Mastering for non-engineers

[odiwgal]

....In conclusion....

Wow, I never thought my thread would generate so many responses with such great information. Thanks. I want to sum up this way. It sounds like if I want the ultimate mix, the best thing to do is turn it over to a mastering house. However, most of what I do is small-ensemble classical recording, many times simple mixes of only soloist, a duo or trio of musicians. So, I am planning to mix and master as best as I can at home, on the computer. That brings me to my question. Should I go with Soundforge, T-racks, or Wavelab? On the face of it, Wavelab looks to offer far more options than T-racks. I have a version of Soundforge that camw with my CD burner, but I suspect that it's a "light" version. Are there other solutions or general suggestions?

I will save the "mastering house" route for my most important projects!

Thanks again to everyone.

--odiwgal (One Day I Will Get A Life)

 

[Rev E]

Odiwgal,

I think that Wavelab is the best bet for what you're doing. While I agree with Bruce, that real mastering is best done by real mastering engineers in real mastering houses, I think that home demos or lower-profile projects would benefit from some home mastering/final tweaking. Wavelab can process 24-bit files (Sound Forge is limited to 16-bit); its internal processing operates at 32-bit (it's important for the processing to be higher than the bit rate of the file that's being processed so the processing doesn't degrade the file before dithering down to 16 bit), plus Wavelab can import plugins to be used with Steinberg's internal processors, among other things.

I haven't tried T-Racks personally, but I've heard nothing overwhelming about it, compared to Wavelab, from anyone whose professional opinion that I trust. So, I'd recommend Wavelab off the cuff.

As long as you are aware that the mastering process is something that engineers spend their entire lives learning and perfecting, you shouldn't have any overestimations of what you will be able to achieve from your home setup. Home mastering (in it's proper perspective) can be a learning tool and a money saver for demos and for outside commercial projects that don't have a budget for mastering. Although, if any project is worth its "salt" in the COMMERCIAL world of music, it WILL have a budget for mastering. But for $300 US, Wavelab is worth a try.

Rev E

 

[dobro]

Rev E, one of two things is the case. Either you're a diamond in the rough, or you happen to be teaching to my level of understanding.

"it's important for the processing to be higher than the bit rate of the file that's being processed so the processing doesn't degrade the file before dithering down to 16 bit"

I never understood this before. A long time ago, Sonusman and RE took on a rep from Alesis over the issue of the quality of the Masterlink, based on its internal processing. So if what you say is true, then you not only need software that processes at a bitrate higher than what you tracked at, but you need a soundcard that will do the same?

And yet I come back to a very relevant question: HOW important is it? Loads of people rave about the sound, the all-important sound, they get from some of these boxes, so if it isn't important to go pro on a project, why not use a box? And just to be a boring fart, I'd like to know if a) it's easier to use a finalizer or masterlink than software, and b) can they do anything Wavelabs can't?

And just in case you're an engineer reading this, I want to say that my ultimate aim is to get into a pro studio with my music. But for the time being, I'm at home, and I don't need or want a homerecording question answered in terms of a pro studio. It just makes me feel stupid and pissed off.

 

[Rev E]

So if what you say is true, then you not only need software that processes at a bitrate higher than what you tracked at, but you need a soundcard that will do the same?

Dobro,

No, you don't necessarily need a soundcard that goes beyond 24 bit. This is because even the best 24 bit Analog/Digital converters currently available - i.e. Apogees - have a theoretical dynamic range of ~120 dB's... and the theoretical maximum of a 24 bit file is 144 dB = 6 x 24. So what sound interface manufacturers need to do is not give us higher bit depth, they need to perfect the A/D converstion process so that we get closer to the 24 bit maximum dynamic range. Truth is, a 16 bit file that is recorded with ALL of its dynamic range is wonderful enough even for many of the most critical ears.

BUT for the sake of argument, when we record a 24 bit file, we want to preserve every bit of the fidelity of that file. So we use processors and processes that operate at a higher level so that the original fidelity is preserved... i.e. we don't lose anything.

And yet I come back to a very relevant question: HOW important is it? ... the sound... they get from some of these boxes, so if it isn't important to go pro on a project, why not use a box? ...I'd like to know if a) it's easier to use a finalizer or masterlink than software, and b) can they do anything Wavelabs can't?

