Friday, December 30, 2011

New Ears Eve

Alright folks - time to learn something about your ears and analog stuff.

In a previous blog post I talked about loudness and how our ears perceive loudness from primarily average levels rather than peak levels. It's pretty interesting how that happens. We hear a certain frequency balance, a certain punch, a certain width (hmm...all words we use when describing analog gear) and it sounds LOUDER to us. More ALIVE.

Part of this is because of a little thing called saturation. Sure, compression helps aide in the squashing of dynamic range in order to bring up RMS level. You compress something 4dB and then bring up the makeup gain 4dB, the "apparent loudness" of the track easily went up 4dB because your peaks stayed the same (thanks to the makeup gain) but your average went up. But what about those times we DON'T want to use a compressor? When it sounds too obvious? When it pumps? What then!??!

Analog desks have multiple benefits. One of which is saturation. Another of which is routing possibilities. Another still is analog summing. A Big trend these days is to use analog summing mixers like the Dangerous 2Bus. This product allows mixers to send out of their DAW into an analog box that sums using components, all analog, like an analog console does. However, these boxes are usually 16 channels. An analog console is usually much larger. 24, 48, 64, 96 input boards are common. This means that you have to do digital summing in the box to get the mix down to 8 stereo groups (i.e., 16 mono outs) anyway. On a real analog console, it's easily one track per fader.

So what makes analog summing better? I used to think it was some mythical thing about width and depth and blah blah. What I've learned (at least what I THINK is happening) is the subtle saturation characteristics from each channel adds what I call a "tightening and heightening" effect. It tightens up the low end thanks to its saturation (i.e., saturation from transformers / circuitry) and it sort of livens it up due to the harmonic content added by the saturation circuitry. It's subtle. You may not notice it on a track by track basis. However, when added up together, it's sort of like the Lombard effect - people in the pub talk louder and louder as they slowly can't hear themselves over each other, and soon enough the loudness of the room is extremely obvious. Little bits add up quick.

To prove my point, take a listen to these two drum clips.



Which one of these sounds more like the "analog characteristics" we talked about? More punch, apparent loudness, and width?

If you put them side by side in a DAW and play them, you will notice a few things. First, Clip A sounds a bit louder, wider, and has more depth. Listen to the room sound on clip A - it's more apparent! You hear more of the snare in the room. Now look at the meters - Clip A is 4dB QUIETER than clip B. On my meters, Clip A is -9.4 dBFS and Clip B is -6.0 dBFS.

Why does Clip A sound bigger, wider, and deeper, but it's over 3dB quieter?


Clip B was the original drum bus render. I ran it through some analog preamps and drove the transformers fairly hot. No EQ or compression, just natural saturation from analog preamps. This decreases the peaks, bringing up the harmonics and saturation, and allows for the natural compression (without pumping) to take place. This in turn yields a fuller, more balanced drum sound.

So, then after you do that, turn up Clip A 3.4dB to match the peak levels of clip B. Now guess which one is still louder? Haha...much more obvious that time.

Crest Factor is a term that means "the different between the highest peak level and the RMS level." So if your mix peaks at -1.0dB and your RMS level is -11dB, you have a crest factor of 10. Most modern standards suggest that anything more than 20dB is a super dynamic mix. Many modern rock mixes have crest factors of around 5-8dB. It is very likely that if you did a mix on an analog console and tracked to tape, your mix (DRY) would have a natural crest factor of maybe 10-12dB BEFORE compression, and a digital one may have 20+dB before compression. It is in this way that digital mixers often have to add more compression than analog mixers to compensate for this.

The solution: If you're wanting the Analog sound (like many of us are), use a healthy balance of saturation and compression. If you only use saturation, your mix will likely be out of control and probably gritty and overdriven. If you use only compression, your mix may have no edge, and instead pump and breathe in weird ways - it will sound smaller and wimpy rather than in your face.

