VPS Latency Guide: Ping, Traceroute, and What “Good” Actually Means
If your ping is under 20 ms, you’re playing on effectively the same footing as a wired connection to a local server. Under 5 ms is close to the physical floor , the point where light in fiber optic cable simply can’t travel any faster. Above 100 ms, most competitive games start to feel sluggish, and past 150 ms even casual play gets frustrating. This guide shows you exactly how to measure your own number using ping and traceroute, what the results actually mean, and why the data center you connect to changes everything.
What Ping Actually Measures
Ping and latency get used interchangeably, but they’re not quite the same thing. Latency is the real-world delay: the time it takes a signal to travel from your device to a server and back. Ping is the tool you run to measure it , a small packet sent out and timed until its reply comes back, expressed in milliseconds (ms).
Every millisecond of that round trip comes from somewhere real. Part of it is unavoidable physics: light in a fiber optic cable travels at roughly two-thirds the speed of light in a vacuum, which works out to about 1 ms of round-trip delay for every 100 km of distance between you and the server. Part of it is routing , how many hops your traffic takes before it reaches the destination. And part of it is processing time at each hop along the way.
This is why “how far away is the server” is the single biggest lever you control. You can’t beat physics, but you can choose a data center close enough that physics barely matters.
How to Run a Ping Test
You don’t need special software. Every major operating system ships with a built-in ping tool.
On Windows:
- Open Command Prompt (search “cmd” in the Start menu)
- Type ping example.com (replace with your target IP or hostname) and press Enter
- Windows sends 4 packets by default and reports the round-trip time for each, plus min/max/average
On macOS or Linux:
- Open Terminal
- Type ping example.com and press Enter
- Unlike Windows, this runs continuously , press Ctrl+C to stop it and see the summary statistics
Reading the output: the number that matters is the time= value on each line, measured in milliseconds. Run it for at least 15–20 seconds and look at the average, not a single reading , a single ping can be thrown off by a momentary network hiccup. Also check for packet loss in the summary; occasional loss is normal on the public internet, but consistent loss above 1–2% points to a routing or ISP problem, not a distance problem.
How to Run a Traceroute
Ping tells you the total delay. Traceroute tells you where that delay is coming from, by showing every router hop between you and the destination and the latency at each one.
On Windows: Command Prompt → tracert example.com
On macOS or Linux: Terminal → traceroute example.com (macOS) or traceroute example.com (most Linux distributions; if unavailable, sudo apt install traceroute on Debian/Ubuntu)
Each line in the output is one hop, with three timing samples. What you’re looking for:
- A steady, gradual climb in latency as hops increase , this is normal; it just means the traffic is physically getting further from you
- One hop where the number jumps sharply and stays high , this usually marks a congested or distant network segment, often where your traffic crosses from your ISP onto a backbone provider
- Asterisks (* * *) on a hop , this often just means that router doesn’t reply to traceroute probes, which is common and not itself a problem, as long as hops after it keep responding normally
If your total ping is high, traceroute is how you find out whether it’s your ISP, an inefficient route, or genuinely just distance to the server.
What Counts as Good Latency
There’s no single universal threshold , it depends heavily on what you’re doing on the connection. Based on published gaming-latency guidance from network testing services and esports resources, the general bands look like this:
| Ping (ms) | Rating | What it feels like |
| Under 5 ms | LAN-like | Effectively the physical floor , the range professional esports players get on local-network hardware |
| Under 20 ms | Excellent | Imperceptible delay; the standard competitive FPS and esports players aim for on regional servers |
| 20–50 ms | Good | Smooth and responsive across nearly every genre; where most professional players sit on regional servers |
| 50–100 ms | Playable | Acceptable for most game types; fast-paced shooters may show occasional, minor input delay |
| 100–150 ms | Degraded | Noticeable lag; still playable for turn-based or slower genres, a real handicap in fast shooters |
| Above 150 ms | Poor | Rubber-banding and missed inputs become common in most real-time multiplayer games |
Note: these are widely-cited general bands, not a Rabisu-measured benchmark. Different games and genres have different tolerances , a turn-based strategy game is far more forgiving than a competitive FPS.
