Mastering philosophy

Jennifer wrote on 6/3/2004, 12:37 PM

I'm mixing audio tracks from a 4track through Sonar Studio and then mastering in CD Architect. Everything's working but I'm new to the process. Looks like I need to make all edits and volume levels in Sonar (could make them in CDA, but am mixing anyway in Sonar), then just make the CD's in CDA. I thought there was a step in there about volume leveling/compression. I can add track compression in either program...that's it? Is that desireable? And do I have to set my volumes manually throughout the CD? I thought a feature did this and made all the songs the same "required" volume. Any advice on this appreciated.



Geoff_Wood wrote on 6/3/2004, 3:37 PM
In CDA you can 'normalise' each event to full volume, and/or apply an level reduction or volume envelope overall or per track. Or apply overall dynamics control , or the same on a per individual track basis.

Not to mention EQ and any other appropriate plugin.

Jennifer wrote on 6/3/2004, 11:31 PM
Okay, thank you. Would you comment on using compression and also how one can apply overall dynamics control? I appreciate your help. Sounds like you're in the UK.
drbam wrote on 6/5/2004, 6:57 AM
>>Would you comment on using compression and also how one can apply overall dynamics control?<<

Do some "compression" and "mastering" searches of these forums. You'll come up with enough material to keep you reading for weeks. Then start experimenting and practicing - using compression in a truly *musical* way is a high level skill that takes a long time to master.

Jennifer wrote on 6/5/2004, 12:34 PM
Thank you.

Most of these searches reveal that I can't possibly master my own material since it is an esoteric art that takes years to learn. Not as helpful as one would hope. I can pick up a few tips, though.

drbam wrote on 6/5/2004, 4:52 PM
>>Most of these searches reveal that I can't possibly master my own material since it is an esoteric art that takes years to learn. Not as helpful as one would hope. I can pick up a few tips, though.<<

Although not exactly an esoteric art, it is indeed an art form. I personally do not master my own material. The additional set of trained, experienced and fresh ears a good ME brings to the process can literally bring an "ok" sounding mix to an extraordinary one. This of course, assumes that the performance is there and a basically good quality mix is already in place. MEs are artists but not miracle workers. Also, its extremely rare that a project studio would possess the acoustic properties for a decent mastering environment.


Jennifer wrote on 6/6/2004, 3:54 PM
My initial research indicated I could not afford to hire someone to master (or mix) my recordings. If I could, I would not be messing around with this software; I'm a musician, not an engineer. I'm in a 'do it yourself' mode purely due to finances. If there is some group that masters music for free or trade, I'd sure like to know about it.
drbam wrote on 6/7/2004, 6:54 AM
You have a lot of company. The primary reason people master their own material is lack of $. At any rate, with lots of research, practice and experimenting you should be able to create something that you feel good about - which is the main issue I suppose. You might try some other forums as well. is probably as good as they get and it has a specific forum for mastering issues.

Good luck,

Jennifer wrote on 6/7/2004, 1:34 PM
Thank you, I'll check it out.

Geoff_Wood wrote on 6/10/2004, 6:17 PM
The most basic method is put something like WaveHammer (or Waves L1, L2, whatever...) on each track, and tweak the preset and threshold on each to get a good balance between all the tracks.

rpmbassman wrote on 6/17/2004, 3:23 PM
Well, I am a mastering engineer in Austin, TX, mastering projects for clients here in Austin, NY, NJ and even Spain.

My rates are $35/hr, so you can find reasonably inexpensive mastering these days.
Softy wrote on 6/29/2004, 6:55 PM
Want to tell me a little bit about the style of material? I might be interested in working on it.
jeff112464 wrote on 8/4/2004, 5:18 PM
hello so i master my audio as weel in Sonar 3 producer pack and i find it to be mush cleaner in sonar compression i do alot of mastering in the studio
but im having troubles with my DVD when I try to burn it it prepares but it doesn burn
Rednroll wrote on 8/20/2004, 4:05 PM
Don't let mastering intimidate you. I do mastering, mixing, recording, editing, basically the whole works as far as audio goes. You'll see a lot of posts discouraging you from doing your own mastering. These gloomy discouraging posts come from people that just don't understand mastering. To me mastering is much easier than mixing a large project. I find my skill for mastering is much better than my mixing. The key to mastering is learning the general tools used in mastering, then you need to learn those tools and EVERY parameter in them and what they do, to better achieve what you want. There is no one button push or single plugin you can use for mastering for every song, just like there isn't one for mixing. That's why you need to learn the general tools used, and know what each does, and then choose the proper ones for EACH particular song. I use a very basic setup in my mastering and feel I can master with the best of them.

Here's my list of basic tools. Note I don't use everyone, everytime. I just know what each one does and am able to hear the effects if I start to overuse it.