Find a balance between saturation and compression and you will surely improve your mixes.

So the verdict on analog's depth? I don't think it's some magical "stereo widening" effect or some crazy "depth creator" even though we use the terms like "more width" and "more depth" from analog. Why is that though? I truly believe it's because of the aforementioned "tightening and heightening" effect of analog. Single tracks, entire mixes, it was obvious in the drum clip I send you above.

Another thing to think about - these days we are often using smaller rooms for recording (as many large studios have gone out) and then having to create ambience later. By running through analog gear post mix and driving it, we are essentially compressing it via saturation. When you compress something with ambience, the ambience is brought up. Note the drum example! Those were done in a big room with lots of compressed room mics. When you run it through analog, the room sound gets compressed too, making it much more obvious. This I believe is what makes a lot of "analog" mixes better. We used to record more room mics, more "real space" and now we don't, we use it on FX sends.

Another one of my solutions for this? Don't do aux sends for your room sounds. Plates, chambers, delays, sure --those have very often been aux sends. But for room sound - if you want the recording to sound REAL like it was REAL ambience recorded on those drum tracks, put the reverb ON THE TRACK. Come on, your computer is powerful enough to do that. That way when you compress them, it will react like it really would have in the analog days. Real room sound to a compressor. Sending a compressed drumkit to a room is nowhere close to compressing a big drum room stereo pair and blending it in. TRY IT.

In my opinion, the inserted reverb sounds more ALIVE. More REAL. I've been working on a book / DVD about recording and mixing for ITB and this is one of the biggest chapters in there - understanding ambience! It's not just "reverb," it's supposed to be PART of the sound. Sending the mics to an aux does very little for us. It doesn't make the room part of the recording. It makes it an effect. It doesn't make the sound quality of a space apart of what affects our EQ and compression decisions, which I 100% believe it SHOULD.

Doing it this way also allows you to be more tactile with mix. You need more verb? Turn up the ROOM MICS. Not "reach for an aux." It's more natural. Easier to remember, also.

Take these things into consideration, and have a safe new year.

Tuesday, December 27, 2011

Mixerman Quote

LOVE these tidbits from Mixerman.

"Don't feel obligated to use a part everywhere it's been recorded could be pure genius at a momentary break in the song, and pure crap everywhere else."

"So if you bring up a part and it does nothing good for the track, provides no worthwhile counter-melody, serves only to distract the listener's focus from what's important, or offers no valuable rhythmic support, it's a stranded part, one that should have long ago been removed from the session. Underdub it

"Since the vocal (or melody) is the most important part in a mix, you should make the preponderance of your tough decisions based on how a part affects the vocal"

"Remember, at all times, you're attempting to push the listener forward, manipulate their focus, and cause them to have an involuntary physical reaction to the music this will cause the listener to sing the song, which consequently will lead them to want to hear the track again."

from Zen and the Art of Mixing

Seriously, buy this book:

Saturday, December 24, 2011

Review of the URS Classic Console Strip Pro v2.0

click to enlarge I picked up this baby yesterday and I must say, Whoa.

This plugin is NUTS. The idea behind it is to model the EQ / "coloration" characteristics of all these different types of input stages, compressors, and EQs, and put them all into one plugin that is a channel strip of all three. Enter URS Console Strip Pro. Though it doesn't look all that fancy, it's a great sounding plugin that does a LOT for only $300.

Jeez. Where to start.

I heard about this plugin from two main producers. One was Charles Dye, who is the first engineer to have a number one record that was completely mixed in the box. He's sort of a plugin legend.

