"Soundscape" - Another great book with plenty of inspiration!

 "Soundscape: The School of Sound Lectures 1998-2001" is, similarly to "Film Sound", a collection of reflections and experiences of sound practitioners and academics. There are few really interesting chapters in the book that gave some well needed inspiration boost for my Dissertation Project.

 In the chapter entitled "Touch of Silence" Walter Murch, the pioneer sound designer, talks in detail about creation of sound design for a cult film "Apocalypse Now". The 5.1 surround standard was just been invented and the film helped shape and popularise it greatly. Murch takes about a few interesting concepts, like 'worldizing'. This is a concept that Murch helped to develop during his career. The idea behind is to take the sound out of the studio and into the field, and rerecord it capturing the reverberant sound. This technique has been used to great advantage in "Apocalypse Now", for example by recording Wagner's 'Ride of Valyries' through megaphones attach to military helicopters. This technique produces a very realistic sound of the space is trying to capture and has been used a lot since it was created. With the advent of digital technology and high-quality realistic convolution reverb plug-ins it would be really interesting to see how the results between 'worldizing' a sound and using a plug-in compare with each other.

Sick Note and Assignments Extension

Unfortunatelly, I have fallen ill. I have been to the doctors and managed to get an extension fo my assignments. I will need to take a good few days off rest.

"Film Sound" and why it might possibly be the best book I read this semester!

 "Film Sound: Theory and Practice" is a collection of editorials, articles and essays that broadly talk about different concepts around film sound design. The editors of the book complain that film sound design is vastly under researched subject and claim that simply not enough have been written about the subject, especially compared to film msuci scoring. This general consensus seems to repeat itself in other publications on film sound as well. This makes a very interesting point, because it provides me a very strong rationale on why to do a film sound specific Dissertation Project!

 There are many great texts in this book, but possibly the next biggest breakthrough came after reading a chapter entitled "Fundamental Aesthetics of Sound in the Cinema". This is the exact idea that I want my Dissertation to be! In this chapter David Bordwell and Kristin Thompson present us with some very engaging writing. They claim that sound can actively shape the way we interpret images in film. They give an example of an obscure film 'Letter from Siberia' from 1957. The director Chris Maker demonstrates how powerful sound can be and he it affects the way we interpret scenes in the movie. In the film, we see a sequence of scene that repeats three times, each time with a different sound accompanying the picture. A full detailed analysis per-scene is presented in the book and provides a very interesting concept to explore further in the Dissertation.

 Furthermore, in ths chapter, there are presented criteria according to which we can analyse sound in films. Authors make a classification of properties of sound, like Acoustic (loudness, pitch & timbre), and also Rhytm, Fidelity and Space. They also talk about sound diegesis, which is explored in an even more detail in "Audio-vision" by Chion.

All in all, I am very happy that I started reading this book. This is the first time all semester that I really feel like I am making some progress!

Final Reply from Kenny. Back full circle?

 So, I received final feedback from Kenny yesterday and after a discussion this is one of the last emails:

"Why not just study the principles and aesthetics of cinematic sound design and apply them to film?" - Wow, we are back full circle here!

 Yes. It seems that first ideas that I had about film sound design being the best topic for the dissertation for me were indeed right. The shame is that it took me almost 3 months of work only to arrive at the zero point, the point where I was at the start of semester in September. I am really running out of time, and have to think quick. I decided to start reading "Film Sound" to hopefully get those ideas going. 

Feedback on Proposal Draft Ideas

 So, I got a reply from Kenneth yesterday with a very detailed feedback to the Proposal Draft that I sent in. I present his comments, below with a few lines of commentary from me. Overall, he didn't liked my idea that much. Read on to find more.

Your initial aim (...) looks okay, although I would still like to understand more about the context and the research problem that you’re trying to address (why investigate sound design principles like this?). You go on to give an expanded version: Create a VST like software (or plug-in) to generate customised ‘bigger-than-life’ Hollywood style punches and hits out of a library of pre-recorded and synthesised sounds and based on a set of user defined parameters, to demonstrate sound design techniques and principles in the title. This isn’t really a project aim, this is really a task, I think - it’s about creating a VST plug-in. One question, though - what sort of experience do you have in software development? Could you realistically expect to create a functional VST plug-in by yourself? If the answer to that is no, then why give yourself the added complication?