For context outside gaming, the ITU-T G.114 standard , the long-standing international recommendation for real-time voice and interactive traffic , puts the ceiling for transparent, unnoticeable interactivity at 150 ms one-way delay, with up to 400 ms still classed as “acceptable” for less time-sensitive applications. That’s a useful reference point for anything beyond gaming: remote desktop sessions, VoIP, or SSH into your VPS all follow roughly the same physics.
Why Region Choice Changes Everything
Here’s the part most guides skip: your ping number isn’t really about your internet connection quality , it’s overwhelmingly about the distance between you and the server, converted into milliseconds by the speed of light.
A rough working rule: every 100 km of fiber distance adds close to 1 ms of round-trip latency, before you even account for routing overhead, network congestion, or the number of hops in between. That means a player in Istanbul connecting to a server in Los Angeles is fighting roughly 9,000+ km of unavoidable cable distance , no amount of “premium” hosting closes that gap. The only real fix is picking a server region closer to the players actually connecting to it.
This is exactly why Rabisu operates across 13+ global data center locations spanning Europe, North America, Asia-Pacific, and Australia , the goal isn’t to be everywhere for its own sake, it’s to put a physical hop closer to wherever your players or your trading infrastructure actually sit. If your Minecraft community is split across regions, the right move is picking the location closest to the majority of your players, not the one closest to you as the server owner.
The same physics applies directly to Forex trading infrastructure , see the [forex VPS guide] for how the same distance-to-latency math plays out when milliseconds separate a filled order from a requote.
The Proof Hook: Test It Yourself
Don’t take a published latency table’s word for it , including this one. Here’s a three-step check you can run against any provider you’re evaluating, including Rabisu:
- Baseline your own connection first. Ping a well-known, geographically close target (your ISP’s own DNS server is often listed on their support page) to get a sense of your local network floor before adding any distance.
- Ping and traceroute the actual data center location you’re considering, not just the provider’s homepage. Rabisu’s [Looking Glass tool] lets you run live ping, traceroute, and MTR tests directly against each of the 12+ location endpoints before you commit to a plan.
- Test at the time of day you’ll actually use it. Latency to a server is mostly stable, but congestion on shared network segments can shift slightly during regional peak hours , test during your typical play or trading window, not just whenever is convenient.
Run these three checks against any hosting provider’s marketing claims before you buy, ours included. A provider that’s confident in its network will hand you a testing tool and let you verify it yourself before you pay for anything.
Why Rabisu
Rabisu operates data centers across 13+ locations on four continents specifically so the region-to-latency math works in your favor, whether you’re running a Minecraft server for a community split across Europe, a Rust server for a North American player base, or a Forex EA that needs to sit close to a broker’s matching engine. Every location runs on the same current-generation hardware baseline , AMD Ryzen processors and NVMe storage , so the network hop is the variable you’re optimizing, not inconsistent hardware between regions. You can verify latency to any location yourself with the Looking Glass tool before choosing a plan.
Quick Answers
Under 20 ms is considered excellent and matches what professional esports players target on regional servers. Under 50 ms is good for nearly all game genres. Above 100 ms, most fast-paced multiplayer games start to show noticeable input delay.
Latency is the actual delay data experiences traveling to a server and back. Ping is the diagnostic tool used to measure that delay , the terms are often used interchangeably in casual conversation, but ping technically refers to the measurement method.
Most of your ping is fixed by physical distance, but a small portion can shift with network congestion on shared segments, especially during regional peak usage hours. A consistent 5–10 ms fluctuation is normal; large swings suggest a routing or ISP issue worth investigating with traceroute.
Yes. Wi-Fi adds variable delay from radio interference and retransmissions that a wired Ethernet connection doesn’t have. This is separate from server distance , it affects the first hop from your device to your router, before your traffic even leaves your home network.
Yes, and you should. Reputable providers offer a “Looking Glass” or similar public testing tool that lets you ping and traceroute their actual data center locations before you commit to a plan, rather than relying on marketing claims alone.
Sources Referenced
- ITU-T Recommendation G.114 , One-way transmission time (international standard for acceptable interactive delay)
- Published gaming-latency threshold guidance (esports resources, network testing services)
- High Performance Browser Networking (O’Reilly) , fiber optic propagation speed reference
- Fiber latency rule-of-thumb calculations (~4.9–5 μs/km one-way, ~1 ms per 100 km round-trip)