Last Updated: July 2026
Pulse width modulation (PWM) is the most common brightness control method in modern monitors, laptops, and phones. Instead of reducing the backlight's power output, PWM rapidly switches it fully on and fully off — sometimes hundreds, sometimes thousands of times per second. Your eyes average the flicker into perceived dimness. Your brain does not. Between 10% and 30% of the population experiences headaches, eye strain, or nausea from PWM flicker they cannot consciously see, according to research compiled for the IEEE 1789-2015 standard on LED flicker risk.
If you have ever felt a dull pressure behind your eyes after an hour of screen work — even though your prescription is current, your posture is fine, and you take breaks — the problem might not be your eyes. It might be your monitor's dimming method.
This post explains what PWM is at the electrical level, why it exists, which devices use it, how to test your own screen, and the three realistic ways to stop the flicker. If your headaches happen specifically on a MacBook, we wrote a dedicated guide to fixing screen flicker headaches on MacBook that covers the Apple-specific hardware behaviors. This piece is broader — it applies to every monitor, laptop, phone, and LED panel you interact with.
What Is Pulse Width Modulation and How Does It Work?
PWM controls screen brightness by switching the backlight between fully on and fully off at a fixed frequency, where the ratio of on-time to off-time determines perceived brightness. At 50% brightness, the backlight spends half each cycle on and half off. At 20%, it is on for one-fifth of the cycle and off for four-fifths.
The physics are straightforward. An LED backlight has two states: emitting light or not emitting light. To produce brightness levels between 0% and 100%, the display driver applies a square wave signal — a rapid alternation between full voltage and zero voltage. The frequency of this square wave is fixed (it does not change with brightness). Only the duty cycle changes: the proportion of each cycle the LED spends in the "on" state.
At 100% brightness, the LED is continuously on — duty cycle is 100%, and there is no flickering at all. At 75% brightness, the LED is on for 75% of each cycle and off for 25%. At 10% brightness, it is on for just 10% of the cycle, spending 90% of the time dark. The lower the brightness, the more time the display spends off. And the more time it spends off, the more aggressive the flicker pattern becomes.
The fundamental problem: your retina does not average light the same way a photometer does. A 2010 paper by Wilkins, Veitch, and Lehman presented to the IEEE found that headache incidence roughly doubles when comparing 100 Hz flicker to steady illumination, even when subjects reported they could not see the flicker. Your visual cortex responds to the modulation. Your conscious perception does not register it. The headache is the only signal your body has to tell you something is wrong.
Why Do Monitors Use PWM Instead of Just Dimming the Light?
Manufacturers choose PWM because it preserves color accuracy across all brightness levels — a priority that engineering teams value more than flicker sensitivity, which affects a minority of users. The alternative, DC dimming (reducing the actual current to the LED), causes color temperature shifts and reduced color gamut at low brightness. For a display manufacturer marketing "color accuracy," that is unacceptable. For the 10-30% of users who get headaches, PWM is worse.
There is a second engineering reason. LED efficiency is nonlinear. At low currents, LEDs become less efficient and their light output becomes harder to predict precisely. PWM avoids this entirely by always running the LED at its optimal current during the "on" phase — the brightness control happens in the timing, not the power level. The LED is either at peak performance or completely off. No ambiguity.
The trade-off is invisible to most users and plainly visible to a significant minority. Display manufacturers have optimized for the majority. If you are in the minority, you need to understand what you are dealing with.
How Does PWM Flicker Cause Headaches and Eye Strain?
PWM triggers headaches through involuntary pupil constriction cycles: the rapid light-dark alternation forces your iris muscles to contract and relax continuously, fatiguing the ciliary body and generating referred pain behind the eyes and across the forehead. The mechanism is neurological, not psychological, and it operates below the threshold of conscious perception.
The IEEE 1789-2015 standard, developed by the Power Electronics Society's committee on LED health effects, classifies flicker risk into three tiers based on frequency and modulation depth. Below 90 Hz with high modulation is "high risk." Between 90 Hz and 1,250 Hz is "low risk but not no risk." Above 3,000 Hz with low modulation is "no observable effect." Most consumer monitors fall squarely in the middle zone.
