# How do I test the latency to my JadeKVM server?

> Measure the round trip to a JadeKVM server with ping, mtr, pathping or Test-NetConnection, and compare it with public city averages.

- URL: https://jadekvm.com/help/test-latency-to-your-server/
- Last updated: 2026-11-16
- Publisher: JadeKVM (https://jadekvm.com)

**Quick answer — How do I measure the latency from my computer to a JadeKVM server?**

Copy the server's IPv4 from the **JadeKVM** dashboard and run `ping -c 20` with it on macOS or Linux, or `ping -n 20` on Windows; the average is your round trip. Then run `mtr`, or `pathping` on Windows, to see each hop and its loss. Compare with public averages, such as 14 ms from Hong Kong to Taipei.

**In brief**

- **Round trip**: ping 20 times and read the average, not a single reply.
- **Path**: mtr or pathping shows each hop, its delay and its loss.
- **Port check**: `nc -vz` or `Test-NetConnection` confirms SSH answers on port 22.
- **Benchmark**: JadeKVM's latency index lists public averages from each city to 19 destinations.

*macOS and Linux*

```bash
# 20 pings; read the avg value of the last line
ping -c 20 203.0.113.10

# hop-by-hop report, 100 cycles (Debian/Ubuntu: apt install mtr-tiny; macOS: brew install mtr)
mtr -rwc 100 203.0.113.10

# does SSH answer on port 22?
nc -vz 203.0.113.10 22
```

*Windows (PowerShell)*

```powershell
# 20 pings; read Average at the end
ping -n 20 203.0.113.10

# hop-by-hop with loss per hop (takes a few minutes)
pathping 203.0.113.10

# does SSH answer on port 22?
Test-NetConnection 203.0.113.10 -Port 22
```

## What round trip should I expect from my area?

Latency figures on JadeKVM pages are public averages measured by WonderNetwork between its own servers on 16 November 2026; they show how far apart the cities sit on the network, not a promise about any one JadeKVM server. Your own figure adds the path from your home or office network to the nearest of those test servers.

*Public average round trip from each JadeKVM city to eight user areas, WonderNetwork, 16 November 2026*

| User area | Hong Kong | Tokyo | Singapore | Nearest JadeKVM city |
| --- | --- | --- | --- | --- |
| Taipei | 14 ms | 31 ms | 43 ms | Hong Kong |
| Seoul | 90 ms | 35 ms | 85 ms | Tokyo |
| Manila | 18 ms | 187 ms | 28 ms | Hong Kong |
| Bangkok | 47 ms | 80 ms | 23 ms | Singapore |
| Jakarta | 53 ms | 79 ms | 15 ms | Singapore |
| Hanoi | 62 ms | 86 ms | 50 ms | Singapore |
| Mumbai | 102 ms | 134 ms | 55 ms | Singapore |
| Sydney | 125 ms | 99 ms | 92 ms | Singapore |

## How do I read the ping and mtr output?

**Read the average and the loss, and look at the last hop first.**

- **avg (or Average)**: the round trip in milliseconds. Under 40 ms, an SSH session feels local; from about 100 ms, each keystroke echoes with a visible delay.
- **Loss on the last hop**: real loss on the path or at the server. Repeat at another hour before drawing a conclusion.
- **Loss on a middle hop only**: usually a router that answers test packets slowly; if the last hop shows 0%, traffic is fine.
- **A jump between two hops**: where the path gets long, often a sea cable or a hand-off between carriers.

JadeKVM read the WonderNetwork pages on 16 November 2026 for the table above; it runs no measurement of its own.

## What if my figure is far above the table?

**Test again from another network, then compare cities.** Mobile and Wi-Fi links add delay. A VPN on your computer sends the test through its own exit first.

