Elimination of Retinal Magnocellular Flicker Fatigue and Cortical Mach Band Contours in the Android Display Pipeline via Zero-Hertz Surface-Locked Spatial TPDF Blue-Noise Quantization

Modern mobile and television displays operating under the Android operating system (Android 15, 16, and 17) exhibit a fundamental neuro-ergonomic conflict at the final quantization boundary between high-precision internal composition buffers (FP16) and physical display panels (8-bit or 10-bit OLED/Mini-LED). Currently, System-on-Chip (SoC) Display Processing Units (DPUs)—such as Qualcomm Snapdragon DSPP, ARM Mali DPU, and Google Tensor display pipelines—rely on hardware Temporal Dithering (Frame Rate Control / FRC) oscillating at 60 Hz to 120 Hz to mask bit-depth truncation. This paper demonstrates that while retinal photoreceptors (cones and rods) integrate rapid luminance fluctuations biochemically via cyclic GMP (cGMP) hydrolysis (acting as an RC low-pass filter with a time constant τ ≈ 30 to 50 ms), temporal dithering below 90 Hz bypasses conscious perception to continuously stimulate fast magnocellular parasol ganglion cells (M-cells) in the human retina. This unmitigated sub-threshold M-cell firing projects to the superior colliculus and primary visual cortex (V1), inducing oculomotor instability, visual fatigue (asthenopia), and migraines in flicker-sensitive populations. Conversely, disabling hardware temporal dithering without spatial compensation exposes hard quantization steps (contour banding), which activate retinal center-surround lateral inhibition and flood visual cortex V1 with false edge action potentials (Mach bands). We present a complete, zero-overhead architectural remediation for the Android SurfaceFlinger (Skia Graphite Vulkan backend) and HWC3 pipeline: Hardware Temporal FRC Suppression: Disabling temporal register rotation inside the SoC DPU / DRM-KMS display driver (0 Hz temporal modulation). Surface-Locked Spatial TPDF Blue-Noise Quantization: Injecting a single-cycle Triangular Probability Density Function (TPDF) spatial blue-noise pass into SurfaceFlinger's RenderEngine, phase-locked 1:1 to physical screen pixel pitch and advected by view/scroll translation vectors d(x, y) with analytical sub-pixel bilinear variance compensation. This architecture eliminates both magnocellular temporal flicker (0.0 Hz retinal modulation during static viewing and smooth pursuit eye movements) and cortical Mach band contours, requiring only an 8 kB L1-resident lookup texture and fewer than 6 ALU instructions per pixel.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-10-16
DOI
https://doi.org/10.5281/zenodo.23406317
Primary Topic
Image and Video Quality Assessment
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

Elimination of Retinal Magnocellular Flicker Fatigue and Cortical Mach Band Contours in the Android Display Pipeline via Zero-Hertz Surface-Locked Spatial TPDF Blue-Noise Quantization

Michal Mazgal
Zenodo (CERN European Organization for Nuclear Research)
Image and Video Quality Assessment
preprint

Elimination of Retinal Magnocellular Flicker Fatigue and Cortical Mach Band Contours in the Android Display Pipeline via Zero-Hertz Surface-Locked Spatial TPDF Blue-Noise Quantization

Michal Mazgal
preprint en

Abstract

Modern mobile and television displays operating under the Android operating system (Android 15, 16, and 17) exhibit a fundamental neuro-ergonomic conflict at the final quantization boundary between high-precision internal composition buffers (FP16) and physical display panels (8-bit or 10-bit OLED/Mini-LED). Currently, System-on-Chip (SoC) Display Processing Units (DPUs)—such as Qualcomm Snapdragon DSPP, ARM Mali DPU, and Google Tensor display pipelines—rely on hardware Temporal Dithering (Frame Rate Control / FRC) oscillating at 60 Hz to 120 Hz to mask bit-depth truncation. This paper demonstrates that while retinal photoreceptors (cones and rods) integrate rapid luminance fluctuations biochemically via cyclic GMP (cGMP) hydrolysis (acting as an RC low-pass filter with a time constant τ ≈ 30 to 50 ms), temporal dithering below 90 Hz bypasses conscious perception to continuously stimulate fast magnocellular parasol ganglion cells (M-cells) in the human retina. This unmitigated sub-threshold M-cell firing projects to the superior colliculus and primary visual cortex (V1), inducing oculomotor instability, visual fatigue (asthenopia), and migraines in flicker-sensitive populations. Conversely, disabling hardware temporal dithering without spatial compensation exposes hard quantization steps (contour banding), which activate retinal center-surround lateral inhibition and flood visual cortex V1 with false edge action potentials (Mach bands). We present a complete, zero-overhead architectural remediation for the Android SurfaceFlinger (Skia Graphite Vulkan backend) and HWC3 pipeline: Hardware Temporal FRC Suppression: Disabling temporal register rotation inside the SoC DPU / DRM-KMS display driver (0 Hz temporal modulation). Surface-Locked Spatial TPDF Blue-Noise Quantization: Injecting a single-cycle Triangular Probability Density Function (TPDF) spatial blue-noise pass into SurfaceFlinger's RenderEngine, phase-locked 1:1 to physical screen pixel pitch and advected by view/scroll translation vectors d(x, y) with analytical sub-pixel bilinear variance compensation. This architecture eliminates both magnocellular temporal flicker (0.0 Hz retinal modulation during static viewing and smooth pursuit eye movements) and cortical Mach band contours, requiring only an 8 kB L1-resident lookup texture and fewer than 6 ALU instructions per pixel.

Zenodo (CERN European Organization for Nuclear Research)
Image and Video Quality Assessment
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.