Jump to content

How can I make the music louder and keep the good quality?


girldairy

Recommended Posts

11 hours ago, Larry Jones said:

+1 to @abacab and @Brian Walton. If you just want to get started fast (and have LOUD mixes), buy iZotope Ozone Elements for $129 and use it. In time you'll want to be more sophisticated in your mastering, but this plugin will get you going right away, with almost no thinking (not that there's anything wrong with thinking).

I got Ozone Elements for free at Plugin Boutique last year. It was free with any plugin purchase. So I bought a $2.00 plugin, and that was all it took!

So pay attention to those "deals"!

  • Like 1
Link to comment
Share on other sites

6 hours ago, Kurre said:

So...no one cares about overtones. Interresting.

Actually one reason why I referenced LUFS in my previous post about normalization is because it is about perceived volume, and that includes frequency-domain elements.

There's much more to "loud" than normalization or compression. For example, the human ear is most sensitive in the 3 to 4 kHz range. If you boost frequencies in that range, the perceived level will be louder (although a little goes a long way - if you go too far, the music will sound harsh, and lead to listener fatigue).

Also, I have a technique called "micro-mastering" for producing loud masters that retain dynamics by doing the following to an uncompressed/non-limited stereo mix.

1. Locate all peaks above a certain level, like -3 dB. Hopefully there will be only a couple dozen (that's why it's important the mix not be compressed).

2. Normalize the level of individual cycles, or even individual half-cycles, to -3 dB (yes, it's time-consuming :)).

3. Now you can raise the level of the entire track by 3 dB. 

Because you're changing the level of individual half-cycles, they go by so quickly that nothing seems limited. Also, because you're doing this with DSP, there's zero pumping, breathing, artifacts, etc.

If you then add compression or limiting, you won't need to add as much as if you hadn't used this technique.

  • Like 1
Link to comment
Share on other sites

Coming from a traditional tape recording background, I still find it tempting to record as loud as possible. This made sense when tape was noisy and we needed to minimise the signal to noise ratio.

The thing is, in the digital world this is probably the worst thing you can do, as you're completely destroying any headroom and severely limiting your dynamic range.

If your mixes sound good, but they're quiet, then that's probably a good thing. 24 bit digital recording has a huge dynamic range, but this is extremely limited if everything is close to peak.

If you need it louder during mixing, turn the volume knob on your amp/monitors up.

Now obviously when you get around to distributing your music, you don't want it that quiet, hence all the advice on mastering.

I'd treat mastering as a separate process. Take your "quiet" mix as a stereo pair, put it in either a new project in Cakewalk, or use an audio editor such as SoundForge or Audacity and use something like Ozone to master it.

For interim mixes (i.e. when my project isn't finished, but I want others to listen to it), I'll quite often temporarily stick Waves L2 on the master bus and wack it up as far as it'll go without "pumping" (you can use Cakewalk's Boost11 as an alternative).

Edited by msmcleod
  • Like 1
Link to comment
Share on other sites

5 hours ago, Craig Anderton said:

ud masters that retain dynamics by doing the following to an uncompressed/non-limited stereo mix.

1. Locate all peaks above a certain level, like -3 dB. Hopefully there will be only a couple dozen (that's why it's important the mix not be compressed).

2. Normalize the level of individual cycles, or even individual half-cycles, to -3 dB (yes, it's time-consuming :)).

3. Now you can raise the level of the entire track by 3 dB. 

Enabling snap to zero crossing makes this easier.

Link to comment
Share on other sites

5 hours ago, Craig Anderton said:

True indeed, but trust me...it's still time-consuming :)

Yes, but since you taught me, I have made a practice of it. Not every file needs it.

The other technique I picked up from Bapu was using clip gain to control what you feed in to the fx chain. An example of where this makes my process much faster is for bouncing sermons (tedious job and doesn't have to be perfect). Because I do this every week (sometimes more than once) I have a fairly fixed process and I'm quick.

1. import file to an audio track.

