TranceAddict Forums (www.tranceaddict.com/forums)
- Production Studio
-- Re doing your sounds while your mixing

Re doing your sounds while your mixing

Hey guys, I just wondering if any of you re do your sounds while your


mixing. Lately for me with everything that I have been working on

I will make the song and then get to the mixing stage and start messing

with my sounds on the synths and I just end up with a bunch of crap

that I throw away. Or I will halfway make a sound saying I will fix the


sound or the automation parameters when I enter the mixing stage. Well,

it's not working out for me too good so I was hoping for some advice

on the matter. Sequencing music can be sort of confusing when you have

a compressor, eq, distortion, reverb, ect on the synth AND the channel

mixer... I was hoping for a run down on the proper steps I should be

doing things. Thanks for listening

Yes, sometimes.

1) learn how to write really good music (if you're only going to be doing remixes of other people's songs, skip this step).

2) get to know the sounds you have at your disposal, and which of them work well together. Use a really standard chord progression /lead, and spend a few months just combining sounds (this will also give you a handle on the primary mixing tool, the level faders). Do a few a day, don't spend too long on a given mix. When you have 50-100 put together, you can listen through them and notice how good some combinations sound, even without eq or other effects. You'll also notice how bad some combinations sound, even though the individual sounds in the mix might sound great on their own. This process will also refine your ears, so you can quickly work out when a sound fits well into a mix, lock it in, and not be tempted to change it later on.

3) learn about eq/filtering and how it can further refine an already very decent sounding mix, and apply it.

4) learn about all the other tools you have at your disposal to enhance a mix, and apply these.

5) know the difference between learning and doing, and be clear about which one you're doing on a given day. If you're learning about the sounds you have, work on that and don't worry about trying to create a song. If you're getting a song done, don't spend timelearning about the sounds you have - use what you know, and get the song done. Don't focus on "perfection" - focus on getting it done to the best of your ability. If that ends up short of where you'd like it, schedule some more learning sessions to improvewhere you need to improve.

quote:

Originally posted by clay always changing everything but the mixer faders, thats my go at it.

Which is how 0.00000001% of all producers and engineers work.

Derail shared some useful tips but my take is it's about decisions and being good with commitment. If you're at the mixing stage and start messing with sounds then your sounds weren't right in the first place and you need to perfect that element of your work before moving on to the mixing stage.

It's also good practice to finish songs even if you're not 100% happy with the sound selection as at least you will be able to look back and think "I should do XXXX differently next time" but if you never finish it, you'll never know.

It also helps you then make better selections when choosing your sounds and simply put, that just comes by doing it over and over again. You want to know why some composers are successful? It's not really even musical talent in most cases. It's because they've done it over and over again to the point they just know what works. That then gives you scope to experiment.

IN this way, you experiment to do something different, and not just endlessly tweak sounds throughout the entire process resulting in something useless.

Stop getting bogged down with all the other FX and crap you can do. As with your VST's/Synths, just pick a select few and learn them inside out. Now make a tamplate that has certain tracks, auxes, busses, synth s already loaded; this "limitation" will actually speed up your workflow.

Choose sounds, stick with them, move on. once you've got your sounds, commit to them and only then move on to mixing.

Sure, if one thing is not working, fine than change it but don't get into revisiting everything. If somehting else begins to sprout as part of the creative process and the tracks goes in to a different direction, then save as so you either work in this new strain or go back to the old when at your convenience.

I think you'll find that most/many electronica/EDM producers tweak sounds, levels, FX, etc. throughout the whole production process. It's more experimental in that regard, compared to traditional/mainstream recording processes, where the song is composed and practiced, the majority of the song is then tracked (recorded), and finally the tracks are mixed (usually with some overdubs and/or minor additions, as needed). In EDM, the composition, recording/tracking (if any), sound design, and mixing phases are pretty fluid for many people and a lot of people don't really seem to approach it very methodically, but that's what works for them. So, if your current approach isn't working for you, try a more/less methodical approach and see if you can find that sweet spot. And, definitely see derail's advice above.

Personally, I tend to take a pretty methodical approach because I come from a more traditional musical background. But I find that I'm usually a lot less methodical and more fluid in my approach when I'm creating an EDM/electronica track than I am when I'm working on rock, pop, ballad, etc. projects.

Well thanks for the feedback guys.

I guess I'll figure something out but I see what you mean

quote:

Originally posted by clay always changing everything but the mixer faders, thats my go at it.

LOL

quote:

Originally posted by clay always changing everything but the mixer faders, thats my go at it.

If I'm correct this ain't the first time you mention this.

And imo it is uber stupid.

It's not a DJ approach though; what you're doing is akin to adjusting the output on your turntable. The gain knob IS the fader in a DAW mixer. And sure, I know you;re going to say "no, the gain knob in the instrument is the gain" but that's not the case.

You're meant to dial in a suitable "hot" level (call this the "100% mark", which is the optimum level of output from the device) from the instrument itself, then control it's volume with the fader, and leave the gain where it is. That is your gain staging right there.

In the same way you set the channel gain via the knob on a DJ mixer then bring it in with the upfader (only noobs use the xfader), and that's where the misconception is in this case: Your first and fixed point of gain is the gain on the instrument, then the fader is meant to be your variable.

