No normalisation?

[jaggedarray]

I use this workaround : open Audacity, "Effect" => "Amplify".

You beat me to it, I was gonna say Audacity, it's free, does almost everything one would need. An invaluable addition to our toolset, and did I mention free?

[blouaap]

Mmm. Normalising, just because of the wave form image? So drag up the noise with it? Not particularly a sample pack I would use if the samples were all normalised to 0dbfs, just for visual aid. Audio is about hearing, not seeing...

I think what you want is a Vertical Zoom, like Wavlab or Audacity, etc, that don't affect the actual signal.

[frenchboy]

Mmm. Normalising, just because of the wave form image? So drag up the noise with it? Not particularly a sample pack I would use if the samples were all normalised to 0dbfs, just for visual aid. Audio is about hearing, not seeing...

I think what you want is a Vertical Zoom, like Wavlab or Audacity, etc, that don't affect the actual signal.

You obviously don't know what normalizing means. Most sample packs are normalized and they should. Normalizing is a completely non destructive process, it's just for ease of use so each sound has the same peak volume.

[Ogopogo]

Normalization (or any other process) is destructive if the file is saved that way so you can't undo, right? That's the definition of destructive, isn't it? So normalization could be nondestructive if it's applied that way in your daw (like a sort of extra fader applied to the item to make sure the peak hits 0db that you could turn off at any time), but normalization isn't inherently destructive or nondestructive.

[gravy]

The type of critique of normalising above seems like something you hear about the recording technique of compressing everything in sight.

Normalising is not compression and will only be destructive in the sense of introducing some very minor rounding errors. Of course it would destroy the volume relationships if applied separately to multiple stems but why couldn't BWS allow selection of a range acroas multiple stems and analyse the volume across those stems to determine amplification level.

Most bouncing is also destructive if you delete the bounce source. But the technique gets used to simplify and speed up the creation process and consolidate work done.

All I'm suggesting is a bounce-normalise function.

Are their any sensible objections to that?

I think a non-destructive normalise-amplifier might be useful but it seems implenting it logically would be a lot more complicated especially in the bws editing paradigm. I'm not convinced their is a way to do it well.

The whole point of having visual waveforms is to be able to actually see them and make use of the visual info. Since we're all here working with sound I think it's safe to assume we use our ears sometimes in the process.

[obx777]

I am not sure if normalizing is right or wrong, however, in my workflow of tracking I have been using it for years. Be it laziness on my part or not.

Example, I track a left guitar part with a 57 on axis and then do a right with a different cab and a 421 off axis the option to quickly "normalize" these two tracks then throw my levels to around -7 and have them in the ball park is handy. I was perturbed that Pro Tools did not have a "right click" normalize option.

I have been pelting the feature requests for some right click editing features (cut, cross fade, fades, etc.) Dom told me they are aware of these sort of editing/ tracking feature desires, but for now some other things are on the front burner.

I'm a linux user who dumped pro tools so I am in panic mode trying to get a DAW I can live with. Bigwig is very, very close. Really I think there are a few of us Linux people who wish Reaper would just hurry up and get a native version out. But I digress. I second the original posters feature request and would like to see it on a right click drop down. Better yet give us the ability to customize our own right click options, now that would be cool, at least to me.

[tommyzai]

I normalize all digital audio. It's a good sound practice, and there is no rational argument against it assuming peaks are maxed out at a safe level of -1dB. It's not compression for Pete's sake (not really sure who Pete is). This is a MUST feature for all DAWs! Having to drag audio into another application works against the entire reason for using a DAW. If normalization is not in Bitwig at this time, I'm sure it will be at some point. Back to teaching far, far away.

[Coockie1176ln]

Normalize means bringing the entire audio on wav down or up to match loudest peak to 0db. Meaning it will either be boosted or lowered.
example if you have a 32 bit floating point wav that is +3 db it will be set to 0 db. If having -4 db audio will be raised so loudest peak match 0 db. In generall that is what normalize have set as value.

