Contrast ratio
More actions
The contrast ratio of a display is the ratio between the luminance of the brightest white it can show and the luminance of its darkest black, usually written in the form C:1 (for example 1000:1).[1] In a 2017 analysis in Optics Express, Haiwei Chen, Guanjun Tan and Shin-Tson Wu describe it as a key display metric for image quality, especially for enabling high dynamic range (HDR).[2]
In virtual reality (VR) headsets the figure depends mainly on the panel type and on the lenses. Self-emissive OLED and micro-OLED panels can switch pixels off and reach very high ratios, while LCD panels leak some backlight in the dark state.[2][3] Stray light inside the headset optics then lowers the contrast that reaches the eye.[3][4] In optical see-through augmented reality (AR) displays, light from the real world adds to both the bright and the dark parts of the virtual image, so the relevant quantity there is the ambient contrast ratio rather than the dark-room figure.[3]
Definition
Expressed as a ratio, contrast is the maximum luminance divided by the minimum luminance, C = Lmax / Lmin. The same range can also be given in orders of magnitude (log10 of the ratio) or in photographic stops (log2 of the ratio, where one stop is a doubling or halving of light).[1] Luminance is measured in candelas per square metre, a unit also called the nit.[3]
Display measurement also uses Michelson contrast, computed as (Lhigh - Llow) / (Lhigh + Llow). It is a fraction rather than a ratio, and the U.S. Food and Drug Administration (FDA) uses it in its measurement method for AR headsets described below.[5]
Published figures come in several forms. Chen, Tan and Wu use the term "static contrast ratio" for the dark-room contrast of an LCD or OLED panel.[2] On LCDs with local dimming, the backlight changes with image content, and a 2025 paper by BOE Chuangyuan engineers uses "dynamic CR" for the ratio between the full-white (L255) and full-black (L0) luminance measured within the same test pattern, which the industry usually takes from a checkerboard.[4]
Measurement
Two common test patterns are described in a 2001 National Institute of Standards and Technology (NIST) report by P. A. Boynton and C. Fenimore on digital cinema projectors, which followed the ANSI/NAPM IT7.228-1997 and ANSI/PIMA IT7.227-1998 electronic projection standards and the VESA Flat Panel Display Measurements Standard version 2.0 (the table also includes the FDA headset method described below):[6]
| Method | Test pattern | Calculation |
|---|---|---|
| Full-screen contrast | A full white field, then a full black field | Center-screen white luminance divided by center-screen black luminance[6] |
| Checkerboard contrast (ANSI contrast) | A 4x4 checkerboard of white and black boxes, measured at the 16 box centers | Average of the white-to-black luminance ratios of the eight checkerboard pairs, with veiling glare compensated as necessary[6] |
| Michelson contrast on a headset (FDA method) | Checkerboards of a low and a high gray level, each box about 1/10 of the field of view | (Lhigh - Llow) / (Lhigh + Llow), averaged across the display and repeated at several ambient light levels[5] |
In the checkerboard method, bright and dark boxes share the screen at the same time, and the NIST procedure calls for compensating for veiling glare.[6] Test equipment for near-eye displays keeps the same distinction: Gamma Scientific's NED V-Series tester, for example, lists "FOFO or Sequential Contrast" as a measurement separate from checkerboard contrast and ANSI contrast.[7]
In a head-mounted display the user views a virtual image formed by the optics. The FDA's regulatory science tool RST24MX04.01 for optical see-through AR headsets, published on 6 June 2024, aligns a calibrated imaging light measuring device with its entrance pupil at the headset's eye point, focuses it on the virtual image plane, and refers to IEC 63145-20-10 and section 19 of the Society for Information Display's Information Display Measurements Standard (IDMS) for the instrument requirements and alignment. The box size of about one tenth of the field of view is meant "to avoid substantial veiling glare and optical aberration that could affect the contrast measurement".[5] The manual's example data come from a Microsoft HoloLens 2 and an Epson Moverio BT-300, measured in the dark and with an ambient luminance of 381 cd/m2; the manual takes its contrast formula and example data from a 2022 SID paper by Chumin Zhao, Ryan Beams, Matthew Johnson and Aldo Badano.[5][8]