The Plugin

Alright so let's talk about the plugin. According to the URS website, this is what all the plugin plugin models in the input stages:
  • Three Class A American input transformers
  • Class A British input transformer
  • Class A German input transformer
  • Two Class A tube input stages
  • 15ips 2" tape electronics/head bump
  • 30ips 2" tape electronics/head bump
  • 30ips 1/2" tape electronics/head bump
  • Fifteen input transformer & tape combos
  • Five Console Input/Summing Buss stages
I really do love the input stage feature on this plugin. If you start pushing it it can sometimes give you the only EQ you need - I find this to be true a lot on bass. I will sometimes push one of the transformer input stages and it beefs up the low end really nicely. You can lock the input stage (which I often do) because when you change the compressor preset, it changes the input stage to match the appropriate type of compressor.

The EQ is laid out with 5 different models of EQ across 4 bands. The models are:

1951 - Pultec Style
1967 - API Style
1970 - Neve 1073 Style
1972 - Neve 1081 Style
1980 - SSL Style

The characteristics of these EQ curves really are noticeable when boosting or cutting a lot. If you set a boost in the lows or low mids to 4 or 6 dB and then switch through the different presets, you can hear the difference. It's subtle, but it's definitely there. The Pultec has a nice big bottom and smooth top, the Neves have a really tight low end and upper midrange boost, and the SSL has the classic edgy top. API doesn't really sound like the API to me, but it's still a good EQ.

You can arrange the EQ and compressor in a few ways. EQ first, Compressor first, Filters first, or all different combinations of those three. You can also bypass parts completely, or individual EQ bands only. You can link all the EQ "types" together using the Link button on the HF band. In addition to the EQ section you have the filters section for HPF and LPF, which I usually put pre compressor, and I love using them to tighten up the sound and remove any extraneous sounds I don't want to compress.

The compressor side is really what makes this thing cool. It has SIXTY presets for compression. Yes, SIXTY. This includes things like LA2, 1176, Distressor, API 525, Fairchild, VariMu, etc. I personally think that is too much, but it does allow you to get quick sounds. Apparently these are not only modeled after the EQ curves of the piece of gear, but when you open a preset it adjusts the ratio / attack / release / release type (manual or 7 different auto modes) to fit the sound of the compressor. In addition to these, you can adjust for the "faster attack" switch, adjust the knee, and the gain makeup. The one thing I dislike is that it is totally dependent on the types of models, and there is no ratio control, so you sort of have to know what you're wanting in order to get a sound without tweaking many knobs. However, some would argue this makes you use your ear more. I can see that.

Probably my favorite part about the compressor section is that in V2 of this plugin (which I have) there is a Wet/Dry mix control for the compressor. This allows you to do parallel compression in real time with no phase issues from DAWs that don't have automatic delay compensation.

Some side notes - I LOVE that this plugin has input and output meters separately, as well as a gain reduction meter. I hate having to select what I want to see on the meter, I want to see all three! MAN that makes it easy for gain staging. The plugin operates very smoothly with your mouse and keyboard. Holding control will give you large knob turns as you scroll your mouse wheel. Releasing it will give you normal precision, and holding shift will give you fine precision.

The saving grace of this plugin to me was the presets. Once I got it I was like "oh man, I have no idea where to even start." The first five presets in the plugin are modeled by year, which select all the EQs to the same year, and then select a compressor / input stage that would match that year. Then they also have tape and tube compression presets. This is a nice place to start. I don't ever get into the presets about specific sources (e.g., "Rock Kick" or "Drum Bus" presets) because they NEVER work for me. Every source is different so I just use my ear.

Overall, this plugin is a REALLY great deal for $300. It comes with a lot of amazing tools in it. I think it beats out the Waves SSL channel for me, particularly because of all its features. The compression takes a little bit to get used to, but after you find a couple of models you like, you will find it really addicting. I just wish it wasn't ugly as hell.

Review of the Sonalksis Stereo Tools

(click to enlarge)

Alright, so I'm new to this brand called Sonalksis, (I believe pronounced Son-aulk-sis), which is this smaller plugin company that makes a handful of really need plugins. I recently found out about them from reading an interview with a producer I like, and I decided to head over to their website. After many hours, I ended up getting the 14 day trials for almost all of their plugins, and I ended up purchasing the Sonalksis STEREO TOOLS plugin.