Research presented to the IEEE committee documented these health effects from invisible LED flicker:
- Headaches and migraines — the most commonly reported symptom. A study published in the Journal of Environmental Health Science and Engineering found that participants exposed to high-frequency PWM screens reported significantly more headaches than those using DC-dimmed displays over the same work period.
- Asthenopia (eye strain) — the medical term for the fatigue, blurred vision, and burning sensation that screen workers often attribute to "too much screen time." In many cases, the screen time itself is not the problem. The flicker is.
- Increased migraine frequency — Dr. Arnold Wilkins at the University of Essex demonstrated that even subliminal flicker (above 100 Hz) measurably increases cortical excitability in migraine-prone individuals.
- Cognitive fatigue — electroretinography studies show measurable cortical response to flicker well above the 80 Hz conscious perception threshold. Your brain is processing the flicker whether you see it or not. Over hours, that processing load accumulates as fatigue.
The critical variable is not just frequency — it is modulation depth. A monitor running PWM at 1,000 Hz with 100% modulation (backlight goes completely off during each cycle) is more problematic than one running at 500 Hz with 30% modulation (backlight dims to 70% but never fully extinguishes). Most consumer monitors use 100% modulation because partial modulation adds circuit complexity with no benefit the average user notices.
Which Devices Use PWM? A Frequency-by-Frequency Breakdown
Nearly every display you own uses PWM at some brightness level — but frequencies range from 60 Hz (dangerous) to 30,000 Hz (effectively harmless), and the worst offenders are often the most expensive screens.
| Device Category | Typical PWM Frequency | PWM Active Range | IEEE 1789 Risk Level |
|---|---|---|---|
| OLED smartphones (Samsung Galaxy, Pixel) | 60–480 Hz | Below 50% brightness | High risk (below 240 Hz) |
| iPhone 15 Pro / 16 Pro (OLED) | 480 Hz | Below ~50% brightness | High risk at >40% modulation |
| MacBook Air (M2-M4) | ~500 Hz | Below ~40% brightness | Moderate risk |
| MacBook Pro 14/16" miniLED | 14,800 Hz | All brightness levels | Low risk (but always active) |
| Budget LCD monitors (generic) | 200–400 Hz | All brightness below 100% | Moderate to high risk |
| Gaming monitors (flicker-free branded) | DC dimming / none | N/A | No risk |
| BenQ / ASUS / LG "flicker-free" models | DC dimming | N/A | No risk |
| External OLED monitors | 60–240 Hz at low brightness | Below 30-50% brightness | High risk when active |
| LED room lighting (non-display) | 100–400 Hz typical | When dimmed via switch | Moderate to high risk |
A few patterns emerge from this table. First, OLED displays are the worst offenders at low brightness — their PWM frequencies are often below 500 Hz, which falls in the range where most sensitive individuals experience symptoms. Second, "flicker-free" branding on external monitors is not marketing fluff — those monitors genuinely use DC dimming, and TUV Rheinland certification independently verifies the claim. Third, the MacBook Pro's 14,800 Hz frequency sounds safe, but it runs at every brightness level including 100% — so the total flicker exposure time is longer than devices that only activate PWM below a threshold.
Data sourced from NotebookCheck's PWM database and RTINGS display measurements.
How to Test If Your Monitor Uses PWM
You can detect PWM on any monitor using nothing more than your phone's camera and a dark room. The test takes sixty seconds and gives you a definitive answer.
Here are three methods, from simplest to most precise:
Method 1: The Smartphone Camera Test
Open your phone's camera app. Set your monitor brightness to 30% or lower. Display a full white screen (open a blank document or a white webpage). Point your phone camera at the monitor from about 30 centimeters away.
