Your group's Minecraft server runs in Mumbai, or you are about to rent one there, and the reports already disagree. A friend in Pune says it feels smooth, a friend on mobile data says it lags, and nobody knows whether to blame the server, the route or a phone on weak Wi-Fi. A number on a website cannot settle that, because every player reaches Mumbai over their own connection. This guide shows you how to measure yours: in Minecraft, with ping and a route test, on cable, Wi-Fi and mobile data, at the hours you actually play. What runs in Mumbai and what it costs is in our overview of Minecraft hosting in India.
Quick answer: Measure from the device and connection you play on, to your own server's address. In Java Edition, hover over the bars in the server list to see your ping in milliseconds, and press F3 + 3 in game for a live ping graph. Then send 100 pings (ping /n 100 on Windows, ping -c 100 on Linux) and run a route test (pathping or WinMTR on Windows, mtr on Linux) to see where delay or loss starts. Repeat on cable, on Wi-Fi and on mobile data, at the hours your group plays and at a quiet hour. Judge the worst values and the loss, not only the average.
Start With Your Server's Address
Every test below runs against the address your group joins, from the device and connection you play on. If your panel shows 203.0.113.10:25565, Minecraft uses all of it, but ping, pathping and mtr take only the part before the colon. The examples here use that documentation address, so put in your own. If your group joins with a domain name, test the IP address from your panel as well, so you know which machine answered.
Step 1: Read Minecraft's Own Numbers
Java Edition
- The server list. Hover over the bars next to your server and a tooltip shows your ping time to it. The bars alone are too coarse: the Minecraft Wiki lists full bars for anything between 0 and 150 ms, so two friends with full bars can still have very different pings. Read the number.
- The ping graph. Since Java Edition 1.20.2, F3 + 3 shows a ping graph on the right side of the debug screen. It keeps the last 240 ping times and shows the minimum, average and maximum on top, so you see the spikes during play that one number in the server list hides.
- The Tab list. Hold Tab to see a bar icon for every player. The server works out each value from its keep-alive messages and smooths it: in the wiki's formula, a new measurement counts for a quarter of the value shown. A short spike barely moves the bars, and a problem that lasts does. If one friend's bars sit lower than everyone else's, look at that friend's connection first.
Bedrock Edition on phones
Phone players see the server they added in the Servers tab with a colored ping indicator. The wiki's Play screen page gives green for 0 to 150 ms, yellow for 150 to 300 ms and red for 300 to 600 ms. That tells you when something is badly wrong, but little more, so step 3 shows how to get real numbers for mobile data. How phone players join a Java server through Geyser is in our Java or Bedrock guide for Indian friend groups.
Step 2: Run Ping and a Route Test
Minecraft's numbers tell you how long the round trip takes. Ping measures the same round trip with a simpler packet, and a route test measures every network hop on the way, which shows where delay or loss starts.
Windows
Open Command Prompt and send 100 pings instead of the default 4 (Microsoft's ping reference):
ping /n 100 203.0.113.10
At the end, ping prints its statistics: packets sent, received and lost, then the minimum, maximum and average round trip in milliseconds. Next, measure the route:
pathping /n 203.0.113.10
/n skips name lookups, which can make it faster. Pathping first lists every router on the route, then sends 100 queries to each one and computes the loss per router and per link. How long that takes depends on the number of hops, so let it finish. If you prefer a window to a command prompt, WinMTR is a free Windows program that combines traceroute and ping in the same way.
Linux (Ubuntu 24.04 and Debian 12)
sudo apt install mtr-tiny iputils-ping
ping -c 100 203.0.113.10
mtr -rwzc 100 203.0.113.10
ping -c 100 sends one ping a second and ends with the minimum, average, maximum and mdev. The ping manual describes mdev as an average of how far each ping is from the mean, so a high mdev means your ping jumps around. In mtr, -r prints a report, -w keeps host names whole, -z adds the network (AS number) each hop belongs to and -c 100 runs 100 cycles of one second each. The mtr-tiny package gives mtr's helper raw network access when it installs, so mtr runs without sudo.
Step 3: Compare Cable, Wi-Fi and Mobile Data
Run the same 100 pings three ways, a few minutes apart, and keep the results side by side:
| Test | How | What it measures |
|---|---|---|
| Cable | PC connected to the router with an Ethernet cable | Your provider's route to Mumbai, without Wi-Fi |
| Wi-Fi | The same PC on Wi-Fi, where you normally play | That route plus your home Wi-Fi |
| Mobile data | A laptop on your phone's hotspot | Your mobile operator's route |
Minecraft's own multiplayer troubleshooting article suggests an Ethernet cable instead of Wi-Fi on a stationary device, and closing programs that use your bandwidth. Your three results show whether that applies to you. If cable and Wi-Fi differ, test the Wi-Fi link on its own by pinging your router. On Windows, ipconfig lists its address as the default gateway:
ipconfig
ping /n 100 192.168.1.1
Put in the gateway address ipconfig shows instead of 192.168.1.1. Microsoft uses the same test to check a base station, modem, router or network cable. If replies from your own router already come back late or go missing, the delay starts at home, and no server location changes that.