Contrast in a VR headset also varies across the field of view. Zhao, Beams and Badano reported in 2023 that the virtual image is blurred toward the edge of the field of view, mainly by the lenses stacked in front of the display, which lowers resolution and contrast there. They argued that methods adopted from flat-panel evaluation do not account for these spatially varying artifacts, and developed test patterns that measure Michelson contrast as a function of radial position, including at different interpupillary distance settings.[9]
Ambient contrast ratio
A dark-room contrast ratio ignores room light reflected from the screen. Chen, Tan and Wu define the ambient contrast ratio (ACR) of a direct-view display as:[2]
- ACR = (Lon + Lambient × RL) / (Loff + Lambient × RL)
where Lon and Loff are the on-state and off-state luminance of the display, Lambient is the ambient luminance and RL is the luminous reflectance of the panel. Ambient light is usually given as illuminance in lux, which the authors convert to luminance by dividing by pi. Their modelling of phones, TVs and public displays found that a high static contrast ratio matters most in low ambient light, while display brightness becomes the dominant factor as ambient light increases.[2]
Display technologies
Emissive and non-emissive displays differ by orders of magnitude. Chen, Tan and Wu state that the contrast ratio of an OLED "can approach 1,000,000:1 or even higher", while an LCD's is limited by depolarization in its thin-film transistors, liquid crystal layer and color filters.[2] Yuge Huang, En-Lin Hsiang, Ming-Yang Deng and Wu give typical LCD values of about 5000:1 for the multi-domain vertical alignment (MVA) mode, 2000:1 for fringe-field switching (FFS) and 1000:1 for twisted nematic (TN).[10] For the small light engines used in AR, Xiong and colleagues list micro-LEDs and micro-OLEDs at above 106:1, laser beam scanners at up to 106:1 because the laser can be switched off completely, and the reflective LCoS and DLP (digital micromirror device) engines at about 2000:1 to 5000:1.[3]
| Display type | Typical contrast ratio cited | Source |
|---|---|---|
| OLED (direct view) | Approaching 1,000,000:1 or higher | [2] |
| LCD, MVA mode | About 5000:1 | [2][10] |
| LCD, FFS mode | About 2000:1 | [2][10] |
| Micro-LED and micro-OLED microdisplays | Above 106:1 | [3] |
| Laser beam scanning | Up to 106:1 | [3] |
| LCoS and DLP light engines | About 2000:1 to 5000:1 | [3] |
Local dimming
Local dimming raises the contrast of an LCD by splitting its backlight into zones that can be dimmed independently, so dark parts of the image get less light. Huang and colleagues describe it as dual modulation: a segmented, low-resolution mini-LED backlight behind a high-resolution LCD panel. With enough zones, they write, the halo effect and clipping can be suppressed to an unnoticeable level.[10] A dual-modulation design of this kind was described in a SIGGRAPH 2004 paper by H. Seetzen, W. Heidrich, W. Stuerzlinger, G. Ward, L. Whitehead and colleagues, whose two high dynamic range display systems paired an LCD panel with either a DLP projector or a custom-built LED panel. The project page reports a dynamic range "well beyond 50,000:1" and maximum intensities of 2700 cd/m2 and 8500 cd/m2 for the two prototypes, and states that the LED design was later sold commercially by BrightSide Technologies.[11][2]
VR headsets
Several VR headsets use self-emissive panels. Sony lists the display method of the PlayStation VR2 as OLED, at 2000 x 2040 pixels per eye,[12] and Apple's Apple Vision Pro "uses micro-OLED technology to pack 23 million pixels into two displays", according to Apple's announcement of 5 June 2023.[13] Pimax, which develops both QLED (LCD with quantum dots) and OLED headsets, writes that OLED pixels can be turned completely off, giving "effectively infinite" contrast, and that mini-LED local dimming "cannot fully match OLED's pixel-level control", with some blooming or elevated blacks in extreme-contrast scenes.[14]
The Meta Quest Pro uses two 1800x1920 LCD panels with quantum dots and more than 500 mini-LEDs in the backlight; Meta claimed the mini-LEDs give 75% more contrast than the Meta Quest 2.[15] Local dimming on Quest Pro was opt-in for each app, and UploadVR reported in March 2023 that the Public Test Channel build of system software v51 enabled it by default for Link and Air Link PC VR streaming. The same report noted the trade-off of blooming: with far fewer backlight zones than pixels, dark pixels next to bright ones are also lit.[16]
Effect of the optics
The panel's contrast is not what the user sees. Xiong and colleagues note that image contrast in a VR headset depends on both the display panel and stray light, and that pancake (catadioptric) headsets usually show darker images with lower contrast than headsets with conventional refractive lenses, because surface reflections and polarization leakage in the folded light path create stray light and ghost images.[3] The grooves of Fresnel lenses also cause diffraction artifacts and stray light that degrade image quality.[3]