If you see dark horizontal bands rolling across the screen in the camera viewfinder, your monitor uses PWM. The bands are the camera's sensor capturing the backlight's off-phase during each scan line. The thicker and darker the bands, the lower the PWM frequency. If the screen appears perfectly uniform in the camera, either your monitor uses DC dimming or its PWM frequency is too high for your camera's frame rate to capture (above ~30,000 Hz).
For more precision, switch to slow-motion video mode (240 fps on most modern phones). Slower PWM frequencies become dramatically more visible at high frame rates. If bands appear in slow-motion that were invisible in standard mode, your monitor's PWM frequency is in the 1,000–10,000 Hz range.
Method 2: The Pencil Test
Hold a pencil or pen vertically in front of your monitor displaying a white screen at low brightness. Wave the pencil rapidly back and forth (like a windshield wiper). If the pencil appears as a smooth blur, the backlight is continuous (DC dimming). If you see a stroboscopic effect — the pencil appears to exist in multiple positions simultaneously, like a freeze-frame sequence — the backlight is flickering via PWM. The more distinct the "frozen" positions, the lower the frequency.
This test exploits the same principle as a strobe light at a dance club. The intermittent illumination "freezes" the moving object at each flash. The pencil test is less precise than the camera method but requires zero technology beyond a writing implement.
Method 3: Online Flicker Test Tools
Sites like The FPS Tester's flicker tool and TestUFO display patterns specifically designed to reveal PWM artifacts. These cannot measure your monitor's PWM frequency directly — they test whether PWM interacts with on-screen motion in visible ways. If motion test patterns show "ghosting" or stroboscopic artifacts that disappear at 100% brightness, PWM is the cause.
For definitive measurements, check your specific monitor model on NotebookCheck or RTINGS. Both test PWM with oscilloscopes and publish exact frequency and modulation data for hundreds of monitors and laptops.
The Three Real Fixes for PWM Flicker
There are exactly three ways to eliminate PWM: buy a monitor that does not use it, keep your current monitor at maximum brightness, or use software that bypasses the hardware dimming entirely. Everything else — blue light glasses, night mode, screen protectors, 20-20-20 rule — does not address the mechanism.
Fix 1: Buy a Flicker-Free Monitor (DC Dimming)
The cleanest solution. Most major manufacturers now offer DC-dimmed monitors marketed as "flicker-free." BenQ, ASUS, LG, Dell, and ViewSonic all have product lines with TUV Rheinland flicker-free certification. These monitors reduce backlight brightness by lowering the current to the LEDs rather than switching them on and off. No flicker at any brightness level.
What to look for: the TUV Rheinland "Flicker-Free" or "Eye Comfort" certification badge. It is an independent lab test — the manufacturer cannot self-certify. If the box says "flicker-free" but has no TUV Rheinland seal, verify with RTINGS or NotebookCheck before buying.
The trade-off: DC-dimmed monitors can show slight color temperature shifts at very low brightness (below 20%), because reducing LED current changes the emission spectrum. In practice, this shift is minor and most users never notice it. The monitors cost about the same as PWM equivalents — flicker-free is standard in most 2026 monitors above $200.
This solution works for desktop setups. It does nothing for your laptop's built-in display.
Fix 2: Lock Brightness at 100% (Eliminate the Duty Cycle)
At 100% brightness, the backlight runs continuously — 100% duty cycle means the LED never enters the off state. No off state means no flicker. This works on most devices, with two important exceptions:
- MacBook Pro miniLED models: local dimming zones use PWM even at 100% brightness. The backlight for each dimming zone independently modulates. You cannot escape PWM on these machines through brightness alone.
- Some OLED panels: certain OLED implementations run PWM across the entire brightness range for pixel-level uniformity.
The obvious problem with this fix is that a display at maximum brightness is uncomfortably bright in most environments. Working at 100% brightness in a dim room will cause different eye strain (light fatigue rather than flicker fatigue). You trade one problem for another.
Fix 3: Software Dimming (Gamma Table Override)
The best solution for laptops and devices where you cannot swap the display hardware. The concept: lock the hardware backlight at maximum (eliminating PWM), then reduce perceived brightness through the display's gamma lookup table — a software-level adjustment that scales pixel output without touching the backlight.
