Measure it, then fix it.
One question each, end to end: what the symptom looks like, what it means, and what to change next.
Hardware guides, from symptom to fix
These 48 guides each answer one hardware question end to end: what the symptom looks like, how to measure it with the free browser stress test, and what to change once you have the number. They cover thermal throttling, telling a processor limit from a graphics one, reading a benchmark result honestly, improving sustained performance, and how path tracing works.
The guides sit between the other parts of the site. The fix articles start from a single complaint, such as a laptop that is slow on battery, and rank its causes; the diagnosis pages explain what the test shows for each symptom; the glossary defines every term a guide uses; and the methodology page sets out what each stage of the test measures. A guide is the place to understand the idea, and the test is the place to see it on your own machine.
Every guide here starts from something a machine is doing that you can see, and ends with something you can change. None of them ask you to install a monitoring suite first, and none of them recommend a hardware purchase as the opening move: most sustained-performance problems on a laptop are a cooling or power-limit problem, and several of those are free to fix.
The order they are worth reading in depends on the symptom. If the machine is fast and then slowly stops being fast, start with thermal throttling. If you are trying to work out whether the processor or the graphics card is holding a game back, start with the bottleneck guide. If you already have numbers from somewhere and want to know whether the difference between two of them is real, start with reading benchmark results.
Each guide says plainly what a browser-based measurement can support and what it cannot. Temperatures, fan speeds and power draw are not readable from a web page on any browser, so where a cause is named it is inferred from the shape of a throughput curve, with the confidence that inference deserves.
Good places to start
Memory latency vs bandwidth: what a random-walk test shows
SystemCheck · 6 min read · updated 29 October 2026
Single-thread vs all-thread memory bandwidth
SystemCheck · 5 min read · updated 29 October 2026
What "not measured" means in your report
SystemCheck · 5 min read · updated 29 October 2026
Reading the confidence interval on every figure
SystemCheck · 6 min read · updated 29 October 2026
Is 80C safe for a CPU? What a browser can tell you
SystemCheck · 5 min read · updated 29 October 2026
Mac in Safari: why only 8 threads show
SystemCheck · 4 min read · updated 29 October 2026
Is a browser stress test as good as Prime95 or Cinebench?
SystemCheck · 5 min read · updated 29 October 2026
Why a PC slows down over time, and how to tell which kind
SystemCheck · 6 min read · updated 29 October 2026
How to check a used laptop before you pay
SystemCheck · 5 min read · 5 steps · updated 29 October 2026
PC specs and performance: what a test's numbers mean
SystemCheck · 6 min read · updated 29 October 2026
GPU stress test: how long is long enough?
SystemCheck · 5 min read · updated 29 October 2026
How to benchmark a PC for gaming without installing
SystemCheck · 6 min read · updated 29 October 2026
Frame pacing vs FPS: why smooth is not the same as fast
SystemCheck · 5 min read · updated 29 October 2026
How to reduce a GPU bottleneck once you have measured it
SystemCheck · 6 min read · updated 29 October 2026
What CPU throttling means, read off one chart
SystemCheck · 5 min read · updated 29 October 2026
What the WebGPU compute test measures, and what it does not
SystemCheck · 5 min read · updated 29 October 2026
Testing an SSD's sustained write speed in a browser
SystemCheck · 6 min read · updated 29 October 2026
Responsiveness test: input delay your PC adds under load
SystemCheck · 5 min read · updated 29 October 2026
Recovery test: how fast a machine comes back after load
SystemCheck · 5 min read · updated 29 October 2026
How to document laptop throttling for a warranty claim
SystemCheck · 6 min read · updated 29 October 2026
Before and after: measuring a repaste or a fan clean
SystemCheck · 6 min read · updated 29 October 2026
Online exam system check: what each check actually tests
SystemCheck · 6 min read · updated 29 October 2026
The webcam, microphone and connection lines we use for calls
SystemCheck · 5 min read · updated 29 October 2026
Battery or plugged in: what changes in a laptop's run
SystemCheck · 6 min read · updated 29 October 2026
Heat, a power limit or a busy machine: what a curve shows
SystemCheck · 7 min read · updated 29 October 2026
Bufferbloat: why your ping rises when the line is busy
SystemCheck · 6 min read · updated 29 October 2026
Jitter: what an uneven ping does to calls and games
SystemCheck · 5 min read · updated 29 October 2026
Bluetooth headphone delay: measure it, then cut it
SystemCheck · 5 min read · updated 29 October 2026
Check 5.1 and 7.1 surround on Windows and macOS
SystemCheck · 5 min read · updated 29 October 2026
Is hardware acceleration on? A check for each browser
SystemCheck · 5 min read · updated 29 October 2026
WebGPU vs WebGL: what each is and which one you have
SystemCheck · 6 min read · updated 29 October 2026
sRGB, Display P3 and Rec. 2020: what a browser can tell
SystemCheck · 5 min read · updated 29 October 2026
How to find your mouse's real DPI
SystemCheck · 5 min read · updated 29 October 2026
How to check if your GPU decodes AV1 in hardware
SystemCheck · 6 min read · 3 steps · updated 29 October 2026
How much VRAM can a browser use, and why not all of it
SystemCheck · 6 min read · updated 29 October 2026
Can my laptop run a local LLM? The memory arithmetic
SystemCheck · 7 min read · updated 29 October 2026
How to spot a fake USB drive or SD card
SystemCheck · 6 min read · 3 steps · updated 29 October 2026
How long will my laptop battery last under load?
SystemCheck · 6 min read · updated 29 October 2026
What controller polling rate means, and what a browser sees
SystemCheck · 6 min read · updated 29 October 2026
What resolution is my webcam really sending?
SystemCheck · 6 min read · updated 29 October 2026
What a reaction time test actually measures
SystemCheck · 6 min read · updated 29 October 2026
How path tracing works, from one ray to a finished image
SystemCheck · 7 min read · updated 16 September 2026
Path tracing vs ray tracing: what the difference actually is
SystemCheck · 6 min read · updated 16 September 2026
How to diagnose and fix thermal throttling
SystemCheck · 9 min read · 8 steps · updated 29 October 2026
How to read a benchmark result without fooling yourself
SystemCheck · 8 min read · updated 29 October 2026
How to tell whether your CPU or GPU is the bottleneck
SystemCheck · 8 min read · 7 steps · updated 14 September 2026
How to improve sustained performance, not just peak scores
SystemCheck · 9 min read · 8 steps · updated 12 September 2026
How accurate is a browser benchmark, really?
SystemCheck · 10 min read · updated 12 September 2026
Questions
It depends on the symptom. If the machine is fast and then slowly stops being fast, start with thermal throttling. If you want to know whether the processor or the graphics card holds a game back, start with the bottleneck guide. If you already have two numbers and want to know whether the difference is real, start with reading benchmark results.
No. None of them ask you to install a monitoring suite first; the measurement is the free browser test. A browser cannot read temperatures, fan speeds or power draw, so where a guide names a cause it is inferred from the shape of the throughput curve and says so.
The standard run takes 3 minutes and loads the processor, memory and graphics in turn. The deep run takes 9 minutes and adds a display check, a combined processor and graphics load, and a hot re-test against the cold run.
The guides are practical: one symptom, one measurement, one change. The blog holds the longer arguments behind how SystemCheck measures, such as why a peak score misleads.