> ## Documentation Index
> Fetch the complete documentation index at: https://shareclick.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Latency

> Where the microseconds actually go — ShareClick's ~6 µs one-way transport overhead and the design decisions that keep input lag minimal.

Lowest input lag is ShareClick's headline feature. Every design choice defers to
it, and we keep a benchmark in the repo so regressions get caught immediately.

## Measured overhead

Measured on loopback with `shareclick bench --encrypted`:

<CardGroup cols={2}>
  <Card title="~6.5 µs one-way" icon="bolt">
    Transport overhead: serialize + seal + socket + open + deserialize.
  </Card>

  <Card title="~20 ns" icon="lock">
    Extra cost of encryption (ChaCha20-Poly1305 on tiny packets).
  </Card>
</CardGroup>

```bash theme={null}
$ shareclick bench --count 20000
METRIC rtt_median_us=~12     # loopback round-trip
METRIC oneway_us=~6          # one-way transport overhead
```

<Note>
  **\~6 µs one-way** means our code is not the bottleneck. On a real LAN,
  end-to-end input lag is dominated by the network RTT (\~0.2–1 ms) and OS event
  injection — not ShareClick.
</Note>

## Why it's this fast

<AccordionGroup>
  <Accordion title="UDP for input, TCP for bulk" icon="network-wired">
    A dropped mouse move is irrelevant a millisecond later, so UDP avoids
    head-of-line blocking. The fast path never waits on retransmits of the slow
    path. See [Decision #1](/develop/decisions).
  </Accordion>

  <Accordion title="Dedicated blocking socket, not async" icon="microchip">
    An async runtime adds scheduler jitter on the hot path. A dedicated thread
    with a blocking socket has the most predictable latency — `transport.rs` is
    intentionally synchronous. See [Decision #3](/develop/decisions).
  </Accordion>

  <Accordion title="Per-tick event coalescing" icon="layer-group">
    High mouse polling rates (1000+ Hz) would flood the network and cause
    "jumpiness" when polling exceeds display refresh. Coalescing a tick's events
    into one packet fixes both. See [Decision #6](/develop/decisions).
  </Accordion>

  <Accordion title="Allocation-light, lock-free hot path" icon="feather">
    `transport.rs` uses immutable cipher state and no per-packet locks or
    allocations, precisely to avoid scheduler jitter.
  </Accordion>
</AccordionGroup>

## Run the benchmark yourself

No OS permissions needed — the benchmark runs entirely on loopback:

```bash theme={null}
cargo run --release -- bench --count 20000            # plaintext
cargo run --release -- bench --count 20000 --encrypted
```

<Tip>
  Latency is a **tracked metric**. Before/after any change on the input path, run
  `bench --encrypted` and confirm no regression. The history is recorded in
  `autoresearch.jsonl`.
</Tip>

## Next

<Card title="Architecture" icon="diagram-project" href="/concepts/architecture" horizontal>
  The full dual-channel design behind these numbers.
</Card>