Take a step back and think about what your project, as expressed above, is actually about - it’s about the principles of good (sound) design. By specifying that you’re aiming to create ‘larger-than-life’ film sounds, what you’re really doing is giving yourself specific  design goal and a framework for assessing your work. What you’re also doing is specifying the context in which those sounds are going to apply - in film. With that in mind, I think you have to take on board the idea that you’ll need to demonstrate/evaluate those sounds in a filmic context.

So am I right in thinking that what you’re planning on investigating is the process of how you design and create sound effects? If so, think about how you evaluate and compare processes. Part of that will be the end result. What else is important? How long the process takes? How much effort is involved? Whether there’s a pattern to the work, or if it’s different every time? It would be an idea to try and find some supporting literature on design in general and see if you can apply some of the thinking to your own project.

Or, are you planning on looking at how you automate adaptive sound effects (which seems to be what you’re suggesting in your rationale)? If that’s the case, it’s not so much about the process of sound design at all, it’s about using/developing sound tools that will take a raw sound as input and manipulate it according to a set of parameters. This is really a project about interactive and adaptive sound, and it would be worth, then, having a look at some of the existing tools that are out there (Wise, FMOD, UDK etc), and then development environments like Pure Data or Max to see if you could develop the sort of thing that you’re after. If this is the path you’re keen to go down, I think you’ll need to adapt your aim to reflect this. Something like:

Many forms of new and interactive media use sound effects as an important method of feedback. Because of the demands of interactivity it is common to use procedural and other automated methods to adapt a small library of sounds in real time for different situations. This project aims to explore the use of real-time adaptive sound as a mechanism for feedback by (for example) creating a series of real-time adaptive patches in Pure Data (or working with UDK, FMOD to create and implement a real-time adaptive sound engine). This will be evaluated by…

That aim has a sentence or two of context, which provides the rationale, a statement of the project aim (to investigate how sound can be adapted in real-time to provide feedback), and states how you’re going to do it (create your own software patch or use an existing engine).

In your expected outcome section, you raise exactly the query that I would have - if, in something like film, you have a need to design specific sounds for specific situations, why would you choose to use a tool like the one you’re describing, which takes the professional control from the sound designer and gives it to a piece of software instead? On the other hand, if you’re planning to use it, as detailed above, in something like a game, why would you redesign a system when there are already tools out there that you can use?

Also in your contingency section, you say exactly what my advice would be: either make this a project about the design principles that underpin good sound design (which would be a perfectly fine project) or make it a project that explores how you can apply adaptive methods to manipulate sounds in real time for interactive feedback. Don’t try and do both.

 Kenny has raised some few interesting issues. He asked can I realistically code a plugin? Well, at the moment I can't, but I was going to use visual plug-in creator tools like MaxMSP anyway. Kenny then went on to suggest that I possibly use FMOD for this task. I'm confused now, when I had a meeting with James he said I shouldn't do a FMOD project... I like the idea that Kenny came up with and the project aim he proposed by my main worry is that all of the research that I done on film sound design will go to waste now. I have sent him an email expressing my worries and I am waiting for a reply.

Proposal Draft sent for Feedback

 So, I spent most of weekend developing my latest idea. I present the results below. This comes from a document that I sent to the tutor for feedback.

This is the main idea that I came up with, which would serve as the topic (title?) of the proposal:

The use of sound design techniques and principles (like layering) and recording techniques in creating bigger than life film soundtracks.

I very well understand that this is a very broad topic which could cover everything relating to film sound. However, I would mainly want to focus on sound effects here, but I want to leave myself a sort of open door here in case something goes wrong during my research, so then I wouldn’t have to change the whole topic. (But more on that later...) (Also, I could substitute the term ‘bigger than life’ with just ‘the film sound’)
 

 Therefore, I would introduce a sub-heading here, which could also serve as the aim, something along the lines of:

Create a VST like software (or plug-in) to generate customised ‘bigger-than-life’ Hollywood style punches and hits out of a library of pre-recorded and synthesised sounds and based on a set of user defined parameters, to demonstrate sound design techniques and principles in the title.

Going back to what I said earlier about not wanting to just simply re-score or re-design clips from movies, my idea here is to take it all a step further. Instead of doing what’s been done a million times over, I came up with a concept of creating a VST plug-in, in which I could dynamically generate sound effects in real-time, and then use this plug-in to re-score scenes from movies in MIDI.

