the preamp and power amplifier stage. However, this is more of a technical distinction since the preamp in these amps was most often run clean. When the preamp runs clean it doesn't make much difference at all whether the reverb appears before or after the preamp in the signal chain. For Leo Fenders's engineers, the decision to put the preamp ahead of the reverb was almost certainly made for reasons of cost.
Architecture three emerged from the studio to appear in stage rigs - probably during the 1980s and 1990s - as guitarists became more attuned to the contribution of power amp distortion, speaker breakup and speaker cabinet resonance. Effects positioning is similar to architecture two except that time-based effects may also be positioned in the FX3 slot, and EQ effects may be introduced into both FX2 and FX3.
Architecture three depends upon either isolation for the mid-chain speaker (followed by a microphone and flat preamp) or a load box. The load box is useful for getting the contribution of power-amp distortion at the expense of having speaker breakup as a component of your tone. On the other hand, a load box is much smaller and lighter than an isolation cabinet.
A modeling rig is an architecture 3 rig where most of the chain is virtualized in algorithms.