Measuring…
The verdict for this game appears once the first full second of packets is in.
Comparing with other tests to this server…
Measuring…
The verdict for this game appears once the first full second of packets is in.
Comparing with other tests to this server…
Are you sure you want to delete all results?
| Date & Time | Duration | Avg Ping | Avg Jitter | Packet Loss | Actions |
|---|
Ultimate goes on cooldown without firing? Hits that land visually but deal no damage?
Marvel Rivals launched on 60-tick servers - the creative director confirmed the figure before release, and it is one of the few numbers NetEase has stated publicly. Sixty ticks means the server reconciles the fight every 16.7 ms, which is fast enough that your own connection, not the server rate, is usually the limiting factor. That matters for a hero shooter specifically: abilities are discrete events, and a discrete event either arrives or it does not.
What this test sends: 60 packets per second of 200 bytes, matching Marvel Rivals rather than a generic ping. verified
NetEase confirmed 60-tick servers at launch.
Run the test above for at least 60 seconds. Here is what each number means for this game specifically.
Loss is the reason an ultimate goes on cooldown without happening. The input packet never arrived, the server never cast it, and your client had already played the animation. This is the metric to watch for wasted ultimates.
Dive heroes live and die on a few hundred milliseconds of timing. Jitter above the 16.7 ms tick interval makes enemy positions arrive out of step, so your dive lands where the target was, not where it is.
Under 40 ms is comfortable. Above 80 ms, reacting to a telegraphed ultimate stops working: the window you are reacting to closed on the server before your screen showed it.
What you see in game, and which number on this page explains it.
| What you see in Marvel Rivals | Look at | What it means |
|---|---|---|
| Ultimate on cooldown, never cast | loss | The input packet was lost. The animation you saw was your own client predicting. |
| Dives land where the target was | jitter | Enemy positions arriving out of step with the 16.7 ms tick. |
| Cannot react to telegraphed ultimates | ping | The reaction window closed on the server before your screen showed it. |
| Bad only at season launch | capacity | Server load. A clean chart during a bad match settles it. |
Marvel Rivals launched on 60-tick servers, and the creative director confirmed the figure before release - one of the few numbers NetEase has stated publicly. It is worth correcting a widespread error here, including one this site made: the game does not run at 64 ticks. That figure gets copied over from Counter-Strike and then repeated. We measured Marvel Rivals at 64Hz until this page was written, and the traffic pattern this test sends has been corrected to 60.
Sixty ticks means the server reconciles a fight every 16.7 milliseconds, which is quick enough that your own connection, rather than the server rate, is usually the limiting factor. That shifts where the useful diagnosis is: not "are the servers good enough", but "does my line keep up with them".
Hero shooters have a specific vulnerability that tracking shooters do not. Abilities are discrete, expensive events. A lost packet in an aim-based game costs a few frames of tracking accuracy; a lost packet carrying an ultimate cast costs the entire ultimate. Your client plays the animation locally because it predicts optimistically, the server never receives the cast, and the cooldown you are looking at is that prediction being corrected. There is nothing to learn from and nothing to play around - which is why loss is the number to watch on this page.
Jitter is the second-order version of the same problem. If your packets vary by more than the tick interval, dive timing becomes unreliable in a way that feels like inconsistent aim, because the target's position is arriving out of step with the ticks it belongs to.
If the chart above shows packet loss or high jitter, work through these in order.
The Symptom: Ping far above what this test shows.
Marvel Rivals lets you pick a region, and the default is not always the closest one. This chart shows what your line can reach - a large gap means a region setting, not a broken connection.
The Symptom: Loss in short bursts.
At 60 ticks a burst of wireless retries blanks out several server updates in a row - long enough for a whole engagement. A cable removes the cause outright.
The Symptom: Spikes while streaming or uploading clips.
Upstream saturation delays the small input packets the game sends constantly. QoS or an upload cap keeps them moving.
The Symptom: Only bad right after a season starts.
Server load at a season launch is not something your connection can fix. A clean chart during a bad match settles the question.
60 ticks per second on the live servers, confirmed by the creative director at launch. Note that this is not 64 - a figure often copied over from Counter-Strike.
The input packet was lost. Your client played the animation locally, the server never received the cast, and the cooldown you see is your client's optimistic prediction being corrected.
It is one reconciliation every 16.7 ms, which is enough that your own jitter usually dominates. If your jitter is above that interval, the server's extra precision cannot reach you.
It shifts what you see relative to the server's truth, so close-range hits get less reliable. Packet loss causes outright misses, which is the more visible failure.