Parallel Processing

Blending processed and unprocessed sound is a classic and effective technique that can provide drastic improvements – and it can be done in every DAW!

What is it?

The difference between processing a sound and parallel processing is simple. Both start out with an original, unaltered sound and signal path. In a processed sound, there is no amount of the original signal left. The sound passes through the processing and is altered before continuing to the output; all you hear is the processed sound. Typically this is what is done during the mixing phase. We will add EQ, compression, saturation, etc. to a sound.

Parallel processing, on the other hand, leaves the original sound unaltered but adds an amount of processed sound alongside it. It’s the blend of these two elements that constitutes the end result. Adding a reverb or delay effect is not regarded as parallel processing. Parallel processing relates more to generating a whole new sound by means of compressing, equalizing, filtering, distorting, re-amping, and generally using and abusing non time-based audio processors.

Parallel processing is a non-destructive technique. The basic process is done this way: the original sound is on one channel. An auxiliary input track is created next to it, leaving us now with two tracks. On the 2nd track we add whatever effect we want to use, i.e. a distortion effect plugin. Using a bus send on the original track, send this to the new auxiliary track with distortion. You can add a lot of distortion if you want! The channel fader for the aux track will most likely not be at unity (0). While playback is engaged, starting with the fader down all the way at infinity, slowly bring up the fader on the aux track until the distortion is heard. Set the fader where it suits you. Doing things this way allows the original signal to go to the main output, and then the parallel distorted signal is also being sent to the main output, but only the amount we desire to have. Blend these two tracks to taste.

In the screenshots below there is a synth bass track that wasn’t coming through the mix very well. It has a lot of energy below 100 Hz. To help bring it out in the mix better I added some distortion using parallel processing. On the original bass track I added a bus send (bus 1), routing it to a mono aux. track which has the distortion plugin.

Notice that on the bus send I set it to pre-fader send and that the volume fader is set to unity (0). It just so happens that on the aux channel that has the distortion plugin, the channel fader is set quite high, -5 dB or so. Because of the type of preset I used on the distortion plugin I could get away with this strong of a mix level. Usually when doing parallel processing the channel fader is much lower. I did start with it all the way down, though, and brought it up slowly until it made a difference in the mix to my liking.

In the next screenshot I am using a saturation plugin. This is adding some harmonics which will emulate some analog equipment.

Again, because of the type of processing I am doing, some of the settings are a little different from “normal” use. I have never set the saturation to 1.0, but because I am using it in a parallel situation I can get away with that. Many times I will put a saturation plugin directly on a track. When I do it this way, the saturation is set no higher than .4. And again, notice that the aux channel level is set to unity. Again, because what I trying to achieve, this was acceptable.

The next screenshot is the same thing but this time I am using a compressor. This is, of course, known as parallel compression.

All settings for I/O routing are the same. Notice on the compressor I am achieving 6 dB of gain reduction, while also adding 6 dB of makeup gain. The 6 dB gain reduction is almost an arbitrary number. I knew since I was doing parallel processing I could afford to hit the compressor a bit harder, thus 6:1 ratio with 6 dB GR. Remember, I can always “dial in” the amount of the compressed signal I desire alongside the unprocessed signal.

While listening through earbuds, all 3 of these effects of parallel processing worked really quite well (saturation, harmonics, compression). The goal was to get the sub-bass synth bass to come through the mix better. This was definitely achieved with great results.

On a rap track I’m currently mixing I used parallel processing on the hook lead vocal. I set up a compressor on one aux. channel, and a doubler plugin on a 2nd aux. track. I then sent two different bus sends from the original vocal track to each of these two aux. tracks (pre-fader, unity send). Each aux. channel fader was then set appropriately. In this case, they were not at unity. The doubler track was set somewhere near -20 or -30 dB. The compressor track was also set close to the same.

Parallel processing is a great tool to use and one of many in any mixer’s toolbox. You are only limited by your imagination! I read about one of the top mixers who uses parallel processing even for EQ. He prefers not to EQ the original track, instead preferring to do it with a parallel track. Many times we don’t want to alter the original track to drastically, but still add an effect. This is a perfect scenario for parallel processing!

I hope you find this information helpful!

And …… HEY! Make it a great day!

Tim

Advertisement

EQ – Different Frequency Bands [20 Hz – 20 kHz]

Learning what different frequencies sound like and the effect they have on the sound of different instruments is an invaluable skill. These are the names we use to classify the bands – the frequencies are approximate, so use your ears!