Ma and colleagues at BOE Chuangyuan modelled this for LCD-based pancake headsets in 2025. Their model gives the system contrast as 1 / (1/CRLCD + K), where CRLCD is the panel's dynamic contrast and K is the stray light factor of the optics, which they put at generally 0.2% to 0.4% for VR systems.[4] Measured results matched the model within 3.8%:[4]
| LCD dynamic contrast | Stray light factor | Calculated system contrast | Measured system contrast |
|---|---|---|---|
| 600:1 | 0.2% | 273:1 | 265:1 |
| 600:1 | 0.4% | 176:1 | 183:1 |
| 1500:1 | 0.2% | 375:1 | 387:1 |
| 1500:1 | 0.4% | 214:1 | 208:1 |
In their tests, a mini-LED backlight module had 2.76 times the dynamic contrast of a conventional (ELED) backlight module, but the advantage fell to 1.7 times in one VR optical system and disappeared in another. As panel contrast rises, system contrast grows more and more slowly, so stray light in the lens sets a ceiling. The authors also reported raising the panel's dynamic contrast by about 90% through changes to the mini-LED backlight films intended to reduce halo (a narrow-light film and a white dam sealant between LEDs) together with changes to the backlight algorithm.[4]
Optical see-through AR
In an optical see-through display, the virtual image is added to light from the real world. Xiong and colleagues give the ambient contrast ratio of a see-through AR display as ACR = (Lon + Lam × T) / (Loff + Lam × T), where T is the see-through transmittance of the optics. They estimate the ambient luminance at about 30 nits in a normal living room (about 100 lux), 150 nits under office lighting, 300 nits outdoors on an overcast day and 3000 nits on a sunny day. A minimum ACR of 3:1 is needed for recognizable images, 5:1 for adequate readability and 10:1 or more for outstanding readability. Ignoring optical losses, an ACR of 10:1 on a sunny day needs a display brightness of at least 30,000 nits, which they describe as a big challenge for both the microdisplay and the optical combiner.[3] They also note that the native contrast of a light engine matters mainly in dark surroundings, while in bright surroundings the ACR is governed mostly by peak brightness.[3]
Lowering the transmittance T is the other way to raise ACR. In 2016, Ruidong Zhu and colleagues proposed an AR system with a guest-host liquid crystal cell whose transmittance could change from about 73% to about 26% with 8 V, responding in about 50 ms; combined with a light sensor, the cell was used to improve the ambient contrast ratio under different lighting conditions.[17] The Magic Leap 2 uses a dimmer for a similar purpose. Magic Leap's developer documentation notes that its devices "have additive displays" and that the headset adds "a new low resolution panel which selectively subtracts photons in the environment". Global dimming tints the whole view behind the content, trading visibility of the environment against the opacity of virtual content, while segmented dimming darkens only selected portions, making virtual content appear more opaque without tinting the whole display.[18] Related lens-tinting technology is covered in Electrochromic dimming.
See also
References
- ↑ 1.0 1.1 Rafał K. Mantiuk (2018-12-05). "HDR, displays & low-level vision". SIGGRAPH Asia 2018 course: Cutting-edge VR/AR Display Technologies. University of Cambridge. https://www.cl.cam.ac.uk/~rkm38/pdfs/SIGGRAPH_Asia_HDR_VR.pdf. Retrieved 2026-10-15.
- ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 Haiwei Chen, Guanjun Tan, Shin-Tson Wu (2017-12-22). "Ambient contrast ratio of LCDs and OLED displays". Optics Express, vol. 25, no. 26, pp. 33643-33656. Optica (Optical Society of America). https://doi.org/10.1364/OE.25.033643. Retrieved 2026-10-15.
- ↑ 3.00 3.01 3.02 3.03 3.04 3.05 3.06 3.07 3.08 3.09 3.10 3.11 Jianghao Xiong, En-Lin Hsiang, Ziqian He, Tao Zhan, Shin-Tson Wu (2021). "Augmented reality and virtual reality displays: emerging technologies and future perspectives". Light: Science & Applications, vol. 10, article 216. https://doi.org/10.1038/s41377-021-00658-8. Retrieved 2026-10-15.
- ↑ 4.0 4.1 4.2 4.3 4.4 Zhanshan Ma, Jingran Niu, Wenhong Tian, Xuelu Wang, Jianyun Xie, Jingfei Mu, Shibiao Wang, Lei Xiao (2025-06). "42-3: Correlation between LCD Dynamic Contrast and Pancake VR Optical System". SID Symposium Digest of Technical Papers, vol. 56, no. 1, pp. 557-560. Society for Information Display. doi:10.1002/sdtp.18220. https://sid.org/Portals/sid/Files/SID-Digest/docs/42-3.pdf. Retrieved 2026-10-15.