Importance is something for YOU, the user, to decide. This will only come from actually using the stuff. Let's see... to use a finalizer or Masterlink, I have to first spend $1500 for a piece of HARDware. Every time I want to master, I still have to transfer the file from my computer workstation (when working with Finalizers; Masterlink has a built in hard disk) to the Finalizer AND do my home mastering on a screen the size of a popcicle stick. Don't forget it's a HARDware piece, so when ever Alesis or TC Electronics comes up with something new.... (e.g. better A/D converters, new processors, better circuitry) I have to plunk down another $1500+ to get another HARDware box. You decide.

Whereas, your friend Ed bought Wavelab which processes files at a higher internal processing rate than both boxes. He gets to edit his master sitting at a 17" monitor. He can buy Waves power pack plugins (regarded as the best in the FX software in the business) for $700 (that's $500 less than the hardware stuff). When Waves upgrades their plugins, He can upgrade them for <$200... (ever tried upgrading a hardware box) PLUS, he can still use those same awesome plugins in his DAW recording/mixing software during the mix process. Like I said, only the USER can decide what's best for him/her. But, to me the decision is clear.

Rev E

 

[LongWaveStudio]

I personally think the processing quality of a Finalizer is better than all the PC stuff ive tried, and Ive tried most of the PC software. Why not get a Finalizer 24/96khz or get a masterlink and just learn how to use it properly. thats an option.

 

[sonusman]

A few clarifications....

Rev E, it would seem that Alesis changes their tune on the Masterlink box. In that thread long ago, I ask the Alesis rep to tell me WHAT the internal processing was. He didn't know, and went to Alesis and came back with a 24 bit fixed answer. Later, on Alesis's website, they claim that their box actually does 40 bit internal processing. Maybe the bit depth "increases" to 40 bit before being dithered back to 24.... Who knows. That being the point. Who knows....

Next. Computers work on a 32 bit floating point dsp scheme. THAT is equal to 24 bit fixed dsp roughly. Some software works on a "double sampling rate" scheme, which is hard to find any information about, but it is something to the effect of producing two samples of every "actual" one for the processing to be applied to, then it is "sampled back down" to the original sample rate. Steinbergs QMetric mastering eq does this, and it is an exceptional sounding digital eq to my ears.

odiwgal, if you are working with classical music ensambles, I would stay away from ANY digital mastering that you would do yourself. That is music that simply does not need "mastering" in the sense that Popular music does.

Okay, let me clarify even further what I mean. There are many components to the modern mastering process. Eq, compression, song to song overall level adjustments, cross fading between two songs, fade outs at the ends of songs, fade in to songs, and of course the old "hidden track" at the end of a CD. Any or all of these components COULD be applied to a collection of songs that will be on the same CD. I say COULD because you may not have to do ANY eq, compression, or song to song overall level adjustments. You may not have crossfading between songs, etc.....It really depends.

The reason I said earlier that maybe you should start doing it and experimenting is that you will not know what you need to do until you see what you CAN do and hear the results.

With classical music, the most I would consider doing to it if I was mastering the project is placing the proper amount of time between the songs, and cleaning up any noise before the piece starts and after it ends. That is IT! No eq, no compression, no song to song level adjustments. ANY of that stuff you do is going to create problems with the original sound if not done with the highest quality equipment. In classical music, it is all about the sound being "natural". The techniques used in recording it are all about making a microphone catch the performance as close to what your ear hears at the performance as possible. Eq and compression are seldomly applied, and usually only because there is a severe problem that needs to be addressed.

So, I guess what I am saying is that of any music style there is, classical is traditionally the one that doesn't get the hard core treatment in mastering. For that style, song order, a few noise problems, and space in between songs is what is usually applied to the original "mixes" of it in mastering.

Hell, for that, you could use Goldwave. It is about $40 and works very well.

Don't go messing with those beautiful sounding acoustic instruments. If it don't sound right at mix, you messed it up while tracking it, and no amount of eq and/or compression is going to make it sound right again. If anything, eq and compression on the instruments in classical music with make things sound worse.

My opinion. Take it for what it's worth.

On a side note about mastering. I believe that mastering is a culminization of the whole recording process. Without a great handle on eq and compression, and keen insight into the benefits/drawbacks of it, and a pretty good handle on obscure stuff like dithering and what not, it is best to just leave your product alone. I have yet to really post much about the whole process of mastering because it is hard to articulate because you don't have to do the same thing twice per se in a mastering process. It is very involved to do it right, and takes a lot of explaining. Any engineer who has done it or seen it done knows what I am saying here.

I believe great results can be achieved with the "less then" equipment, but that of course requires that you really know what you are doing with it. I don't suggest however that a person with no idea what they are doing could produce great results with all the best gear just because they have the best gear.

