Most laptops hide battery condition two or three menus deep, so the fastest reliable check is a built-in report: Windows 11's powercfg /batteryreport HTML file, macOS's System Information power panel, and Linux's upower -i output. All three read the same two numbers — design capacity and present full-charge capacity — and their ratio tells you how much life the cell has actually lost. Anything above 80% of design capacity is normal wear. Below 60% usually means the machine is a desk computer that unplugs. This guide covers the commands, the numbers they spit out, and the failure modes that make a healthy battery look dead.
Why Design Capacity and Full-Charge Capacity Decide Everything
Battery health is a ratio, not a feeling. Manufacturers print a design capacity on the cell (in milliwatt-hours, or mWh) and the battery controller reports what the cell can still hold when fully charged. Divide the second by the first and you get a percentage that maps directly to usable runtime. A 60 Wh ThinkPad battery reading 48 Wh at full charge sits at 80% health, which is exactly the point where most warranties stop covering degradation.
The ratio also explains why runtime complaints outpace health percentages. A battery at 85% health paired with a five-year-old CPU running a modern browser loses more minutes to power draw than the cell itself lost to chemistry. Lithium-ion cells also lose capacity faster in the first year, roughly 2-4% under normal use, then flatten out. Panasonic's cell datasheets and the Battery University reference site both document that curve.
Two failure modes corrupt the reading. First, calibration drift: if the laptop rarely discharges below 30%, the controller's estimate of "full" creeps upward and full-charge capacity reports artificially low. Second, a swollen cell can still report high capacity while physically expanding — that's a safety issue, not a performance one, and no software report will flag it. If the trackpad or chassis bulges, stop using the machine and have the battery replaced regardless of the percentage.
The design capacity figure lives in the battery's firmware, so it doesn't change. If a report shows design capacity that looks nothing like the spec sheet — say 38,000 mWh on a laptop sold with a 52,500 mWh pack — the battery was likely swapped by a previous owner or a repair shop used a cheaper third-party cell. Treat that mismatch as a red flag before trusting any health figure.
Generate the Windows Battery Report in Three Commands
Windows builds the most detailed battery report of any consumer OS, and it takes under a minute. The report is an HTML file you can open in any browser, and it includes a capacity history table that shows how the battery degraded month by month — the single most useful feature Windows has that macOS and Linux lack. Run it as administrator so the history table populates fully.
1: Press Win + X, then choose Terminal (Admin) or Windows PowerShell (Admin). 2: Type powercfg /batteryreport /output "C:\battery-report.html" and press Enter. 3: Open File Explorer, navigate to C:, and double-click battery-report.html. 4: Scroll to Installed batteries for design capacity and full-charge capacity. 5: Read the Battery capacity history table for month-over-month full-charge values.
If the command returns "Unable to perform operation," the most common cause is a missing or disabled Microsoft AC Adapter driver entry, not a dead battery. Check Device Manager under Batteries — a yellow triangle on Microsoft ACPI-Compliant Control Method Battery means Windows can't read the controller. Right-click and reinstall the driver, reboot, and rerun the report. On some corporate-managed machines, group policy blocks powercfg entirely; run powercfg /batteryreport without the output flag and it writes to your user folder instead.
A second failure case: the report generates but shows a single "Installed batteries" row dated today with no history. That happens when a third-party battery lacks the smart-chip data Windows expects. The health percentage still works; the degradation trend doesn't. Lenovo Vantage and Dell Power Manager read the same chip directly and sometimes extract data Windows misses, so cross-check there before concluding the battery is healthy.
On gaming laptops with dual batteries or external battery packs, the report lists each pack separately. Read the row that matches the internal cell's part number, not the combined total. Alienware and Razer Blade models in particular mislead users who add up both rows.
Read the Windows Report Without Misreading the Numbers
The Installed batteries section shows four values that matter: DESIGN CAPACITY, FULL CHARGE CAPACITY, CYCLE COUNT, and the two capacity percentages underneath. The health figure is FULL CHARGE CAPACITY ÷ DESIGN CAPACITY. A 2021 Dell XPS 13 shipped with a 52,000 mWh design capacity; a three-year-old unit reading 41,600 mWh sits at exactly 80% health and roughly 4-5 hours of real runtime, down from 7-8 hours new.
Cycle count tells you how the battery got there. One full discharge from 100% to 0% equals one cycle; two half-discharges also equal one cycle. Lithium cells are typically rated for 300-500 cycles to 80% capacity, so a 400-cycle battery at 80% health is behaving exactly as designed, while a 120-cycle battery at 70% health points to heat damage or a defective cell.
