Your real ping, even when the server lies about it.
Some servers force the tab ping to a flat 1ms. Quasar times a right-click to the moment the chest actually opens, so it knows your true round trip, then scales prediction and socket behaviour to it.
It cannot lower your real ping. It removes the delay the client itself adds, and it never drops, delays, or alters a packet.
v1.1.01
Illustration. Real ping is unchanged; what moves is the delay the client adds.
An illustration of a connection settling: felt latency drops from 350 to 200milliseconds while the real ping stays at 350 milliseconds.
The connection does not change. What you feel does.
Your real ping is the floor. Everything above it is delay added between a packet arriving and you seeing the result, and that stack is the part a client can actually work on.
Same connection in both. The dashed line is the real ping and it is identical on each, because nothing a client does can move it. What changes is everything stacked on top: the wait for the next tick, the freeze between movement packets, the correction you see after the fact. Illustration, not a measured benchmark.
- STABLEavg < 80ms, jitter < 15ms, no loss
- MODERATEavg < 200ms, jitter < 50ms, loss <= 10%
- UNSTABLEanything worse
- HIGH PING (stable)far away, but jitter < 50ms and loss <= 5%
The fourth one exists because a far-away link with low jitter and no loss is not unstable. It is just far away, and the client refuses to call it anything else.
On join, Auto-Tune freezes you for a few seconds, samples the link, and applies a setting overlay for that session only. It never writes your manual config, and it reverts on disconnect.
See how it is measuredA right-click is a stopwatch.
A server can tell your client anything it likes about your ping, and some of the big ones pin it to 1ms. That makes the tab list useless as a measurement, and it makes every setting derived from it useless too.
So Quasar stops asking. It notes the instant you right-click a chest and the instant the window actually opens on your screen. That round trip had to cross the real network in both directions, so it is a real number, and nothing the server reports can fake it.
That measurement is what feeds the rest. Prediction scales at roughly one tick per 50ms of measured ping instead of a fixed guess, which is why a fixed three ticks never covered a 350ms link and players froze anyway.
Culling, with the distances left in your hands.
Entities, block-entities and particles each get their own cull distance rather than one blanket setting, because a storage room and an open field fail for different reasons.
Stop drawing entities the camera cannot usefully see.
- Cull Distance 16-160
Block-entity and particle culling, the two biggest FPS wins in built-up areas.
- Block-entity culling
- Block-entity cull distance 16-64
- Particle limit
- Particle cull distance 8-48
FPS and ping, drawn in the client's own panel style.
- Show FPS
- Show Ping
- Place
- 98% exact
- Break
- 100% exact
- Hits
- ~64% estimate
- Ping
- 187/12 ms avg / jitter
- Rej
- 3 ghost blocks
- Corr
- 4rb 2tp rubber-bands / teleports
- Gaps
- 3t ticks
Place and Break are exact: the server confirms or contradicts every block action, so the percentage is a fact. Hits carries a tilde because a hit is inferred from knockback inside a ping window, and a zero-knockback hit can be missed.
All of it, listed.
Twenty-three modules ship today. The description under each one is the text the client shows you when you hover it in game, copied across rather than rewritten.
Socket-level tuning plus the real-ping measurement that survives a spoofed tab ping.
Client-side compensation so other players stay trackable when the link is long.
Stop drawing entities the camera cannot usefully see.
Block-entity and particle culling, the two biggest FPS wins in built-up areas.
FPS and ping, drawn in the client's own panel style.
Clicks per second, left and right.
WASD and mouse-button overlay.
Position, facing and in-world day counter.
Draw entity hitboxes, with the colour under your control.
Highlights an armour piece in any container that beats what you are wearing.
Raise the gamma so caves and shadows stop hiding people.
Hold to zoom, with an optional eased transition.
The measured reach of your last hit.
Consecutive hits on the same target.
Rich Presence over a named pipe, with no dependency pulled in.
Sends gg once when a Hypixel SkyWars game ends.
Per-session telemetry, including which of your block actions the server rejected.
Live success rates. Two of them are exact, and the one that is an estimate says so.
An optimizer that tunes the safe settings toward your measured hit registration.
See other Quasar users, and their real ping, wherever you play.
Equippable items, visible to other Quasar users.
A branded overlay for recording and streaming.
Undecorated window at desktop resolution, so capture software can actually see the game.
Socket tuning and coalescing change timing and buffers. Nothing is dropped, delayed past its tick, reordered or altered.
Combat priority flushes your attack immediately instead of waiting for the tick. Timing only. A hit the server rejects stays rejected.
Distance is distance. What Quasar removes is the delay added on your side, and what it changes is how the rest of it feels.
Free, and 1.8.9 only.
One installer. It sets the client up on first run, signs you in, and updates itself after that.
Questions go to the Discord.