This is how Sundown's flicker-free mode works on Mac. The hardware backlight stays at full power. Sundown modifies the gamma curve via CoreGraphics APIs to reduce the perceived brightness of every pixel. The result: the screen looks as dim as you want, but the backlight never flickers. Zero CPU overhead (the gamma table is set once, the GPU handles the rest), zero tracking, 398 KB total app size. The mechanism is explained in detail in our deep dive on flicker-free mode.
The trade-off with gamma dimming is a slight reduction in contrast ratio at very low perceived brightness. Because you are capping the maximum output of each pixel rather than reducing total light, the darkest values get compressed. Sundown compensates with a perceptually calibrated gamma curve that preserves shadow detail even at deep dimming levels.
Other tools offer similar approaches: BetterDisplay and MonitorControl for Mac, Iris for Windows, and Dimmer for basic brightness overlay. The difference is implementation quality, resource usage, and whether the tool addresses other display health issues (blue light, dithering) alongside flicker. For a comparison of Mac display health apps, see our best Mac display health app roundup.
PWM on Phones: Why Your OLED Display Is the Worst Offender
OLED phone displays produce the most aggressive PWM flicker of any consumer device — often at frequencies between 60 Hz and 480 Hz, deep inside the range where sensitive users experience symptoms.
The engineering reason: OLED pixels emit their own light (no backlight), and at low brightness the voltage differential between on and off becomes difficult to control precisely. PWM solves this by pulsing each pixel at full voltage for shorter durations. The result is accurate colors and uniform brightness at the cost of flicker that IEEE 1789 classifies as "high risk" at these frequencies.
Samsung Galaxy phones historically used 240 Hz PWM below 50% brightness. Recent models push this to 480 Hz or higher. Google Pixel devices vary by generation. Apple's iPhone Pro models with OLED use approximately 480 Hz PWM at low brightness, which IEEE 1789 rates as high risk above 40% modulation depth.
If you use your phone in bed at minimum brightness — which most people do — you are experiencing the most aggressive PWM flicker of any device you own. Some Android phones now offer a "DC dimming" or "flicker reduction" toggle in display settings. Check Settings > Display > Eye Comfort or similar. On iPhone, no such option exists in iOS as of mid-2026.
This is relevant even if you are reading this because of monitor headaches. PWM sensitivity is cumulative. Hours of low-brightness phone use in the evening can prime your visual system to be more reactive to subtler PWM from your monitor the next morning.
The Brightness Trap: Why Dimming Your Screen Makes Headaches Worse
The most counterintuitive fact about PWM headaches: turning your brightness down — the thing every ergonomics guide tells you to do — makes the flicker more aggressive, not less.
At 50% brightness, the backlight spends half its cycle in the off state. At 20% brightness, it spends 80% of its cycle off. The lower you go, the higher the contrast ratio between the on-phase and off-phase becomes. Your retina is exposed to a sharper, more abrupt light-dark transition at low brightness than at high brightness.
Think of it this way. A ceiling light flickering between 100% and 90% output is barely detectable. The same light flickering between 100% and 0% — even at the same frequency — is a strobe. Low-brightness PWM is closer to the strobe end of that spectrum. The duty cycle shrinks. The dark phase gets longer and darker. The neural impact intensifies.
This is why many flicker-sensitive individuals report the worst headaches when working in dim rooms with low screen brightness — exactly the scenario that ergonomics guidelines recommend. The advice is correct for glare reduction and blue light management. It is actively harmful if PWM is your primary trigger.
California's Title 24 building code now specifies "reduced flicker operation" requiring less than 30% amplitude modulation at frequencies below 200 Hz for LED lighting installations. No equivalent standard exists for display panels. The monitor on your desk is held to a lower flicker standard than the overhead light in your office.