If any of you really feel you need to learn the art of mastering, it is quite simple, find a mastering engineer who you feel does a great job and hire him/her to master a few projects for you and sit in and watch what it entails and hear what they hear. This will be an expensive learning process, but you will learn a lot from it. Tis' how old sonusman learned to do it.....

Anyway. Software or hardware.......Software for sure. Wavelab on a PC for sure. Waves plugin's work nice. Hardware for the same price? Well, be advised that you will wind up running that mix back into the computer to author a Red Book Standard disk complete with PQ timesheet, so why bother with $1000 or less hardware? That is money better spent on plug in's that will do that same thing. THAT is what I didn't like about the Masterlink. Also, like RevE said, plugin's can be upgraded, and new ones can be bought.

Top end analog gear yields the best results in mastering for eq and compression. Anyone can argue all they want, but there is a perfectly good reason why most mastering for top artists is done with analog gear.....you know what I mean? I have heard the difference with my own ears too, and the analog processing route is the way to go for the best results. But it is very very expensive. The next best route is software plugin's. Some great one have already be named.

Good luck.

Ed

 

[odiwgal]

I think you're right there. No matter what you are using, you must learn to use it properly. I could sit behind the console at a mastering house, and all I would do is make a mess. But a mastering engineer could probably use Wavelab or a finalizer in MY studio and come up with something decent. I talked to a friend earlier today who has been doing home recording professionally for many years. He does alot of commercial work, and has done soundtracks for several major US public TV documentaries. He has always done his own mastering, armed only with a finalizer, a finely tuned ear, and a clear idea about the sound he's after. So I guess it's possible, as long as you are "one" with your equipment. Thanks for all the discussion.

 

[dobro]

Rev E - thanks for your excellent reply, this thread's just about nailed it for me.

"So what sound interface manufacturers need to do is not give us higher bit depth, they need to perfect the A/D
converstion process so that we get closer to the 24 bit maximum dynamic range."

If this were the case, would it then be much more difficult to clip when tracking?

Your description of the downsides of mastering with a box was ace. For me, the bottom line is the sound, but ease of use is really important.

Ed - it takes a lot to get you going, but when you do, it's worth reading. What you said about classical music not needing mastering probably applies to loads of folk music as well, right? Or does close miking a number of instruments bring up a need for mastering more often than classical?

 

[guhlenn]

Thanks for answering Ed,

Do you know of any good books on mixing/mastering as there no talented mastering angineers in my neighbourhood. Well, just no mastering engineers, just some people who call themselves that way.

Again : thanks,

Guhlenn

 

[Rev E]

Ed,

Thanks for your reply. I found it invaluable.

Dobro,

"So what sound interface manufacturers need to do is not give us higher bit depth, they need to perfect the A/D
converstion process so that we get closer to the 24 bit maximum dynamic range."

If this were the case, would it then be much more difficult to clip when tracking?

To my knowledge, it would be just as easy to clip as before. The benefit is that when you have greater dynamic range at your disposal, you are able to record at a lower level (SAFEly lower than the clip point) and still preserve the beauty of the sound. This is because higher bit rates give higher dynamic ranges so that a 16 bit recorder recording sound with a maximum peak at about 80% of the recorder's meter will get about 80% of 16 bits.... ~13 bits of info (0.80 *16 = 12.8) or about 78 maximum dBs of sound (13 *6 = 78) . When a 24 bit recorder records that same sound at 80% of its scale, the result is.... ~19 bits of info(.80 * 24 = 19.2) or about 114 maximum dbs of sound (19 * 6 =114). Remember that CDs ultimately have 16 bits or 96 dBs of info on them.

So when you record at a higher bit rate/dynamic range, you are able to capture more information than a lower bit rate medium at a LOWER recording level. When you dither down to 16 bit you capture ALL of the first 16 bits of info leading to a cleaner and more accurate sound. Sample rate affects sound also, but that's another discussion.

Rev E

 

[Meshuggah]

The results from what I did very sloppy in Sound Forge can now be found at http://www.mp3.com/meshworks

It's a song called hit the bottom, and both the original version aswell as the "remastered" version can be found there.

 

[camn]

I think to "master" you have to send it out.. And at home you can only hope to "journeyman" your mix.

xoxo

 

[LI Slim]

I apologize for coming late to this thread, but...