This is a very simple plugin for adjusting stereo width in sources, similar to the Waves S1 or one of my personal (and free) favorite plugins, the Omnisone stereo imager by jb. However, this has a couple features that really make it shine.

First of all, there is a phase invert switch on the top left of the plugin for the Left, Right, or Both. This can be useful when dealing with any stereo pair of mics as a quick check before and after you do stereo image processing. Sometimes making the stereo image narrower will make imaging problems more obvious. You also have Mid-Side input or standard left and right, as well as a "flip left and right" button. In the main part of the plugin we have Balance (like a pan pot) and then WIDTH. Our main focus! This width has a really cool feature about it in that you can select "Zero Width Below" and you basically have a high pass filter. This is AMAZING for putting this across the stereo bus, because you can select the filter down to maybe 80 or 100hz, and nothing below that will be shifted to the outsides, they will be shunted to mono. Now this is usually not a problem because the only things that are in that region are already mono! (kick, bass, etc).

However, this becomes really cool on things like acoustic guitars, which can start to sound really weird if you make them too wide. What I have done so far is put the HPF on to maybe 100-200hz, and just let the width stay at 100%. That way, I have my normal wide stereo image for acoustic, but my low end is solid in the middle.

There is also a stereo panning section that allows you to pan the left and right channels yourself rather than just having a single balance fader. In addition, there is a Mono button that simply allows you to keep it mono.

One of my favorite parts of this plugin is that is had a great Phase scope, Frequency Analyzer, and Lissajous meter. These three things make it really useful for gauging the size and shape of the mix when placed on the stereo buss (which honestly I find that I like a whole lot!) I don't always do stereo widening on the 2bus, but I find that digital doesn't always have the width and depth that analog has, so you have to be a lot more creative with your mixing to create

My only qualm about the plugin is that sometimes it seems to take away some low end from the source, and there is no readjustment for adding low end back, which I think should be on a new update.

Aside from the Stereo Tools plug, my favorites were the Sonalksis SV314 Compressor and the Sonalksis Creative Filter. Some of their plugins I found to be difficult to use, very hard to read, too much on the screen, like their two multiband dynamics plugins. I did really like their gate plugin, but there was no Floor adjustment, which bothers the heck out of me; I never gate to zero because it sounds too unnatural.

So this is a great plugin! If you need a solid stereo imager and want something useful, check out the Sonalksis StereoTools. If you don't have the money for it, get JB Omnisone.

Monday, December 19, 2011

Balance is a Virtue

Alright so what's up with the title of the post - let's talk about balance for a minute. This upcoming RECORDING LOUNGE PODCAST episode is about getting things to fit in a mix using reverb, compression, and EQ. Figured I'd share a story about this.

The time was last week, some late weeknight, and I'm working on a mix. At that time I had drums and bass grooving together, guitars starting to work, and the vocals in the verses. I kept looping the verse and working on that and it sounded awesome. Eventually I got to a point where I said, man I love this verse. Time to move on to the chorus. The chorus had a different selection of certain parts (different guitars, different pads) so I was trying to mix those all in with the already solid drums and bass.

The problem I began having was that the chorus just did - not - work. I kept blaming the parts. This guitar part isn't working, so I'd mute it and then the chorus would sound empty. Then I thought, okay maybe it's just too loud, so I turned it down. Then that didn't work. So then I started messing with the EQ. The part I was really having trouble with was this guitar part that was subtle but added a lot to the chorus live. When it was recorded, it just was not working.

I decided to take a different approach. I muted the guitar part and just listened to the chorus. Nice pretty highs, nice powerful lows, but the mids seemed to be a bit empty, and it was primarily in the 400-1k range, which we generally think of as boxiness. It wasn't that I had filtered out too much on other instruments, it was just like someone in the band wasn't playing that was supposed to be - that's literally how it sounded!

