RECORDING
Dajaun Martineau is a producer / engineer / writer based out of Toronto who spent the better part of a decade as a senior staff engineer at Phase One Studios before going freelance . For his full discography and more information , visit Dajaun . com .
By Dajaun Martineau
The Practicalities of Parallel Processing
Parallel processing in a mix has eluded many engineers for years . I am going to attempt to demystify what it is and how to use it effectively in a mix .
In its simplest form , a parallel process is anything that doesn ’ t directly affect your signal path . Any signal that you send out and return on a new channel can be considered parallel processing . The process itself can include compression , equalization , distortion , delays , reverbs , etc .
Reverb and delay returns are two of the most commonly used forms of parallel processing ; conversely , any process that isn ’ t run in parallel would be directly inserted and is considered to be running in series in the signal path .
Drums Drums are a great place to start with parallel processes . Engineers started bussing all of the drum tracks to a stereo set simply to have volume control over the entire drum kit without moving every fader . Later , someone had the brilliant idea to insert a compressor across the drum bus in order to get the level under control . All of this would be considered serial processing as the compressor is affecting all of the sounds heard from the drums .
As the tone of overly compressed drums became more popular , the amount of compression applied grew in tandem . Over time , the compressors became so slammed that the actual sound of the drums was getting lost . The original unprocessed signal was reintroduced into the mix and the idea for parallel drum compression was born .
To do so , the engineer can simply engage a second set of busses and return the unprocessed sounds to the mix bus , allowing them to blend between the processed and unprocessed sounds .
When deciding if your drum processing should be parallel or serial , you should stop and ask yourself the following question : Are you trying to control the sound or are you trying to creatively add tone ? If you want to create tone while maintaining the original integrity of the drums , then the process should be parallel ; if you ’ re trying to control the drums , then the process should be in series .
Vocals Vocals are another interesting place to use parallel processing . When you approach the vocals in the mix state , you should no longer be concerned about making the vocals sound “ correct ”; vocal compression , equalization , and microphone selection should have been carefully considered during the recording process .
Once you have the mix built up around the vocals , you may need to boost or cut with an EQ just to make sure that it fits in the mix . Personally , the levelling compression I would apply at this stage is more just to make sure I ’ m not losing any words . After you have the vocal sounding correct , you can start to consider tonal augments . You may choose to slam a vocal with a compressor using an “ All Buttons In ” method to add some harmonic distortion and overall saturation , but all of that would be done in series . There are two main examples of parallel processing in vocals that you should reach for based on what you are hearing .
If you were to find that you have a vocal where some words are just going missing in the mix but the client specifically states they don ’ t want a compressed sound for their vocal , then you can run a parallel channel with compression on the vocals . This should allow the quieter syllables to be brought up and emphasized while anything that is going to sound compressed can be kept quiet enough that the unprocessed vocal should overshadow it in the mix , thus giving the illusion of an uncompressed vocal sound while remaining perfectly level .
Another great trick is for a client who wants an ultra hyped vocal . Boosting high end on any EQ can get harsh very fast ; in fact , boosting frequencies in general is a dangerous thing to do . Running a parallel chain with the vocals ultra hyped and heavily compressed can allow you to add sheen to the vocal while still relying on the unprocessed vocal to maintain integrity .
Mix Bus When treating your overall mix , you want to consider what it is you are trying to accomplish before deciding whether your processing should be parallel or serial . If you are simply trying to level the mix and work towards getting the overall volume up , then you will want to run the process in series . While utilizing a series process , you must be cautious of making any drastic moves . Subtlety is key in serial processing on the mix bus ; however , if you ’ re looking to saturate your mix and get a really grungy sound or just add overall size , then you could run a parallel mix bus that would feed back into your main mix . On the parallel mix bus , you can make extreme moves and generate interesting tones that can be fed back into the main mix at an appropriate level .
A good rule to go by when considering parallel vs . serial processing is to consider what you want to accomplish . If your goal is to level an instrument through compression or equalize it to compensate for tonal inadequacies , then serial processing is probably what you are looking for . If you are looking to excite , distort , or manipulate a sound to create something new while retaining the original content , then parallel processing is a great option .
WWW . CANADIANMUSICIAN . COM CANADIAN MUSICIAN • 59