Monthly Archives: March 2022

AMA001: Why print sounds through plugins? Why not just stack up plugins? Why not run it all live into the mix?

No one directly asked me this question, but I came across it a while ago in an interesting blog article by Mauricio Ruiz after he attended a seminar: Craig Henighan on the sound design of Stranger Things

Here is a quote from that article, but it’s worth reading the whole thing.

“Another good insight Craig shared was about his workflow. He prefers not to “stack” plugins, one over the other, but “print” the effects, and then process the sound again, and print again, until he finds the sound he wants. He didn’t state why he prefers to work this way…”

 

 

This is also how I have always worked, and I know many other sound designers and sound editors do the same. I am happy to explain why I work this way, but as always these are just my reasons and they may or may not apply to your work.

First is about resource management.
Most established sound designers have taken many years to develop their skills and have also lived through many tech changes. While even free DAWs now ship with a huge collection of plugins, some of us are old enough to remember a time before plugins and for example when Waves first released Q10 and L1, it seemed revolutionary!
During the evolution of computers we have also had to learn to carefully manage resources, and even when using PCI card based DSP, resources were still limited and had to be managed, same as the voice count still has to be managed on large sessions. Making the best use of CPUs, RAM and storage were critical. While we now can take much for granted, the truth is that resources still have to be managed, and as a simple example I have been really enjoying exploring a new resonator plugin I got recently: Moodal by Tritik. But it places heavy demands on my current Mac, such that running three instances in a 24/96kHz design session makes it grind to a hault. Every plugin you stack is consuming resources, and if you haven’t yet bumped your head on that ceiling, then know it exists and that you will, sooner or later.
It is also worth saying, when you take your work to a dub stage for premixes and final mix, they will not want to know about what wacky plugins you used to create the work. They also very likely do not have them installed on their dub stage DAWs, as they have to be 100% reliable. They will expect properly presented audio tracks, which will play correctly every time without fail, and without being dependent on some version of some plugin.

 

Second is about version control.
Any sound you design will go through evolution. Maybe your first instinct and attempt is perfect and makes it into the final mix. But it may be version 27. Or the director might like the first half of version 1, but the ending of version 22.
It is an essential part of your job, to be able to move seamlessly between versions. But not just versions of your own sounds, also different picture cuts and if there are VFX, different versions of VFX where sync and content may change on a daily basis.

There is no issue with making complex plugin chains to process and create a sound. But can you manage that plugin chain such that you can revisit it exactly as it was when you output version 1? How about version 12? Will you save settings of each plugin and document the order, every time you output a new work-in-progress version for feedback and approval? That seems messy and unreliable.

Depending on the project it may be best to work with seperate small sessions for designing sounds and a master edit session for all your material in reels as you work towards the predub deadlines. Most sound editors work 24bit 48kHz, but maybe you want to create that monster vocal in a seperate 24 bit 96kHz session. So keeping that messy work, with plugins and many VFX versions seperate from your main edit session may be wise, so it does not clog up your main work.

By documenting versions (picture cut, VFX and your own naming system) you can safely work to create and output versions, transferring the results into your main edit session for final placement and context. Creating ‘fixed’ versions of work in progress is crucial to how films get made, for every department. It needs to be a part of your normal workflow.

 

Third is about creating new source material.
By printing versions you create new media to work with. So that example the director likes (the first half of version 1, with the ending of version 22) is easy to cut together and try, because you are not wrangling multiple plugins with their settings. You can simply grab the audio you printed from version 1 and from version 22, and do a quick edit and try it. Maybe its not perfect, but because you documented your work with versions, you can revisit the sessions that created version1 and version 22, and reprint them with the necessary tweaks (eg maybe version 1 now needs a clean tail, so version 22 can take over)

Every project you work on, you solve new creative problems and gain new skills. For example, I remember the first time I had to design sounds for a timelapse sequence in a film – The Locals in 2002 (which was also my first film collab with Dave Whitehead, fun times!!) The ideas I pursued and developed for those sequences become part of my own techniques, but so does all of the sounds I created for it. After I finish every project, I trawl through all of the development work, and create a seperate sound library of that material, tagged with the project name, so I can find it easily.

Lets take another example: you are working on a short film and right at the end a monster appears and destroys everything. The monster opens its mouth only once and lets out a huge roar. The work you do to create the final creature vocal may involve working with hundreds of ideas, lots of source material and new recordings. And maybe while doing that work, you create lots of new material which doesn’t get used in that one shot in the short film. Should you throw that material away? No way! It will be useful in the future, in ways you can not imagine. But what will trigger the idea to use that material will be your memory of that project.

Just as every sound designer should have their own field recording kit, every sound designer should be constantly building their own personal sound library as it is your most valuable resource. And those sounds that you designed, that only you have? They are invaluable!

