Free while we’re in preview

Get SystemPulse

Times one command thirty times over and tells you when it gets slower.

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, SystemPulse is included in every plan along with the rest of the range.

Your command, not your CPU

A benchmark tells you the machine is fast. It won’t tell you your app’s startup went from 340ms to 390ms between two commits. SystemPulse runs one command many times, discards the warmups, and reports the whole distribution: min, median, p90, p95, p99 and standard deviation. Save that as a baseline, then have compare re-measure later and exit 2 when the median or p95 regressed past your threshold.

systempulse bench `--runs` 50 `--label` v1.4.0 `--save` base.json -- ./myapp `--version`
systempulse compare `--baseline` base.json `--runs` 50 `--threshold-pct` 8 -- ./myapp `--version`

Every percentile is a run that happened

Nearest rank, no interpolation. That is deliberately not numpy’s default, and the difference is a definition mismatch rather than a bug. The formula and the raw per-run nanoseconds go into every baseline file, in execution order, so you can recompute the arithmetic and check it.

Small print, and it matters

For any run count of 20 or fewer, p95 collapses onto the single worst run; the tool warns you when that happens. Failed runs are still timed and still counted, so a command that starts failing fast will look faster. --timeout defaults to off, meaning a hung command hangs SystemPulse. There’s no shell either, so wrap pipes in sh -c yourself.

Reading on the job itself

Is my computer actually getting slower? Time the same thing thirty times and keep the median, not the average. One slow run is noise; a median that...

The rest of PC Performance Console

How fast this machine really is, and whether it stays that way — one core, the whole machine, a score you can compare, and a soak test.

CorePilot

Simple Ready

How much faster is your machine with every core than with one?

PerformanceDeck

Pro Ready

One benchmark score, plus the breakdown showing which part of the machine moved.

ThermalFlow

Team Ready

Benchmarks on a schedule and flags the day your machine got slower.

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

A working CLI prototype of the latency profiler: it times a specific command across many runs and reports the whole distribution — min, median, p90/p95/p99, stddev — then compares against a saved baseline to catch regressions, with a per-run timeout so a hung command can't fabricate a result. Verified: sleep 0.2 and sleep 0.5 land where they should, and every statistic matches an independent recomputation exactly. It measures wall-clock time; per-process CPU and memory attribution are on the roadmap.

Naming status

Working name only — brand verdict **AVOID** (Very High). Collision: SystemPulse. Rename completely.

What it draws on

Existing paid software whose best ideas shaped this program: Process Lasso Pro, AIDA64 Extreme, HWiNFO Pro, Argus Monitor.

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.

systempulse 1.0.0 - command latency profiler (Techlosoft PC Performance Console)

USAGE
  systempulse bench   --runs N [--warmup N] [--label NAME] [--save FILE]
                      [--timeout DUR] [--json] -- <command> [args...]
  systempulse compare --baseline FILE --runs N [--threshold-pct P]
                      [--warmup N] [--timeout DUR] [--json] -- <command> [args...]
  systempulse show    --baseline FILE [--json]
  systempulse help | -h | --help

COMMANDS
  bench     Time <command> over N runs and report the full distribution.
  compare   Re-measure <command> now and diff it against a saved baseline.
            Exits 2 when the regression exceeds --threshold-pct.
  show      Print a previously saved baseline.

FLAGS
  --runs N            Measured runs (default 30, must be >= 1).
  --warmup N          Discarded runs executed first (default 3).
  --label NAME        Free-text label stored in the baseline.
  --save FILE         Write the bench result as a baseline JSON file.
  --baseline FILE     Baseline JSON to read.
  --threshold-pct P   Regression threshold in percent (default 10).
  --timeout DUR       Per-run wall-clock limit, e.g. 2s, 500ms.
                      Default 0 = no limit (a hanging command hangs the tool).
  --json              Machine-readable output, including raw per-run samples.

EVERYTHING AFTER -- IS THE COMMAND UNDER TEST.

EXIT CODES
  0  success / no regression
  1  usage or runtime error
  2  regression detected above threshold

METHOD
  percentiles: nearest-rank: rank = ceil(p/100 * n) over ascending sorted samples, 1-based, clamped to [1,n] (IEEE-754 double arithmetic)
  stddev     : sample standard deviation (Bessel-corrected, divisor n-1); 0 when n < 2

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

SystemPulse

Preview