I know I already posted this song, but I remixed and remastered it to go along with Black and She, so I’m reposting it. Plus its really good. So there. The old link will no longer work. But this one will!
Something in The Way – Nirvana Cover
=Z
I’m not sure I should admit this, but Green Day’s Dookie is actually the first album I ever bought for myself in a record store with my own money. So here is cover number 2. This is a pretty drastic re-arrangement. Its all acoustic, its in 3/4, and I changed the melody of the chorus. All with the utmost respect and love, given the countless times I spun this record when I was 13.
=Z
So I’m doing 5 covers as a kind of back up to my LP that I released in Janurary. Five songs from five bands I listened to a lot in the 90s, that had a big influence on my songwriting in general, and my LP specifically. The first one I finished is Pearl Jam’s Black, so here it is.
=Z
I’m trying to learn how to hear it. I figured since I had some success visualizing it with the waveforms last time around I could work on hearing it. That way I wouldn’t have to bounce half a dozen .wavs during my mastering process. If I could hear the compression I could just put the plugins in the Master Out in the arrange window. Then its just one bounce to get from arrange to .mp3. So I tried it with one of my cover songs last night and the problem I had was the differences in speakers. Below you can see a comparison between the frequency response of my expensive studio headphones (the blue line) and a pair of stock Apple iPhone earbuds (the red line). See how the earbud line falls off below 300 and above 10,000? Also see all that boost Apple gives you between 1,100 and 10,000? Apple has targeted the frequencies most people care about and ignored the ones they don’t. The problem is when I bust my ass making minute changes in the EQ below 300 and above 10,000, the iPhone earbuds squash them out.
So compression is about decibels (loudness), not frequencies. But each instrument sits in a different range of frequencies. A baritone singing voice (like mine) has a fundamental range of 100 – 400Hz, but produces harmonics well above that. You can see on the graph that my studio headphones and the iPhone earbuds have the same response in that range, but Apple boosts the signal above 1,100hz, whereas my headphones drop it a little. Apple is trying to give you brightness, while my headphones are trying to produce a more natural sound. When I sing a loud part and those frequencies get a little louder, I don’t notice them on my studio headphones because the response is more natural. I hear a nice even mix with lots of room in it. When I switch to the earbuds, and those frequencies are amplified, the mix sounds uneven, the vocals sound tinny, and the whole thing sounds muddled.
So I guess the trick is I should mix on iPhone earbuds? Just kidding. I’ve tried that and it is a nightmare. What I really need to do is hit the dynamics hard. When a track sounds good on my studio headphones, I just need to hammer down on the compression a little bit more. Take instruments in the mid range and give them maybe 10% more compression than I think they need at first. And also ease up on the Master EQ, because I’m sweating over stuff most people can’t hear and don’t care about.
=Z
So there were four things I didn’t like about that last version. Three were mix related, one was mastering related.
1) The synth bass was too loud and panned too far left
2) The handclaps were too loud
3) The drums were not loud enough
4) The high frequencies were squashed because I got too aggressive with the Multipressor
So this time I really did go back to square 1. I went back to the original arrangement and tweaked the mix how I wanted it. But instead of waiting until I had a master .wav to address the compression issues, I took a look at what I had going on at this stage. Previously I had slapped a default compressor with no gain boost on every track in the mix. I realized this time around that wasn’t enough compression for the hand claps and the guitars. When I bounced all the tracks in place I could see that the compressor had done almost nothing to the waveform. So I just very gently started pulling back the threshold on those tracks. This made the mix sound incredibly tight, and brought the range of the whole song under control. You can see the difference between the two raw mixes below (screencaps rescaled from previous post).
Original Mix:
ReMix:
With the song under such better control at this stage, I didn’t have to mess around with the Multipressor. In fact I didn’t end up using it all (thankfully). The last time around I spent hours with the Multipressor pushing down on the high frequencies that were giving me overloads. What I got was something that sounded overly bassy and squashed up top. This time around I didn’t need the Multipressor at all because I had addressed the problem frequencies by putting compression on the problem instruments at the mixing stage. This is an important and very basic thing for me to remember going forward. If there is a problem, treat the source not the symptoms.
Next I used the Linear Phase EQ to get some clarity up top, and some punch down low. Then I added the Adaptive Limiter to punch up the gain and make sure we stayed at 0.0db. I took some time with the Silver Compressor to add a final layer of overall compression. I don’t completely understand this one, but with the Adaptive Limiter in the chain above it, it is a really nice tool for adding some final tightness and another little kick of volume. And then finally I put the Expander on to bring the quieter parts out a little more. Give both versions a listen and see if you can hear the difference.
Best of All Possible Worlds DEMO 2
Best of All Possible Worlds DEMO 3
Niv & Chris liked this latest version, and although I never think anything is perfect, I think this one is done.
=Z
The track I’m using as my testbed is called The Best of All Possible Worlds. That’s the first time I’ve typed that title out, and it seems kinda stupid now that I’m looking at it. Might have to rethink that one. Anyway, the reason I chose this one is because my “production team” (Niv & Chris) both thought it was the tightest and most complete song I’ve put together in the last couple months. Its easy to see why they think that, with three choruses and one verse its a pretty straightforward pop track. There are two synths, two guitars, a bass, three vocal tracks, and one drum track. If I can’t make this one sound good, I’ll be screwed with some of the more complex songs.
