User:Scarpy/Image quality perception

70mm feels better than 4K should allow are:

Acuity vs. suprathreshold perception (sharpness isn’t a threshold thing). Visual acuity and pixels/degree arguments are about detection at the limit. Most movie watching is suprathreshold: you’re judging texture, clarity, snap, dimensionality, and realism well above detection. Perceived sharpness tracks things like edge steepness (acutance) and mid-frequency contrast more than "can I resolve the tiniest line pairs?"

MTF and microcontrast. Two images can have the same nominal resolution and look very different if one preserves more contrast at mid spatial frequencies. That’s why lens quality, print/scan pipeline, and projection optics can change clarity even when you’re ast 4K.

Spatiotemporal vision (motion changes everything). Your effective resolution drops with motion because of eye tracking limits, integration time, and display/projector temporal behavior. Film’s motion portrayal (shutter, blur characteristics) and digital's (sample/hold vs impulse-like, judder, motion processing) can change perceived detail and comfort independent of static pixel count.

Foveation + saccades + temporal integration. You don’t stare at one point; you sample the scene with rapid eye movements. The brain integrates information across fixations. If a format preserves subtle detail/contrast, your repeated foveal “glimpses” can accumulate into a stronger sense of realism even if any single instant wouldn’t let you count pixels.

Peripheral vision and crowding (not just it’s blurrier). The periphery is limited by crowding and pooling, not only resolution. Big screens/IMAX-like FOV can feel more “present” because peripheral stimulation drives presence/vection—even though you can’t resolve fine detail out there. That’s a separate axis from 4K sharpness.

Masking, noise, and dithering (grain can help perception). The visual system has internal noise and is subject to contrast masking. Film grain (or well-shaped noise) can act like a perceptual “dither,” reducing banding and sometimes making textures feel more natural. It can also mask artifacts. This is a real reason “cleaner” isn’t always “better.”

Adaptation and local contrast normalization. Your visual system continuously adapts to luminance/contrast and effectively auto-levels scenes. Small differences in highlight roll-off, flare, and local contrast can change the perceived depth or pop more than raw resolution.

Dynamic range, glare, and flare. Perceived image quality is strongly affected by black level, stray light, veiling glare, and highlight handling. A theater with better optical cleanliness/less flare can look higher resolution because contrast is preserved-especially in shadows—without adding a single pixel.

Color appearance and chromatic detail. Color has its own spatial sensitivity and interactions with luminance. Differences in primaries, spectral power distribution (laser vs xenon vs print dyes), and metamerism can subtly affect realness, skin tones, and separation, again independent of 4K.

Psychophysics of preference (expectation, context, and presence are real variables). People’s judgments are influenced by theater context (screen size, sound, seating, branding, rarity). it’s part of the cognitive experience. Includes top-down perception alongside optics.

quality as a multidimensional space-mid-frequency contrast + temporal behavior + FOV/presence + flare/black level + noise/grain structure often move the needle more than "can the fovea resolve 4K pixels from this seat."

Subjective video quality - explicitly about video quality as experienced by humans and psychophysical testing

Video quality - perceived degradation through processing/transmission

Quality of experience - the broader delight/annoyance umbrella, explicitly tied to cognitive science and subjective experience

Mean opinion score -

Structural similarity index measure

Presence (telepresence) as a concept page that explicitly treats being there as a media-perception phenomenon

what higher spatial resolution (and wide field-of-view) does to subjective experience

The key point: 4K is enough because the eye tops out at X pixels/degree is an oversimplification. Visual limits depend heavily on contrast sensitivity - not just pixel counting. Humans are most sensitive around ~3-4 cycles per degree and less sensitive at very low and very high spatial frequencies, and acuity numbers are typically best-case, high-contrast targets - not natural images.

acutance (perceived sharpness not equal to resolution)

