Final Mixing for Soundcloud (newbie vs the loudness wars again...)

wmalan

wmalan

Member
So for the first time, I'm attempting to put a few of my songs up on SoundCloud. Low and behold virtually every other song I compare to there has been pumped up. So without going into the loudness wars... What do ya'll do prior to uploading. My tracks are all around -15db (Reaper). On the stereo mix buss do I crank the gain to 0db? Add some compression? I know it's very vague but I'm thinking there is a basic approach for the final cut (I won't use the word mastering as I'm just a humble songwriter that happens to record his own songs).

Basics to follow? Or a well worn thread topic that I should see?


Bill
 
what does it sound like before you loudness normalize it? why is it necessary to take part in the loudness wars?
 
wmalan said:
My tracks are all around -15db (Reaper).
Click to expand...

Average or peak? If that's an average then you're a bit lower than the slammed stuff at -10dB average. If that's a peak level for the whole track then you have plenty of room to simply raise your level.

You might as well start now with the traditional procedure of bringing your mix file into a new project and process it there. Processing may be as simple as normalizing peak levels or as complicated as you like. A typical basic mastering chain is an eq into a mastering limiter.
 
bouldersoundguy said:
Average or peak? If that's an average then you're a bit lower than the slammed stuff at -10dB average. If that's a peak level for the whole track then you have plenty of room to simply raise your level.

You might as well start now with the traditional procedure of bringing your mix file into a new project and process it there. Processing may be as simple as normalizing peak levels or as complicated as you like. A typical basic mastering chain is an eq into a mastering limiter.
Click to expand...

I guess I'm a bit more vague than I thought. I try to record each track at -15db as my understanding is unlike the days of tape, I don't need to "saturate" the signal (correct?). As far as average or peak - that would be peak as Reaper shows me the max db level recorded on each track. I do sometimes use a light compression on some tracks at the input side to help it be more of an average without squashing it.

Thinking about this, I'll try raising the final output to say -3db keeping in mind I'll need to make mix changes relative to the output.

Creating a new project just for tweaking the stereo output sounds like a great idea! (My humble attempt at mastering...).

Thanks all!

Bill
 
If your mix is really peaking at -15 then you can raise it as much as 15db. Do this in your mastering project. Best practice is to leave a tenth or three of a dB headroom.
 
If you don't already know, the limiter for the most part you just want it to sort of dance, lightly touching your peaks ever so often to get maximum volume without slammed compression. For me, if I see some light compression (just where I see the meter move), I know after that point, an increase in gain, the compression will be more and I start to losing dynamics.
 
Personally I tend to use an intersample peak aware limiter on my master, last in chain

I don't master in my DAW, I do that elsewhere... but I still use a limiter to catch any anomalies

I stay at -1 - not -.1

Reason being is taking into account intersample peaks - where two or more concurrent samples at 0 can equal more than zero (clip)

Now, I use an intersample peak aware limiter that will catch those (Waves WLM Plus), but, still at -1, because you still want to account for potential clipping, or added level when passing through D/A converters... if you have a true -.1 source, good chance you will get some distortion while passing through D/A's.

That said, I output my session files at -15 LU, -1db limiting...
Then, in mastering, I'll boost everything to -10 LU if I am going for loud/commercial type 'standard' ..... again, at -1db limiting
 
ShortySMC said:
Personally I tend to use an intersample peak aware limiter on my master, last in chain

I don't master in my DAW, I do that elsewhere... but I still use a limiter to catch any anomalies

I stay at -1 - not -.1
Click to expand...

I use the SSL X-ISM meter to check things. I haven't had a problem with -0.3dBFS peaks in my files, but any measurement of intersample peaking is approximate as it ultimately depends on the actual converter used. I don't think this is the thread for getting all detailed about that stuff.
 
As has been said, recording needs to take place at a lower level to allow headroom for safety and to also permit mixing and effects to happen in the best range.

However, once you've completed your mix, there's nothing wrong with bringing your levels up to approaching 0dB(FS). It's probably best to stay half a dB or so under 0 since metering isn't always perfect and, frankly, nobody will hear half a dB anyway.

The thing is, when people talk about "loudness wars" really they're usually talking about "dynamic range wars". It's not having a maximum peak at nearly 0 that causes the problem. Rather, it's compressing the life out of your music so EVERYTHING, not just peaks, sits at nearly 0. As I said, this sucks the life out of your mix (at least in my opinion) and things don't sound as good.

But it's not a simple as saying "don't compress". If you don't and your original has a wide dynamic range, people will either think the quiet bits are too quiet to hear properly...or if they adjust for the quiet bits, the peaks will blast their ear drums. I think a lot of this problem stems from the way people listen to music now. Instead of being on speakers in a quiet domestic room, listening happens in noisy cars or via ear buds on a noisy street or train or bus.

For this reason, you'll need to find a compromise that suits you in terms of keeping enough dynamic range that your music has some "air and space" but not so much that the quiet bits get lost in the background hubbub. There's no easy formula...what's right depends on your tastes and your musical style.

But, back to the original query...put your peaks as near 0dB(FS) as you dare. After that, where the quiet parts go is up to you.
 
DM60 said:
If you don't already know, the limiter for the most part you just want it to sort of dance, lightly touching your peaks ever so often to get maximum volume without slammed compression. For me, if I see some light compression (just where I see the meter move), I know after that point, an increase in gain, the compression will be more and I start to losing dynamics.
Click to expand...

