Audio Buffer Latency (samples to ms)
Enter buffer size in samples and sample rate in Hz for the one-way delay in milliseconds: samples / rate x 1000, so 256 at 48 kHz gives 5.33 ms.
Latência (ms)
—
Audio buffer latency: from samples to milliseconds
To get audio I/O latency from the buffer size, use latency_ms = buffer_samples / fs · 1000. A 512-sample buffer at 44.1 kHz comes out to ~11.6 ms. Run the same buffer at 48 kHz and it drops to ~10.7 ms, and a tighter 128-sample buffer at 48 kHz gets you down to ~2.7 ms. As a rough guide, anything under 10 ms feels great for live performance, under 20 ms is fine for tracking, and once you go over 30 ms the delay starts to show up in your monitoring and can trip up a singer or instrumentalist. Keep in mind that round-trip latency stacks the input and output buffers on top of the driver overhead, so the real number usually runs two to three times this single-buffer value. And the audio driver pulls its weight here too: ASIO on Windows, CoreAudio on macOS, JACK or PipeWire on Linux can matter as much as the buffer setting itself.
Applications: music production and live recording
Producers in Logic, Ableton Live, FL Studio and Pro Tools tend to shrink the buffer while tracking, since the low latency keeps things feeling responsive, then bump it back up for mixing so each cycle has more CPU to spare for plugins. Live engineers chase low latency for in-ear monitors. When someone is editing with virtual instruments and CPU-heavy plugins and isn't playing live, a larger buffer is no problem. With ASIO drivers and a dedicated USB audio interface, hitting 5 ms round-trip is fairly routine.
FAQ
Why does smaller buffer cause crackles? Each block has to be ready before the next one comes due, and a smaller buffer leaves the CPU less time to get there. If a plugin runs long, the audio glitches. That's a buffer underrun.
Is round-trip latency just double the buffer? More or less. You're stacking the input and output buffers and adding a few samples of driver overhead on top, which often works out to 2.5–3× the single-buffer figure.
ASIO vs WDM/WASAPI? ASIO skips the Windows mixer and talks to the hardware directly, which is how it gets under 5 ms. Standard WDM/WASAPI tends to sit around 20–40 ms.
Related Tools
Henderson-Hasselbalch Buffer pH
Compute buffer pH: pH = pKa + log10(base/acid).
Fixed Income CDI Percentage Yield Calculator
Computes the annual gross yield of a fixed income product contracted as a percent of the Brazilian CDI from CDI rate and offered percent.
Mandacaru Cactus Pot Size
Suggests pot size for Mandacaru cactus by height.
Stack Heat Loss (Siegert)
Calculate the heat loss through the exhaust gases by the Siegert formula, loss = K × (T_gas − T_air) ÷ CO₂, from the fuel factor K (~0.5 for natural gas, ~0.6 for oil), the gas and combustion air temperatures (°C) and the CO₂ percentage in the gases. The result, in %, is the largest energy loss of a boiler or furnace — the heat escaping hot through the stack. Lowering the gas temperature (with economizers and preheaters) and adjusting the excess air (which dilutes CO₂) minimizes this loss. The combustion efficiency is approximately 100% minus this loss. Enter the K factor, the temperatures and the CO₂.
Tabata Time Calculator
Compute total Tabata workout time: 8 rounds × (20s on + 10s off) by default. Configurable rounds, on, off, blocks.
Digital Call (Cash-or-Nothing)
Computes the price of a cash-or-nothing digital call option: it pays a fixed amount if the asset finishes above the strike, and nothing otherwise. The price is the discounted payout multiplied by the risk-neutral probability of finishing in the money, Q·e^(−rT)·N(d2). It's the purest form of a binary bet and the building block of many exotic structures. Enter the spot price, the strike, the interest rate, the term, the volatility and the payout.
The results provided by this tool are for general informational and educational purposes only and do not constitute professional, financial, medical, legal, tax or accounting advice. Always confirm important decisions with a qualified professional and official sources.