 

Fourth, I will add to this as I think of other motives for working this way.
I am always interested to hear how others work, so if you have insight to share please do!

 

 

 

First I’ll describe this backwards, as though it was created from a concept or idea (it wasn’t)

1. Imagine assembling a drum kit made of metal.
2. Imagine designing an algorithm to play that metal drum kit.
3. Imagine recording 14 hours of multitrack generative metal beats.
4. Imagine sifting through 14 hours of recordings and finding the gems.

This is exactly what I have been doing for a while now, but the recording process has been limited to only two days due to the nature of the algorithm: this drum kit is played by raindrops!

Now this idea is not unprecedented. If you remember AMB018 STRANGE RAIN I recorded multitrack textures of rain, by placing microphones inside objects such as metal bowls, pipes, waterphones, gas tanks etc… In that example, the algorithm is direct rain and as the objects were fully exposed, even light rain created a continuous texture or ambience.

For this upcoming library, the algorithm is sparser and the objects were chosen for deeper percussive impact, as you can see and hear in the video below.  I’ll explain how I came to discover this idea, but before I do that I’d like to present a challenge:

RAINDROP BEATS – SOUND DESIGN CHALLENGE

Using a short multitrack loop from the upcoming RAINDROP BEATS Library, make something inspiring and send me a link. If it strikes me as ‘interesting’ I will reward your creativity with a free copy of the library.

Rules: There aren’t any. Do whatever you like. Write a symphony! Design a dystopian ambience! Make bleepy techno! Design an evil sludgecore machine! Make synthpop!  But… be aware if it has no relationship to the source material then it will be difficult to appreciate the ideas and skill involved, in the context of this challenge.

DOWNLOAD:
>RAINDROP BEATS SD CHALLENGE Source
(10 x 24bit 96kHz WAV files – 86.8MB download – 110.2MB decompressed)

This SD challenge is now finished – 20220402.
Fantastic work in every entry!
Results will be shared when the new library is released.
Date TBC

OK I lied, there is one rule: the deadline is 11.59pm March 31st (in your local timezone!)
Email a link (soundcloud, youtube etc) of your creation to: EDU at hissandaroar.com

Also please take a screenshot or photo of the session you created it with.
I’d love to share the winning projects and how you achieved what you did.

I did not find the idea for RAINDROP BEATS, it found me!

I discovered the idea for this library during the Xmas holidays, December 28^th to be exact. We had an unexpected Summer rainstorm sweep through, causing flooding and damage. But it also, by accident created an interesting sound. From my studio, I could hear this steady pounding drum beat. Were my neighbors having a party?? I went to investigate…

Turned out I had left a trashcan upside down in the porch, and by pure coincidence, it was directly underneath a crack in the clearlight roof. Somehow this drip was very steady, so I grabbed my recorder and put an MKH8050 inside the inverted trashcan, and also set up my MKH8040 pair to capture the general ambience.

Listening on headphones while recording, it really struck me how musical these sounds were. The steady pounding pulse from inside the trashcan provided a rhythmic reference, making the drips falling onto other junk and props feel like granular percussion. So I quickly set up all of the metal objects I had close at hand (bucket, milk can etc) and I left it recording.

After the rain ended, I loaded the 3.5 hours of raindrop recordings and listened to all of the variations. Immediately after a heavy shower passed the ambience would quieten down, but the pounding trashcan beat continued, presumably fed by the build-up of water on the roof. At this point I decided the idea was worth pursuing – I loved the bassy sound from inside of the trashcan.

A few comments on the drums. When I think of an old school trashcan, they tend to be very solid, but the one I had was a cheap import, made from fairly lightweight steel. I first bought it to capture IR sweeps inside, for the Metal IR Library. And it was the light gauge of the skin (base) of the trashcan that provided the initial resonance, as it was thin enough to act like a drum skin. But I discovered another useful aspect of this particular trashcan: it had a small lip or ridge around the outside, maybe 5mm high. Listening to some of the long takes, I noticed that the pitch of the trashcan would change over 30 minutes. As the water slowly pooled on its surface, the pitch would become deeper!

Next step: I wanted polyphony! I needed more trashcans! I headed off to the hardware store and after returning home spent the next hour arranging the three trashcans, and many more metal props. My metal drumkit and percussion ensemble were ready to play! Now I just needed some rain.

I waited. And waited. January passed with no rain – I was hand watering my vegetable garden every day. Six weeks later the weather forecast predicted major rain!

SEVERE WEATHER WARNING

HEAVY RAIN warning for Kāpiti-Horowhenua, Wellington 11:00 pm Friday 4/02 -11pm Saturday. Heavy rain may cause streams and rivers to rise rapidly, surface flood and slips also possible. Driving may be hazardous.

