Most laptop hardware doesn't fail without warning. The warning signs are often there — unfamiliar sounds, unexpected behaviour, performance that changes under specific conditions — but they're easy to dismiss as software glitches, background processes, or just "the laptop being slow." Understanding what those signals actually mean can make the difference between catching a fault early, when more repair options are available, and dealing with a complete failure that arrives at the worst possible moment.
This article covers the most common hardware failure indicators, what each tends to suggest about the underlying cause, and what a reasonable response looks like in each case.
Storage Drive Failures
Storage faults tend to give a lot of warning before a complete failure, but the warnings can be easy to attribute to software. They're worth taking seriously because a failing storage device poses a data risk regardless of what else is going on.
Signs to watch for:
Slow, sluggish boot times that weren't present before and haven't changed with OS reinstalls or disk cleanup suggest the storage device is taking longer to complete operations — either because of sector remapping on a failing HDD, or because of controller slowdown on a degraded SSD. Most SSD controllers preserve read performance longer than write performance, so if writes feel particularly slow, that's meaningful.
Files that open incorrectly, applications that crash at launch, or an OS that fails to boot despite appearing to start correctly are all consistent with storage errors during read operations. Each error the drive encounters internally may or may not be reported to the operating system; many are silently corrected using spare sectors until the spare pool is exhausted.
The single most useful check is running CrystalDiskInfo (Windows) or smartmontools (cross-platform). These tools query the drive's internal S.M.A.R.T. diagnostic data, which includes reallocated sector counts, pending sector counts, uncorrectable error counts, and power-on hours. Elevated counts in the first three categories warrant immediate data backup and usually drive replacement.
For SSDs, wear level and remaining spare block counts are the key indicators. Drives with very high write totals relative to their rated TBW (total bytes written) endurance spec deserve attention even if they show no errors yet.
Memory (RAM) Problems
RAM faults are among the more disruptive hardware failures because memory is used by every process running on the system. A single bad memory cell will cause errors in whatever program happens to use that address at that moment — which means the symptoms can look completely random and unrelated.
Signs to watch for:
Random application crashes without a clear pattern are a classic early indicator. If your browser crashes, then your office application crashes an hour later, then the system itself bluescreens with a different error code each time, and this happens across a clean OS installation, RAM is a reasonable suspect.
Blue screen errors (on Windows) with codes like MEMORY_MANAGEMENT, SYSTEM_SERVICE_EXCEPTION, IRQL_NOT_LESS_OR_EQUAL, or PAGE_FAULT_IN_NONPAGED_AREA are strongly associated with memory faults, though they can have other causes. The key is whether the error codes are consistent across multiple crashes or vary each time — variation is more indicative of RAM problems.
Corrupted files — documents that won't open, images that display with artefacts, archives that fail integrity checks — can also result from RAM errors if the corruption happens while data is being written to disk, since the write goes through memory first.
The standard test is MemTest86, a bootable utility that tests RAM independently of the operating system. A clean pass across all test patterns provides reasonable confidence. Errors, particularly errors in the same address range across multiple passes, confirm a fault. Note that MemTest86 can take several hours to complete thoroughly; a single quick pass is not sufficient to rule out subtle faults.
GPU and Display Problems
Graphics hardware faults manifest visually, which makes them easier to recognise than some other failure modes — but they're also easier to confuse with display hardware faults or driver issues.
Signs to watch for:
Visual artefacts — random pixels, coloured blocks, horizontal lines, or distorted geometry — that appear during graphics-intensive tasks like gaming or video playback typically point to the GPU rather than the display. If the artefacts appear on an external monitor connected to the same port, the display itself is ruled out. If they only appear on the built-in screen, the fault could be either the GPU, the display cable, or the panel.
Crashes or driver errors that occur specifically when the GPU is under load suggest either a failing GPU chip, inadequate thermal performance in the GPU cooling path, or a power delivery issue on the GPU power rail. The correlation with load is the key distinguishing factor — a fault that only manifests under load almost always has a thermal or power component.