So below you should see a link to the first raw mix of the track, along with a screenshot of its waveform. This has compression on the drums and the bass, and that’s it. Its basically just my first attempt at the song, bounced so Niv & Chris could give me their opinion. When you look at the waveform you can see how quiet it is, and the range in the dynamics. When you listen to it you can hear how “wide” it sounds. Its kind of all over the place, and not very tight.
Best of All Possible Worlds DEMO
To punch it up from here I started by putting compressors on every instrument in the mix. Then I bounced each track in place to audio files so I could see the effect the plugins were having (I’m a visual learner). In the new arrange window I tweaked the mix a little bit and added more compression to the guitars using the canned “bright guitar compressor” setting in Logic’s Multipressor. In retrospect things were already getting kind of squashed at this point, but here is where I really went nuts. I bounced the entire song to a .wav and started playing with the Multipressor. I was trying to learn the settings but what I was really doing was screwing up the range of the song. After each iteration of the Multipressor I added a layer of Adaptive Limiter to punch it up. Unfortunately I don’t have every version of the song saved, because it was crazy to see the minute changes I was effecting with each layer of compression and limiter. I went through about 20 iterations of this process and ended up with something that was way too loud and very squashed.
So back to square 1. Or rather square 2. I scrapped what I had and went back to the .wav of the original mix. I did not change anything in the mix itself. I put one multipressor on it – targeting the guitars again, because they are the most troublesome part of this song, sonically – and bounced it to see what it would look like. From there I added the default Expander, Silver Compressor, and Adaptive Limiter, to get this:
Best of All Possible Words DEMO 2
Its loud, and it has presence. Altogether it sounds really good. But – there’s always a But, isn’t there – it IS still kind of squashed at the loud parts. And having it this loud showed me some weakness in the mix that I decided to go back and fix. So I’ll post the results of that process in a few days.
=Z
I know my first LP just came out, but I’m working on new material because that’s what I do. I compose and I write lyrics, and then I HAVE to hear what it sounds like – not just in my head or on the acoustic guitar but with a full arrangement. Imagine having an itch that you can’t get rid of unless you go through a specific series of steps to scratch it. That’s what this feels like. So its not out of disrespect for what I’ve done that I plod on mercilessly without regard for life or limb or family. Its just a dangerously pathological compulsion.
My work to date has lacked something I hear in commercial recordings that I will call “presence.” The mixes have been just fine, but altogether kind of loose. It may just be because I know each part by heart, but when I listen to my music I hear each part individually, not the sum whole. And they are not nearly loud enough. When I listen to professional releases it sounds like a coherent “one” thing. I can pick out the individual parts by listening closely, but they don’t stand out to me. I used to think having the mix right would fix this problem but that’s not enough. Its not the mix that provides the cohesion and presence I’m looking for (though having a good mix is obviously an essential step along the way), its the mastering process.
Each song is composed of unique parts, each composed of a unique arrangement of tones, each vibrating at its own frequency with its own harmonics. The range of frequencies that a human ear can hear is limited, but the number of ways those frequencies can combine is. . .beyond my simple level of comprehension. Simply put, that means that what works for one song likely WON’T work for another. Even if it has the exact same instrumentation. This is in stark contrast to the recording process itself, which I have been focusing on for the last four years. When you record an instrument you figure out what works for that instrument given your equipment and that’s that. That’s what has allowed me to do the tracking for 10 new songs in weeks instead of months. I know what kinds of sounds I want and how to get them out of the instrument and on to the hard drive. Now I have to move on and figure out how to present those sounds in the best possible light to the listener. And there is no one answer. The tools are all the same, but the settings will differ.
So these are the plugins I’m working with in Logic to create a uniform and loud presence: Silver Compressor, Expander, Multi-Band Compressor, Adaptive Limiter
I’ll post more over the next week or so detailing my process and giving some examples.
=Z
This is something I did a few weeks ago as an experiment while I was traveling on business. It was done entirely on my laptop by clicking in place the synth notes and drum hits (since I didn’t have my keyboard). I used the laptop mic for the vocals and layered on the effects to cover up the how terrible the signal is. Then I tested out some of the mastering techniques I’m experimenting with to punch it up a little. Turned out sounding pretty good.
As always the cover art is by Niv.
=Z
I’ve been wanting to cover this one for a while. I mean since I was 13. I tried to stay as true to the three guitar arrangement on Superunknown as possible, but that was pretty difficult since I’m not Kim Thayil. So at the end it drifts closer to the Chris Cornell acoustic arrangement. Having learned this song and recorded it in six hours though, there are definitely some parts missing. There are sections on the album where the third guitar drops off, but to keep the sound full I kept the third guitar playing the bass notes wherever I thought it was necessary.
As for production, I’m working on beefing up my post-processing to get a fuller, more “commercial” sound. I took it a little too far here, as you can hear some distortion from the limiters in a few places. If I had been smarter during the recording process that probably wouldn’t have happened though. I think those might be clips from the original recording that were just accentuated by the mastering process. Again I did this in six hours. Its a learning process.
Anyway, here it is. Big influence on me and my recent album. Just like the last two covers I posted.
Like Suicide – Soundgarden Cover
=Z
My lovely wife Stacie and I danced our first dance to this song, so I recorded a cover of it for her for Valentine’s day.
Someone Like You – Van Morrison Cover
=Z
