Normalizing audio in DP10?

[mikehalloran]

The signal still has to be hot enough to be usable in a mix. Normalize, when needed, does not fail at this. 4 clicks including scaling back on Bite gain and I'm ready. I used to have to do it a lot, nowadays, it's rare.

Most of my work these days does not involve tracking. Many of my collaborators do not understand the tools they have. One person may give me tracks that are clipped and others where he had the output nearly all the way down (same guy). When I'm on weekly deadlines, I don't want to be talking them through how to export tracks againnnnn! Ok, I do it but prefer to do Zoom meetings with my Performer Lite, GarageBand and Logic users.

I'm ok with texting, "Hey, half the tracks are blank, could you please export again?" That's down to once a month or so, nowadays.

[monkey man]

I don't understand why normalising would be preferential to bumping up the gain with Trim.

All my mixer channels have Trim inserted in the first slot.

[stubbsonic]

I use normalization often.

The first case is where I know a track is a bit low and I need to boost it, but I don't want my boost to introduce clipping anywhere. Normalizing safely boosts without any risk of adding clipping.

The second case is when I've done a mix and I wasn't riding levels super attentively to get max gain before clipping. I'll normalize before exporting. I think the L.A.M.E. export might normalize before converting to mp3, but I'm not sure.

And third, and to me, most importantly, if I'm processing samples, I always normalize samples. Even, and especially velocity-switched samples. It's a boring explanation, but long story short, Volume is ramped by MIDI velocity, with samples switching as you go up smoothly from 1 to 127.

[HCMarkus]

I use normalization often... The second case is when I've done a mix and I wasn't riding levels super attentively to get max gain before clipping. I'll normalize before exporting. I think the L.A.M.E. export might normalize before converting to mp3, but I'm not sure.

If normalizing to peak at full digital code, the potential for inter-sample peaks exceeding digital zero will be present. This can cause audible distortion in some circumstances. The old standard was to leave .3dB headroom. It is currently recommended to leave 1dB headroom. Of course, many simply say F it and even go for a wee bit of digital clipping.

A less obvious issue when setting gain for digital masters can occur on playback. Whether it's a compressed file like an AAC file or an uncompressed file such as a CD, digital data goes through several processes to be converted to an analog signal for playback. One common process is called oversampling. This upsamples the digital data at four times the original sample rate to improve the quality of the digital audio signal being converted to analog. If the original digital audio data is at 0dBFS, oversampling can result in undesirable clipping. And if the original was already clipped, oversampling can make it worse. Our recommendation is to leave at least 1 dB of headroom in order to avoid such clipping.

https://www.apple.com/itunes/docs/apple ... asters.pdf

[monkey man]

And third, and to me, most importantly, if I'm processing samples, I always normalize samples. Even, and especially velocity-switched samples. It's a boring explanation, but long story short, Volume is ramped by MIDI velocity, with samples switching as you go up smoothly from 1 to 127.

Yup; that's standard practice I would have thought, Stubbsy. I certainly did it back in the day when making sample-based instruments. It provides a consistent baseline.

[stubbsonic]

If normalizing to peak at full digital code, the potential for inter-sample peaks exceeding digital zero will be present. This can cause audible distortion in some circumstances. The old standard was to leave .3dB headroom. It is currently recommended to leave 1dB headroom. Of course, many simply say F it and even go for a wee bit of digital clipping.

That's what I like about DSP Quattro's "normalize to..." (which I use for batch work). It allows you to set a non-zero max. I wish DP would offer a version of that.

And third, and to me, most importantly, if I'm processing samples, I always normalize samples. Even, and especially velocity-switched samples. It's a boring explanation, but long story short, Volume is ramped by MIDI velocity, with samples switching as you go up smoothly from 1 to 127.

Yup; that's standard practice I would have thought, Stubbsy. I certainly did it back in the day when making sample-based instruments. It provides a consistent baseline.

I'm glad to hear you say that. I've seen quite a few libraries over the years that give you non-normalized samples where their goal is to "preserve the original dynamics". The result is maybe 6-20 velocity levels, with some version of jerky jumps in level across the velocity range.

[monkey man]

And third, and to me, most importantly, if I'm processing samples, I always normalize samples. Even, and especially velocity-switched samples.

Yup; that's standard practice I would have thought, Stubbsy. I certainly did it back in the day when making sample-based instruments. It provides a consistent baseline.

I'm glad to hear you say that. I've seen quite a few libraries over the years that give you non-normalized samples where their goal is to "preserve the original dynamics". The result is maybe 6-20 velocity levels, with some version of jerky jumps in level across the velocity range.

Ridiculous!

Internal waveform dynamics are preserved during normalisation, and as you say, MIDI velocity determines which layer/s is/are triggered.

The overall volume is determined by the velocity->amp parameter, so their thinking is antiquated at best, but not without good intentions, obviously.

[funkyfreddy]

Interesting thread!

When I first got started using DAW's (many moons ago) I used to normalize almost everything. Now I use it mainly for effects or creating samples. I NEVER apply it to a mix.....

[stubbsonic]

This was probably already mentioned in this thread, but with 24 bits, if you have reasonably good level, you don't have to be as fussy about getting close to zero with tracks. However, with mixes, it is probably a little more important to try to get fairly close-- in part to just be a bit more on par with a basic level.

The loudness wars used to inspire mastering engineers to maximize perceived volume through aggressive limiting, but fortunately that is falling away. People are beginning to appreciate (contrasting) dynamics in music more widely.

[monkey man]

Indeed they are, Stubbsy, thank God.

It may not be fully-mainstream yet, but at least it's become "a thing".

[funkyfreddy]

I still wish it was as easy to normalize in DP it used to be.

When I'm playing around with samples and/or doing sound design I like to quickly get them all at the same level. This was much easier in DP when the normalize command was available outside of the waveform editor.

DP's waveform editor has always seemed a bit clumsy to me. I always preferred BIAS Peak.....