I actually finish something!
After yet even more work, I am happy to say that I have finally finished the mastering on my CD.
It seems as if every stage is more difficult than the last, and only time and feeback, if I get any, will tell if I have done the job well. Certainly I have learned a lot of things along the way:
As you have read everyplace, it’s a great idea not to try and master too many things at one time. I wasted a lot of time on a couple of days thinking that I had the one preset that would work with every song. This might be true if you only wrote one style of music and used the exact same instrumentation for everything………nah, probably not even then.
In my case, my songs are all different from each other. Different synths, drums, basses and wind instruments plus my guitar – there was no way I could master them all with one mega-preset in Ozone.
As I mentioned earlier, I did take a sampling of the eq of a commercial CD and used the matching feature in Ozone to put a basic eq on each track.
From there everything was different for each song. I found that, as Rick Snoman in the Dance Music Manual says, you have to do mastering quickly so that your ears don’t get used to the sound. What I ended up doing was building on each Ozone preset the tweaking, tweaking, tweaking, mostly with the loudness and multiband dynamics and multiband harmonic exciter.
I made a project in Sonar and had tracks for each un-mastered song, and then a track below each song with the mastered version, once I had it.
In some cases, I had to make many, many changes and so I would 1) save the preset in Ozone and 2) dump different versions so that I could easily hear the difference.
This was a great help to me as the process moved along. Each day I would dump the complete CD to a rewritable disk, listen to it on my portable CD player and then on my headphones at work and my crappy computer speakers at work.
What I found was that while I could not hear significant differences in overall volume from track to track on my studio speakers, the much poorer quality of the other devices I listened on made the differences quite apparent. This got pretty frustrating at times, when I thought I had it all done and bam, found out that some of the songs just didn’t cut it compared with the others.
That is fine however, since it would have been stupid to spend so much time in the writing and mixing of these songs to give up right then.
In some cases I had to go back to the actual mix to add or usually subtract something using eq, or on some of the tracks cutting a part of the stereo mix wav boosting a portion of the file. In one song, I dropped out a synth part during part of a verse, and the entire song took on a better, clearer sound. Its difficult, but its always better to cut parts of a song than add them; it always seems to make everything much more focused, especially if you are like me and are melody mad. Restrain yourself!
One last test was to hear the CD on my regular stereo, and for the first time ever I was able to listen to my own stuff in my living room and not cringe at the results. That was the final, final test and it passed. How cool is that!
Now I can move onto the last things, which are getting the artwork complete and getting it pressed, and then selling a million copies. Well, probably not that last part. I would be quite happy just to sell enough to make back my costs, which are not great, but if I can then I can look forward to making my next CD, and starting to work with my friend Jeff on our new project.
Another benefit of this is that now I can start learning some of the new tools, like Project 5 and Reaktor 5, that I have had to put off until I finish the CD. Exciting stuff!
And last but not least, to perhaps finally get these finished songs dumped into my XL7, and start practicing to perform. Isn’t that what I have been trying to do, all these years sitting in the studio, trying to get back on stage? YEAH!
I haven’t had much chance to work on the masters, but what I have done has been productive.
What I am realizing is that mastering itself can highlight many problems on the basic mix, that I wasn’t aware of. Its interesting to me how differently I think of these songs when they are just stereo wav files, as opposed to the project files will all the tracks. Maybe its partially visual: I am not looking at all the sliders and virtual mixer, etc etc in Sonar, but just this little wave form.
In any event, I hear things differently than when inside the project file.
What this has lead to is having to go back and adjust things within the project. For instance, in one my songs there is a lot of midrange instruments, and it can make things muddy. I realized the ‘mud’ was coming from this marimba/vibraphone patch that I was using, that is quite integral to the sound of the piece.
I thought of eq’ing it, but that didn’t work, since there was a certain amount of fuzz that is part of the sound. My next thought was to simply replace the sound with more a more traditional vibraphone sample. I called up Kontakt and grabbed a vibe patch from a sample CD. It was good, but not quite was I was looking for, although it is always difficult when you are not working in the context of the material.
I dropped into the song, keying it off the original midi notes, and muting the original instrument. That was better, in that it cut off the muddiness, but it lost some of the character I was looking for. Just for the heck of it, I unmated the original part and – bam! – the sound I was looking for.
I did have to change the volume on both parts, so that the fuzzy original sound was down in the mix, and I panned both parts together so there would be no problem with taking too much of the mix up. It really sounds good now, makes all the difference in the world.