If the path stays slow, the city may be the wrong one for your area: the [latency index](https://jadekvm.com/asia-vps-latency-index/ "What latency and price does an Asian VPS have in November 2026?") shows the nearest city for 19 destinations, and the [city guide](https://jadekvm.com/hong-kong-tokyo-or-singapore-vps/ "Should I host my VPS in Hong Kong, Tokyo or Singapore?") explains the choice. Loss on the last hop over several runs is a case for support: follow [server down or not delivered](https://jadekvm.com/help/server-down-or-not-delivered/ "What should I do if my JadeKVM server is down or not delivered?") and send the mtr report.

## Can I test from mainland China?

**You can run the same commands, but expect high figures: JadeKVM sells no China-optimized route.** JadeKVM sells standard international transit in all three cities. No plan is sold on a China-optimized route such as CN2 GIA, CMI or 9929, so traffic to mainland China takes the ordinary path between carriers. WonderNetwork listed 145 ms from Tokyo and 325 ms from Hong Kong to Shanghai on 16 November 2026. A Hong Kong server is outside mainland China, so no ICP filing applies to it; what visitors in mainland China can reach still depends on their own network and its filtering, which JadeKVM does not control.

## Latency test questions

### Why does ping time out while SSH works?

A firewall on the server may drop ICMP. Test port 22 with `nc -vz` or `Test-NetConnection` instead.

### Should I test from my computer or from the server?

From where your users are. A test from the server to a far city mirrors the same path in reverse.

### Is mtr installed on JadeKVM images?

Not necessarily. On Debian and Ubuntu, install it with `apt install mtr-tiny`; on AlmaLinux and Rocky Linux, with `dnf install mtr`.

### Does JadeKVM promise a latency?

No. Latency figures on JadeKVM pages are public averages measured by WonderNetwork between its own servers on 16 November 2026; they show how far apart the cities sit on the network, not a promise about any one JadeKVM server.

### Which city is nearest to Korea?

Tokyo, at 35 ms from Seoul in WonderNetwork's figures. JadeKVM sells no Seoul server.

**Verdict**

- **Ping 20 times and run mtr** from your users' network to the JadeKVM server's IPv4; read the average and the last-hop loss.
- **Compare with the public averages**: a figure far above them points to your own link, or to a better JadeKVM city for your area.

## Sources

1. [mtr manual page](https://manpages.debian.org/bookworm/mtr-tiny/mtr.8.en.html) — Debian (read 2026-11-16)
2. [pathping command reference](https://learn.microsoft.com/en-us/windows-server/administration/windows-commands/pathping) — Microsoft (read 2026-11-16)
3. [Test-NetConnection](https://learn.microsoft.com/en-us/powershell/module/nettcpip/test-netconnection) — Microsoft (read 2026-11-16)
4. [Global ping statistics: Hong Kong](https://wondernetwork.com/pings/Hong%20Kong) — WonderNetwork (read 2026-11-16)
5. [Global ping statistics: Tokyo](https://wondernetwork.com/pings/Tokyo) — WonderNetwork (read 2026-11-16)
6. [Global ping statistics: Singapore](https://wondernetwork.com/pings/Singapore) — WonderNetwork (read 2026-11-16)

**Test a city for 3 months** Bead: 1 shared vCPU, 1 GB RAM, 20 GB NVMe, 1 TB a month, $22.50 for 3 months, the shortest term. [Order Bead in Hong Kong](https://jadekvm.com/app/checkout/?plan=bead&term=3m&location=hk-hkg&from=help-test-latency-to-your-server&lang=en) · [All help guides](https://jadekvm.com/help/ "Where are the JadeKVM help guides?")

---

JadeKVM: JadeKVM sells Linux KVM VPS in Hong Kong, Tokyo and Singapore, paid in crypto with no ID document, and publishes the measured latency from each city to the rest of Asia. 5 plans from 1 to 16 GB of RAM cost the same in all three cities and, on the 12-month term, sit at least 25% under a named comparable no-KYC crypto plan read on 16 November 2026. The operator is DomaineGoat LLC, Panama.
