5 ms·

Typical DAW already run in floating point, as do most digital mixing consoles. It's only input/output to physical audio cards that's quantized to 24bit. If you

by cnvogel 12y ago

Typical DAW already run in floating point, as do most digital mixing consoles. It's only input/output to physical audio cards that's quantized to 24bit. If you want to "render", i.e. produce a final .wav file from your input tracks, with applied effects, filters, gain-changes,... most DAW let you output a .wav-file with floating point data (not that it would make sense). https://imgur.com/MJvf3cJ https://imgur.com/MJvf3cJ But, you probably guessed it, there are people already discussing the merits of 64bit floating point over 32bit floating point (e.g. "double" over "float")... [already in 2007] https://www.gearslutz.com/board/music-computers/117203-does-64-bit-sonar-really-sound-better.html https://www.gearslutz.com/board/music-computers/117203-does-...

zamalek 12y ago
> there are people already discussing the merits of 64bit floating point over 32bit floating point In my very humble opinion (haven't had the extremely fortunate fortune to be formally educated on DAC and in general the math involved there), I think that discussion is greatly valuable. When you are compounding information into information you can never become accurate enough. Error accumulates - that is the woe if digital composition. I would love to accelerate all this stuff on a GPU - given workable knowledge on how to specifically turn all those crazy mathematical formulas into code; which given only a rip-off degree is practically impossible.
- emn13 12y ago
  Even if you add 1 unit of interference at each processing stage (and since rounding tends not to be malicious, you may well do better than that), you'd need 128 poorly-implemented processing stages for a 32-bit float to be reduced to mere 16-bit integer precision - but in practice, likely more. When it comes to clipping or loss of data on the lower end, well, 32-bit floats have an 8 bit exponent (254 reasonable values); that means that the loudest full-precision unclipped signal is 765 dB (!) louder than the softest un-quantized signal. Even with mediocre centering, that's more than enough. I don't think 64-bit audio is likely to be noticable, even for processing purposes, outside of really specialist kind of niches.
- cnvogel 12y ago
  > "all those crazy mathematical formulas" Most of it is convolution or multiply/accumulate, nothing crazy at all.
  - zamalek 12y ago
    Really, you don't understand how much of a marketing scheme my degree was. An absolute disgrace in terms of what education should be, I'm an idiot for falling for it. I learnt convolution in my own time, so if you have resources (that don't include integration) I would absolutely love to read them. DSP is something that seems hard and I'd love to get into it.
    - diydsp 12y ago
      Heya, I keep track of a large number of resources for learning DSP over at http://diydsp.com http://diydsp.com Take a look at the Theory, FAQs and Books sections. There are some recommended books, links to university course lectures, etc.
      - zamalek 12y ago
        This is why HN is great: industry experts on-hand. Thank you so much, I'm going to dig right into it this weekend.

by cnvogel 12y ago

zamalek 12y ago

> there are people already discussing the merits of 64bit floating point over 32bit floating point In my very humble opinion (haven't had the extremely fortunate fortune to be formally educated on DAC and in general the math involved there), I think that discussion is greatly valuable. When you are compounding information into information you can never become accurate enough. Error accumulates - that is the woe if digital composition. I would love to accelerate all this stuff on a GPU - given workable knowledge on how to specifically turn all those crazy mathematical formulas into code; which given only a rip-off degree is practically impossible.

emn13 12y ago
Even if you add 1 unit of interference at each processing stage (and since rounding tends not to be malicious, you may well do better than that), you'd need 128 poorly-implemented processing stages for a 32-bit float to be reduced to mere 16-bit integer precision - but in practice, likely more. When it comes to clipping or loss of data on the lower end, well, 32-bit floats have an 8 bit exponent (254 reasonable values); that means that the loudest full-precision unclipped signal is 765 dB (!) louder than the softest un-quantized signal. Even with mediocre centering, that's more than enough. I don't think 64-bit audio is likely to be noticable, even for processing purposes, outside of really specialist kind of niches.
cnvogel 12y ago
> "all those crazy mathematical formulas" Most of it is convolution or multiply/accumulate, nothing crazy at all.
- zamalek 12y ago
  Really, you don't understand how much of a marketing scheme my degree was. An absolute disgrace in terms of what education should be, I'm an idiot for falling for it. I learnt convolution in my own time, so if you have resources (that don't include integration) I would absolutely love to read them. DSP is something that seems hard and I'd love to get into it.
  - diydsp 12y ago
    Heya, I keep track of a large number of resources for learning DSP over at http://diydsp.com http://diydsp.com Take a look at the Theory, FAQs and Books sections. There are some recommended books, links to university course lectures, etc.
    - zamalek 12y ago
      This is why HN is great: industry experts on-hand. Thank you so much, I'm going to dig right into it this weekend.