> 20 – 60 Hz – Sub-Bass: Gives boom, depth, and richness – too much sounds flabby and out of control. Small speakers don’t reproduce this.

> 60 – 150 Hz – Bass: ‘Thump’ and punch in drums, especially kick and snare, and richness in bass and guitars. Too much sounds woolly.

> 150 – 1 kHz – Lower mid: Important for warmth, but too much sounds thick and congested. The 500 Hz – 1 kHz region especially is crucial for a natural vocal tone, but too much sounds boxy and nasal.

> 1 – 3 kHz – Upper mid: The most sensitive area of the ear, important for edge, clarity and bite, but too much will sound harsh and tinny.

> 3 – 8 kHz – Low Top: Provides fizz and sizzle; and edge and aggression in guitars – too much sounds thin and brittle.

> 8 – 12 kHz – Top: Gives openness, air and clarity – too much sounds over-bright and glassy.

> 12 – 18 kHz – Very high top: These frequencies can add sheen and sparkle and sweeten things up, but too much sounds unnatural, gritty and forced. [FYI – I have the Kush Clariphonic parallel EQ hardware. I add these frequencies on my mixbuss or sometimes use it for vocals. It really opens up that top end. A little goes a long way.]

Tip #1: Don’t solo an instrument when EQ’ing. Set the EQ when playing the instrument in context with the rest of the track. You can solo to quickly check things, but be sure to take out of solo mode fairly quick.

Tip #2: Sometimes when soloing a track or instrument, the EQ we add makes that instrument sound worse! But in context of the whole mix it sounds great. That is what matters. Part of the time you can expect this to happen.

Tip #3: If there are two parts that are fighting in the mix because they occupy the same frequency range, it can sometimes help to boost the EQ on one of them and cut the other at the same frequency, then reverse the strategy and boost the second sound in a different place while cutting the first. This emphasizes the contrast between the two parts, with gentler boosts, and helps stop things sounding unnatural.

Tip #4: In regard to Tip #3 above, this can sometimes be called ‘masking.’ Masking is when two instruments are fighting for the same frequency or frequency space. For example, kick and bass guitar. If when the kick hits, the bass is obscured some, this is masking. Using Tip #3 above will help get rid of this problem. Make sure to ‘cross-EQ’ both ways. In other words, boost instrument 1 and cut instrument 2 in same place. Then boost instrument 2 and cut instrument 1 in same place.

Tip #5: Do a ‘boost & sweep.’ When searching for a frequency that you want to get rid of, use a bell curve EQ (band, or parametric), boost @12 dB with a somewhat narrow bandwidth (high Q). Sweep up and down in frequency until you find (hear) the unwanted or annoying frequency. Then set that band for a cut instead of a boost. How much you cut depends on the specific situation, it might be a little or a lot.

As always – I hope this helps!

And ……. HEY! Make it a great day!

T

The Different EQ bands and What they mean (part 2)

Dyn3 7-band EQ (Avid Pro Tools free plugin)

If you look toward the bottom of the EQ pictured above, you will notice 5 different bands: 1. LF, low frequency, red; 2. LMF, low-mid frequency, orange; 3. MF, mid frequency, yellow; 4. HMF, high-mid frequency, green; 5. HF, high frequency, blue.

In today’s blog I will talk about these five bands. I want to start with band 1 and 5. These are typically used and referred to as “shelves.” Band 1, low frequencies, is the low shelf, and band 5, high frequencies, is the high shelf.

But these two bands each have two different settings. The small left icon, next to the LF and HF, is called a bell-type EQ. It kind of looks like -o-. This will either boost or cut a section of frequencies set by you with the frequency knob. The ‘Q’ knob will determine how wide or narrow the bell curve will be. A low Q setting will give you a wide band of frequencies, and a high Q will render a narrow band of frequencies. A good rule of thumb is wide when boosting and narrow when cutting.

The typical use for this is to, say, boost the lower frequencies to bring out a kick drum or synth bass. On the high end, with the HF knob, we can boost upper ‘air’ frequencies to make guitars or vocals stand out or sound brighter. Of course, we can also cut in these frequency ranges as well.

The other icon setting is called a ‘shelf.’ This is the more common use for these two bands. Typically we use a boost here (low or high). When boosted, it looks just like a “shelf.” If on the low shelf, we set the frequency knob to 125 Hz, then everything from 125 on down (to 20 Hz) is boosted the same amount. On the high shelf, we might add a shelf for vocals starting at 6 kHz. In this case everything from 6 k up will have a boost. Of course, we can also cut using a shelf, but this happens less often then a boost.

