Buffer Size vs. Audio Latency Explained

Understand the difference between buffer size and latency in professional broadcast audio, including ms calculations for 128 samples at 48kHz.

Essential for diagnosing lip-sync drift or configuring live monitors. For instant results, use our audio latency calculator. Explore more broadcast engineering tools for IP video and media networking.

What is Audio Latency?

Audio latency is the time delay between an audio signal entering a system and emerging from it. In broadcasting, live events, or studio recording, this delay accumulates as the audio travels through ADC converters, computer processing buffers, digital network protocols, and finally the DAC interface.

Whether you're doing live audio over Dante and AES67 networks, or trying to synchronize lip flaps to a production cameras output via our A/V Sync Tool, understanding how digital audio latency works is paramount for a professional engineer.

How to Calculate Audio Latency

Inside a digital audio system (like OBS, vMix, Pro Tools, or Reaper), the latency is primarily defined by the buffer size and the sample rate.

The Formula:
Latency (ms) = (Buffer Size / Sample Rate) * 1000
Example: (128 / 48000) * 1000 = 2.67 milliseconds.

To eliminate manual calculations in the heat of a production, we built a professional audio buffer latency calculation tool. It instantly provides the exact delay for any combination of samples and frequency, ensuring your 128 samples at 48kHz latency ms are 100% accurate.

Examples: 128, 256, and 512 Samples

Here are the precise latencies for standard buffer sizes operating at the broadcast standard of 48kHz:

128 samples = 2.67 ms: Excellent for in-ear monitoring (IEM) and real-time processing. Requires a fast CPU.
256 samples = 5.33 ms: The sweet spot for live production processing. A safe balance between CPU overhead and acceptable vocal delay.
512 samples = 10.67 ms: Noticeable to a performer wearing headphones, but excellent for stream encoding buffers or mastering chains.
1024 samples = 21.33 ms: Often used in purely post-production environments where rendering stability is more important than latency.

Driver Architecture: ASIO vs. WDM

On Windows-based broadcast systems (vMix, OBS, Wirecast), the driver type you choose is more important than the buffer size itself:

WDM / WASAPI: The standard Windows driver. It routes audio through the Windows Kernel Mixer, which adds significant, unpredictable latency (often 30–100ms) regardless of your software settings. Avoid for live production.
ASIO (Audio Stream Input/Output): A protocol that bypasses the Windows OS mixer entirely, allowing the software to talk directly to the hardware. This is mandatory for achieving sub-10ms latencies in professional broadcast.

Pro Tip: "Zero-Latency" Monitoring
Many professional interfaces (Focusrite, RME, Universal Audio) feature "Hardware Direct Monitoring." This routes the input signal to the output headphones before it hits the computer's buffer. If a performer complains about delay, use hardware monitoring rather than software-based monitoring to achieve true near-zero latency.

Total Round-Trip Latency (RTL)

Keep in mind that buffer latency is only one layer of the cake. Total system latency (or glass-to-glass latency) involves:

A/D Conversion: ~1ms (microphone to digital)
Input Buffer: (e.g., 128 samples @ 48kHz = 2.67ms)
Processing: VST plugins, EQ, Compression (variable)
Output Buffer: (e.g., 128 samples @ 48kHz = 2.67ms)
D/A Conversion: ~1ms (digital to speaker)

A software mixer running at 128 samples will have a minimum RTL of roughly 8–10 ms. Any "Look-ahead" limiters or heavy noise reduction plugins will add significantly more delay to this chain.