The capacity history table is where the report earns its keep. Look for sudden drops — a battery that held 45,000 mWh in March and 38,000 mWh in April lost 15% in one month, which almost always means a firmware issue, a bad charger, or a swelling event. Gradual declines of 1-2% per month are normal. Windows also prints estimated battery life at full charge and at design capacity, giving you a comparison of current runtime against the day-one specification.
Ignore the "Battery life estimates" section if it shows absurd figures like 11 hours. Windows derives those from recent discharge sessions, so a day of light word processing inflates the estimate. The capacity ratio is the durable number; runtime estimates fluctuate with every session. If you need a second opinion, Lenovo Vantage, HP Support Assistant, and Dell Power Manager each display vendor-calculated health that sometimes differs from powercfg by 2-5 percentage points.
Find Battery Condition on macOS in Two Places
Apple splits the information across System Settings and System Information, and the two don't always agree. System Information reads the raw controller values; System Settings applies Apple's own "service recommended" thresholds and battery-health management logic. For a true capacity ratio, use System Information.
1: Hold Option and click the Apple menu, then choose System Information. 2: Select Power in the left sidebar. 3: Read Cycle Count, Condition, Maximum Capacity, and Design Capacity under Battery Information.
Maximum Capacity in macOS is already a percentage of design, so a MacBook reading "Maximum Capacity: 87%" has a health ratio of 87%. The raw milliwatt-hour values appear further down in the same panel if you want to compute the ratio yourself. Apple's own documentation says a MacBook battery is designed to retain 80% of its capacity at 1,000 complete charge cycles for models from 2010 onward, which is a much higher cycle rating than most Windows laptops carry.
System Settings → Battery → Battery Health shows a simpler status: "Normal," "Service Recommended," or "Battery Health Management" active. The wrinkle is that Apple's Battery Health Management can pause charging at 80% in daily use, which depresses "Maximum Capacity" temporarily without any real degradation. If the number drops and then recovers after a full discharge, the management feature caused it, not the cell.
On Apple Silicon Macs, the Power panel also lists "Battery Temperature" and "Battery Voltage." Temperature above 35°C while idle suggests a failing thermal path or a charger issue. According to Apple's battery support documentation, ambient temperatures above 35°C can permanently reduce capacity, so a MacBook that lives on a radiator or in a hot car shows degradation that the cycle count alone won't explain.
Check Battery Health on Linux With upower and sysfs
Linux exposes the battery through the kernel's power-supply class, so every value a manufacturer's tool would show is readable from /sys/class/power_supply/. The friendly way to read it is upower -i, which most desktop distributions include. The root-level way works even on minimal installs without a desktop environment.
1: Open a terminal and run upower -i /org/freedesktop/UPower/devices/battery_BAT0 (replace BAT0 with the device name from upower -e). 2: Read energy-full (present full-charge capacity), energy-full-design (design capacity), and capacity (the vendor's health percentage). 3: If upower isn't installed, run cat /sys/class/power_supply/BAT0/energy_full and cat /sys/class/power_supply/BAT0/energy_full_design, then divide.
The kernel reports energy in microjoules (µWh) on most systems, so a 52,500 mWh design shows as 52,500,000 if the unit is µWh instead of mWh. Check the label printed next to the value; misreading the unit produces a health percentage off by a factor of 1,000. Ubuntu, Fedora, and Arch all ship upower by default, but headless server installs and some minimal Debian configurations don't.
Failure handling matters here because Linux ACPI implementations vary wildly by laptop brand. Some Sanyo and third-party replacement batteries report energy_full equal to energy_full_design permanently, which reads as 100% health regardless of actual wear. Cross-check with cat /sys/class/power_supply/BAT0/cycle_count, which many firmwares also fail to populate. If both cycle count and full-charge capacity are missing or static, use a live runtime test: note the time, unplug, run a fixed workload like a 1080p video loop at 50% brightness, and see how long until the machine shuts down.
ThinkPads and Framework laptops have the best Linux battery reporting in the industry, because their embedded controllers expose full ACPI battery data. Cheap consumer models from Acer and Asus often expose only voltage and current. In that case, charge to full, note the voltage, and watch how fast it sags under load — a healthy 3-cell pack holds above 11.1V for at least half its rated runtime.
Use a Live Runtime Test When Software Reports Lie
A runtime test bypasses software entirely and measures what actually matters. The method: set the display to 50% brightness, disable sleep and automatic dimming, unplug the charger, and run a repeating workload — a 1080p video on loop is the usual pick because it stresses CPU, GPU, and display consistently. Record the start time and note exactly when the machine shuts down or hibernates at the critical-battery threshold.
