› Forums › Music › Sound Engineering › Sample rates
- This topic has 31 replies, 8 voices, and was last updated December 11, 2011 at 11:08 am by dubstep_joe.
-
AuthorPosts
-
November 21, 2011 at 2:17 pm #1245718
does floating point (32 bit float) mean that you can clip in the digital domain and as long as the mixer faders aren’t clipping it’ll not effect the sound? (I read that somewhere but am not sure how true it is).
November 22, 2011 at 9:03 pm #1245724@noname 458332 wrote:
Hope that all makes some sort of sense – gonna stop now cos it got a lot longer than I planned it to be… :sign0020:raaa:lol_crash
How long did it take you to write that? :bounce_p:
Thanks to all for the explanations I’ll try figure it out … Eventually xD
November 22, 2011 at 11:05 pm #1245719@noname 458332 wrote:
gonna stop now cos it got a lot longer than I planned it to be… :sign0020:raaa:lol_crash
You should of cheated used loads of pictures like I did. 😉
December 11, 2011 at 8:28 am #1245703@DaftFader 458349 wrote:
does floating point (32 bit float) mean that you can clip in the digital domain and as long as the mixer faders aren’t clipping it’ll not effect the sound? (I read that somewhere but am not sure how true it is).
32 bit float just means the resolution is 32bit and uses floating point representation (in accordance with IEEE 754 ).
As it’s just a way for computers to represent information, and perform calculations, and doesn’t relate to a physical property of the audio I would suggest that it would have no effect on clipping (which is a physical property of the audio). If it’s clipping in the digital representation then it will be clipping when that representation is converted back into the audio represented…
@dubstep_joe 458487 wrote:
How long did it take you to write that?
A wee while (memory trawling gets more difficult the older you get – especially if your youth was properly mis-spent 😉 )
@DaftFader 458349 wrote:
You should of cheated used loads of pictures like I did.
I tried, but the single picture there took me about as long to draw as it took to write the text :laugh_at:…:crazy_fre
December 11, 2011 at 8:58 am #1245720@noname 460577 wrote:
32 bit float just means the resolution is 32bit and uses floating point representation (in accordance with IEEE 754 ).
As it’s just a way for computers to represent information, and perform calculations, and doesn’t relate to a physical property of the audio I would suggest that it would have no effect on clipping (which is a physical property of the audio). If it’s clipping in the digital representation then it will be clipping when that representation is converted back into the audio represented…
I never clip my faders … I ment when running stuff thru vst’s … would it matter if they were all red lining at 32bit if the final faders were not clipping, or have I totaly misunderstood that part of 32 bit floating point?
@noname 460577 wrote:
I tried, but the single picture there took me about as long to draw as it took to write the text :laugh_at:…:crazy_fre
See you need to cheet even more and steal pics from google 😛
December 11, 2011 at 9:33 am #1245704@DaftFader 460579 wrote:
I never clip my faders … I ment when running stuff thru vst’s … would it matter if they were all red lining at 32bit if the final faders were not clipping, or have I totaly misunderstood that part of 32 bit floating point?
Yes it will matter – if you are redlining any part of the signal chain it will distort whether your mixer is digital or analogue. As I say, 32 bit floating point is just a way of representing in numbers the audio you are working with. If you redline the mixer channel with a VST instrument you are driving it up to a point where there isn’t enough space at the upper end to represent it properly – the result will be clipped audio and distortion (which can obviously be introduced purposefully for a particular audio effect – if you then ensure the main out channel isn’t overdriven it will sound fine assuming you want that overdriven sound for that particular VST – I would suggest you would be better off using an overdrive effect instead of doing this, but if it gets the sound you want doing it this way then it’s all good :wink:)…
@DaftFader 460579 wrote:
See you need to cheet even more and steal pics from google 😛
Normally I would’ve, but I couldn’t find one that showed what I was trying to explain properly :crazy_dru
December 11, 2011 at 11:08 am #1245721@noname 460581 wrote:
Yes it will matter – if you are redlining any part of the signal chain it will distort whether your mixer is digital or analogue. As I say, 32 bit floating point is just a way of representing in numbers the audio you are working with. If you redline the mixer channel with a VST instrument you are driving it up to a point where there isn’t enough space at the upper end to represent it properly – the result will be clipped audio and distortion (which can obviously be introduced purposefully for a particular audio effect – if you then ensure the main out channel isn’t overdriven it will sound fine assuming you want that overdriven sound for that particular VST – I would suggest you would be better off using an overdrive effect instead of doing this, but if it gets the sound you want doing it this way then it’s all good :wink:)…
see that’s what i thought, but I read somewhere that 32 bit float has no clipping point when in the digital domain (or something like that)
-
AuthorPosts
- You must be logged in to reply to this topic.
› Forums › Music › Sound Engineering › Sample rates