2. find the sermon, split around the sermon with a little music leading up to it.

3. normalize the clip to -3dB

4. use clip gain envelope to notch down coughs or other noises. Perhaps bring up the quiet sections or down the loud.

5. Bounce to clip to see if it looks nice.

6. Normalize again to -3dB

7. Bounce to track through compressor and expander so noise is reduced and spikes are compressed. I use CA2A and MSpectralDynamics for this. High pass filter on the eq.

8. Normalize resulting track to -.2dB and see if the peaks are fairly consistent. If something stands out, return to step 4 and notch it.

9. limit a couple of dB, increasing the overall volume a little more. I use MMultiBand limiter for this.

I can get through exporting this to MP3 end-to-end in about 15 minutes, including recording the intro and posting it to the web.

Edited by Gswitz
Link to comment
Share on other sites

On 4/8/2019 at 7:55 PM, Craig Anderton said:
  • Normalization doesn't affect the sound or the dynamics. It's no different from turning up the level. If you plan to add limiting or compression, turning up the processor's input level, or lowering its threshold, is functionally equivalent to normalizing the input signal to a higher value.

+1, I am baffled at times why people blindly slam normalization when its function is fairly simplistic and isn't going to hurt anything so long as you don't clip. As compressors/limiters are built with "normal operating bands/ranges" in them (software not so much, but hardware for sure), what normalization does achieve is the ability to use the follow-on compressor without having to use extreme values for expected results (and potentially undesired side-effects). FX chains (in general) lend themselves best to processing somewhere close to unity.

  • Like 2
Link to comment
Share on other sites

1 hour ago, mettelus said:

+1, I am baffled at times why people blindly slam normalization when its function is fairly simplistic and isn't going to hurt anything so long as you don't clip. As compressors/limiters are built with "normal operating bands/ranges" in them (software not so much, but hardware for sure), what normalization does achieve is the ability to use the follow-on compressor without having to use extreme values for expected results (and potentially undesired side-effects). FX chains (in general) lend themselves best to processing somewhere close to unity.

+1 backatcha, and to add to what you said, presets for dynamics processors are more or less useless unless the input level at which the presets were created is known. The presets I make for myself assume peaks of -3 dB, so being able to normalize to that level prior to dynamics processing  saves time doing final tweaks on the presets.

  • Like 2
Link to comment
Share on other sites

You may need to take a minute and check out a few videos on mixing/mastering. If you turn up the volume on your track and it crackles, then there is probably a problem with your mix. For a quick start, cut out all frequencies in your track under 30hz for a more clear low end and then cut all above about 14000hz so there isn't too much high end. Also one big tip,  before you boost the overall volume of your track, make sure the mix isn't already peaking. You want to achieve an initial mix that doesn't peak past something like -6db so that when you make your track louder in the master process it doesn't distort. Check out the tutorial topic in this forum and others on youtube, that is just a small part of mixing/mastering. 

Link to comment
Share on other sites

On 4/9/2019 at 5:54 PM, S.L.I.P. said:

I guess the OP is busy milking cows...

My exact thought's too.  :) 

But it did dig up a good topic that never gets boring. On Gear slutz thois whould have become a 6 pager with 1000 "opinions" So far it's been quite civil with only the very small disagreement about Normalizing. I love that tool. It just might be my long standing favourite for many reasons and situations. Distructive? Every thing in digital is thus. 

Link to comment
Share on other sites

🤦‍♀️

So far we have the Waves plug-in recommendation, an assumption that someone with the handle "girldairy" is male, the "time travel" solution (where you go back and re-record everything, except better)....

Also an Amish rakefight on the topic of normalization and, something that baffles me no end, multiple recommendations to purchase commercial FX whose functions are duplicated by plug-ins that ship with Cakewalk. Like having a pizza delivered to the free all-you-can-eat buffet.

Boost11 anyone? Sonitus MultibandFX? Both quite good, both quite capable of serious enloudening, depending on the source material. As the saying goes, if you can't get it done with these, buying "better" ones will be of little help.

