LagScope — The Connection Test Built for Gamers

Speed test says fine, you still lag? 55 ms means nothing without the game — fine for Fortnite, dead for Valorant .

CALIBRATING…

Warming up the snitch — first real numbers in a few seconds.

MEASURING

SHARE

Latency — last 3 minutesauto · grows with data

3m<br>5m<br>10m<br>15m

latency<br>baseline<br>spike threshold<br>lagspike<br>no data

Ready for

✓ good<br>⚠ borderline<br>✗ risky

per-game readiness from your measured baseline, jitter & loss ⓘ

Weaponize this result — card, link, Discord, Reddit, X.

Download card<br>Copy image<br>Copy link<br>Ask your AI

Share to

Discord

WhatsApp

Reddit

Telegram

Challenge a friend — beat my score<br>Post your verdict in our Discord

Call out your ISP

Forum signature badge

Create live badge

Copy BBCode<br>Copy image URL<br>new badge

Stays live while this tab measures. Goes OFFLINE after 2 h idle — honesty is the brand.

Share

Challenge someone back →

Latency now

—ms

round-trip time

Jitter

—ms

stability

Packet loss

—%

last 5 minutes

Lagspikes

this session

Uptime

—%

0 outages

baseline — · link: testing… · uptime — · 0 outages

Lagspike & outage log

Major+<br>All<br>Severe+<br>Clear

No crimes yet — spikes and outages land here when your ISP slips.

Do something with it

⇊Run bufferbloat stress test8 s of download saturation — is your router the hidden lag source?→<br>⬇Download ISP evidencetimestamped proof they can't wave away→<br>↗Share result cardyour score + verdict as an image — Discord, Reddit, X→<br>Join the Discordverdicts, outage alerts, help — discord.gg/C28Kq3qm7e→

Lagspikes per hour — last 12 h

Packet loss — last 10 min<br>failed-probe ratio

30 s buckets · GAME TRAFFIC = true UDP loss · STANDARD = failed-probe ratio (TCP hides some loss)

Last 24 hours<br>minor<br>major<br>severe / outage

spot patterns — e.g. congestion every evening

Bufferbloat stress test

8 s of download saturation while latency probes keep running; compares your latency<br>idle vs under load . A big jump means your router queues packets — lag whenever someone<br>downloads or streams. Uses ~10–30 MB.

Run stress test

Outages

no outages recorded

StartDurationStatus<br>no outages — connection has been up

ISP evidence

Everything recorded — lagspikes, outages, jitter, loss, uptime — bundled as proof<br>for your ISP. Print-friendly report in plain language; the CSV has every event with ISO timestamps.

Download ISP evidence<br>Export CSV<br>Create public proof link

Copy link

anonymized — metrics and ISP name only, no IP.<br>Anyone with the link can read it; deleted after 30 days.

Settings

Probe interval

250 ms — very sensitive<br>500 ms — gaming (default)<br>1000 ms — relaxed

how often latency is measured

Spike sensitivity

Gaming — spikes above ~+30 ms<br>Normal — ~+50 ms<br>Relaxed — ~+80 ms

how far above baseline counts as a spike (adapts to line noise)

Bandwidth test

every 10 s (~90 kbps avg)<br>every 20 s (~45 kbps avg)<br>every 60 s (~15 kbps avg)<br>off

64 KB download + 32 KB upload per test

Target server

measure to a different server (overrides node selection)

Measurement node

Auto — recommended<br>US — St. Louis<br>EU — France

Auto races every node at page load and uses the fastest.<br>Changing this clears measurements and restarts calibration.

Discord webhook alerts

Test

severe-spike & outage alerts to your Discord channel; nothing else leaves the browser

Alerts<br>beep on lagspike<br>reset history

Anonymous stats<br>Contribute anonymous stats<br>to the ISP leaderboard (ISP + country + metrics only — no IP, no tracking)

Stream overlay (OBS)

Copy URL

live RTT + readiness as a transparent widget — add as Browser Source, 360×130

What am I looking at?

Latency — round-trip time, you → server → you. Under 50 ms great, 50–100 fine, 150+ feels laggy. Hover the number for p95/p99 — a low median with a high p99 means a spiky line; gamers feel the tail, not the average.

Game traffic — header toggle: real UDP packets (WebRTC, 128 bytes at 30 Hz, no retransmission) instead of HTTP/TCP. This is how games actually talk, so latency and loss match in-game reality. Falls back to HTTP automatically if UDP can't get through.

Baseline — your normal latency (rolling median). Spikes are measured against it — the monitor adapts to your line.

Jitter — how much latency jumps between probes. Under 5 ms rock solid; 15–20+ ms stutters games and robotizes voice even at a good average ping.

Packet loss — probes that never came back. Above ~1% is noticeable in games (rubber-banding, teleporting).

Lagspike — latency above your baseline, sustained ≥0.5 s — shorter micro-bursts are absorbed by game netcode and never counted. Severity: minor (small jump), major (+60 ms), severe (+150 ms or sustained loss); bands scale up on high-baseline lines (satellite, fixed wireless). Duration counts only time actually above threshold — a one-probe blip shows as<br>Speed test says 500 Mbps but your game still lags?

Bandwidth and stability are two different things. A speed test measures how much data your line<br>can move in a few seconds — the size of the...