A PC that will not boot can fail at several different stages. Pressing the power button might do nothing, the fans might spin without a picture, or the computer might show its logo and then stop. Those symptoms point to different checks. The quickest way forward is to identify the last stage that works, change one thing at a time, and record what happens. This guide focuses on desktop PCs, where you can inspect and isolate individual components.

Identify what “won’t boot” means

Before opening the case, watch a startup attempt carefully. Note whether fans move, lights appear, the monitor wakes, and any message stays on screen. A spinning fan proves that some power is present; it does not prove that the power supply or motherboard is working correctly.

• No power response: Nothing visibly happens when you press the case button.
• Power but no startup display: Fans or lights operate, but you cannot see the firmware screen.
• Firmware works, operating system does not: You can enter setup, or see a boot-device error, but Windows or another operating system never loads.
• Startup begins, then fails: The operating system starts loading before freezing, restarting, or showing an error.

Think about the last change before the failure. A move, new component, changed firmware setting, power interruption, or update provides a useful starting point. Reverse a recent reversible change before dismantling an otherwise working system.

Make the next checks safe and repeatable

Before touching internal parts, shut down if possible, switch off the power supply, and unplug the power cable. Press the case power button briefly to help discharge the system, but do not assume every circuit is fully discharged. Never open the power supply itself. Stop if you notice smoke, a burning smell, liquid, or a scorched connector.

Work on a stable surface, avoid carpet, and handle components by their edges. Follow antistatic precautions and photograph connections before removing them. Disconnect power again before each internal change. Keep the CPU cooler installed during powered tests, and keep loose screws and tools away from the board.

If pressing the button does nothing

Start outside the case. Check that the power cable is firmly connected at both ends and that the supply’s rear switch is on. Try a known-working outlet and, if available, a compatible known-working power cable. Bypass a suspect extension lead or power strip. Do not change a voltage selector experimentally.

Inside a conventional ATX desktop, inspect the main motherboard power connector and the CPU power connector near the processor. Both must be fully seated. A CPU power cable and a graphics-card power cable are not interchangeable, even if they look similar. Use the connector labels and the hardware manuals rather than forcing a plug.

If the supply has modular cables, inspect its cable connections too. Never reuse modular cables from another supply unless the manufacturer explicitly confirms compatibility. The component ends can look identical while the supply ends use different wiring.

Check the case power-switch connection against the motherboard manual. A disconnected or misplaced switch lead can make a healthy system appear dead. If your motherboard has a built-in start button, use it as directed to help distinguish a case-switch problem from a wider power fault. Standby lighting alone cannot clear the supply of suspicion.

If the PC powers on but shows no picture

Rule out the display path first

Make sure the monitor is powered on and set to the input you are using. Reseat the video cable, then try a known-working cable or display if one is available. Temporarily remove docks, adapters, and extra monitors so the signal takes a direct route.

With a dedicated graphics card installed, connect the monitor to that card for the initial test. Motherboard video outputs require a processor with supported integrated graphics and an appropriate configuration. Moving the cable to the motherboard is only a useful isolation test when that graphics capability exists.

After a memory change or firmware reset, allow the board to complete its documented memory-training process. Some systems restart during this process. Repeatedly cutting power can interrupt normal initialization, so consult the manual before treating a longer first start as failure.

Use startup clues to choose a test

Record any persistent diagnostic light, displayed code, or beep pattern, then check the motherboard documentation. These indicators identify where initialization stopped; they do not necessarily identify the defective part. A memory-related indication, for example, can involve seating, settings, a slot, or the processor’s memory connection.

With power disconnected, reseat the memory. Test one module in the slot recommended by the manual for single-module operation. If it fails, try another module in that same recommended slot. Keep notes so you can distinguish a problem that follows a module from one associated with a slot or configuration.

Inspect the graphics card’s seating and required power connections. Release retention mechanisms as instructed before removing anything. If supported integrated graphics are available, testing without the dedicated card can help isolate the graphics path. A successful result narrows the problem but does not automatically prove the card itself is faulty.

Reduce the system to essential hardware

If the initial checks reveal nothing, disconnect unnecessary USB devices, extra expansion cards, and storage drives. Leave the motherboard, processor with cooler, one memory module, power supply, and whatever graphics hardware is needed for a display. Disconnect suspect case USB accessories if they could be causing a short, using the manual to identify their headers.

The aim is to reach firmware setup, not load the operating system. A computer can normally reach setup without a storage drive. If it does, reconnect removed items individually, with power disconnected between attempts, until the symptom returns.

If instability began after a firmware setting change, restore defaults using the motherboard’s documented procedure. Before resetting settings, obtain any drive-encryption recovery key you may need. A reset can change settings that affect access to an existing installation. For a new build, also check the board’s CPU support information and required firmware version; do not flash firmware merely because the screen is blank.

If firmware opens but the operating system will not load

Check whether firmware detects the intended system drive. For a missing SATA drive, inspect both data and power connections. For an NVMe drive, check seating and slot compatibility against the manual. Disconnect power before making adjustments. A missing drive is a detection problem before it is an operating-system repair problem.

If the drive appears, check that the correct operating-system boot entry is selected. Remove external storage that might be taking priority. Avoid randomly switching storage-controller modes, UEFI compatibility options, or security settings: an existing installation may depend on their previous values.

When the operating system begins loading and then fails, note the exact error and use its recovery environment or official recovery media. Startup repair or a safe-mode startup may help distinguish damaged startup files from a driver problem. Do not format, initialize, or reinstall over a drive containing needed files. Protect the data before attempting changes that could overwrite it.

Know when component substitution makes sense

If a minimal configuration still cannot reach firmware, a compatible known-working power supply, memory module, or graphics device can help isolate the fault. Substitute one part at a time, and use the replacement supply’s own cables. Without suitable spares, a repair shop can test these possibilities more efficiently than buying parts on suspicion.

Bring your symptom notes, recent changes, and isolation results to the technician. Once the PC starts again, reconnect remaining devices gradually and confirm that it can restart normally. That final check helps establish whether you found the cause or simply disturbed an intermittent connection.