Resampling/Dither
Resampling/Dither
I'd rather not start a topic on what I feel like is the lower end of the scale of important things to music making. However that's just how I feel, it might be way more important than I think, it might not.
This a byproduct of the days I read too much online, still do probably but this stayed with me and it's something I don't bother thinking much about now, but feel like it would be good to get some solid info on and some knowledge on the board from people with experience with what I'm talking about.
So - I read a while ago that whenever resampling, and making any changes at all, you are creating truncation errors blaah blah, good practise is to dither every time resampling. I can't be fucked with that to be honest and feel like just a single dither at the end of the chain should cover any nasties up, though I don't know. Not particularly sure what to listen out for... Maybe this matters more if your resampling many times,I don't know.
I almost wish I never read it as it's not the kind of shit I want to think about when making music, however knowledge on these things can always help and I'm open to knowing what the crack is with it and just put it to rest in my mind and perhaps help my music. If your really new to production, my advice would be read manuals and do some basic digging but don't spend too much time reading online, it can be a plethora of total bullshit that is hard to get out of your mind once there.
So basically these are my questions - resampling practices, is it wise to dither each time or just once at the end/not at all if sending to mastering. If in Ableton, I change the resample bit rate to 32bit, do I not need to care about this at all and it will just use a bit more disk space for recording audio as it runs at 32bit floating point?
Part of me does not give a fuck about this and knows there is much bigger picture things going on, part of me thinks it'd be wise to learn about it or at least feel like I know what I'm doing with it. Clearly a major case of cognitive dissonance going on as it's stayed with me quite a while now, I'm hoping some light can be shed on the subject and thus some resolve and forget about it. I don't think this has been covered on the board from what I could find, perhaps it has but if not, it could lead to some valuable info.
This a byproduct of the days I read too much online, still do probably but this stayed with me and it's something I don't bother thinking much about now, but feel like it would be good to get some solid info on and some knowledge on the board from people with experience with what I'm talking about.
So - I read a while ago that whenever resampling, and making any changes at all, you are creating truncation errors blaah blah, good practise is to dither every time resampling. I can't be fucked with that to be honest and feel like just a single dither at the end of the chain should cover any nasties up, though I don't know. Not particularly sure what to listen out for... Maybe this matters more if your resampling many times,I don't know.
I almost wish I never read it as it's not the kind of shit I want to think about when making music, however knowledge on these things can always help and I'm open to knowing what the crack is with it and just put it to rest in my mind and perhaps help my music. If your really new to production, my advice would be read manuals and do some basic digging but don't spend too much time reading online, it can be a plethora of total bullshit that is hard to get out of your mind once there.
So basically these are my questions - resampling practices, is it wise to dither each time or just once at the end/not at all if sending to mastering. If in Ableton, I change the resample bit rate to 32bit, do I not need to care about this at all and it will just use a bit more disk space for recording audio as it runs at 32bit floating point?
Part of me does not give a fuck about this and knows there is much bigger picture things going on, part of me thinks it'd be wise to learn about it or at least feel like I know what I'm doing with it. Clearly a major case of cognitive dissonance going on as it's stayed with me quite a while now, I'm hoping some light can be shed on the subject and thus some resolve and forget about it. I don't think this has been covered on the board from what I could find, perhaps it has but if not, it could lead to some valuable info.
Re: Resampling/Dither
So, dithering from what I remember is adding low volume noise which frequency is independent of the source material, but depends on sampling rate.
If so, you dont need to care about it if you're not downsampling.
Also, if you would dither twice, you would basically just increase the volume of that noise.
Then, if you would add eq, you would eq the dither as well.
So if I'm correct, I would leave it for the last export, on the mastered track (and send your mix without it in native sample rate)
If so, you dont need to care about it if you're not downsampling.
Also, if you would dither twice, you would basically just increase the volume of that noise.
Then, if you would add eq, you would eq the dither as well.
So if I'm correct, I would leave it for the last export, on the mastered track (and send your mix without it in native sample rate)
Re: Resampling/Dither
P0607r0n wrote: ↑Wed Oct 03, 2018 1:12 amSo, dithering from what I remember is adding low volume noise which frequency is independent of the source material, but depends on sampling rate.
If so, you dont need to care about it if you're not downsampling.
Also, if you would dither twice, you would basically just increase the volume of that noise.
Then, if you would add eq, you would eq the dither as well.
So if I'm correct, I would leave it for the last export, on the mastered track (and send your mix without it in native sample rate)
Thing is and the reason for my post, was that it is apparently a good thing to add dither when resampling because your either adding low level noise, or truncation errors, the former being more conically pleasing. So your stacking one or the other. Some DAWs do it automatically. Thing is, I don't really know and like I said in the post, it's a byproduct of reading into it. I understand what dither is, just wanted to hear some experienced views as to whether it's actually worth it or not. Like I said, I feel like it's way down on the scale of things to be concerned over, much bigger things to focus on but perhaps some worth in discussing it.
Surprised this hasn't had much response at all. I think if I use the 32bit setting in Ableton for recording, and then just dither at the end this is the best solution and then I don't have to give it any thought.
Re: Resampling/Dither
From what I gathered, dithering when resampling is obsolete, unless you're downsampling to smaller bitrate then native. In case of downsampling, you might benefit from dither (but don't quote me on that). If your want to add dither while resampling I think you should do it only once, when you change from native sample rate to smaller.
Re: Resampling/Dither
I think your right yeah, if I change the record settings in Live to 32 bit then it won't change the rate thus avoiding the truncation errors. I am not super clued up on it but I think that's correct. Hopefully that doesn't use way more disk space.
Appreciate your replies good sir.
Appreciate your replies good sir.