As important and useful as much of the above is, mastering, either by design or inevitably, remains a mystical process. So let me try to approach it this way:

Let's assume that I want to, as Camn says, "journeyman" a song (let's put aside the parts of the mastering process involved in putting a bunch of songs together on a cd). (A popular as opposed to classical song.) Let's also assume that I'm going to do it at home using software plug ins, and I understand that I'm sacrificing the clear advantages of having a professional listen to it with fresh ears using great monitors and kazillion-dollar equipment. And let's assume that I'm committed to taking the advice that I'm about to ask for as a starting point, and try various permutations of it and engage in trial and error and listen really well and not expect miracles.

First, what sorts of compression settings should I start with to, for example, try to bring up the volume and power without squashing the sound. Second, what eq boosts/cuts might I start experimenting with to improve the mix, in order to, for example, add power without mud, clarity without tinniness, etc.?

Finally, if my mixing software already has all the features the the mastering software plug-ins have, why do I need aseparate mastering plug-in?

Master, your humble grasshopper asks that you speak to his listening. The sound of one hand clapping is the sound of one hand clapping.

 

[dobro]

Nah, I'm not the master, just another mix... uh, grasshopper. But I'll make some educated guesses. First, I just listened to your 'God is in the Internet' song. My guess is you could do a whole CD of songs like that, and need very little, if any, mastering. That kind of music is so simple - guitar plus voice (and maybe voice 2). If you mixed it right (level plus EQ plus reverb), it's probably mastered already. My guess is that good mastering is most needed for complex mixes.

Well, that's what I think, anyway. (dobro scans the horizon anxiously) - no sign of sharks yet.

 

[Rev E]

Originally posted by LI Slim First, what sorts of compression settings should I start with to, for example, try to bring up the volume and power without squashing the sound.

Second, what eq boosts/cuts might I start experimenting with to improve the mix, in order to, for example, add power without mud, clarity without tinniness, etc.?

if my mixing software already has all the features the the mastering software plug-ins have, why do I need a separate mastering plug-in?

LI Slim,
I'll say the same "irritating" thing that most everyone else says with these kind of questions... "it depends". Without actually hearing a collection of mixes to be "journeymanned", it's difficult to comment on exact settings. I will say this, the kind of compression that should be used shold be "practically invisible" (unless used for an effect-processors for an "effect" shold be mostly done in the mix stage). On the level of home recording equipment available, there are few hardware AND software products that can legitimately be used as true "mastering processors" (comparable to commercial mastering houses). So the take-home message is that your mastering processes should not "drastically" affect the mix. If it does, then the song(s) was/were not mixed correctly to begin with or you have bad technique.

The age that we live in will someday be looked back upon as the "Age of Compression". I think that everyone is too overexcited about compression. But aside from my rants, the amount of compression used in mastering is little (1:1.5 to 1:20 USUALLY with high to moderately high thresholds, slow attacks and song-dependent releases). Remember, mastering is a subtle art. But IT ALL DEPends on the MIX. To answer one of your questions, mastering processors are different than mixing processors, because they are designed to do whatever they do and do it "practically invisibly".

Rather than search for exact numbers, do this.... do your best mixes and then when mastering/journeymanning, do whatever it takes to make all of the songs (1) have a similar "average" level. Remember not peak level, the human ear is mostly sensitive to average levels (Compression), (2) crispy and clear (EQ), (3) punchy (compression/multiband compression), (4) have a wide and even stereo image (whatever mastering stereo image processor you have), (5) Not overly sibilant (de-esser/EQ), (6) operate below digital zero (limiter) (7) have smooth fades and cross fades, if needed (mastering editor), (8) converted to the 16 bit/44.1 kHz CD standard (dither for bit reduction and sample-rate conversion for different sample-rates), (9) have the "right" distance between each song (this is a song by song "artistic" decision), (10) have noise removed from silent sections and whaterever else it needs. As to the order of these processors, much of this is song-dependent, but make sure that your "high quality" mastering limiter is last in the chain so that whatever processes you do that may change the overall gain, the limiter will catch and stop "invisibly".

Better yet, sit in on a real mastering session so you can get a sense of where to start. You spent all of this money on gear, the least you should do is find out how people in the commercial music world actually use it.

Rev E

[Edited by Rev E on 01-19-2001 at 12:37]

 

[LI Slim]

Thanks, Rev.....

....that actually is somewhat helpful. One follow-up: when utilizing eq in the mastering process toward getting that "crisp" quality -- does it make sense to experiment with narrow bands, ie, generally narrower than one is working with at the mixing stage?