High ping is a large round-trip time (RTT) between your device and a server, while packet loss is when some sent packets never arrive. The usual causes are a congested link, Wi-Fi, ISP routing, or an overloaded server. Diagnosis comes down to finding the hop where latency rises or packets start dropping.
This guide covers what ping values are normal, the main causes of loss, how to measure and isolate the problem with ping and traceroute/mtr, and what you can fix on the user, ISP, and server side.
What counts as high ping
| Ping (RTT) | Rating | Where it shows |
|---|---|---|
| < 30 ms | Excellent | Gaming, video calls with no lag |
| 30–80 ms | Normal | Comfortable for most tasks |
| 80–150 ms | Elevated | Noticeable in online games and VoIP |
| > 150 ms | High | Lag, stutter, sluggish UI |
Even 1–2% packet loss is felt in games and calls; for TCP traffic (websites, downloads) loss triggers retransmissions and drops throughput.
Causes of high ping and packet loss
- Wi-Fi and local network: weak signal, interference, congested 2.4 GHz band, old router.
- Link congestion (bufferbloat): when the uplink is saturated (torrents, uploads) ping spikes due to bloated buffers — see bufferbloat.net.
- ISP and routing: a congested hop, a suboptimal path, or trouble at a network peering point.
- Destination server: CPU overload, rate limits, geographic distance (~1 ms physical delay per 100 km).
- MTU and fragmentation: a wrong MTU causes loss and stalled connections.
How to measure and isolate the problem
Basic ping and loss measurement (20 packets):
# Windows
ping -n 20 example.com
# Linux / macOS
ping -c 20 example.comTo see where latency rises or packets drop, use a continuous trace:
# Linux / macOS
mtr example.com
# Windows
tracert example.comWatch which hop starts showing Loss% or rising latency: early hops are your network/ISP, middle hops are transit networks, final hops are the server's network. ICMP is defined in RFC 792. Quickly test ping and port reachability to your server with enterno.io's free ping and port checker, or trace the full path with traceroute.
How to fix it
- User side: use Ethernet instead of Wi-Fi, reboot the router, free up the link, enable SQM/QoS against bufferbloat, update firmware.
- ISP (support ticket): if loss begins on their hops, attach the
mtr/tracertoutput and the time of the test. - Server owner: check load and limits, move infrastructure closer to the audience (CDN, nearer region), set a correct MTU.
Frequently Asked Questions
Ping is high only to one site — is that normal?
The problem is probably not on your end. If ping to other hosts is fine but high to one specific site, the cause is that server, its geography, or the route to it. Compare traces to a good and a bad host to find the diverging hop.
Why does ping fluctuate (jitter)?
Unstable ping (jitter) usually comes from link congestion, Wi-Fi interference, or buffering at an overloaded hop. Check whether ping rises during an active download — a classic bufferbloat sign, fixed by enabling SQM/QoS on the router.
How do I tell an ISP problem from a server problem?
Run mtr or tracert. If Loss% and rising latency appear at early hops, it is your network or ISP. If they appear only at the final hops near the target, the issue is on the server side or its network.
Does packet loss affect download speed?
Yes. TCP treats loss as congestion and shrinks its send window, causing retransmissions. Even 1–2% loss noticeably lowers throughput and increases how long pages and files take to load.
Is it normal for the first hop to lose packets in a trace?
Not necessarily meaningful: routers often deprioritize ICMP to themselves, showing false loss at intermediate hops. Only loss that persists all the way to the final destination across subsequent hops is significant.