On other songs, using advice from books I have, I have been carving out specific frequency ranges on certain instruments. Also changing the pan has a great positive effect in some cases.
Another cause of problems has been the relative volume and sound between the different pieces of music for the CD. One of my kinks is that I don’t use the same instrumentation, or at least not much of it, in the different songs. This makes mixing and mastering much more difficult, since one song may have screaming guitars and the next just synth sounds.
As an example, the song that I think will start the CD is without question the loudest and busiest arrangement of the bunch: vocals, several guitars, lots of drums and tons of synths. Oh yeah, and background vocals too.
I have working like crazy on this one. It has a pulsing synth bassline that runs thru the whole thing, as well as a second bassline on parts of it too. Sure, it sounds big and full, but again, it may just be too big for its own good.
A couple of things moved me in the right direction: I did some eq carving on the synth bassline. I also reduced, several times, the volume of this repeating bass. What is helping more is reducing the gain even more where the bass is running under the vocals, and competing for with space the rhythm guitar this helps. Still, I think it may require even more cutting: one of the hardest things to do is get rid of a part of a composed piece of music. I don’t think I will bag the bassline totally, but may cut it out when the guitar starts. There actually can be too much of a good thing. I know for a certainty that the song I plan to have on the CD after it has much more clarity, due to the reduced number of parts on the song, and a different structure.
And so it goes: one idea I have been working with: I grabbed the eq for a commercially released CD, and matched it in Ozone, sort of as a starting point for mixing. I have then been building up on this for my first preset.
After that, I have based my presets around this eq and subsequent additions to Ozone, like the loudness maximizer, harmonic enhancer, and all the rest. I know this may not be the best way to do things, but its not like all the presets will be the same at all, just a bit of commonality to begin with. This isn’t nearly done yet, and all I have been checking the mixes on are the junk speakers at work, or even worse headphones, so I will need to do the car stereo thing at some point.
Honestly, its still very confusing and most difficult to try and get these songs sounding like they are at least somewhat as good as a commercial CD, but hey, I don’t have a choice!
I happpily burned a CD with all the tracks I mastered and brought it to work to listen to it on a truly crappy pair of headphones and multimedia speakers – and I said “Yeech!”
After doing a lot of head scratching, I realized some things.
For instance, when I was mixing all the tracks, the one I didn’t touch at all was the song called 120 Ways, which I mixed and mixed and mixed to death, as chronicled here in so much detail. When I played it here, it sounded like mud, and my vocals were way off the distance, way too much reverb……….
It took me some time to figure out why. I was using a slightly modified Ozone preset, and the damn thing has reverb as part of the setup!!!!! Why, oh why didn’t I realize this before? That’s the last thing I want on an entire mix, since I do that all on the individual tracks.
As I continued to listen to the album, things weren’t all this bad, and I started making notes to help me remember what was good and bad. I can’t stress this enough (and I am saying it here so I remember): keep detailed notes on the mixes, what you have done and what you hear. No one can keep this all straight throughout the course of an album, and that’s what I believe you really need to do: listen to the whole thing, many times, to get the feel for the whole record, including the song order. I got pretty close on the song order first time round, but not quite there.
Going back to 120 ways, last night I worked on it some more, which, trust me, after all the work I have already done on it, I was loathe to do. I am not a perfectionist, but this just needs help, and I guess that is what I get for working on mixing so much, at least hopefully, my ears and skills have improved.
One thing I did that is still not right, was reducing the levels on the tracks. I know this from examining the wav file after I dumped it from the mix. Its very strange, the level is just too hot, even though it never peaks the meters in the mix. I mean, its even hotter than a commercially made song, and so that by default means its gonna distort.
When I listened to it at work, at least part of the problem finally dawned on me, or least I think so. This song features some jacked up vibraphone like sounds, and I have applied a ton of effects on them. They sound just great by themselves, but when combined with all the other tracks, the sound gets all muddy. So the next step at home is to try and keep the tones I like but do apply some EQ to keep them from stepping on everything else. Certainly, when the vibes drop out the whole tonal spectrum gets clearer, so I really have to do something with them.
I already worked on the vocals, and changed the processing quite a bit. I have a low voice and very midrangey, so I played a lot with HF exciters and other EQ enhancements to see if I could overcome the difficulties. As I have read, the best practice is to cut rather than boost frequencies, since as we all know, louder always sounds better.
