Game server ping by region from Germany

A game server in Germany gives players in Central and Western Europe a round trip of roughly 5-30 ms, most of the rest of Europe 20-60 ms, Turkey, the Caucasus, Israel and North Africa about 55-95 ms, the US East Coast 85-110 ms, and anywhere in Asia, South America or Oceania well over 150 ms. Those bands decide who can play what. A survival or building game is comfortable up to about 120 ms; a competitive shooter starts to feel unfair to the far player somewhere past 60-70. The rest of this post is the detail behind those numbers: the table by region, what each band feels like by genre, why the number in the scoreboard rarely matches a ping test, and what you can and cannot do for the player who is far away.

For the physics behind latency and how to pick a location for a group, read where your game server should live - it has a city-by-city table to Frankfurt. This post looks at it the other way round: the server is in Germany, and the question is what each player will get.

Ping to Germany by region#

Germany is a good place to measure from because Frankfurt hosts DE-CIX, one of the largest internet exchanges in the world, and most European networks either peer there or pass through it. The figures below are typical round trips from residential connections on a wired link, in the evening, to a well-connected datacentre in Germany. Your players' numbers will differ by ISP, by city and by time of day; treat the table as an expectation to test against, not a promise.

Two things are worth noticing. Distance sets a floor - roughly 1 ms of round trip for every 100 km of fibre - but routing decides how far above it a player lands. That is why Tbilisi, about 3,000 km away, can measure worse than New York at more than twice the distance. And the jump from "Europe" to "everyone else" is not gradual. There is a step at the edge of the continent where the path stops being direct, and most of the world sits on the far side of it.

What those numbers feel like, by genre#

Ping is not good or bad in itself. It is good or bad for a particular kind of game, because games hide latency in different ways.

Shooters are the strictest because the server has to decide who hit whom. Source-engine games use lag compensation: the server rewinds other players' positions by the shooter's latency to check a hit. It works well, but its side effect grows with ping - the low-ping player gets shot after reaching cover, because the high-ping shooter saw them in the open. At 30 ms nobody notices; at 120 ms everyone does, and it is the low-ping players who complain.

Survival and building games are forgiving because most actions are not contested in real time. Placing a block or chopping a tree at 150 ms is fine. Combat feels worse than it should, and Valheim adds its own twist, because each area is simulated by the first player to enter it: a far-away player can make monsters stutter for everyone standing near them. Valheim performance and lag goes into that.

Lockstep games such as Factorio run every client's simulation in step. The game hides latency for a player's own movement, but actions on the world - placing, mining, opening a chest - take effect after a round trip. A 200 ms player in a Factorio megabase is playable but noticeably sluggish, and a player on a bad connection can drag everyone's catch-up.

Racing is unforgiving in a different way. Cars are predicted between updates, so at high ping other cars warp and contact becomes unfair. Leagues often set a ping limit for exactly this reason.

Why the in-game number differs from your ping test#

Players report a number from the scoreboard and you measure a different one with ping. Both can be right.

• ICMP ping measures the network round trip to the machine. It is handled by the operating system, so it ignores how busy the game is.
• The in-game number is measured by the game over its own protocol, and usually includes time waiting for the next server tick, time in the client's queue and, in some games, a deliberate averaging window. On a 64-tick server, waiting for the next tick alone can add up to about 15 ms.
• Some games show something else entirely. Minecraft's signal bars are buckets, not milliseconds, and are measured with keep-alive packets that are only sent every few seconds. Valheim does not show ping at all unless you open the console and type ping.

A difference of 10-30 ms between the two is normal. A big difference - a ping test of 40 ms and a scoreboard showing 150 - points at something other than distance: a server that is overloaded and late answering, a client dropping frames, or packet loss causing retransmissions in a game that has its own reliability layer. What tick rate actually means covers the server side, and latency, jitter and packet loss covers telling the three apart.

Notes by region#

Central and Western Europe. The easy case. Paths are short and almost every ISP peers in Frankfurt or Amsterdam. If a player here reports 60 ms, it is nearly always their own connection - wifi, a congested uplink, or a VPN they forgot was on - rather than the route.

Nordics, Baltics and Iberia. Distance starts to show, and so does the ISP. Two players in the same city on different ISPs can differ by 15 ms because one hands traffic to a transit provider that takes a longer route. Before blaming the server location, compare two players from the same area.

Eastern Europe and the Balkans. The widest spread in Europe. Good ISPs route directly through Vienna or Frankfurt and land in the 30s; others go via a regional hub and land at 60. Ukraine and Moldova in particular vary a lot by provider.

Turkey, the Caucasus and the Middle East. These routes often go the long way. Caucasus traffic frequently passes through Istanbul or another hub before reaching Western Europe, so physical distance underestimates ping. For a group in Georgia or Armenia, Germany is usually still one of the better options in Europe, at 60-90 ms - good enough for survival, sandbox and roleplay, marginal for competitive shooters.