My batteries were charged, and at 5am I woke to the sound of rain, and quickly set up to record. This time I used three seperate trashcan mics MKH8020/8050/8020, plus the external stereo MKH8040 pair. I set my SD788T recorder rolling, had a quick listen and went back to bed. Every two hours I would swap batteries, and I  recorded throughout the day, until the rain cleared that evening.

From session two I recorded a total of 11.5 hours of multitrack raindrops!

At the time, the rhythmic interplay sounded wild on headphones, but listening in the studio the next day these raindrop beats seemed even more interesting. The three thuddy resonant trashcans were consistent and have a correlation in intensity, but musically their rhythms are decoupled. At times they lock into a beautiful pattern, but it may only last for a few bars before nature and gravity gently alters the pulse.

During very heavy showers the ‘dot would become a line’ as the raindrops would be so heavy that the pulse became a rumble. But after a heavy shower passed and everything settled down, the metal percussion started to sound like a mix of free jazz and gamelan.

Since then I have been playing around with the recordings, analysing them and tempo mapping the recordings, and from within these recordings I plan to output a collection of multitrack raindrop beats, and long loops. These can of course be mixed and used as ambiences, but there is another potential use that intrigues me.

These recordings are also a pattern library.
I have been playing around with using the asymmetrical patterns to trigger other sounds (via slice to MIDI, and via real time triggering) and I think there is much potential to use these patterns to generate ambient elements.

As a simple example, imagine designing a night ambience where each of the trashcan pulses is triggering a cricket chirp. While you ‘could’ spend time placing single chirps, or use a randomised algorithm, the gently random distribution of pulses from the raindrop beats could provide a naturally generated pulse pattern, which would be difficult to manually generate, and unlikely to take this form programmatically.

So an important part of this upcoming library and a motive for this creative challenge is this:

What unusual, inspiring ideas can you come up with? SURPRISE ME!

Some techniques I have been exploring:

– Using MELODYNE app (not plugin) to create a tempo map and apply it to other source material.

– Using resonators to sustain and tune individual hits (GRM resonator, MI Rings)

– Using ENVY to apply the envelope from RAINDROPS to other source material.

– Slice to MIDI and use that MIDI to ‘play’ other sounds and instruments.

– Using plugins such as a drum replacer or ATTACCO to trigger other sounds.

 

But these are just starting points… Have fun!
I look forward to hearing what you can come up with!

I have been working on a long term project of creating a BELL PIANO, and after a year I have collected up about 25 bells. Of course when you buy a bell via Trademe or Ebay, no one specifies the pitch of the bell. So it is a bit of a lottery, guessing based on size as there is no way to check until they arrive. And each time I get a bell I write on it what note it is, and leave it by my piano.
This is the current state of progress:

Earlier last week I won an auction for a little collection of bells that looked like it would be useful to fill in some of the range. I emailed the seller to make payment, and mentioned what I wanted the bells for. The sellers daughter replied, it was her 88 year old Dad who was selling them, due to moving house.
This is what I bid on, and won:

They sent me a tracking # and on the way to do my shopping a few days later a box was sitting beside my mailbox. When I picked it up I thought wow it feels heavy for those 12 bells, but didn’t think too much more about it. But when I got home again I opened up the box and there was bell, after, bell, after bell… All carefully wrapped in newspaper! By the time I had unpacked everything, this is was what it contained:

I counted up 70 bells! Literally a lifetime collection. I was astounded and emailed them to first say a huge thank you, and to check I was ok and this wasn’t a mistake. But I also offered to pay more, as the extra freight cost would have eaten into the purchase price for the 12 bells. They very kindly refused, and said they were happy the bells were going to good home, and for an interesting reason. While I appreciate this sentiment, I am also not one to take no for answer when I have the power to acknowledge such generosity, and since I had their bank details I made another payment and sent a quick email, thanking them but also mentioning how much I appreciated this and that I would like to insist on shouting them lunch when it’s safe to do so.

Since they arrived I have slowly been sorting them into pitch order, and it is fascinating to observe how their tonality varies due to both materials, shape, construction and the clanger. As per my existing collection, I was totally focused on brass and steel bells, so it was interesting to hear a range of porcelain bells which of course have an entirely different tonality, and remind me of some of the tiny fuurin temple bells I bought when in Japan.

So this is another to add to my ‘pitched instruments to deep sample’ collection. And now with the Sanken CUX100k mics I can start the capture process, knowing a 3 octave pitch shift is viable! In this specific case I think a BELL DRONE PAINO will be on the cards too, using some hi rez spectral freezes to achieve infinite sustain.

So thank you to Neil and Shirley Everett. Your bell collection shall be immortal!