That's how I'm using the Kjerhaus Master Limiter, getting the peaks to -0.3db and this seems to translate well onto Soundcloud.
 
wmalan said:
So for the first time, I'm attempting to put a few of my songs up on SoundCloud. Low and behold virtually every other song I compare to there has been pumped up. So without going into the loudness wars... What do ya'll do prior to uploading. My tracks are all around -15db (Reaper). On the stereo mix buss do I crank the gain to 0db? Add some compression? I know it's very vague but I'm thinking there is a basic approach for the final cut (I won't use the word mastering as I'm just a humble songwriter that happens to record his own songs).

Basics to follow? Or a well worn thread topic that I should see?


Bill
Click to expand...

well,
whenever i play back a tune on soundcloud, or any other streaming service,
if it's too soft compared to other ones,
i just turn it up.

it is dirt simple, to turn the volume up.



usually, what i find is that the softer sounding songs, cranked up, sound WAY WAY WAY WAY better than the slammed ones.

they usually have good dynamics, better sounding mixes, and don't hurt my ears after a while, like most of the other LOUD MASTERED SONGS do,
with very few exceptions,
even on pro material.



now if you want your stuff to sound LOUD just to compete with the LOUD GROUP,
then you have to learn to master your material.


you can either pay a pro to do it (highly recommended) or do it yourself (i've heard more bad home mastered material than any other issue that exists).
 
GONZO-X said:
well,
whenever i play back a tune on soundcloud, or any other streaming service,
if it's too soft compared to other ones,
i just turn it up.

it is dirt simple, to turn the volume up.



usually, what i find is that the softer sounding songs, cranked up, sound WAY WAY WAY WAY better than the slammed ones.

they usually have good dynamics, better sounding mixes, and don't hurt my ears after a while, like most of the other LOUD MASTERED SONGS do,
with very few exceptions,
even on pro material.



now if you want your stuff to sound LOUD just to compete with the LOUD GROUP,
then you have to learn to master your material.


you can either pay a pro to do it (highly recommended) or do it yourself (i've heard more bad home mastered material than any other issue that exists).
Click to expand...

Yeah, it's dirt simple to turn up the levels on a quieter track when you're just checking it out in isolation. However, if you have a playlist of 50 songs going in your car or on your iPod, turning things up and down for every track becomes the sort of royal pain in the butt that most people won't want to put up with. Remember, we're not just talking recording enthusiasts here. The "general public" have different thoughts.

Beyond that, while there are excellent reasons to keep your levels down during the recording and mixing process, there is NO reason not to get up near 0dB(FS) in the version that's finally mastered for distribution. You can leave the dynamic range completely untouched if that's your choice.

However, as I said earlier, even on this you probably need a compromise for that dreaded "general public". Somebody listening on earbuds in a noisy environment won't appreciate the subtleties of a 40dB dynamic range if they can't hear that quiet, sensitive introduction.
 
GONZO-X said:
well,
whenever i play back a tune on soundcloud, or any other streaming service,
if it's too soft compared to other ones,
i just turn it up.

it is dirt simple, to turn the volume up.
Click to expand...
Soundcloud is actually a pretty good example of where you absolutely don't want to force your listeners to do that. If you jump to a specific song, SC will auto-play a semi-random song (usually by a different artist) right after. If a user has to crank your volume up and then get his head exploded by a loud-mastered song right after, it's not going to improve his impression of your song.
 
Bobbsy said:
Beyond that, while there are excellent reasons to keep your levels down during the recording and mixing process, there is NO reason not to get up near 0dB(FS) in the version that's finally mastered for distribution. You can leave the dynamic range completely untouched if that's your choice.

However, as I said earlier, even on this you probably need a compromise for that dreaded "general public". Somebody listening on earbuds in a noisy environment won't appreciate the subtleties of a 40dB dynamic range if they can't hear that quiet, sensitive introduction.
Click to expand...

Both statements true - if you really want to preserve your dynamic range, but get it louder, just crank up the track slider, and render it that way.

GONZO-X said:
^^^ this way of thinking is quickly becoming obsolete.
in case you haven't heard......

iTunes Mastering - "Mastered for iTunes" explained | JustMastering.com
Click to expand...

Is that what MFiT actually does - masters everything to the same volume/dynamic range? I hate Apple, and this is another reason - only allow ing music that has been mastered by one of their 'approved' engineers to carry the MFiT label. And the fact still remains - only audiophiles and people like us homerecordists (the recording pros, of course) really care about the sound quality difference.

VomitHatSteve said:
Soundcloud is actually a pretty good example of where you absolutely don't want to force your listeners to do that. If you jump to a specific song, SC will auto-play a semi-random song (usually by a different artist) right after. If a user has to crank your volume up and then get his head exploded by a loud-mastered song right after, it's not going to improve his impression of your song.
Click to expand...

I HATE when that happens -I'm usually fast on the draw to close the window out - wish everyone would embed their SC files here in the MP3 clinic instead of linking to them.
 
mjbphotos said:
Is that what MFiT actually does - masters everything to the same volume/dynamic range? I hate Apple, and this is another reason - only allow ing music that has been mastered by one of their 'approved' engineers to carry the MFiT label. And the fact still remains - only audiophiles and people like us homerecordists (the recording pros, of course) really care about the sound quality difference.
Click to expand...

As far as I can tell it's just a set of standards and procedures aimed at mitigating certain resampling/encoding distortions. The final volume is still in the hands of the listener and you can still crush your stuff to death if you wish. What's the problem?
 
You must log in or register to reply here.
Back
Top