1. TC Finalizer- Multiband Compressor, Digital EQ, Limiter, Deesser
2. SpectraLab- Spectral Analyzer
3. Waves S1- Stereo Imager plugin
4. Sound Forge- VU meter,Normalization,DC offset remover, A/B level comparison between songs, final fade outs.
5. Steinberg Magneto- Tape compression simulation plugin, without the side effects of going to tape.
6. Vegas- Final CD track assembly and burning master CDR.
7. Nero- CD duplication simultaneously recording to multiple CD recorders, adding CDtext information.
StyleSupportMIDI wrote on 8/25/2004, 10:16 AM
Hi Rednroll:
What tools do you focus on FIRST when you are doing your master?
ex: Wavehammer - Could this be a starting place?
I think some need to have a starting place.
You have a long list of impressive tools, but many of us have SF7 CDA, Sonar kinda stuff.
You're Quote:
The key to mastering is learning the general tools used in mastering, then you need to learn those tools and EVERY parameter in them and what they do, to better achieve what you want. There is no one button push or single plugin you can use for mastering for every song, just like there isn't one for mixing.
We hear similar statements to yours often and are somewhat intimidated, because I'm left with a sense of "where to start."
Maybe wavehammer, as in an earilier post, is a good start.
Rednroll wrote on 8/30/2004, 3:18 PM
The last place to start should be Wavehammer for mastering. Wavehammer is a plugin that has no consideration of spectral qualities. That is the majority of time your high frequencies are the highest parts of unmastered material, due to the nature of transients. So what happens when you use a plugin like Wavehammer or Waves L1 it will over compress the high frequencies, causing a harsh sounding loud mastered final product. The first place you should start is with a multiband compressor. With a multiband compressor you can add different amounts of compression to the different frequency bands. To bring up the overall loudness using a multiband compressor, you can do a lot of compression on the mid frequencies, with little noticeable side effects. The high frequencies you compress the least and the low frequencies are somewhere inbetween the mids and the highs.

Here's a snippit from another post I put in the Vegas Audio forum for a starter for using a multiband compressor.

I use 8:1 on low, 4:1 on mid and 4:1 on highs. The thresholds then become more important on how much compression you actually do for each frequency band. You will find you can highly compress(ie lower threshold) the mid range to achieve a higher level. The low's threshold value should be a good +6dB higher than the mid, with the highs a good +10dB higher. You will most likely find you can get about a -6dB reading on the gain reduction meter with the compression on the mids with little side effects. Compress sparingly on the lows, with the least compression on the highs.

I recommend you get a good multiband compressor and start there. Waves has a good one, and the Isotope Ozone one seems pretty good, and also the Ultafunk one is nice, but no longer available except with Cakewalk products. It is probably best to start with a preset and make similar adjustments like I mentioned above. This way it will give good suggested attack and release settings for each bands compressor, to reduce pumping sounds. The next tool is a good EQ plugin. The multiband compressor may alter the spectral curve of your audio, so you may need to make EQ adjustments to compensate. My usual starting point is to have an EQ followed by a multiband compressor.

Like I mentioned though, the Wavehammer should be your last step. If you find you need additional volume after applying multiband compression, then use Wavehammer sparingly. Using Wavehammer alone to achieve an overall competitive level is a very bad idea, and will add a lot of unwanted distortion side effects. I personally use Steinberg's Magneto for this step. Mastering engineers have been using analog tape compression for years to level things out and make the audio louder. Magneto is a plugin which emulates tape type compression without the side effects of adding tape hiss noise. I find this is the type of compression that has the least amount of side effects to my spectral curve, where Wavehammer or Waves L1 will f**k up your high frequencies, making you go back and have to readjust your EQ settings, but you're still left with some possible distortion and harshness.

Here's a good thread to read in the Vegas Audio forum, which shows another users experience with multiband compression:
Rednroll wrote on 8/30/2004, 4:04 PM
Here's a little further tips on mastering with some of the other tools I mentioned above.

VU meter:
A peak meter is only good to make sure you are not going over the 0dB digital limit, therefore adding distortion, but no good for perceived loudness. Your ears respond to loudness very similar to how a VU meter responds. So if you want to make songs match up in loudness a VU meter is a very nice tool. So when using the multiband compression and other loudness maximizer tools, pay close attention to the VU level meters. If you want your material to be as loud as a commercially released CD, then look at the average VU meter reading when you're playing that song back and make a note of where the VU meter is reading. Then start mastering your material and bring your material up to match that VU meter reading. Do this on the rest of your subsequent songs and when you are done, along as the VU meter level is within 1 dB of each other all your songs should match in level.

Spectral Analyzer:
Use this similar to using the VU meter, except now you're matching up frequency response curves. Take some commercially mastered material that you know sounds good and that is similar in style and content to the material you are mastering. Play that material through a spectral analyzer and capture that frequency response curve. Now look at your material and look for drastic differences in certain frequencies compared to the commercial release. Use an EQ to tone down or turn up those frequency differences. You can more easily make EQ adjustments by making sure your Q settings are not too wide or too narrow, and you will SEE the effects on the spectral analyzer of your EQ adjustments. It is easy to pinpoint problem frequencies using a spectral analyzer, especially if you don't have a trained ear and can pick out frequencies by listening.

Be careful though, don't become dependant on the spectral analyzer, your ears should always be the final judge because every song is different. For example, I caught myself relying on a Spectral Analyzer one time too much. I was mastering a song and noticed the mid frequencies where extremely high. So I started to reduce the mid frequencies to better match up with my reference material. I got done and bypassed my EQ to hear the original unmastered material and realized the lead vocals had a phone type of effect on it (ala Nine Inch Nails type of voice). Well that type of vocal effect has a strong mid range boost to it, and when I reenabled my EQ adjustments, I noticed I totally whiped out the intended vocal FX and it sounded like a normal uneffected vocal part.....oops.....wasn't paying enough attention to the original content. Luckily I caught my mistake before I shipped it out the door.

I also had a mastering job, where the client was annoyed with a noise everytime the keyboard bass line played. Through the use of a spectral analyzer I was able to see that everytime the bass played it had a higher harmonic frequency right around 12Khz. Optimally, this is something that should have been caught by the mixing engineer, but wasn't. By looking at the spectral curve I was able to pinpoint the exact frequency the annoying harmonic was at and reduce the bandwidth of my EQ so it only effected that harmonic, and not the surrounding frequencies. So I was able to make a very accurate EQ adjustment on the entire mix, with no bad side effects.

I run the spectral Analyzer and the VU meter together while making multiband compression adjustments and EQ adjustments. They are great visual aids to be able to see the effects of the adjustments I'm making.