- ↑ 5.0 5.1 5.2 5.3 U.S. Food and Drug Administration (2024-06-06). "User Manual: Contrast Measurement Method for Optical See-Through Augmented Reality Head-Mounted Displays (RST24MX04.01)". Catalog of Regulatory Science Tools. FDA Center for Devices and Radiological Health. https://cdrh-rst.fda.gov/sites/default/files/2025-04/Contrast%20Measurement%20Method%20for%20Optical%20See-Through%20Augmented%20Reality%20Head-Mounted%20Displays.pdf. Retrieved 2026-10-15.
- ↑ 6.0 6.1 6.2 6.3 P. A. Boynton, C. Fenimore (2001). "Characterization of Projection Systems for the MPEG-4 Digital Cinema Compression Scheme Evaluation". NIST Interagency Report 6792. National Institute of Standards and Technology. https://nvlpubs.nist.gov/nistpubs/Legacy/IR/nistir6792.pdf. Retrieved 2026-10-15.
- ↑ "NED V-Series Production Benchtop Tester (product sheet, rev. 05.18.23)". Gamma Scientific. Gamma Scientific. https://gamma-sci.com/wp-content/uploads/2023/05/NED-V-Series-rev.05.18.23.pdf. Retrieved 2026-10-15.
- ↑ Chumin Zhao, Ryan Beams, Matthew Johnson, Aldo Badano (2022-06). "18-2: Assessment of Image Quality in Augmented Reality Displays Using a Computational Model of Target Detectability". SID Symposium Digest of Technical Papers, vol. 53, no. 1, pp. 194-197. Society for Information Display. https://doi.org/10.1002/sdtp.15451. Retrieved 2026-10-15.
- ↑ Chumin Zhao, Ryan Beams, Aldo Badano (2023-06). "82-1: Distinguished Paper: Radially Variant Contrast in Virtual Reality Headsets". SID Symposium Digest of Technical Papers, vol. 54, no. 1, pp. 1147-1150. Society for Information Display. https://doi.org/10.1002/sdtp.16777. Retrieved 2026-10-15.
- ↑ 10.0 10.1 10.2 10.3 Yuge Huang, En-Lin Hsiang, Ming-Yang Deng, Shin-Tson Wu (2020-06-18). "Mini-LED, Micro-LED and OLED displays: present status and future perspectives". Light: Science & Applications, vol. 9, article 105. https://doi.org/10.1038/s41377-020-0341-9. Retrieved 2026-10-15.
- ↑ H. Seetzen, W. Heidrich, W. Stuerzlinger, G. Ward, L. Whitehead, M. Trentacoste, A. Ghosh, A. Vorozcovs (2004). "High Dynamic Range Display Systems". ACM Transactions on Graphics, vol. 23, no. 3 (SIGGRAPH 2004). University of British Columbia Imager Lab. https://www.cs.ubc.ca/labs/imager/tr/2004/Seetzen_2004_HDR/. Retrieved 2026-10-15.
- ↑ "PS VR2 tech specs". PlayStation. Sony Interactive Entertainment. https://www.playstation.com/en-us/ps-vr2/ps-vr2-tech-specs/. Retrieved 2026-10-15.
- ↑ "Introducing Apple Vision Pro: Apple's first spatial computer". Apple Newsroom. Apple. 2023-06-05. https://www.apple.com/newsroom/2023/06/introducing-apple-vision-pro/. Retrieved 2026-10-15.
- ↑ "QLED vs. OLED in VR: Technology Trade-offs and Product Strategy". Pimax Store. Pimax. 2026-01-30. https://store.pimax.com/blogs/highlights/qled-vs-oled-in-vr-technology-trade-offs-and-product-strategy. Retrieved 2026-10-15.
- ↑ David Heaney (2022-10-11). "Quest Pro Specs, New Features, And How They Compare To Quest 2". UploadVR. https://www.uploadvr.com/quest-pro-specs-features-details/. Retrieved 2026-10-15.
- ↑ David Heaney (2023-03-28). "Quest Pro v51 Adds Local Dimming To PC Link, Significantly Enhancing Contrast". UploadVR. https://uploadvr.com/quest-pro-pc-vr-link-local-dimming/. Retrieved 2026-10-15.
- ↑ Ruidong Zhu, Haiwei Chen, Tamas Kosa, Pedro Coutino, Guanjun Tan, Shin-Tson Wu (2016). "High-ambient-contrast augmented reality with a tunable transmittance liquid crystal film and a functional reflective polarizer". Journal of the Society for Information Display, vol. 24, no. 4, pp. 229-233. https://doi.org/10.1002/jsid.427. Retrieved 2026-10-15.
- ↑ "Global/Segmented Dimmer". Magic Leap Developer Documentation. Magic Leap. https://developer-docs.magicleap.cloud/docs/guides/features/dimmer-feature/. Retrieved 2026-10-15.