Compare the measured runtime against the manufacturer's rated battery life and against the amount of capacity the report says remains. A laptop reporting 85% health should deliver roughly 85% of its original video-playback runtime. If it delivers 50%, the cell is sagging under load — a common form of failure where surface voltage reads normal at idle but collapses under sustained draw. That pattern doesn't show up in any capacity report.
The test also confirms whether a swelling or heat issue exists in disguise. If the bottom of the laptop gets dramatically hotter than usual during the test, or the battery percentage drops in jumps (91% to 87% to 82% in seconds), the cell can't sustain current. Neither symptom is visible in a powercfg or macOS report. Repeat the test three times, and if results vary by more than 20%, the controller is unreliable.
Attribution matters for the comparison. Apple publishes rated battery life in video hours for each MacBook model; Microsoft publishes UP TO ratings that frequently assume lighter workloads than video playback. According to Apple's MacBook Pro tech specs, the 14-inch model rates up to 17 hours of Apple TV app playback, a figure that drops to 11 hours of browsing in independent testing by outlets like Notebookcheck. Use the video figure for a fair comparison, since the test workload is video.
Interpret Cycle Count, Wear Level, and Temperature Together
A single health percentage hides the story. Cycle count, wear level, and operating temperature each explain part of the decline, and they disagree in useful ways. A battery with 150 cycles and 65% health is not simply aged — it was exposed to heat, overvoltage, or a defective charger. A battery with 700 cycles and 78% health is textbook normal.
Lithium-ion cells degrade by two mechanisms: calendar aging (slow loss over time regardless of use, roughly 2-3% per year at room temperature) and cycle aging (loss from charging and discharging, 20-30% over 300-500 cycles). Panasonic's NCR18650 datasheets — widely used as a reference in the industry — chart both. Heat accelerates both: every sustained 10°C above 25°C roughly doubles the degradation rate, which is why a laptop in a hot climate ages twice as fast as one in a cool office.
Cycle count interpretation also depends on how the laptop charges. Plugged-in laptops that sit at 95-100% most of the time accrue very few cycles but still lose calendar capacity. A five-year-old desktop-replacement machine with 90 cycles can report 80% health purely because it spent four years at full charge and warm. That's why newer laptops from Apple, Lenovo, Dell, and Framework include charge-limit or optimized-charging features that cap the battery at 60-80%.
Temperature is the value most reports get wrong. Windows battery reports don't include it; macOS shows it in the Power panel; Linux exposes it as temp in /sys/class/power_supply/BAT0/ in tenths of a degree Celsius. If the reported temperature at idle exceeds 40°C (104°F), the battery is either failing or the cooling system is compromised. Both accelerate permanent damage, so fix the thermal issue before judging the capacity reading.
Compare Windows, macOS, and Linux on the Same Battery Numbers
Each platform exposes the same underlying ratio but packages it differently, and one clear winner emerges for diagnosis while another wins for convenience. The comparison assumes the laptop is plugged in, fully charged, and has been in normal use for at least a year.
| Feature | Windows 11 (powercfg) | macOS (System Information) | Linux (upower) |
|---|---|---|---|
| Health percentage shown directly | No — compute capacity ÷ design | Yes — Maximum Capacity | Yes — capacity field |
| Design capacity in mWh | Yes | Yes | Yes |
| Cycle count | Yes | Yes | Firmware-dependent |
| Month-by-month degradation history | Yes | No | No |
| Battery temperature | No | Yes | Yes (tenths of °C) |
| Charge-limit feature visibility | Partial (OEM tools) | Yes (Battery Health Management) | Yes (sysfs charge_control) |
| Command required | powercfg /batteryreport | Option-click Apple menu | upower -i |
Windows wins outright on diagnosis because of the capacity history table. Apple wins on convenience because the comparison is already calculated and thresholds match the service policy. Linux wins on raw control for power users with charge thresholds and telemetry scripts, but loses badly on consistency across laptop brands. If you want to understand why a battery died, Windows prints the timeline. If you want to know whether to book a service appointment, macOS tells you in one glance.
Vendor tools aren't neutral referees. Lenovo Vantage, Dell Power Manager, and HP Support Assistant all read the same firmware data as the built-in tools but sometimes apply stricter health formulas for their own warranty policies. A laptop sitting at 78% in powercfg might read 75% in Vantage, because the vendor subtracts a safety margin. That matters when arguing a warranty claim: the vendor's number is what the warranty department will cite.