North America. The East Coast sits at 85-110 ms, which surprises people: closer to Germany in practice than much of the Middle East. A mixed group of Europeans and East Coast Americans can share a co-op server in Germany comfortably. West Coast players at 150-170 ms can share a survival game but will have a poor time in a shooter.

Everyone else. From Asia, South America and Oceania, a server in Germany is a long way. It works for slow games and for players who are used to it, and that is about all.

Measuring it fairly#

Ask each regular player to run the same test, on the same day, and report numbers rather than impressions.

# Windows - 50 pings, then look at the minimum, average and maximumping -n 50 203.0.113.10# Linux or macOSping -c 50 203.0.113.10# The route, with loss per hop, over 100 samplesmtr -rwc 100 203.0.113.10

On Windows, pathping or WinMTR does the job of mtr. Four rules keep the results honest:

1. Wired, not wifi. Wifi adds anything from 2 ms to 50 ms and makes jitter look like distance.
2. Evening, not morning. Test at the hour you actually play. Residential networks are slowest when everybody is home.
3. Look at the minimum and the spread. The minimum is the route; the gap between minimum and maximum is the player's connection and its queues.
4. Use the server's address. Some hosts block ICMP to game nodes; if ping gets no reply, test with the game's own query or by joining. Query ports and A2S has a one-line check.

If one player's numbers are far worse than everyone else's in the same region, reading traceroute and MTR explains how to tell a real routing fault from the harmless loss that routers show on intermediate hops.

VPNs and "game accelerators"#

Players with a bad route sometimes ask whether a VPN or a paid game accelerator will lower their ping. The honest answer is: occasionally, and usually not.

A VPN cannot make light travel faster. It replaces the player's ISP route with a route to the VPN's server plus the VPN provider's route onward, and adds a little encryption overhead. If the player's own ISP sends traffic to Germany on a scenic detour - the Caucasus and parts of Eastern Europe are the usual cases - and the VPN provider has a direct path, the total can come out lower. If the ISP's route was already reasonable, which is true for most of Europe, the VPN adds 5-30 ms and some jitter.

The test is cheap. Have the player run the same 50-ping test with and without the VPN, at the same hour, and compare the minimum and the spread. Keep it if it wins by a clear margin; drop it otherwise. Two side effects are worth warning about: some games and anti-cheat systems treat VPN addresses with suspicion, and a player whose address changes every session will trip any ban or allow list keyed to IP addresses.

Helping the far-away player#

You cannot reduce the distance, but a few things stop a high-ping player having a worse time than they need to.

• Keep the server healthy. A server that misses ticks adds its own delay on top of the network. Every player feels it, but it pushes the far player past their limit first. Check CPU headroom before anything else.
• Leave lag compensation on. In Source games, sv_maxunlag sets how far back the server will rewind for a hit, in seconds. The default is generous; lowering it to punish high-ping players mostly punishes everybody's hit registration.
• Do not lower the tick rate to help them. Fewer updates per second means more waiting between them, which adds latency for everyone.
• Set a ping limit where fairness matters. For competitive matches and racing leagues, a kick above a threshold is kinder than a match decided by geography. Most admin frameworks have one.
• Fix the player's own last mile. A cable, a router with smart queue management, and closing whatever is uploading in the background routinely take 20-50 ms off a bad connection.

Who a server in Germany suits#

Put plainly: a German server is excellent for players anywhere in Central and Western Europe, good for the rest of Europe, workable for Turkey, the Caucasus, the Eastern Mediterranean and the US East Coast, and a compromise for everyone else. RE:NODE runs one location, in Germany, on hardware we run, so this table describes every server we host - we do not offer a choice of regions and will not pretend otherwise. If your group lives in Europe or around its edges, that is a location close to the middle of it. If your group is mostly in Asia or the Americas, a host near them will serve them better, and how to choose game server hosting covers what else to compare.

FAQ#

What ping will I get from the UK to a server in Germany?

Typically 12-25 ms on a wired connection, depending on the ISP and how far from London you are. That is comfortable for every genre, including competitive shooters.

Is 100 ms ping playable?

For survival, building, roleplay and co-op games, yes - most players will not notice much beyond slightly delayed combat. For competitive shooters and sim racing, 100 ms is a clear disadvantage and frustrating for the players on the other end of each fight.

Can Americans play on a European server?

East Coast players usually see 85-110 ms to Germany, which is fine for co-op and survival games. West Coast players see 150 ms or more, which works for slow games and is poor for anything competitive.

Why is my ping to Germany higher than my friend's in the same city?

Different ISPs route traffic differently. One may peer directly at a European exchange while the other sends traffic through a transit provider on a longer path. Run an mtr from both connections and compare where the extra milliseconds appear.

Does a better plan lower ping?

No. Ping is set by distance and routing, not by memory or CPU. A plan that is too small can add delay by missing ticks under load, but once the server has headroom, a bigger plan changes nothing about latency.