PG Music Home Forums PG Music Product Forums RealBand Most basic question

Hi all.

This is a pretty standard question but it's puzzling me a little.

I've created some top notch backing tracks using realband and realtracks and I've loaded them up onto my Ipad so that I can play them as backing for my guitar / vocal act.

Some of them are quieter than others, which is a bit of a pain as it means I have to remember which ones need turning up / which ones down and which ones are just right (Goldilocks levels you could say).

I've tried normalising them but of course, whatever is the highest existing peak on the waveform can restrict the amount of change. E.g., there is invariably a cymbal crash that for some reason wants to be Mount Everest.

Now, I've tried compression (in a DAW), but that is a something of a ticklish business because I've a feeling that the Realtracks have already been mastered (I'm not using the more expensive wave files).

What does everyone else do ?

Cheers
Mike.

www.Blues-House.com

Mike,

A general answer.
When getting into "serious" mixing:
1. Go to the Showcase site and listen, listen, listen to the
music placed there.
This alone is/was invaluable to me anyway. They do use primarily PG Music goodies.
Discount the genre when listening, many of same mix rules would apply across several genre's.

Some amount of mixes, did not care for, but reason was my listening was wrong.
You have to listen to each of the presentors individual tracks as they play thru the music.

Listened to music all my life, hence why re-learn how to listen.
Being able to pick out specific parts of the music is a great guide to your
own mix.

Also based on reading folks on the "Forums", they do give a lot of excellent advice.

Probably not a new word, but been using the word "earring". My version of how to listen
to each song and track within that song.
This is a non-stop ongoing learning process.

Also based on advice from other member purchase a very good set of "Studio" headphones.
Before this, had no idea of the difference in the Hphones.

Also having a very good quality of Monitor speakers is crucial. Based on reading forums.

The work-flow being used by folks would vary from person to person.

Here, create the music in BIAB.

Save As Wav file clicking Normalize, and also all tracks. Normalize here maybe point of contention with some folks.

From Win 7 open window containing all the music tracks, drag/drop them into RB.
You can also do this directly from BIAB.

SaveAs a raw music file upon drag and drop completion.

Then solo melody/dominant/primary track.

Then Highlight 8-10 measure of initial melody or dominant track(will call primary now on)
along with all of the music tracks.
This gives you ability to listen and listen to the mix for initial primary start.

Set primary volume level... This level usually monitor by Hphone and also the
meter on bottom of the screen.

Next add specific track/s you want to listen to one at a time. Usually here,
in initial mix use primary with one support track at a time. Gives me a grasp of
sounds.

In your case with backing tracks, probably no melody(now primary), so pick out dominant track
to mix against.

Once each track is set against the Primary track, then start adding additional
tracks. Listening adjusting levels for those track balances.

My experience in beginning was thought Bass, and drums were to dominant.
Wrong, big time, especially for dance music. That beat is extremely crucial.
Obvious to most folks here, but for newbie mixing person was not.

After repeated track combo listening step, then go to adding FX.
Lucky to own some Cakewalk Sonar products.

Add first VST plugin PG dynamics. Usually use default setting volume level to 2
sometimes bit more.

Then Add Sonitus MultiBand VST which has presets. Lot of folks may frown on this, but
gets me to reasonable sounds. Basically "medium" mastering" setting.

Then add K-meter Stereo VST, using one of its presets.

Then am avid fan of K-meter level metering. IMHO, its an outstanding tool to use.
Use Standard level setting, also peak display. Using this shoot for overall levels
about -12DB.

Adjust final output level use the big RB mixer.
At this point do save SEQ as "FilenameFX.SEQ. Active work file.
Actually do this earlier several times.

Once the sound is decent in the Hphones, take off, turn on monitors and listen some more.
Some times, bonkers on the mix sound, then go back and fix it up.

Then click Audio > Merge audio and vsti/dxi tracks to stereo.

