Enabling Video & Audio Formats E-AC3 & E-AC4 F16 F32 F64 & Bfloats 04:35 05/05/2024 (c)RS

Mission enable E-AC3 & E-AC4, DTS-HD Master Audio, DTS:X, Dolby® TrueHD, Dolby Atmos
i understand that E-AC3 ++ is available for use,

Improving Audio & Video quality without undue costs : Performance enhancing audio production in video, audio & life, in 3D Without a very high performance cost on a processor RS

32Bit & 64Bit processing with 16bit to 64bit audio makes sense from a CPU point of view; because most processors by AMD & Intel have 32Bit & 64Bit processing Integers,

So ideal logic is to use 32Bit & 64Bit Integer +- or 32Bit & 64bit floats, The logic of the 256Bit SiMD AVX does allow a Wave of 16Bit & 32Bit values, Bear in mind!

Dithering the upper end of 64Bit & 32Bit Processor Operations would allow losses to 64Bit to 32Bit & 24Bit to 16Bit output without errors.

Modern processors have F16b & although not as precise on the floats as 24Bit, F16b does constitute a 24Bit integer with a 4Bit remainder!

So ideal Sound format for E-AC3 & E-AC4 & AAC+ is 24Bit with 4Bit remainder F16b Floats,
F32 is fine because most AMD have pure F32 & F64/2 = 2F32 Lanes, GPU has F16,f32 & F16b now!

F32b is ideally 48it integer with 16Bit float remainder (0.00001),

RS

"What is Trigonomic compression & curvature?" Reference
https://drive.google.com/file/d/1fLUcZPZYearBjd0wQIIxFBbL1fDQG2Hp/
https://drive.google.com/file/d/1GtSy9FfF1KjwjytbdFFN3vcBHh17Ehue/

https://science.n-helix.com/2022/10/ml.html

ML tensor + ONNX Learner libraries & files
Model examples in models folder

https://is.gd/DictionarySortJS
https://is.gd/UpscaleWinDL
https://is.gd/HPC_HIP_CUDA

https://is.gd/UpscalerUSB_ROM

https://is.gd/OpenStreamingCodecs

The perfect Proposal RS

https://science.n-helix.com/2021/10/eccd-vr-3datmos-enhanced-codec.html
https://science.n-helix.com/2021/10/he-aacsbc-overlapping-wave-domains.html
https://science.n-helix.com/2021/09/temporal-aliasing-image-shaping-polygon.html
https://science.n-helix.com/2021/12/3d-audio-plugin.html
https://science.n-helix.com/2022/09/audio-presentation-play.html

Enabling Audio Formats:

E-AC3 & E-AC4, DTS-HD Master Audio, DTS:X, Dolby TrueHD, Dolby Atmos - These are all decoding formats.

You likely won't need to directly "enable" them, but rather ensure your audio player or receiver supports them.

Your device settings or software should have options to select the preferred format.

E-AC3++ isn't a widely used standard. It's a variant of E-AC3 with some improvements, but not all devices will recognize it.

Bit Depth and Processing:

32 & 64-bit processing with 16 to 64-bit audio - You're correct. Modern processors can handle this efficiently. Using higher bit depth during processing helps maintain audio fidelity, especially with complex calculations.

32/64-bit integers vs floats - Both have their uses. Integers are simpler for some calculations, while floats offer more precision for wider dynamic range audio; The choice depends on the specific application & processing features..

256-bit SIMD AVX instructions - These allow processing multiple audio samples simultaneously, improving performance; Using a mix of 16 & 32-bit data within AVX instructions might require specific handling.

Dithering - When converting between bit depths, dithering adds low-level noise to help mask quantization errors (loss of information). This helps maintain audio quality during down-conversion (e.g., 64-bit to 16-bit).

Audio Formats and Bit Depth:

E-AC3 & E-AC4 - These are compressed formats, typically using 16-bit or 24Bit audio.

DTS-HD Master Audio & Dolby TrueHD - These are lossless formats, meaning they preserve the original audio quality. They can support various bit depths, including 24-bit.

F16 (half-precision floats) - Offer a balance between precision and storage requirements compared to full 32-bit floats. They might be suitable for some internal processing but may not be the best choice for final audio output.

F32 (single-precision floats) - Standard format for high-fidelity audio. It offers excellent precision but uses more storage than F16.

F32b & F16b Higher integers with lower precision float remainders.

Choosing the Right Format:

For most users, E-AC3 or Dolby Digital (compressed formats) provide good quality for streaming or everyday listening.

For high-fidelity playback (Blu-ray, audiophile content), DTS-HD Master Audio or Dolby TrueHD (lossless formats) are preferred.

The choice of bit depth (16, 24, 32) depends on the source material and desired quality. 24-bit offers a good balance for most applications.

While F16 and F32 have interesting possibilities from a processing standpoint, their use in specific audio formats like E-AC3 or E-AC4 is available for improvement.

The current standards typically utilize established bit depths (16-bit for compressed formats, 16/24/32-bit for lossless formats).

BFloats present an ideal when available; Floats an ideal for processing in an efficient bit depth.

RS