GTS Learn · Saturation · Intermediate

Tape and Transformer Saturation: Why Real Gear Is Hard to Fake

5 min read

Tape machines and transformers both saturate, but not the way a simple curve does. How much they distort depends on the note: iron saturates the low end long before the top, and tape’s response shifts with level and frequency. That frequency and level dependence is a big part of what people mean by “analog”.

Distortion that depends on the note

Computed live in your browser; textbook curves, not a plugin's tuning

Transformers: the Lows Hit First

A transformer moves audio through magnetized iron, and the iron runs out of room first at low frequencies, where each cycle asks it to hold the most magnetism. So a transformer driven hard adds distortion, mostly 2nd harmonic, to the bass and kick while the vocal and cymbals pass almost clean. In the demo, the gold line is a curve with that behavior; the dashed line is the same curve fixed, distorting every note equally.

Tape: Gentle, Level-Dependent Squash

Tape saturates gradually: as level rises it compresses the peaks softly and adds a light 3rd harmonic. Its record and playback process also shapes the frequency response, with a bump in the low end and a top end that softens when it is hit hard. The result is a sound that gets rounder and denser as you push it, rather than suddenly breaking up.

Why One Curve Can't Fake It

A static curve gives the same harmonics at every frequency and every moment. Real gear has frequency dependence (iron), memory (tape’s magnetism and an amplifier’s power supply both respond to what came just before) and level-dependent tone. Modeling it well means measuring harmonics against level and frequency on the real hardware, then fitting stages that reproduce the whole map rather than one slice of it.

In GTS Plugins

Drive Strip is a transformer-drive channel strip built that way: its voicings are fitted to harmonic measurements across level and frequency in the GTS lab, then checked against them. Try the Tape and Hearth voicings on a bass and a vocal and listen to how differently the same drive treats the two.

Try It

Try this: drive the same saturation on a bass and on a hi-hat at the same input level. On a frequency-dependent saturator the bass thickens noticeably while the hat barely changes; on a static curve both change about equally.

Common Mistakes

  • Expecting “tape” to mean one sound. Machines, tape formulations, speeds and alignment all differ hugely.
  • Driving transformer-style saturation on the mix bus without checking the low end. It thickens the bass first, which can turn a mix muddy before anything else changes.
Go deeper: the maths Advanced · equations

The demo's model

A shelf boosts the lows before a fixed asymmetric curve, and the matching cut restores them after, so the overall tone is flat but the lows drive the curve harder. Real transformers do it through the iron's magnetic flux, which for a given voltage rises as frequency falls:

flux ∝ voltage / frequency

so halving the frequency doubles how hard the core is pushed. Tape adds hysteresis (its magnetization depends on its history) and a bias signal that linearizes it until it is driven past its range.

Keep Going

Hear measured, level-matched saturation on your own tracks: Drive Lite is free, and Drive Strip has a 14-day trial of the whole plugin.