The negative part with normalize is this. Imagine having an audio format which loudest peak are at 0 db.
Of you lowcut an audio or make drastic shapes with eq or boost eq, you will change volume level.
This can be a problem if you have a compressor after. Eq will phase the wav. Lowcutting audio can boost with 1-3 db in volume.

imagine this putting a lowcut on a normalized audio can easy be +3 db on loudest peak.
Now if you want to use a compressor afterwards it will be harder to make a final level. Some compressors have saturation
built in so the louder it is the more the saturation will be mixed in etc. Like hot leveling a audio clip. Also the threshold will allready be active since audio is +3 db.
Imagine if you put a limiter it will puch volume down by 3 db attenuation and ofc everything a limiter touches is some kind of distortion.

Normalize is not bad thing, but one need to be aware of the negative aspects to get better use of it (Y)

ps about it being a must have is not correct. Everything doesnt have to be 0 db to get great mixes on each track (Y) however if it is a sought out function for some here it is your right to want it. i dont use this feature in other daws. as i have better controll of getting right levels through mixer channels without. But feel free to feature request what you feel you need for your workflow here support@bitwig.com

People dont complain about what people need, we all have our preferences, no right no wrong

[NukeArts]

When normalizing a integer wav isn't the rounding error getting an issue? I by myself wouldn't do it and would also just opt for the vertical zoom.

[Coockie1176ln]

When normalizing a integer wav isn't the rounding error getting an issue? I by myself wouldn't do it and would also just opt for the vertical zoom.

Normalizing is the same as just dragin fader up and down to fit 0 db. Except that the interall volume will be 0db. Nothing happens. Its nothing else happening. Just means max peak will be 0db. No roundening or change. Dynamic will be the same. The only negative part is what i wrote above. But some prefer mixing this way and let em if they want

[NukeArts]

When normalizing a integer wav isn't the rounding error getting an issue? I by myself wouldn't do it and would also just opt for the vertical zoom.

Normalizing is the same as just dragin fader up and down to fit 0 db. Except that the interall volume will be 0db. Nothing happens. Its nothing else happening. Just means max peak will be 0db. No roundening or change. Dynamic will be the same. The only negative part is what i wrote above. But some prefer mixing this way and let em if they want

That's why I referred to the integer wav. Of course, when you pull up a fader in the floating point world it's the same.

[Coockie1176ln]

Think of having 0 db going into a emulated analoug plugin or a real hardware that is set to work without distortion at -18 db.



That video explains really well why you should have good amount of headroom. Just imagine how normalize can create issue. This is the issue with normalize and why i never use it. But i am not complaining on those who do or want. but for me this is the reason

[Sound And Found]

I realize this would be done manually, but couldn't you just go to the loudest part of your track and just raise the master fader until you see it peak to almost 0db? Then export the track. It should take like 1 minute, tops.

Also I'm curious what you need to normalize your tracks for. I only normalize when I'm working with video or something. When I have a bunch of files at different levels.

I usually don't even bother with my output level until I'm mastering, and at that point I'm just throwing a tool in the device chain to boost the level.

[Coockie1176ln]

I realize this would be done manually, but couldn't you just go to the loudest part of your track and just raise the master fader until you see it peak to almost 0db? Then export the track. It should take like 1 minute, tops.

Also I'm curious what you need to normalize your tracks for. I only normalize when I'm working with video or something. When I have a bunch of files at different levels.

I usually don't even bother with my output level until I'm mastering, and at that point I'm just throwing a tool in the device chain to boost the level.

its more important to reach -0.3 db headroom at mastering not so much mixing

there is the digital to anlooug conversion that will have isp if volume to close 0 creating distortion

[digitalfix]

I normalize everything going into a mix and then get the headroom by dropping the faders.

If you normalize to 0db and then start mixing with your faders set to the same, you're going to redline quickly. It was a good idea of Bitwig to default the faders to -6db.