How about some of those mastering FX chain presets? For basic "loud," they can get pretty close to Ozone Elements, IMO, at least enough to help out a frustrated newbie. We even have an ultra-simple one-knob thing called MAX. Turn MAX clockwise, MAX makes your track LOUDER.

English does not seem to be girldairy's primary language, but they are savvy enough to have tried Audacity as well as "Sonar." It looks like they tried to get their audio to sound louder by increasing its gain, but ran up against digital clipping (the song just "cracked"). No, a change in level won't increase the perception of loudness, that probably calls for the services of dynamic range and EQ processors. A multiband compressor, limiter, and EQ have been mentioned in this thread. We have an excellent multiband compressor, limiter, and parametric EQ at our disposal in CbB, and a nice handful of presets for them that are designed for just this task.

It is not rocket science to get "one acoustic track" to sound LOUD, without taking classes or spending money or understanding new psychoacoustic principles:

1. Load track into Cakewalk.

2. Apply Cakewalk FX presets (or MAX) until sufficient LOUDness occurs.

If 2. fails, report back to forum. We may need to tell you how to either normalize or apply the Gain knob to your track to give the FX more level to chew on.

Link to comment
Share on other sites

18 minutes ago, Starship Krupa said:

...So far we have the Waves plug-in recommendation...

To be fair, I mentioned that I use L2 for interm mixes, and razor7music mentioned OneKnob Louder with some pretty heavy caveats... that's hardly a recommendation.

20 minutes ago, Starship Krupa said:

...Boost11 anyone?...

I believe I mentioned that too...

I do agree with the rest of what you said though. Cakewalk provides more than enough to do the job.

Link to comment
Share on other sites

55 minutes ago, msmcleod said:

To be fair, I mentioned that I use L2 for interim mixes, and razor7music mentioned OneKnob Louder with some pretty heavy caveats... that's hardly a recommendation.

Still, sikk burnz.

Heh, I just put that in my sig so I could pull people's legs. Well, actually also because it started to weird me out: why was it always Waves and not some other plug-in house? Thinking about it, though, it's because they do offer the most comprehensive line in the business. By far, I think.

Someone's always suggesting one of their products because they're the one company most likely to have something that will work, duh.

I dig my Waves stuff except for that oddball shell that hangs up the loading process. I especially like the fact that any old USB stick can be a licensing dongle.

Boost11 is a pretty cool tool! I hadn't even tried it until a couple of weeks ago I decided to check it out as the last thing on a mix I was playing with. Good graphical representation, easy to follow what the limiter is doing, it sounds good, loud, slammin'.

Link to comment
Share on other sites

On 4/9/2019 at 1:23 PM, Cactus Music said:

If you gave us an example of the recording it would make answering this way easier.  I can think of many reasons your recording is quiet 

If the OP ever returns this is probably the best advice in the entire thread. There are plenty of reasons why, and more potential solutions; but what the OP really needs is to understand what is done with specific material... the why and how... and see the end result on something they know well. It would be a more beneficial learning experience to get this advice rather than generic "what if solutions" thrown back at you. Definitely consider this option.

Also bear in mind that you can also choose the option to selectively post/share material rather than for public dissemination.

Link to comment
Share on other sites

Yes. having even a visual screen shot of the wave form might answer the problem. As my guess is the OP has recorded a acoustic guitar either via PU or mike and there are percussive clips. Those few clips result in overs if you try and increase the level of the whole track. Therefore the OP question. But that's only one of many things that we might find. But seems the OP is a 1 post wonder at this point. Hopefully others might find some of the info useful.

Out of all the free stuff I have from Cakewalk the BT brick wall is my go too when I want to catch a few random overs. I set it at -0.4 on my master. But there a lot more to this topic so I won't dive in that deep. We each have our workflow to achieve loud. Mine is a combo of managing each track via editing out random peaks and just the right amount of compression were needed. But it all starts with a properly recorded audio track.

Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...