Optical transfer function optical transfer function / MTF (contrast transfer vs spatial frequency, i.e., microcontrast-ish but in engineering terms)

display motion blur (eye tracking + temporal presentation effects)

visual masking and visual crowding (visibility/recognition limits that differ from simple acuity)

visual adaptation (your normalization point)

veiling glare (contrast loss from stray light)

Color appearance model (viewing-conditions-dependent color perception)

The gaps, relative to your list, are mostly integration gaps and a few missing/underdeveloped concepts:

No “cinema perception” synthesis: There isn’t a Wikipedia article that cleanly connects MTF + motion portrayal + wide FOV/presence + flare/black level + grain/noise + top-down context into one coherent explanation aimed at cinema formats (70mm/IMAX vs digital). The “quality” pages lean telecom/codec/testing, not theatrical exhibition specifics.

Spatiotemporal CSF / suprathreshold sharpness framing is thin: Wikipedia has pieces (masking, motion blur, acuity, etc.), but it doesn’t strongly foreground the spatiotemporal contrast sensitivity function viewpoint as the central “resolution-isn’t-a-threshold” story for natural images and motion (it’s more fragmented across pages).

“Microcontrast” as people mean it in imaging/cinema is basically missing as a concept: You can get to the idea via MTF/OTF, but the perceptual bridge (“mid-frequency contrast drives clarity/realness”) isn’t packaged in an accessible way.

Projection-specific perceptual variables aren’t tied together: Wikipedia has glare/flare concepts, but it doesn’t strongly connect projector stray light, lens flare/veiling glare, screen luminance choices, black level, theater reflections to “it looks higher-resolution because contrast is preserved.”

Top-down/context effects exist, but not “image quality in a cinema” flavored: QoE acknowledges expectations/context as part of experienced quality, but there isn’t a crisp treatment of “rarity/branding/context changes perception” specifically for film formats.

On “any language”: French Wikipedia has an article explicitly about perceptual image quality (Qualité perceptuelle d’images), which is closer in spirit to what you’re doing than most English pages—but it still isn’t a “70mm vs 4K perception” synthesis.

If you want a Wikipedia-style article that would fill these gaps, it would likely need to sit at the intersection of: cinema exhibition technology + visual psychophysics + subjective quality/QoE—and then explicitly treat resolution as only one axis.

Contrast sensitivity is how small a difference in brightness you can detect as a function of spatial frequency (how fine the pattern is) and it’s the reason the whole pixels per degree argument is incomplete. Your eye/brain isn't equally good at all scales: it's typically most sensitive to mid-scale structure (a few cycles per degree) and much less sensitive to extremely fine detail and extremely broad, slow gradients. That means two displays (or a 70mm print vs a 4K digital presentation) can have the same nominal resolvable detail and still feel very different if one preserves contrast better at the frequencies where your visual system is most responsive. Practically: the thing you experience as clarity, texture, depth, snap, and realness is heavily driven by mid-frequency contrast-skin pores, fabric weave, hair separation, subtle shading cues. So improvements in the optical chain (lens MTF, flare control, grain/noise behavior, scan/print contrast) can be very visible and subjectively meaningful even when you're nowhere near the limit of resolving individual pixels or film grain.

Higher resolution can increase "realness" / higher-order impressions, even in practical viewing conditions.

controlled subjective experiments and found that increasing resolution (e.g., HD → UHD) boosted “lower-order” impressions (sharpness, color vividness, glossiness, 3D-ish impressions) which then boosted higher-order impressions—and they report the effect can show up even beyond the nominal design viewing distance. ^[1]

Resolution can affect perception even when you can't consciously detect the resolution change. in stimuli built from luminance/contrast gradients, and crucially: they found the effect even when the resolution difference was undete4ctable. "it feels more dimensional / more present even if I can’t point to a crisp edge and say "that’s sharper."^[2]

Wide field of view (big screen / closer seat) independently increases presence.

