What Your Ping Number Actually Means — Latency, Jitter, and Packet Loss Explained

The number from a ping test has different meanings depending on what you're trying to do

50ms latency feels fine for loading a web page. It makes online gaming difficult. It makes VoIP calls noticeably degraded. It renders high-frequency trading impossible. The same number means different things in different contexts — because what "responsiveness" requires varies enormously by application type.

Understanding what latency actually measures, how it compounds across a network path, and what numbers matter for different use cases turns a ping result from "good" or "bad" into actionable information.

What ping measures — and what it doesn't

A ping sends an ICMP echo request to a host and waits for an ICMP echo reply. The reported time (RTT — round-trip time) is the elapsed time for the packet to reach the destination and return.

What RTT captures:

• Transmission delay across each physical link
• Propagation delay (the speed-of-light travel time)
• Processing delay at each router/switch
• Queuing delay when network segments are congested

What RTT doesn't capture:

• Throughput — how much data can be moved per second (bandwidth)
• Packet loss — whether packets are being dropped
• Jitter — variability in latency between packets
• Performance under load — a path may show 20ms ping but have high latency during actual data transfer due to bufferbloat

A clean ping result shows one dimension of network performance. Latency, throughput, packet loss, and jitter are four different measurements — all four matter, and a good result on one doesn't guarantee anything about the others.

Latency by physical location: the speed-of-light floor

Light travels approximately 200,000 km/second through fibre (slower than vacuum due to the refractive index of glass). This sets an absolute minimum for any latency measurement:

• London to Frankfurt: ~1,000 km → minimum ~5ms one-way, ~10ms RTT
• London to New York: ~5,570 km → minimum ~28ms one-way, ~55ms RTT
• London to Singapore: ~10,840 km → minimum ~54ms one-way, ~108ms RTT
• New York to Los Angeles: ~4,500 km → minimum ~22ms one-way, ~45ms RTT

Real-world latency exceeds these minimums because cables don't run in straight lines, routing adds hops, and infrastructure processing adds delay. Transatlantic latencies of 70–90ms RTT are typical; trans-Pacific 150–200ms.

These numbers matter when evaluating cloud provider region choices, CDN configurations, and remote server performance expectations.

Latency thresholds for different applications

Under 1ms: local network, same datacenter, shared memory.

1–10ms: local area network (wired); same-region datacenter connections. Essentially imperceptible for all applications.

10–30ms: same country or nearby region. Transparent for web browsing and streaming. Imperceptible for most users. Adequate for competitive gaming (though esports players prefer under 10ms).

30–60ms: inter-regional connections (e.g. US East to US West). Streaming and web browsing: fine. Video calls: fine. Competitive gaming: noticeable, playable. Real-time collaboration: generally acceptable.

60–100ms: international connections within the same continent or to nearby continents. Web browsing: acceptable. Video calls (Zoom, Teams): slightly degraded audio quality but functional. Gaming: noticeable lag, disadvantageous in competitive play.

100–200ms: intercontinental connections. Web browsing: perceivably slower for interactive sites. Video calls: noticeable delay in conversational cadence. Gaming: significant lag, frustrating for fast-paced games. Tolerable for turn-based games.

200ms+: very long distance or heavily loaded paths. Web pages feel sluggish. Video calls develop audio degradation and awkward conversation overlap. Real-time applications become problematic.

Jitter: the hidden performance killer

Jitter is the variance in latency between consecutive packets. A connection with 40ms average latency but ±30ms jitter is often worse for real-time applications than one with 80ms average latency and ±5ms jitter.

Why jitter matters:

VoIP and video calls require a steady stream of audio packets arriving at regular intervals. High jitter means packets arrive with inconsistent timing — the receiving system's jitter buffer must be large enough to absorb the variation, which effectively increases perceived latency. Excessive jitter causes choppy, robotic-sounding audio and frozen video frames.

Gaming is similarly affected — jitter makes network behaviour unpredictable, causing desynchronisation between what you're doing and what the server registers.

Typical acceptable thresholds:

• VoIP/Video calls: under 30ms jitter
• Gaming: under 20ms jitter
• Streaming (non-interactive): tolerates higher jitter; buffering compensates

Packet loss: worse than high latency

TCP retransmits lost packets. A connection with 1% packet loss continuously causes TCP to retransmit, which multiplies effective latency and reduces throughput far more than the same amount of pure latency increase.

Packet loss thresholds:

• 0%: expected on a healthy wired connection
• 0–0.1%: typically acceptable; occasional loss rarely noticed
• 0.1–1%: noticeable quality degradation in video calls; TCP retransmits cause throughput reduction
• 1–5%: significant degradation; VoIP becomes choppy; file downloads slow substantially
• 5%+: connection is broken for practical purposes

Ping with packet loss shows as timeouts or missing reply lines. A host that drops all ICMP (some firewalls block ping) appears as 100% packet loss even when fully reachable via other protocols — always verify with a TCP-based check if ping results look suspicious.

Bufferbloat: why your connection feels slow under load

Bufferbloat is a specific latency problem caused by excessive buffering in network equipment. When a link is congested, routers buffer excess packets rather than dropping them — causing latency to increase dramatically under load.

The symptom: a connection shows 15ms ping when idle but 400ms+ latency when a download is running. The download saturates the buffer; all subsequent packets queue behind it.

The test: run a ping while downloading something. If latency shoots up by 100ms+, bufferbloat is present.

The fix: Quality of Service (QoS) configuration on your router, or upgrading to a router with active queue management (fq-codel, cake algorithms).

How to use the Ping tool on sadiqbd.com

1. Enter the hostname or IP — a domain, server name, or IP address
2. Run the ping — the tool sends ICMP requests and measures RTT
3. Read the results:
   • Average RTT — typical latency to this host
   • Packet loss percentage — any packets that didn't return
   • Min/max/avg — the range shows jitter

Compare against the distance-based minimums: if your RTT to a nearby server significantly exceeds the physical minimum, the additional latency is from routing inefficiency, network load, or infrastructure issues.

Frequently Asked Questions

Why does ping show timeout but the website loads fine? Many servers (and some network devices) block ICMP to prevent flood attacks or simply to reduce attack surface. The server is reachable and serving HTTP/HTTPS requests, but the firewall drops ICMP echo requests. This is very common — a ping timeout isn't evidence of unreachability, only of ICMP being blocked.

What's a good ping for gaming? Under 30ms to the game server is excellent. 30–60ms is good. 60–100ms is playable but noticeable in fast-paced games. 100ms+ is problematic for competitive play. More important than average latency: low jitter and zero packet loss.

Does latency affect video streaming quality? For buffered streaming (YouTube, Netflix), no — content is buffered ahead of time. High latency delays the initial buffering but doesn't affect playback once it starts. For live streaming, latency affects how far behind real-time the viewer is, but doesn't typically cause quality degradation.

Is the Ping tool free? Yes — completely free, no sign-up required.

A ping number is a single data point — useful, but most valuable when you know what you're comparing it to and what application context makes it significant.

Try the Ping tool free at sadiqbd.com — check reachability and measure latency to any host instantly.