Drop a HWiNFO64 CSV sensor log to compare GPU, memory, +12V rail, CPU, and NVMe peaks locally against cited AMD, Intel, NVIDIA, and ATX 3.1 limits.
Self-serve measurement workspace: Your CSV log is parsed locally in this browser tab and never uploaded. Each reading is placed beside a dated manufacturer specification link—or marked unavailable when no primary vendor document states a limit (such as NVIDIA GPU Hot Spot temperature). This instrument compares numbers to published limits; it does not diagnose hardware health or review individual logs by email.
1 · Load or Paste a HWiNFO64 Sensor Log CSV (Celsius [°C])
2 · Hardware Sensor Case Sheet vs. Published Vendor Limits
Standard GPU telemetry sums total board power; per-pin current balance requires per-pin shunt hardware (ROG Astral Power Detector+ or WireView Pro II).
Compare your own HWiNFO64 log against primary vendor specifications
Forum threads routinely mix software-reported GPU board power with wall watts, compare 30-second benchmark runs against 20-minute heat-soaked runs, and cite folklore thermal ceilings that appear nowhere in AMD, Intel, or NVIDIA documentation.
Where Do I Sit? separates three things cleanly on one printable case sheet:
What your own HWiNFO64 CSV actually recorded (peak, median, minimum +12V input voltage under load, row-by-row core-to-hotspot delta where the sensor exists, and total log duration).
What the manufacturer’s own specification page states—with a direct link and capture date, or unavailable when no primary document states a figure.
Where the reading sits in our open hardware censuses—printing exact sample counts (n = 0 counted reader rows until the n = 5 exact-model floor is reached) alongside our first-party ASUS ROG Astral RTX 5080 OC bench dataset.
Does Where Do I Sit upload my HWiNFO64 CSV log to a server?
No. File reading and CSV parsing run entirely in your browser tab using the local FileReader API. Nothing leaves your machine unless you click through to GitHub to submit a summary row to a public census.
Why does the GPU Hot Spot limit say ‘unavailable’ for NVIDIA cards?
On GeForce specification pages, NVIDIA publishes Maximum GPU Temperature (for example 88 °C on RTX 4080 Super and RTX 5080) rather than a Hot Spot ceiling, and RTX 50-series Blackwell removed the user-visible Hot Spot sensor altogether. Where a manufacturer publishes no primary number, this registry leaves it unavailable instead of repeating forum estimates.
Why did my CSV fail closed when temperatures were logged in Fahrenheit?
Every published vendor thermal limit in the registry is specified in degrees Celsius. Rather than silently guessing unit conversions across custom column layouts, the parser requires explicit [°C], [W], [V], and [MHz] unit headers and names the exact column that failed.
Updates and corrections
Last updated
Methodology
See our methodology for research and review standards.
Update log
2026-11-21 — Initial public release: fail-closed browser-local HWiNFO64 CSV parser, CC BY 4.0 vendor thermal/voltage/power limits registry (vendor-limits-registry.json), 10-minute sustained-load conformance check, and printable 1-page case sheet.
Corrections
Spotted an error? Email a correction with the affected line and what you found. The draft includes this page’s address. Verified changes are recorded here. Contact details.