I had just enough time to look at and dump a couple of more songs. I went thru some brain spasm the other night with my track routings in Sonar – let me just say one thing: ALWAYS route the outputs of the tracks to the sound card, not to one of the busses, even if you using the busses, cause the sound, at least on my system, just goes down the drain otherwise.
Actually, I never even thought about this before last weekend, when I was looking at the stereo wav outputs of some tracks. I saw on a couple of tracks that one side would be noticeably louder than the other, I mean from looking at it. The audio difference wasn’t that much, but it bothered me. I spent some, lets say wasted some time dumping split mono tracks (bye-bye my cool panning!) and doing various other tricks to get it right, but no dice. Me gonna leave it like that, I am the only person in the world who would know if the tracks are exactly the same volume on each side.
Of the other tracks I dumped, there is at least one that is really, really close to being a good mix, so I am going to look at this one hard and see what the difference might be.
I should also mention that I am listening to the un-mastered tracks here, unlike last week when I just dumped them and started mastering. MISTAKE! The tip that I read in Computer Music is correct: listen to all the tracks un-mastered first, since you can’t fix it in the mix, eh? The things you learn……
In short, I have more work to do, but I feel I am getting closer. I am rereading the sections in the Dance Music Manual about mixing and mastering – the principles of mixing are the same for everything, so its good stuff.
Also – I must send a big shout-out to my man Jay at Lost Studios NYC for the endless tech advice and encouragement – really, HE ought to be writing this mixing stuff, since he actually knows how to do it!
and back on track
I just spent the better part of the last two days dumping my mixes to wav files and mastering the mixes – I can't believe how much work this last bit has been!
I suppose it was a perfect bit of timing, being away for two weeks made sure my ears were fresh and that I wasn't mentally fatigued either.
The day I got back from my trip, I spent an hour or so just dumping the mixes to stereo wave files, and trying not to pass out from being exhausted. Needless to say, this didn't work out.
When I checked the files, many of them were too hot, and while they didn't distort, I didn't like that at all so I had to do them again.
And again. On a couple of tunes I did some semi major shifting of the mixes, and even doubling some parts so that they would sit better.
My motto: when in doubt, reduce the volume!
Strangely enough, some of the stereo outs from the mixes weren't balanced very well from side to side. Weird: I have never encountered this before, and while I fixed it as well as possible, I never did find the cause of this imbalance.
Today I mostly spent my time mastering the songs. I used the iZotope Ozone 3 – this is a fine mastering plugin, capable of way more than what I am using for. I will admit that the many presets are pretty much all that I need to get the sound I am after, although of course I have to use several depending on the material. As always I am mixing on my trusted piece of junk monitoring system: two Fisher speakers with the added benefit of Radio Shack midrange replacements since when I was given these speakers, those were blown. Oh yeah, and my pawn shop special $35 amplifier. I have about $75 total in my monitoring system.
Yes I know its junk but its what I have and if I can make it sound good on here, I can make it sound good anywhere.
Here is a promise: if I start making money on this CD, first thing I get is good monitors!!!!!!!!!! I will certainly deserve them at that point.
I have given some thought to the order of the songs on the CD. I am working under the old adage, put your best foot forward. In my case, this is going to be the most up tempo of my songs, Passion. Its also the busiest, with lots of guitars in addition to all the synths. Is it commercial? Beats me, but I hope it is.
I am not going to spend weeks and weeks deciding on the order of the songs, I just don't have the temperament or the time for that, and in the end, will I even know what difference it will make? I guess I am going to follow the model that I have found on many records, put the most lively stuff first, because that will make you want to play the record again. Case in point, put on Matisyahus' debut CD: the most catchy song is first, and by the time it ends its kinda mellows out.
But I always want to hear that first song, ya know? Of course I think all my songs are the best things you will ever hear, right? But I will judge them anyway. The one other minor point is that I am not going to put all the say songs with a flute together, I want to mix the songs with the live instruments all over the record.
Couple of other things, way out of chronological order: Friday I took some time and went to a couple of CD manufacturing firms, and have decided to go with CD Forge: they were recommended by CD Baby, where I am going to publish my CD, they have a good deal for really short runs, the people there were cool and easy to deal with, and all I have to do is show up with artwork and a CD and that's it.
I am also working on getting the CD art together, or actually having someone do it for me, and will comment on that as it happens.
Damn, this is a lot of work! And all I want to do is just write more songs…….