Skip the third-party battery apps that flood search results. Most pull the same numbers from the same APIs and add advertising. The two exceptions are BatteryInfoView on Windows, which adds log-over-time functionality, and coconutBattery on macOS, which displays historical capacity readings alongside the current one. Neither reveals anything the built-in tools can't.
Know When a Reading Means Replacement Is Due
Numbers alone don't justify a new battery. Purchase decisions follow three thresholds, each with a different cost and risk profile. The first is convenience: below 60% health, a laptop that once ran eight hours now runs under five, and daily users will feel it. The second is safety: any swelling, unusual heat, or rapid unexplained capacity drops justifies replacement immediately. The third is warranty timing: most major-brand warranties cover battery defects within one year or 300 cycles, per Lenovo, Dell, and HP published policies, and replacement inside that window is free.
Replacement costs vary by form factor. A standard 3-cell or 4-cell internal battery for a mainstream 14-inch laptop typically runs $50-90 in parts, with labor folded in if you do it yourself using vendor teardown guides. Apple's flat battery service fee is $129 for MacBook Air and $159-249 for MacBook Pro models depending on year, per Apple's official service pricing page — a significant premium over the parts cost but including the labor and the warranty on the replacement.
The economics flip when the laptop is older. A five-year-old machine with a 55% battery that still meets your speed needs is worth a $70 cell. A five-year-old machine with a 75% battery and a failing keyboard is not worth the labor. Judge the battery against the rest of the machine, not in isolation.
There's a fourth scenario worth knowing: batteries that report healthy but fail spectacularly. This usually happens after a drop or liquid contact. The cell reads 90% health one week and shuts off at 40% the next. Any sharp discrepancy between the software report and real-world runtime is a signal to replace immediately, regardless of what the percentage says.
FAQ
How accurate is the Windows battery report?
Windows reads the battery controller's own values, so design and full-charge capacity are as accurate as the firmware allows, typically within 2-3% of a lab measurement. The history table and runtime estimates are less reliable, since they depend on how often the machine was used on battery. Trust the two capacity figures and treat the rest as a rough guide.
Does a battery health reading include charger wear?
No, which is why a healthy battery can still charge slowly. The power brick and charging circuit degrade separately from the cell. If capacity reads 90% but full-charge time has doubled, test the charger on another device or measure the charger's output with a USB-C power meter — a worn charger caps at 45W instead of the rated 65W, stretching charge time by roughly 30 minutes.
Can I reset a battery's wear level?
No, and anyone claiming a software reset can is wrong. "Calibration" procedures that drain to 0% and recharge simply refresh the controller's estimate of full charge; they don't restore lost capacity. Capacity loss is chemical, driven by lithium inventory depletion and electrode degradation. A calibration can shift the reported percentage by a few points but never recovers real runtime.
What does it mean if cycle count is blank or zero?
Blank means the firmware doesn't report cycle data to the OS, and that's common on third-party replacement cells and cheap consumer laptops. Zero is more suspicious: either the machine was reset or someone replaced the battery with a unit lacking a smart chip. Confirm by checking whether the design capacity in the report matches the laptop's published spec sheet, and if it doesn't, the battery may be a replacement.
Is a battery at 75% health worth keeping?
Yes, for most users. At 75% health a laptop that originally ran eight hours now runs about six, which is adequate for desk use and short trips. Replacement becomes sensible below 60%, when runtime drops under five hours, or immediately if swelling, heat, or sudden shutdowns appear. Cost decides it: a $60-80 part vs. a $1,000+ laptop.
Why does my Mac show "Service Recommended" at 82% capacity?
Apple's threshold sits at 80% capacity or 1,000 cycles, and "Service Recommended" appears when either is reached, according to Apple's published battery service guidance. An 82% MacBook hitting 1,000 cycles triggers the message even though capacity is technically above the floor. Battery Health Management pausing charges can also cause the message to appear and then clear.
How much runtime loss corresponds to a given health percentage?
Roughly linear: 80% health means about 80% of new runtime under identical conditions. Real-world runtime also depends on power draw, so an older battery paired with updated software that draws more current loses more minutes than the percentage alone implies. Workload changes, not battery chemistry, usually drive the biggest runtime surprise.
Can a laptop battery explode without warning?
Swelling, not explosion, is the realistic risk, and it produces visible or tactile signs — a bulging trackpad, a chassis seam that won't close, or a slight chassis bend — before anything dramatic happens. Any swollen battery is a fire risk and should be removed from the laptop and taken to a recycling center or a repair shop, never punctured. No software report detects swelling.