Then play result thru the Monitor speakers.
Sometime go back to RB and redo offtrack parts. Pun intended.

As you already know with your work, there are many, many in between steps that
will be needed.
One example is again IMHO, add Reverb to "Some" individual tracks as wanted.
Happen to use Sonitus Reverb with short and sweet preset. Especially for piano
and picked guitar.

This answer is long, but should be some tidbits that may help.

Flaw learned from this forum and also hard way is....Each device music is played on
and listened too will sound different.

An Old guy who loves music...Good Luck.

Now you all know the ??? methods I use.

Mike, you may or may not be right about the individual RealTracks having some processing but I don't think the question applies for the way you're using them and your comment about uneven song levels. You may need to look at each of your audio files in a different way and make some adjustments based on what you see and hear.

Normalizing adjusts the overall gain of a file to a preset gain level but doesn't change the overall dynamics within a file much at all. A quiet lullaby stills sounds quiet (low energy) while a heavy metal file will sound like ... heavy metal (high energy). The apparent loudness of the heavy metal file over the lullaby file is due to many factors but the most important factor in this case is the rms value, or energy level, within the two files.

Normalizing works on peak levels. Most gain meters also display peak values. A rms meter calculates the energy level inside a waveform. Try using a rms meter across your stereo master. Play a song that you know is recorded at the level you want then play another song that is quieter than you want. Watch the levels for each and use your ears to lower the gain for the first or raise the gain of the second until both sound like they are at a similar sound level. It's pretty normal to have to change the gain between songs or even between different sections of a song until they are properly mixed and mastered.

The cymbal crashes may require multiple solutions. You could need to lower the overall volume level of the drums for that bar, use a low pass filter to remove some of the higher frequencies of the cymbals or edit out the cymbal crashes altogether.

Welcome to the wonderful worlds of mixing and mastering.

Wow ...

Thanks Seeker and Jim for those very useful and comprehensive answers. That has certainly given me food for thought and I'll try some of those techniques.

Thanks again for taking the time to put those posts together.

Regards.
Mike.

Maybe try using PGPeakLimit (a very simple compressor/limiter),
followed by a multimeter tool like ParMeter or RMS Buddy
http://www.kvraudio.com/product/dfx_rms_buddy_by_smart_electronix/details

Not recommending the link above, and have not tried the above; it just popped up in a search. I like ParMeter, but using any tool that also measures in a way other than the current peak value on the VU meter may be of help.

Using the RMS value of each song over time would likely get them closer pretty quickly.
Play with a combination of the track volume and the boost in PGPeakLimit to get the RMS of each song fairly close without having to overly crunch the signal in PGPeakLimit.

What that actual RMS number should be depends on your recordings, genre, etc.
How close is 'close' depends on how high the target number is; the higher the number, the closer the range should be.

At -22 you could have songs with a range of 6 and it would probably pass.
At -12 you'd aim for more like 2 or 3 differential. I'm sure there is a mathematical explanation for this, that someone like Mac could supply, but just knowing these kinds of tools are available may be of help.

Do not assume that the real tracks are mastered. Mastering is a process done to a group of songs, although we tend to apply the term to single songs.... the better term is "polish".... and no, the real tracks are recorded well but there is always room for improvement.

To get consistent levels, you need to start from the tracking stage.

As you move through your process, be sure the levels are good and strong but not clipping. On the export, shoot for a nice level with not a lot of massive compression. Just knock off the big spikes so the wave looks like this:



I have it in my audio editor WAVE PAD where I trim it and normalize it.

What you see there is the result of the export from Sonar before I normalize it. There is headroom there that normalizing will fill.

In Sonar I use Ozone to polish the mix and add some light compression to knock down and tame the bigger spikes.

Generally, at the end of this process, most of the songs are all at the same general level and nearly equal to radio ready levels.

Attaining a consistent level takes some time and practice but is not that hard to achieve.

Thanks Rharv and Hacker for those great posts.

Cheers.
Mike.