Avisoft Bioacoustics conducted a microphone comparison test focusing on self-noise levels in quiet recordings. Using a soundproofed chamber, they measured the residual noise floor of various popular field recording microphones and recorders, including models from Sennheiser, Sound Devices, Zoom, and others. The results, presented as audio samples and spectrograms, reveal significant differences in noise performance between devices, highlighting the importance of microphone selection for capturing quiet sounds in nature recording and acoustic monitoring applications. The test demonstrates that some seemingly similar microphones exhibit drastically different noise characteristics, emphasizing the value of empirical testing.
Avisoft Bioacoustics conducted a comprehensive comparison of microphone input noise levels across a range of audio recording hardware, primarily focusing on devices suitable for bioacoustic research. Their meticulous testing procedure aimed to provide researchers with objective data to inform equipment choices for capturing subtle sounds in nature. The test involved recording silence in a quiet room using various combinations of recorders and microphones, then analyzing the resulting audio files to determine the inherent noise floor of each setup.
The core of the experiment revolved around measuring the equivalent input noise (EIN) – a standardized metric representing the noise generated by the recording system itself, independent of the connected microphone. This was achieved by calculating the signal-to-noise ratio (SNR) using a calibrated sound source and subsequently utilizing this SNR value, alongside the sensitivity of the microphone, to derive the EIN. This approach allowed for a direct comparison of the noise performance of different recorder preamps. They meticulously documented the precise methodology, including the specific calibration equipment and software employed, fostering transparency and reproducibility.
The results were presented in a detailed table showcasing the EIN values for a multitude of recorder and microphone combinations, categorized by recorder brand and model. This extensive dataset enabled researchers to readily assess the relative noise performance of different setups and identify the most suitable option for their specific requirements. The data reveals a considerable range in EIN values across the tested equipment, highlighting the significant impact of recorder choice on the overall noise floor. Furthermore, Avisoft Bioacoustics offered insightful observations and commentary on the performance of certain devices, discussing factors contributing to noise levels and potential implications for bioacoustic research. Their analysis underscores the importance of carefully considering recorder noise performance, particularly when dealing with faint sounds of interest in natural environments. The study provides a valuable resource for researchers striving to optimize their recording setup for maximal sensitivity and minimize the interference of unwanted noise in their data.
Summary of Comments ( 2 )
https://news.ycombinator.com/item?id=43602652
HN users discussed the methodology of the Avisoft microphone comparison, pointing out that the self-noise measurements weren't standardized (using different gain settings and potentially different preamps) which makes comparisons difficult. Several commenters wished for more expensive microphones to be included in the test, like the Sennheiser MKH series and Sound Devices recorders. Some questioned the value of the SNR measurements given the uncontrolled variables. Finally, a few users offered alternative methods for comparing microphone noise, such as using a quiet, controlled environment and normalizing the recordings. Overall, the consensus was that while the data is interesting, it's not scientifically rigorous enough for definitive conclusions about microphone performance.
The Hacker News post titled "Microphone Input Noise Comparison – Avisoft Bioacoustics" has generated several comments discussing the linked article's methodology and findings regarding microphone noise levels.
Several commenters focus on the importance of considering the entire recording chain when evaluating noise. One user points out that the noise floor measurements might be dominated by the preamplifier's noise rather than the microphone itself, especially with low-output microphones. They suggest that using higher-gain, lower-noise preamps could significantly alter the results. Expanding on this, another commenter emphasizes that the cable used between the microphone and preamp can also be a significant source of noise, particularly in longer cable runs. They advocate for testing with short, high-quality cables to minimize this factor.
Another key point raised is the specific application of these microphones. While the tests focus on low-level sounds, one commenter notes that for louder sounds, the self-noise of the microphone becomes less critical compared to its maximum sound pressure level (SPL) handling capability. This highlights the importance of choosing a microphone based on the intended recording environment and expected sound intensity. Another user echoes this sentiment, suggesting that for many nature recording applications, the low-frequency rumbling from wind or handling noise is a bigger concern than the minute self-noise levels measured in the tests.
The methodology of the testing also draws some scrutiny. One commenter questions the use of A-weighting in the measurements, arguing that it is designed for human hearing and may not be appropriate for evaluating the full spectrum of noise relevant to bioacoustic recordings, where ultrasonic frequencies can be important. They suggest that a flat frequency response or a different weighting curve might be more suitable.
Finally, some comments delve into specific technical details, like the importance of impedance matching between the microphone and preamp, the role of phantom power, and the challenges of measuring extremely low noise levels accurately. One user mentions the possibility of thermal noise within the microphone itself being a limiting factor.
Overall, the comments offer valuable perspectives on interpreting the microphone noise comparison, emphasizing the importance of considering the entire recording system, the intended application, and the nuances of the measurement techniques. They provide context beyond the raw data presented in the article and highlight the complexities of achieving truly low-noise recordings.