Free while we’re in preview

Get SensorDeck

Loads your machine for five minutes and shows whether it holds its speed.

The Mac button is for Apple Silicon. On an older Intel Mac, get this one instead.

One file, both ways. Double-click it for the window. Run the same file from a command prompt with arguments and it behaves as the command-line tool, because the engine is inside it. Nothing else to download and nothing to keep beside it.

Early preview. The window has been built and run, but not yet on a real Windows PC or Mac, so expect rough edges. The engine underneath it is fully tested.

One file — no installer
Runs on your machine, offline
Source code published below

Free while in preview. It isn’t signed yet, so Windows or macOS will ask you to confirm the first time you open it. At launch, SensorDeck is included in every plan along with the rest of the range.

There is no thermometer in this program

SensorDeck will never print a temperature or a fan RPM. Getting those needs WMI or a signed kernel driver on Windows, undocumented SMC keys on macOS, hwmon on Linux, all privileged and all vendor-specific. Plenty of tools claim the numbers and quietly return zeroes. This one measures the effect instead.

The same work, over and over

A machine that throttles gets slower the longer you keep it busy. So sensordeck soak --minutes 5 calibrates an exact operation count once, freezes it, and then performs exactly that many primality checks per interval over exactly the same narrow band of numbers. Only the elapsed time varies. You get the opening baseline, the closing figure, the decay between them as a percentage, and a sparkline so the shape is visible.

Read the shape, not just the number

A steady slide that flattens out is a machine settling at its sustainable clock. One deep notch is almost certainly a background process. Under 5% decay is STABLE, 15% or more is SIGNIFICANT.

Throughput decay is indirect evidence though. A battery profile, a virus scan waking up, or a burstable cloud instance that is designed to slow down after a few minutes all draw that same curve. Run the soak twice before you open the case.

Reading on the job itself

How to find out what is inside a computer Every operating system will tell you the model, the processor, the memory and the storage without any extra...

The rest of Hardware Health Hub

What this machine is and how it is holding up — a plain report, a sustained-load test, folders being watched, and a fleet at a glance.

HardwareLens

Simple Ready

One page describing this computer, honest about everything it cannot see.

DevicePulse

Pro Ready

Logs folder sizes on a schedule and shouts when one jumps more than 25% since last time.

PCHealth

Team Ready

Forty workstations, one sorted table, and the odd one out flagged for you.

Advanced & technical details Command-line builds, source code, checksums and a recorded session — for developers and IT teams

A working command-line build of the soak engine: it does an identical, fixed amount of work in every interval — verified constant, so a change in the time taken is a change in the machine and not in the measurement — then reports the throughput curve, a decay figure against the opening baseline, and the interval where the fall began. It reads no temperature sensor and says so: sustained throughput decay is indirect evidence, and background work, power profiles or a shared host can all produce it. One quiet run in four on this test machine produced a mild false positive, which is why it tells you to repeat a soak before believing a borderline verdict.

Naming status

Working name only — brand verdict **RENAME** (High). Collision: SensorDeck. Rename before launch.

What it draws on

Existing paid software whose best ideas shaped this program: AIDA64 Extreme, HWiNFO Pro, Hard Disk Sentinel Professional, PassMark PerformanceTest.

Command line

There is nothing extra to install. The program you download is the command-line tool as well: give it arguments instead of double-clicking it and it runs as one, with its output on your terminal. The standalone builds below are the same engine on its own, for machines where you would rather not ship a window at all.

SensorDeck 1.0.0 - sustained-load soak test: detect throttling by its effect

Usage:
  sensordeck soak    [--minutes 5] [--interval 10s] [--workers N] [--save run.json] [--json]
  sensordeck compare --before a.json --after b.json [--json]
  sensordeck show    --run run.json [--json]

Commands:
  soak     Run one fixed, identical unit of CPU work over and over for the
           whole duration, with NO idle time between units, and record the
           throughput of each unit. Reports the first-window baseline, the
           final-window average, the percentage decay between them, the
           interval where decay first crossed the threshold, min/median/p95
           throughput, an ASCII sparkline of the curve, and a verdict.
  compare  Diff two saved soak runs - the "did cleaning the fans / repasting
           the heatsink / moving the machine off the carpet actually change
           anything" command. Reports the change in baseline throughput and
           the change in decay.
  show     Re-print a saved run's full report without re-running anything.

Flags:
  --minutes N      soak only: how long to sustain the load, in minutes.
                   Fractional values are allowed (e.g. 1.5). (default 5)
  --interval dur   soak only: how long ONE unit of work should take, roughly.
                   SensorDeck calibrates an exact operation count to this
                   target once, up front, then holds that count fixed for
                   every interval of the run. Go duration syntax. (default 10s)
  --workers N      soak only: number of parallel worker goroutines.
                   (default: the machine's logical core count)
  --save FILE      soak only: write the full run, including every per-interval
                   measurement, to FILE as JSON. Writability is checked BEFORE
                   the soak starts so a bad path fails in a second, not in
                   five minutes.
  --before FILE    compare only: the saved run from before the change.
  --after FILE     compare only: the saved run from after the change.
  --run FILE       show only: the saved run to print.
  --json           Emit structured JSON instead of a text report.
  -h, --help       Show this help.

Verdicts: decay below 5% of baseline is STABLE, 5%-15% is MILD DECAY,
...

Recorded from the shipped binary, not written by hand.

Windows (amd64) macOS (Apple Silicon) macOS (Intel) Linux (amd64)

Source

Every file the program is built from:

console.go · console_test.go · guided.go · main.go

SHA-256 checksums · build instructions & scope notes · full build plan

SensorDeck

Preview