A blank screen at boot — where the device powers on but the display remains dark — is worth distinguishing from a complete no-power fault. If the device makes sounds, LEDs illuminate, and you can hear (or infer from the timing) that it's booting but the display stays dark, the problem may lie in the backlight circuit, the display cable, or in the GPU's failure to initialise the display output.
Power and Charging Irregularities
Power-related symptoms are sometimes intermittent in the early stages, which makes them easier to dismiss. Intermittent faults in power circuits tend to progress; early attention often means a simpler repair.
Signs to watch for:
Unexpected shutdowns — not crash-to-blue-screen, but sudden power loss as if the battery was removed — without any thermal warning suggest a power delivery problem rather than a software fault. This can be a failing battery that's losing capacity under load, a voltage regulator that's behaving erratically, or a connection fault in the power path.
Charging that works only in certain cable positions, or stops working then resumes, points toward the physical charging connector or the cable — both wear items that are relatively straightforward to address. If charging stops working regardless of cable or position, the fault has moved somewhere on the motherboard side.
Battery percentage that jumps unexpectedly — dropping from 60% to 20% immediately, or reading 100% one moment and shutting down for low battery the next — indicates either battery cell degradation or a calibration problem in the battery management circuit. Either way, the battery should be assessed before the pack fully fails, as a swollen or damaged cell is a safety concern.
Unusual Sounds
Mechanical components — the fan, and on older devices the hard drive — are the only parts of a laptop that should produce audible sound during normal operation. Changes in those sounds are worth paying attention to.
Signs to watch for:
Fan noise that is louder than it used to be, particularly if it has a grinding, scraping, or rattling quality, indicates bearing wear in the fan motor. A worn fan doesn't just become noisier — it moves less air, which affects cooling performance. At some point the bearing may seize, at which point the laptop has no cooling at all and will shut down rapidly under load.
Hard drives (where still fitted) produce a regular low-level mechanical sound during operation. A clicking sound — often described as a regular "tick tick" pattern — during read or write operations is a serious warning sign. This is the head actuator attempting and failing to read a sector, a sound sometimes called the "click of death." A drive producing this sound should be considered at risk of imminent failure, and data should be backed up without delay.
The Distinction Between Hardware and Software
Many of the symptoms described above can also result from software problems — driver corruption, malware, OS file system errors, or a botched update. The key tests that distinguish hardware from software are:
First, whether the fault persists across a completely clean OS installation. A true hardware fault exists independent of what software is running; a software fault does not survive a reinstall. Second, whether built-in hardware diagnostics (available on most platforms through the BIOS setup menu, or through Apple Diagnostics on Mac hardware) report any faults. Third, whether the fault is reproducible in a specific, predictable way that correlates with a physical condition — temperature, physical position, power state, storage access pattern.
Faults that survive a clean reinstall, trigger hardware diagnostics, or correlate with physical conditions are almost always hardware problems. Faults that disappear with a reinstall, or that only occur with specific software, are almost certainly software.
Why Early Diagnosis Matters
A laptop that intermittently crashes due to a failing RAM chip is usually straightforward to address if caught before the chip fails completely. A laptop that develops occasional GPU artefacts may need a rework of the GPU BGA at some point, but an early rework is more likely to succeed than one attempted after the chip has been thermally stressed through hundreds of partial failures.
The same principle applies across most hardware fault categories. Early diagnosis gives more options: more time to back up data before a storage drive fails completely, more repair approaches before secondary damage accumulates, and more information to work with because the failure hasn't fully developed.
If your device is showing any of the symptoms described here — particularly if multiple symptoms are occurring — it's worth having it properly assessed rather than waiting to see whether it gets worse. The cost of a diagnostic is much smaller than the cost of recovering from a complete failure.
Concerned About Your Device?
Describe the symptoms you're seeing and we'll give you an honest view of whether it warrants a professional assessment.
Contact the Workshop