Menu

We Eliminated EQ Frequency Cramping Without Oversampling — Here's How (DAFx26 Paper)
🖼️

We Eliminated EQ Frequency Cramping Without Oversampling — Here's How (DAFx26 Paper)

DEV Community: cpp·Gary Doman/TizWildin·5 months ago
#paOU3lE9
#dev#sample#frequency#oversampling#topology
Reading 0:00
15s threshold

We Eliminated EQ Frequency Cramping Without Oversampling

If you've ever pushed a parametric EQ boost to 16 kHz and wondered why the curve looks wrong — you've hit frequency cramping . Traditional biquad filters using the bilinear transform exhibit 199% Q distortion at 16 kHz (44.1 kHz sample rate). The filter becomes almost 3× narrower than intended.

The industry solutions? Brute-force 4× oversampling (adds latency, burns CPU) or proprietary analog-matching curves (FabFilter's approach, closed-source).

We took a third path: the Simper State Variable Filter (SVF) .

The Results

Topology Magnitude at 16 kHz Error

RBJ @ 44.1 kHz +0.73 dB −5.27 dB (cramped)

SVF @ 44.1 kHz +6.00 dB 0.00 dB (exact)

RBJ @ 4× Oversampling +4.82 dB −1.18 dB

The SVF achieves exact gain at the center frequency — no oversampling, no proprietary tricks.

How It Works

The SVF pre-warps the cutoff frequency:

g = tan(π · fc / fs)

Enter fullscreen mode

Exit fullscreen mode

This single line eliminates the bilinear transform's…

Continue reading — create a free account

Join HashtagPLUS to read full articles, follow hashtags, vote, and join the conversation.