The Q factor is a bit more complicated and will have to be reserved for another post.

Bands 2, 3 and 4 allow for bell curve settings only. These are the same as the bell curves on bands 1 and 5. These are used for low-mid, mid, and high-mid frequencies. There are only three knobs: Frequency, Gain and Q. Frequency, of course, sets the frequency that you want to work with. Gain is volume (loudness) and can be plus (positive) or minus (negative). We might say boost 2 kHz 2 dB (2 dB) which is a positive gain. Or cut 1200 Hz 3 dB (-3 dB) which would be a negative gain.

As stated above, Q determines the amount of frequencies being altered by the EQ.

As always, I hope this helps!

And, HEY! Make it a Great day!

Tim

Compression Dos & Don’ts

To wrap things up regarding compressors, I will offer 3 Dos and Don’ts as my final word for now. These are things to always keep in mind when working with compressors. Some may have been previously stated in an earlier blog post.

DO          Avoid using extreme settings to begin with, if you are just trying to control the dynamics.

DON’T   Add compression to every channel by default. Start off with minimal compression, and carefully choose where to add compressors.

DO         Experiment with different types of compressors – hardware and software. There can be differences in how they sound. Compressors can and do sometimes sound different from one another.

DON’T  Forget to bypass the compressor occasionally while setting to check the results.

DO         Remember to balance the output gain so the level doesn’t change when engaged and bypassed. This way you can accurately compare before and after. Also, typically compression is added AFTER the mix has been balanced. So you don’t want to alter levels with either compression or EQ.

DON’T  Be afraid to experiment. Some of the greatest sounds in the history of recorded music came from misused and abused compressors.

Compressors 201 – Threshold

A compressor has a lot of knobs and settings. They can be confusing at first. In this blog I am going to talk about one of those knobs – Threshold.

A compressor is an automatic volume control. We try to make the signal somewhat the same, static. Without a compressor, we would have to do it manually, turning the signal down, then up, then down, etc. But with a compressor, it can do that job for us.

When a signal gets loud (and crosses the threshold), it turns it down. If there’s any makeup gain, it will turn the softer signals up making them louder (along with everything else, of course).

The threshold setting tells the compressor when to start working. Put a compressor on, say, a vocal track. Pay attention to the input signal on the compressor. Let’s say the input signal is -10 dB. Now set the threshold knob to -16 dB, and the Ratio 2:1. What we’re telling the compressor to do is this: Any signal that is stronger than -16 dB, I want it to compress (lower) the signal. When the signal crosses the threshold it will get cut (attenuated) in a 2 to 1 ratio. So in this scenario, the signal is coming in at -10, with the threshold set to -16 and a 2:1 ratio. This means 6 dB is going to get compressed in a 2:1 ratio. The signal will get cut down to 3 dB over. You can think 2 becomes 1, 4 becomes 2, 6 becomes 3, etc.

If the ratio was 10:1, the signal would get compressed more. The higher the number, the more the compression. If the signal crossed the threshold by 10 dB, then it would be reduced to 1 dB.

The threshold determines how much of the signal the compressor is going to affect. You can change where the threshold is set usually in two ways. Using the Dyn3 compressor/limiter, grab the orange arrow on the signal led and slide up or down. Or to the far right at the bottom, grab the threshold knob and set up or down.

TIP: If Pro Tools is your DAW, use the Dyn3 Compressor/Limiter (free). It helps to make understanding what and how a compressor works easier.

I hope this helps!

and HEY! Make it a great day!

Tim

Organize, Organize, Organize! (Sessions, Folders, Files)

Yes, more organization! 🙂

I think I have always been somewhat organized. But I have learned to be very organized since starting my studio. Things can get very unorganized, confusing and messy very quickly.

If you want to save yourself some headaches down the road, grab 20-30 mins to take some time when you’re not very busy and think about how you want to organize folders, files, sessions, and clients. 

Being able to find a client’s session quickly becomes key. For instance, I have one client who came in a few years ago with just one project. As the months and years have gone by she has done a dozen or so different types of projects. I made the mistake of thinking what she was bringing me was a one-time project, so I just threw it in with another of her sessions. I have now done that so many times that when she calls and asks if I have so and so I don’t know the answer. She is the exception for me. But her different projects are a MESS!!

