Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: [TOS 2 - AI generated] Why do some encoders use really specific lowpass values? (Read 455 times) previous topic - next topic - Topic derived from Why do some encoders ...
0 Members and 1 Guest are viewing this topic.

[TOS 2 - AI generated] Why do some encoders use really specific lowpass values?

Quote from: fred-hy on
Some encoders seem to use really specific lowpass values. For example, the EZ CD converter's version of FDK AAC uses a lowpass value of 15.7khz for 128kbps CBR, and 17.2khz for 160kbps. Why are these specific values used, instead of just 16khz and 17khz? You also see this in LAME too. They use 15.1khz for 96kbps instead of just 15khz.

https://www.poikosoft.com/ezcd-fraunhofer-fdk-aac-encoder

Were these values picked algorithmically? Is the lowpass filter not 100% accurate? or is there another reason?
The cutoff frequencies for low-pass filters in audio codecs such as FDK AAC or LAME are not random. They are chosen for the optimal balance between sound quality and compression efficiency. The main reasons why values like 15.7 kHz or 15.1 kHz are used, rather than exactly 16 or 15 kHz, are ?
Algorithmic tuning: Filters in audio codecs are tuned to the characteristics of human hearing and the capabilities of the coding algorithm. For example, for a particular bitrate, the optimal cutoff frequency depends on the trade-off between preserving detail and removing frequencies that are less important to hearing.
Inaccuracy of filters: Real filters do not have perfect slope. This means that frequencies above the specified threshold may be partially retained. Setting the cutoff frequencies slightly lower, for example, 15.7 kHz instead of 16 kHz, compensates for this inaccuracy by providing more reliable removal of unwanted high frequencies.
Psychoacoustic models: The human ear is less sensitive to certain high frequencies, especially at low bitrates. Instead of a precise threshold (e.g., 16 kHz), the filter is adjusted to maximize the perception of sound within the available bit budget.
These factors explain why encoders use non-standard cutoff frequencies. This is an optimization based on technical and psychoacoustic principles.