Best short mixing tips

[faun2500]

Normalize

to -6db!

[faun2500]

Get it right in mono.

[Fritze]

Listen.

Think and decide what you want.

Then:

Volume fader is king.

[ksandvik]

Less is more.

[Xenobt]

Subtractive eq, and if a part doesn't knock you out, mute it!

KVR/eSoundz: Xenobt

[GFunk]

Volume fader is king.

Agreed. No need to normalize.

[DuX]

The only thing we need to normalize is our brains.

ASR-10 is a 16-bit sampler. With 16-bit you should kinda normalize the samples to get more dynamic range. With 24-bit and 32-bit audio of today you can just forget about it. Yes, that big is the difference. Bits matter, a lot! So much more than any sampling frequency.

I saw a video on Youtube of a man demonstrating a recording at 192kHz, but 16 bit... I can't find the words to describe how stupid that is. Sorry, I forgot the link. No wonder...

People are people, as one of my favourite bands would say.

Knowledge is god. Read and learn! Read manuals, if there's nothing else! Read Sound On Sound, it's a library of audio knowledge! If you don't, there's always a vacant place down at Mr. McDeath's where you can flip shit for *more* money. But much less enjoyable work, of course...

[ducter]

Normalize

Why?

If you normalize the audio of a sample thats low in gain, it saves you keep having to boost the fader or apply massive amounts of gain through distortion or compression plugins. When I used to sample with an ASR10 the first golden rule of sampling I was told to Normalize, Normalize, Normalize. Just saves messing about getting your levels up, and helps maximize your dynamic range.
.

Yes I used to do that as well, but when you have 45 "Normalized" tracks hitting the Master fader the risk of distortion is pretty great.

There is a lot of debate whether or not the software Master bus of different software companies are capable of taking on such massive levels.

I have stopped "normalizing" my tracks and started strapping an Gain insert as the first element in the signal path so I can set a relative level with my channel fader at nearly unity gain before starting my automation moves.

I was told that all the software Mixers are the same if the faders are at unity gain and there is no Master fader.

But as soon as you raise or lower a fader or create a Master, you are using DSP and the "quality of your mixer is based on the excellence of the DSP programming of the software engineer. And your mixer's "Sound" is based on his efforts.

To my ears a Master that is not hit too hard by all the channels feeding it gives better results than one that is getting a pounding by channels coming in to the bus at their absolute max.

[skipscada]

best tip I can think of:

don't mix; arrange!

+1

[Trakstar]

Normalize

Why?

If you normalize the audio of a sample thats low in gain, it saves you keep having to boost the fader or apply massive amounts of gain through distortion or compression plugins. When I used to sample with an ASR10 the first golden rule of sampling I was told to Normalize, Normalize, Normalize. Just saves messing about getting your levels up, and helps maximize your dynamic range.
.

Yes I used to do that as well, but when you have 45 "Normalized" tracks hitting the Master fader the risk of distortion is pretty great.

There is a lot of debate whether or not the software Master bus of different software companies are capable of taking on such massive levels.

I have stopped "normalizing" my tracks and started strapping an Gain insert as the first element in the signal path so I can set a relative level with my channel fader at nearly unity gain before starting my automation moves.

I was told that all the software Mixers are the same if the faders are at unity gain and there is no Master fader.

But as soon as you raise or lower a fader or create a Master, you are using DSP and the "quality of your mixer is based on the excellence of the DSP programming of the software engineer. And your mixer's "Sound" is based on his efforts.

To my ears a Master that is not hit too hard by all the channels feeding it gives better results than one that is getting a pounding by channels coming in to the bus at their absolute max.

The point was individual samples and not entire tracks. If you have for instance a kickdrum thats at-6db its a lot easier to normalize it. I guess this is where common sense prevails...

[Trakstar]

Mute individual clips instead of whole tracks. You can audition several clips on several tracks all at once from the arrangement view fast and efficiently.