How to Know If PWM Is Causing Your Symptoms
PWM sensitivity produces a specific symptom pattern that distinguishes it from general eye strain, dry eyes, or visual fatigue: symptoms worsen at low brightness, improve at maximum brightness, and are absent on DC-dimmed displays.
Run this diagnostic sequence:
- Set your monitor to maximum brightness and work for one hour. Note your symptoms (headache, eye pressure, fatigue).
- The next day, set brightness to 30% in the same room, same duration. Note symptoms again.
- If symptoms are markedly worse at low brightness — more eye pressure, faster headache onset, worse fatigue — PWM is likely your trigger.
- Confirm with the camera test. If your monitor shows PWM bands at 30% brightness and none at 100%, the correlation is established.
- Final confirmation: work on a known flicker-free display (check RTINGS or use an external DC-dimmed monitor) at the same low brightness. If symptoms disappear, PWM is confirmed as the cause.
This diagnostic matters because not all screen headaches are PWM headaches. Blue light exposure, accommodation fatigue (your focusing muscles overworking), dry air, or uncorrected astigmatism all produce similar symptoms but have different solutions. If your headaches are equally bad at all brightness levels, PWM is unlikely to be the primary cause. Look at the full spectrum of Mac eye strain causes instead.
Beyond PWM: Temporal Dithering and the Second Invisible Flicker
PWM is not the only source of invisible display flicker. Every Apple Silicon Mac also produces temporal dithering — a pixel-level color alternation at your display's refresh rate that most PWM-focused articles never mention.
Temporal dithering is a separate mechanism from PWM. Where PWM flickers the backlight, dithering flickers individual pixels. To display colors that fall between the panel's native bit depth, the GPU rapidly alternates each pixel between two neighboring color values — frame by frame, 60 or 120 times per second. Your eye blends the alternation into a perceived intermediate shade.
The result is smoother color gradients. The cost is a second independent source of visual instability, layered on top of PWM. If you have ever pushed a blue light filter to deep warm tones and noticed a shimmering, sand-like noise across solid colors, that is temporal dithering becoming visible.
We wrote a detailed explainer on Apple Silicon display dithering — it ranks on Google's first page for its keyword because no other resource explains the phenomenon with this level of technical specificity. If you are sensitive to PWM, you may also be sensitive to dithering. Sundown is the only Mac app that addresses both mechanisms: flicker-free mode for PWM, anti-dithering for temporal dithering.
Making the Right Fix for Your Setup
The best fix depends on your hardware, your sensitivity level, and whether you are willing to spend money on new equipment. Here is the decision framework.
| Your Situation | Best Fix | Cost | Effectiveness |
|---|---|---|---|
| Desktop with budget monitor using PWM | Replace with DC-dimmed/flicker-free monitor | $200-500 | 100% — eliminates the source |
| Desktop, happy with current monitor | Software dimmer (lock brightness + gamma override) | $0-30/yr | 95% — slight contrast trade-off |
| MacBook Air, brightness above 40% | Nothing needed — DC dimming is active | $0 | 100% at high brightness |
| MacBook Air, prefer low brightness | Sundown flicker-free mode | $29/yr | 100% — gamma dimming bypasses PWM |
| MacBook Pro miniLED (any brightness) | Sundown flicker-free mode | $29/yr | 100% — locks backlight, gamma dims |
| OLED phone at low brightness | Enable DC dimming toggle if available | $0 | Varies by phone model |
| Multiple devices, all causing issues | Combination: flicker-free monitor + laptop software + phone settings | $230-530 | Near-total elimination |
The cost column is worth noting. A $200 flicker-free monitor or a $29/year software license is a trivial investment if you are losing an hour of productive work per day to headaches. Over a year of work days, one hour per day at even a modest billing rate is thousands of dollars of lost output. PWM sensitivity is not a minor annoyance — for the people it affects, it is a productivity problem with a straightforward engineering solution.
Start your 7-day free trial of Sundown at trysundown.com. It is 398 KB, uses zero CPU, collects zero data, and you will know within two days of flicker-free mode whether PWM was your problem.