PerformanceDeck
One benchmark score, plus the breakdown showing which part of the machine moved.
Free right now — no account, no card
Preview
1 interface design, then a screenshot of it running. Drag, scroll or use the arrows.
Free while we’re in preview
Get PerformanceDeck
One benchmark score, plus the breakdown showing which part of the machine moved.
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, PerformanceDeck is included in every plan along with the rest of the range.
Four measurements folded into one number
Single-thread CPU, multi-thread CPU, memory copy bandwidth, and sequential disk write with the fsync inside the timed window. Each raw measurement is divided by a fixed reference constant compiled into the build and printed by --help, giving a subscore where 100 means you matched the reference exactly. Weights are integer points, 30/30/20/20. The arithmetic is published, and an independent reimplementation reproduces the overall score bit for bit.
Skip --disk-dir and the disk test doesn’t run. The row vanishes from the output entirely, no default is substituted for it, the three remaining weights renormalise to 0.375/0.375/0.25, and compare refuses to diff a cpu-mem run against a full one in either direction.
Before you trust a drop
Run-to-run noise is real and it’s the main source of false alarms. On the containerised host this build was tested on, idle cpu-mem scores moved by up to 7% between consecutive pairs, and idle disk throughput ranged from 66 to 890 MB/s. On virtualised storage set --threshold-pct to 25 or more or you’ll chase regressions that were never there.
There’s no GPU benchmark of any kind. Core temperature and thermal throttling are invisible to it, so a throttling machine just posts lower numbers with no explanation attached.
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...
Advanced & technical details Command-line builds, source code, checksums and a recorded session — for developers and IT teams
A working prototype of a composite benchmark: it measures CPU single- and multi-thread, memory bandwidth and disk throughput, then combines them into one score with a fully transparent breakdown, so when the number drops you can see which part of the machine did it. Verified here that the weights sum to exactly 1.0 and the contributions sum bit-for-bit to the reported score. Its own tests caught a float-weight bug that made renormalised weights total 0.9999999999999999.
Naming status
Working name only — brand verdict **RENAME** (High). Collision: PERFORMANCEDECK.COM. Rename unless counsel clears and value is compelling.
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.
performancedeck 1.0.0 - composite weighted benchmark suite (Techlosoft PC Performance Console)
USAGE
performancedeck run [--disk-dir DIR] [--seconds N] [--mem-mib N]
[--disk-cap-mib N] [--save FILE] [--json]
performancedeck compare --baseline FILE [--current FILE] [--disk-dir DIR]
[--seconds N] [--mem-mib N] [--disk-cap-mib N]
[--threshold-pct P] [--json]
performancedeck show --result FILE [--json]
performancedeck help | -h | --help
COMMANDS
run Run every subsystem benchmark and print the weighted breakdown.
compare Diff a saved run against another saved run (--current) or against
a fresh measurement taken now. Exits 2 when the overall score drop
exceeds --threshold-pct.
show Print a previously saved result file.
FLAGS
--disk-dir DIR Directory to benchmark disk I/O in. One temp file is
created inside it and ALWAYS deleted again. Omit this
flag to skip the disk test (see PROFILES).
--seconds N Wall-clock budget per subsystem (default 2, max 600).
--mem-mib N Size of EACH of the two memory buffers (default 64).
--disk-cap-mib N Maximum bytes written by the disk test (default 256).
--save FILE Write the run as a JSON result file.
--baseline FILE Saved result to compare against.
--current FILE Saved result to use as the current side (default: measure
now).
--threshold-pct P Overall-score DROP that counts as a regression (default 10).
Disk throughput on shared or virtualised storage can vary
by 2x between runs; widen this for full-profile gating.
--json Machine-readable output.
SCORING - NOTHING HERE IS MAGIC
Every subscore comes from a real measurement normalised against a fixed
reference constant compiled into this build:
subsystem measures reference = 100 pts
cpu_single 1-thread xorshift+LCG kernel throughput 200.00 Mops/s
...Recorded from the shipped binary, not written by hand.
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