Now, this guitar part was sort of a chimey guitar that sounded amazing on its own, but in the mix just didn't work. What I decided to do was reamp the guitar part through a wah pedal and a dark tube amp and got this dark murky filtered sound with virtually no top end and very little bottom end. When put back in the mix, the chorus sounded amazing.

Why was that? Well, when I listened to the chorus without the guitar part and liked the balance but it seemed to miss something in the mids. The guitar part was chimey so lots of pretty highs and beautiful bell like high mids. But no body. Darkening it up with the wah did change the band's "intent" of that part, however, it made the chorus really gel together. It taught me something really great about balance.

Moral of the story: balance is not an individual track by track feat. It's not all about volumes (loud and soft) and pans (left to right), it's also about depth (front to back) and microdynamics (percussive performance vs. smooth performance) as well as tonality (EQ and part selection). When you listen to songs or sections of songs, try to listen to what is "missing" in the section's makeup. If nothing is missing, you are very likely to love that section. If something is missing, it may be some gap in the highs or in the lows, maybe the bass part changed or the guitar part is weird and now there is a gap in the low mids and mids.

Try not to listen to "guitars" and "bass" and "drums" but instead listen for the EQ CURVE of the mix. Listen to what "high frequencies" are coming through as well as what "low frequencies" are coming through. Try to think of it in that way artistically and you may find your mixes changing in a great way. If you ask yourself "what's going on in the mids" you could easily list almost every track, but if you start saying "what's going on in the 500-1k region" and if there's a deficit (or a surplus) then it might sound very off.

It's not about making a mix where every frequency is represented the same volume as every other frequency. This is where the age old quote "follow the song" comes in. Each section will offer a different dynamic balance, tonal balance, and spatial balance. The more you learn to recognize these things as an artistic means to an end, the better your mixes will sound because you will mix quicker, less analytically and more creatively. Also, you will probably find yourself using less processing and EQ!

Good luck. Listen for the next podcast tonight - search iTunes for RECORDING LOUNGE

Wednesday, December 14, 2011

Loudness and Average Level

An excerpt from a favorite book of mine:

"LOUDNESS [as a term] is used specifically and precisely for the listener's perception. Loudness is much more difficult to represent in a metering system, in fact it's best presented as a series of numbers rather than as one overall figure of "loudness." Two pieces of music that measure the same on a flat level meter can have drastically different loudness[es]. A true loudness meter makes a complex calculation using SPL, frequency content, and duration. Exposure time also affects our perception; after five minute rest, the music seems much louder, but then we get used to it again--another reason why it is wise to have an SPL meter around to keep us from damaging our ears.

Contrary to popular belief, the levels on a digital peak meter have (almost) nothing to do with loudness. Here is an illustration. Suppose you are doing a direct two-track recording and you've found the perfect mix. Leaving the faders alone, you let the musicians do a couple of takes. During take one, the performance reached -4dB on the meter, and in take two, it reached 0dB for a brief moment during a snare drum hit. Does that mean that take two is louder? No: because in general, the ear responds [most] to average levels, not peak levels when judging loudness. If you raise the master gain of take one by 4dB so that it too reaches 0.0dBFS peak, it will sound 4dB louder than take two, even through they both now measure the same on the peak meter.

An analog tape and digital recording of the same source peaked to full scale sound very different in terms of loudness. If we make an analog tape recording and a digital recording of the same music, and then dub the analog recording to digital, peaking at the same level as the digital recording, the analog dub will have about 6dB more intrinsic loudness than the all digital recording. Quite a difference! This is because peak to average ratio of an analog recording can be as much as 12-14dB compared with as much as 20dB for an uncompressed digital recording. analog tape's built in compressor is a means of getting recordings to sound louder. That's why pop producers who record digitally may have to compress or limit to compete with the loudness of their analog counterparts."

-Bob Katz, Mastering Audio: The Art and the Science