For mobile data, turn on your phone's hotspot, connect a laptop and run step 2 there. The laptop then reaches Mumbai through the phone's mobile connection, which gives you numbers instead of a colored dot.
Step 4: Test at the Hours You Play
A morning test does not tell you how the evening goes. Run the 100 pings and the route test at the time your group usually plays, and once at a quiet hour, from the same device and connection. Write down the date, the time in IST, the connection type, the average, the maximum and the loss. If evenings are clearly worse while mornings are fine, something on the way is busy at that hour, from other devices on your home connection to your provider's links. Pathping's loss per link, which Microsoft's reference reads as congestion, shows where. Ask two or three friends in other cities, or with other providers, to run the same test at the same time, so you compare routes instead of impressions.
How to Read the Results
Look past the average. The average says what is normal, the maximum (Wrst in mtr) shows the worst moments, and the spread (mdev in ping, StDev in mtr) shows how much the times differ from each other. Loss counts the packets that never came back. Akamai's MTR guide and Microsoft's pathping reference read reports the same way:
| What you see | What it means | Where to look |
|---|---|---|
| Loss at one router in the middle, none at the hops after it | That router limits pings addressed to itself; the traffic it forwards is fine | Nowhere, ignore it |
| Loss that starts at one hop and continues to the last | Real loss on that part of the route | The network that hop belongs to |
| Time jumps at one hop and stays high after it | The long or slow stretch of the route starts there | Your quiet-hour report, to see if it only happens at peak times |
| Your router ping is already bad on Wi-Fi | Your home Wi-Fi | Your setup: a cable, or closer to the router |
| Evenings much worse than mornings | Something on the way is busy at that hour | Home use first, then your provider |
When hops show different amounts of loss, trust the later hops, as Akamai puts it. The AS numbers from mtr -z tell you whose network a hop is in: your provider's, one in between, or the one the server sits in. Akamai also advises reports in both directions, because some loss comes from the return route, and from home you can only run your side.
What Ping Does Not Show
A clean ping does not mean a smooth game. If blocks snap back and mobs freeze for everyone at the same moment, the server is falling behind on its ticks. An operator can open the TPS graph with F3 + 2, and our MSPT guide explains the numbers and the fixes. If only your own screen stutters, the FPS graph on the left of that view points at your device. Ping measures the road between you and the server, nothing else.
Testing Before You Order
Our Minecraft India page does not quote a ping figure, and its plan selector does not measure India at the moment. We do not publish a test address in Mumbai either, so your first real measurement is with your own server. Start on monthly billing and run steps 1 to 4 in the first week, at the hours your group plays. If you are still choosing between Mumbai, Singapore and Europe, our location guide for players in India compares public ping data between those cities.
If the Numbers Look Bad
- Only on Wi-Fi: use a cable or move closer to the router, then ping the router again.
- Only one friend: that friend repeats steps 2 and 3 on cable and on mobile data. If mobile data beats their cable, the weak part is their home provider's route.
- Loss or a lasting jump inside your provider's network: send your provider the report, with the time in IST and your connection type.
- Loss or a lasting jump near the server's end: send the report to us. Our support team answers live chat, Discord and email from 13:30 to 01:30 IST. Paste the report as text and add when you ran it, your city, your provider and whether you used cable, Wi-Fi or mobile data.
FAQ
Why does Minecraft show a different ping than the ping command?
Both measure a round trip, with different packets. The server list sends Minecraft's own ping request over a connection to the game port and uses the server's reply to calculate your latency, as the wiki's Server List Ping page describes, while ping uses ICMP echo requests. Compare each tool with itself over time, not with the other.
Ping says "Request timed out", but we can all join. What does that mean?
Windows prints Request timed out when no reply arrives within the time-out, 4 seconds by default. If Minecraft connects while every ping times out, something on the way does not answer ICMP while the game's traffic gets through. Trust Minecraft's numbers for the round trip. On Linux, mtr can send TCP packets to the game port instead. Java servers use TCP port 25565 by default (Minecraft Wiki), so use the port from your panel:
mtr -T -P 25565 -rwzc 100 203.0.113.10
Only one friend lags. Is it the server?
Probably not. If everyone else's bars in the Tab list stay full and one friend's drop, the server reaches everyone else fine. Ask that friend to run steps 2 and 3 and compare their report with yours: the hop where theirs goes bad shows whose network it is.
Does a VPN lower my ping to Mumbai?
Test it instead of assuming. Minecraft's troubleshooting article lists a VPN among the things that can block or slow your connection. Run the same 100 pings with the VPN on and with it off, at the same hour, and keep the setup with the lower maximum and less loss.
Sources
All opened on 30 October 2026:
- Minecraft Wiki: Server list, Player list, Debug screen, Java Edition 23w32a, Play screen, Server, Server List Ping
- Minecraft Help: Troubleshoot Multiplayer Game Issues in Minecraft: Java Edition
- Microsoft: ping, pathping, Check Microsoft Broadband Network Adapter
- Debian: ping(8), mtr(8), mtr-tiny package, mtr-tiny install script
- Ubuntu 24.04 packages: mtr-tiny, iputils-ping
- Akamai: Diagnosing Network Issues with MTR
- WinMTR: project page on SourceForge