So I had to develop a “system” of sorts to stay organized. Here’s what I do:

I put the clients’ last name first, then first name. So Tom Smith becomes Smith Tom. Bands are simply listed by the name of the band. I always capitalize the main client folder and the main folder for a song (i.e. Smith, Tom).  I usually put a sub-folder inside the client folder with the song title – and I do one folder per song. I do not put multiple songs in the same folder! If they have three songs, they get three folders. The folder for the song I put in caps, like “The Setting Sun”. Then the session in that folder becomes settingsun. I always do ‘Save As’ with the sessions as I get something done, incorporating a number scheme. So the session I put settingsun_01 xxx. With the xxx being what was done during that session, i.e. vocal tracking, eq, rough mix, etc. Then settingsun_02,etc.
Although I don’t always do it this way, it’s good to know that ‘year/month/day’ formatted dates sort alphanumerically; ‘day/month/year’ (UK) and ‘month/day/year’ (USA) standards do not. Think about it: under the UK system, the 1st of December sorts in front of the 2nd of January because 1 comes before 2. And in America Jan the 1st 2008 comes before the December the 1st 2007 because 01 comes before 12.

This works for session files too. Versions of the same day get suffixed a, b, c etc. Dates are more logical than descriptives like… ‘final mix’, ‘final final mix’, ‘final final mix THIS ONE’, new final mix, ‘new new final mix unmastered reverb +EQ’ etc., etc. I don’t use descriptives this way.

If you use the YYYYMMDD format, they’ll sort alphanumerically on a computer and you’ll be organized!

I hope this helps!
Peace- And remember – make it a great day!

Tim

Compressors 102 (More of the Basics)

After learning the basics about compressors (see Compressors 101 earlier blog entry), then you can use this general guide of the type of overall effect you are going for.

If you want a Natural sound (the compressor is not noticeable):

Use a slower attack (longer than 75 ms) and gentle ratios (less than 2:1). Always allow the compressor to “relax” back to zero several times a measure.
For a Punchy Response:

For a harder, punchier sound, use higher ratios and thresholds, but keep an ear out for any distortion.

If you want a Thick and Dense sound:

For a thicker, denser sound use faster attacks, medium ratios, and lower thresholds. There will be much more gain reduction though.

If you want a Pumping Effect (for EDM, for instance):

For an overstated pumping effect use fast attacks, high ratios, and a longer release time.



DO – Avoid using extreme settings to begin with. This is especially true if you are just trying to control the dynamics.

DON’T – Add compressors to every channel just because you think you’re supposed to! Start with minimal compression and carefully choose where, when and why to add a compressor.

DO – Experiment with different kinds of compressors. There can be some big differences!

DON’T – Don’t forget to bypass the compressor occasionally to check that you’re getting good results.

DO – Remember to balance the output gain so the level doesn’t change when you engage the bypass. In other words the before and after volume level should be the same. We hardly ever use compression without changing the output makeup gain. If you add 3 dB of gain reduction (GR), then you should be able to add 3 dB or so of make up gain for the output.

DON’T – Don’t be afraid to experiment. Some of the greatest sounds in the history of recorded music came from misused and abused compressors!

The next blog about compressors I will talk about the “Knee” of a compressor! I really do hope this helps. It helped me!

As always – Make it a GREAT day!

Tim

Creating a Click Track (in Pro Tools)

Using a click track during recording is, of course, imperative. We can’t do our work if we can’t play to the beat! Luckily for us Pro Tools makes it easy to set up a click track! Just simply go to Track on the menu bar, scroll all the way to the bottom and choose Create Click Track. Pro Tools will create an auxiliary track and automatically put a metronome plugin on the track. The metronome will automatically set to the tempo (bpm) of the song.

The metronome is customizable. You can change the sounds used for beat 1 and all other beats (2, 3, 4, etc.). The volume for beat 1 can be set and the volume for all other beats can be set to something different. I usually have mine set so that beat 1 is louder than other beats and is a different sound. That makes it easy to find the downbeat while tracking.


When I set up my templates, I already have the click track set up and ready to go! Easy! And since I organize and colorize my tracks, for me, the click track is all the way to the left in the Mix window and a bright lime green. I always know where it is in the session, no matter working in the mix or edit window.

Lastly, you can save a preset of the type of click you like. On the click plug-in, select the drop-down arrow next to Preset, select Save As, and name it! That’s it!

Peace! And HEY! Make it a great day!