I can't be certain, because I'm not privy to the coding of reason, but you may actually be sacrificing some bit depth resolution by mixing the way you do.

quote:

Originally posted by clay I only use the mixer to watch meters and never let any of them (including master) go above "green". green means ok in my world, everything else tells me im messing up. it might seem stupid for someone trained in a studio but as i only work in a computer i dont see any reason for why this is stupid. i have total control over every instrument, im absolutely sure that my channels isnt clipping in any separate channel or my master and i can easily copy instruments, efex, presets etc or other things between songs and they almost fit instantly (+/-3dB adjustment at max).

With 32bit float audio, you don't need to worry about individual channels clipping. You only need to be concerned about your final (master) channel prior to DA.

http://www.soundonsound.com/sos/jan...es/qa0108_3.htm

quote:

In most practical applications, floating point is still used within a 32-bit framework, but with 24 bits allocated to the mantissa and eight bits allocated to the exponent. If you do the maths you'll find such an approach provides the utterly ludicrous theoretical dynamic range of 1500dB, which means you will never run out of headroom inside the processing, and never lose signal in noise floor. So, as you found, you can increase or decrease the level to the most ridiculous extremes inside a floating-point system, and as long as you restore the gain to something more appropriate to feed the output converter's dynamic range, you will not suffer from the noise or clipping that a more conventional fixed-point or analogue system would. Which is very impressive.

However, some OLD AS FUCK audio plugins still use 24bit fixed audio internally and you do need to worry about levels in those cases. The most notable example of this is the default EQ in Logic (even the latest version). Not only do you lose a lot of resolution, plugins like this also use unnecessary CPU because the computer has to convert every sample from a 32bit or 64bit float to a 24bit integer. Long story short, you shouldn't be using plugins like this anyway.

quote:

Originally posted by echosystm However, some OLD AS FUCK audio plugins still use 24bit fixed audio internally and you do need to worry about levels in those cases. The most notable example of this is the default EQ in Logic (even the latest version). Not only do you lose a lot of resolution, plugins like this also use unnecessary CPU because the computer has to convert every sample from a 32bit or 64bit float to a 24bit integer. Long story short, you shouldn't be using plugins like this anyway.

I'm not sure that's true about Logic EQ being 24bit. From what I know, all the plugs of logic are 32 bit float, it't the output of logic which is fixed at 24bit and you don't ever need to worry about that as you should never clip the master anyway.

I don't know of any currently sold plugins that are still 24bit.

quote:

Originally posted by DJ RANN I'm not sure that's true about Logic EQ being 24bit.

There was a thread on Gearslutz about it, but I can't find it now. Some guy noticed a lot of aliasing from the EQ, which obviously shouldn't happen in an EQ since it's not adding harmonics like a compressor etc. In the end, they worked out the aliasing is caused by truncation noise.

They guessed Apple kept it like this for legacy reasons; lots of people have projects that would sound different if they changed it.

Just FYI... Keeping your mixer faders at unity doesn't mean you aren't hitting the red between your components. Example...

Instrument -> +3db -> Insert -> -3db -> Mixer reads -3db and you think all is well.

Maybe you should put a level meter between every single component in your project. That way you can be 100% sure you aren't going into the red.

quote:

Originally posted by clay the mixer would read 0dB in that case.... i still dont think im doing anything wrong really, perhaps i would agree with Rann if i was working with hardware to keep the instruments dynamic range into the mixer to get better noisefloor etc. but in a software mannor i cannot for the hell of my life understand that what im doing is wrong. i tried raising all the instruments by +15dB and lower all the mixer channels with equal -15dB. not a tiny bit of change on the master, nor the individual meters. i think this is bullshit. i bet all values are calculated at reals (aka floating point) and therefore not matter at all.

I'm not sure in the digital domain what happens, I will ask this to my old Audio teacher, but in the analog domain the instrument gain stage should affect the "timbre" and/or "character" of the instrument as we use to say here in my country "signal gets hotter".
Maybe you can try a simple test using a good spectrum analizer with a high FFT and sending a simple sine at let's say 1Khz with gain at zero and fader at zero (nominal states) and then change volumes first in the gain and then in the fader, take some screen captures and compare if the sound gets distorted or altered in some way. I can try to do this test but I'm not in the studio right now.

by the way what bit resoulution and sample rate are you working?

truncation is different with 32 bit float but i mean it is something you are never going to hear or notice unless somoeone tells you it does and you start imaging shit like those people that buy really expensive cables to run a connection 1 metre.

Here's the thing; Reason (as most DAW's are) is 32bit floating point, except at the output stage which is 24bit becuase for the physical output it has to be 24 per your dac.

So in practical use it is impossible to clip 32BFLP between faders or gain stages so long as you don't clip the master.

Now in theory, lowering one fader or gain stage by x and then raising another by that same amount x should result in the same thing, but here's where it gets complex;

Many VST/AU/instruments/effects do different things in terms of dynamics and even sound different depending on the volume or how hot the signal is. In this respect, working the way you do is not how they were designed to be used so you may get unintentional results or not the full performance from the device in question.

The only way to be sure is to do a null test - record a sound from an instrument the way you currently do it (adjust gain on the instrument then have the channel fader at 0) then the normal way of gain staging (have the instrument at unity and set the fader to match the previous master level output).

Bounce both to audio in the same way, import them both in to a fresh project, one on each track, invert the phase on one and you should have silence.

If not, then you know there's a problem.

Powered by: vBulletin
Copyright © 2000-2021, Jelsoft Enterprises Ltd.