(they cite classic SMPTE-era psychophysics on wide-field "sensation of reality" and related presence measures). So if your 70mm experience also correlates with larger screens / different seating / different theaters, your brain is getting extra “presence” input that has nothing to do with 4K alone.^[1]

Even if a 4K projector could meet a simplistic "pixels per degree" target from some seats, 70mm (especially 5/70 and 15/70 IMAX) can still win in ways your visual system cares about: not just more captured detail, but different MTF/contrast behavior, different noise/grain structure (which can change perceived sharpness/texture), and often a very different exhibition context (screen size, optics, brightness choices, steadiness, etc.). The industry fact that 65/70mm is scanned with extremely high-resolution scanners (multi-K well beyond 4K) is at least consistent with “there’s potentially more than 4K worth of spatial information in the source,” even if the effective detail depends on stock, lenses, print generation, and projection.

^[3]

^[4]

• Keller, Dominik; Rao, Rakesh Rao Ramachandra; Göring, Steve; Raake, Alexander (2024). "Assessing Quality Differences for 8K/UHD-2 and 4K/UHD-1 HDR Video Based on Viewing Distance". TechRxiv.
• Ashraf, Maliha; Chapiro, Alexandre; Mantiuk, Rafał K. (2025). "Resolution limit of the eye: how many pixels can we see?". Nature Communications. doi:10.1038/s41467-025-64679-2.
• "Is your ultra-HD TV worth it? Scientists measure the resolution limit of the human eye". University of Cambridge. 27 October 2025.
• Armstrong, M.; Flynn, D.; Hammond, M.; Jolly, S.; Salmon, R. (2008). High Frame-Rate Television (PDF) (Technical report). BBC Research & Development White Paper. BBC Research & Development. WHP 169.
• Cottaris, Nicolas P.; Wandell, Brian A.; Rieke, Fred; Brainard, David H. (July 2020). "A computational observer model of spatial contrast sensitivity: Effects of photocurrent encoding, fixational eye movements, and inference engine". Journal of Vision. 20 (7): 17. doi:10.1167/jov.20.7.17.
• Hatada, Toyohiko; Sakata, Haruo; Kusaka, Hideo (August 1980). "Psychophysical Analysis of the "Sensation of Reality" Induced by a Visual Wide-Field Display" (PDF). SMPTE Journal. 89 (8): 560–569. doi:10.5594/J01582.
• Shishikui, Yoshiaki; Sawahata, Yasuhito (March 2022). "Quality of 8K Ultra-High-Definition Television Viewing Experience in Practical Viewing Conditions" (PDF). IEEE Transactions on Broadcasting. 68 (1): 2–12. doi:10.1109/TBC.2021.3105031.

• Lombard, Matthew; Ditton, Theresa (September 1997). "At the Heart of It All: The Concept of Presence". Journal of Computer-Mediated Communication. 3 (2): JCMC321. doi:10.1111/j.1083-6101.1997.tb00072.x.

• Johnson, Garrett M.; Fairchild, Mark D. (April 2002). "On Contrast Sensitivity in an Image Difference Model" (PDF). Proceedings of the PICS Conference: Image Processing, Image Quality, Image Capture, Systems Conference. Portland, OR: IS&T. pp. 18–23. - achromatic CSF peaks around ~3–4 cycles/degree, i.e., why pixels/degree is not the whole story.

• Triantaphillidou, Sophie; Jarvis, John; Psarrou, Alexandra; Gupta, Gaurav (July 2019). "Contrast sensitivity in images of natural scenes". Signal Processing: Image Communication. 75: 64–75. doi:10.1016/j.image.2019.03.002. - CSF discussion with an explicit ~4 cpd peak but in a more specialized natural scenes context.

• O'Carroll, David C.; Wiederman, Steven D. (February 2014). "Contrast sensitivity and the detection of moving patterns and features". Philosophical Transactions of the Royal Society B: Biological Sciences. 369 (1636): 20130043. doi:10.1098/rstb.2013.0043. PMC 3886331. PMID 24395970.{{cite journal}}: CS1 maint: article number as page number (link)