Tim

Compressors 101 – the Basics (part 1)

Compressors seem to confuse a lot of people in the beginning, they certainly did me! Here is some helpful information concerning using a compressor in your mixing to help get you started. I will have other blogs on compression, so keep a look out!

1.  Decide what you want to achieve. There are really only 4 reasons for using a compressor – control a dynamic signal, add punch or impact, change the sound, create an unusual effect. Make a decision on what your goal is, which one of the four you would like to achieve. Keep listening with your final goal always in mind. Here is a neutral starting point: 2:1 ratio; 75 ms attack; 100 ms release.

2.  Overdo to begin with. Pull down the threshold until it starts working. It can be helpful to start with exaggeration. If you’re having to turn the threshold way down – boost input level instead. Exaggerating can help get settings right.

3.  Listen. Fine tune settings keeping end goal in mind. Once you get close, adjust the threshold.

4.  Listen again and balance different settings against one another. Higher ratios usually need higher thresholds. Lower ratios usually need lower thresholds.

5.  Experiment. Don’t be afraid to change a setting. Just keep listening! Radical amounts are common: 15-20 dB for electric guitars, room mics, drums and even vocals.

For a smoother sound – Use faster attack and higher ratio (But don’t lose energy & excitement)

To reduce ‘bounce’ – Use shorter release time & ease off threshold, or use a lower ratio. Bounce is when you hear the level ducking as the compressor kicks in and then springs back up when it releases.

To add punch – Use a higher ratio, slightly longer attack and shorter release times, but watch out for pumping. Pumping is where the end of the note is louder than the start. Also when adding punch, be careful not to introduce any distortion.

If you add stereo buss compression – be gentle – 1.5:1 and only 2 – 3 dB of gain reduction.

Don’t be afraid of using compressors. Experiment with them until you understand them. Try this experiment: print a bass track with heavy compression. Compare the original audio track with the compressed audio track. This will help you understand just exactly what the compressor is doing. You will see a visual representation of what your ears are telling you.

Compressors are a vital part of making music. We use them while tracking, mixing, and many times both tracking and mixing.

I hope this helps!

Peace – and as always – make it a GREAT day!

T

Calculating File Sizes (How much hard drive space does it take to record a song?)

So . . .  you want to record a song and you’re running out of space on the computer or external hard drive? Wondering if you have enough room? Here’s how to figure out if you do have enough space:

The sample rate and bit depth of the audio you record are directly related to the size of the resulting files. In fact, you can calculate file sizes using these two parameters:

— Sample Rate x Bit Depth = Bits per second

Or, stated another way:

— Sample Rate x Bit Depth x 60 = Bits per minute

In the binary world of computers, 8 bits make a byte; 1, 024 bytes make a kilobyte (KB); and 1,024 KB make a megabyte (MB). Therefore, this equation can be restated as follows:

— (Sample Rate x Bit Depth x 60) / (8 bits per byte x 1,024 bytes per kilobyte x 1, 024 kilobytes per —  megabyte) = Megabytes (MB) per Minute

Reducing terms gives us the following:

— Sample Rate x Bit Depth / 139, 810 = MB per Minute

A lot of folks are recording these days at 44.1/ 24. That’s a sample rate of 44,100 with a bit depth of 24 bits. Here is the calculation:

— 44,100 x 24 / 139,810 = 7.57 MB per minute.

Here is a basic chart of different sample rates and bit depths:

44.1/16 bit  =  5.04 MB/minute
44.1/24 bit  =  7.57 MB/minute
48/  16 bit   =  5.49 MB/minute
48/  24 bit   =  8.24 MB/minute
88.2/16 bit  = 10.09 MB/minute
88.2/24 bit  = 15.14 MB/minute
96/  16 bit   = 10.99 MB/minute
96/  24 bit   = 16.48 MB/minute

If you figure a normal song of 3 1/2 minutes recorded at 44.1 sample rate and 24 bit, you can plan on it taking roughly 26.50 MB of disk space. I am starting to run a lot of my sessions now at 96/24 bit. So a 3 1/2 minute song is costing me 57.68 MB of hard drive space per song.

Considering that terabyte hard drives are now running close to $50 these days, all this math stuff is not nearly as important as it was just a few years ago. But I know a lot of guys who still aren’t purchasing a whole lot of TB hard drives! It’s still useful information if it’s needed in a crunch!

Hope this helps!
HEY!! Make it a great day!!

T