[Trakstar]

The only thing we need to normalize is our brains.

ASR-10 is a 16-bit sampler. With 16-bit you should kinda normalize the samples to get more dynamic range. With 24-bit and 32-bit audio of today you can just forget about it. Yes, that big is the difference. Bits matter, a lot! So much more than any sampling frequency.

I saw a video on Youtube of a man demonstrating a recording at 192kHz, but 16 bit... I can't find the words to describe how stupid that is. Sorry, I forgot the link. No wonder...

People are people, as one of my favourite bands would say.

Knowledge is god. Read and learn! Read manuals, if there's nothing else! Read Sound On Sound, it's a library of audio knowledge! If you don't, there's always a vacant place down at Mr. McDeath's where you can flip shit for *more* money. But much less enjoyable work, of course...

Mine was an ensoniq ASR10 and indeed was 16bit audio because I got it mixed up with the 24bit FX engine, my mistake. But for a one word answer and all the "shit" Im getting from it makes you wonder what the real problem is.

http://en.wikipedia.org/wiki/Ensoniq_ASR-10

and for what its worth I liked working with it those 11 years ago. And if normalize is such a stupid idea why would they keep it in the DAW and wave editors.

[Trakstar]

I used to never use reverb at all, because I wanted everyone to hear every last drum edit and cymbal hit. It turned out, even people who didn't know what reverb was were asking me to put it in. I think it can really help music out, the thing is, it doesn't have to overtly sound like "reverb" all the time, just a trace to "thicken" a part and give it dimension works wonders.

+1 good advice on the reverb, it adds dimension and interest to certain sounds, especially for audio concerning ambient FX and strings. They tend to sit better in the mix as well

[ducter]

Normalize

Why?

If you normalize the audio of a sample thats low in gain, it saves you keep having to boost the fader or apply massive amounts of gain through distortion or compression plugins. When I used to sample with an ASR10 the first golden rule of sampling I was told to Normalize, Normalize, Normalize. Just saves messing about getting your levels up, and helps maximize your dynamic range.
.

Yes I used to do that as well, but when you have 45 "Normalized" tracks hitting the Master fader the risk of distortion is pretty great.

There is a lot of debate whether or not the software Master bus of different software companies are capable of taking on such massive levels.

I have stopped "normalizing" my tracks and started strapping an Gain insert as the first element in the signal path so I can set a relative level with my channel fader at nearly unity gain before starting my automation moves.

I was told that all the software Mixers are the same if the faders are at unity gain and there is no Master fader.

But as soon as you raise or lower a fader or create a Master, you are using DSP and the "quality of your mixer is based on the excellence of the DSP programming of the software engineer. And your mixer's "Sound" is based on his efforts.

To my ears a Master that is not hit too hard by all the channels feeding it gives better results than one that is getting a pounding by channels coming in to the bus at their absolute max.

The point was individual samples and not entire tracks. If you have for instance a kickdrum thats at-6db its a lot easier to normalize it. I guess this is where common sense prevails...

Is that -6dB relative to the fader or is it referring to dBFS or Full Scale?

They are not the same animal.

24 bit resolution gives a huge potential dynamic range and properly used puts the noise floor very very low.
Even being-6dBFS is fine, But being right at the top on a lot of samples or tracks that happen at the same instant can stress a Master or the plug-ins inserted on the Master, that is my point.
Making sure that the "clip" indicators never flash is good practice but are they totally accurate?

And some of that distortion is hard to hear as sometimes you don't get the time to check and back down on the levels.

You know you start a mix, one thing sounds a bit quiet, then it is another, and another then the whole mix is frying the master. But you are tried and a bit fried yourself.

Sometimes in ProTools I select all the faders and trim all them done a dB or Two just to make sure, It is the only way sometimes to notice the mix is over the top and getting edgy.

Analog was much more forgiven and even pleasing as saturated.

Digital goes harsh and unpleasant.

[GFunk]

Compression 1st, then eq.