Basically, I would focus here on punches (maybe gunshots even), but my concept is as much about the software itself, but also about the sound design principles and recording techniques behind it as well.

Now, I am almost 100% sure something like this doesn’t exist yet. The closest thing is the Kontakt Foley Collection which you can see here: http://www.foleycollection.com/sample-page/ but that is mainly footsteps. As I said, I want to take it that one small step (excuse the pun) further.

This is where things start to look interesting to me, because, as I said earlier, with previous project ideas I was REALLY struggling to come up with my aims and objectives, and mostly the rationale. With this new idea however I can see the project clearly, from conception to the end. Also, I think it helps that this is a more practice-focused project, and as much as I love sound design theory, I think I’d rather focus on doing, or rather creating something. I am going to take a bit of a side-stepping route describing this in detail, so please excuse me if it seems backwards, but it’s proving the rationale which has held me back from all of my previous projects, so let’s start with it:

PROJECT RATIONALE:

I want to create a simple to use plug-in/app/software for beginning sound designers, film makers, animators, video-casters... anyone who does any sort of video projects. Instead of relying on stock sounds (and most libraries don’t have many variations on each sound either) I want to develop a tool which will create on-the-fly professional-sounding effects. Every punch or hit would be different from the other because of randomised sample selection and could range in scale from small hits to big cinematic knock-outs. The plug-in could also be used in low budget film productions and or video games, where there are budget limitations in hiring a professional foley stage and again, to not to rely on stock sounds.

EXPECTED OUTCOME:

Now, I do understand that there are reasons why software like this doesn’t exist. It could simply be a matter of that it just wouldn’t work! Especially in linear visual medium, where every effect like a punch or  hit is hand-crafted to match exactly what is needed in that shot. However, I believe that the principles that I will apply in creating this plug-in could easily be translated into the world of video games for example, and therefore I will have...

CONTINGENCY PLAN:

That is why I would have the main topic of the dissertation being more broad and general, so in case if something goes wrong I could change the sub-heading and re-focus my existing ideas into a different medium like video games. Heck, I could even just scrap the idea of creating the plug-in altogether if the results are sub-par, and just focus on more theory based approach and write about sound design principles (like layering) to get that ‘film’ sound. Half the job (i.e. research and recording) would be done anyway so I could then possibly look into differences of authentic sounds with their Hollywood counterparts. As an extreme example, I could multi-mike and record someone punching me, and then use this recording to compare it with how a punch sounds in film. (I could also do the same with a gunshot, explosion... the possibilities are there but the important thing is, with all this I have the Plan B, C... and so on!)

Now, let me backtrack a bit and talk about the idea itself:

As I said, this project would be as much about the theory and principles of sound design, but also about creating something. However, the difference in this project is that the ratio would be split into a more even 50-50, instead of a more theory-based approach of my previous ideas. I can clearly see the steps needed to carry out this project: from research about sound design techniques, learning to develop the plug-in, carry out recording session to gather the samples, then develop the plug-in and then use it in action.

Some more ideas about the plug-in itself in bullet points:

• Simple VST plug-in that creates a sound effect each time a MIDI key is pressed.
• Combine elements of real-life recorded sounds (newspaper hits, vegetable abuse, face slaps etc) with synthesised sounds (like bass hits and claps for adding low end and transient respectively)
• All samples would be phase aligned and matched to remove possible unwanted cancelation
• Samples would be played randomized each time creating a slightly different variation to the sound (so that for example Fruit 1 would have 10 samples playing in random order, Fruit 2 would have 10 samples and so on...)
• User can decide the volume and other parameters of each layer, therefore creating unique combinations of sounds
• A simple GUI where the user can make all the selections.
• My main area of focus would be to create presets based on a certain film or a genre (so could have something like a ‘kung-fu’ preset, or ‘Hollywood’ preset...)
• At the output stage there would be a limiter, also a stereo-width tool..

I have started prototyping in Ableton to see what can be done. If I get this project OK’ed, then the prototype will go into the Concept Development.

OBJECTIVES:

• Discover the principles behind sound design
• Research recording techniques
• Develop the plug-in that uses the discovered principles
• Demonstrate the use of the created software on a series of clips from existing films, and re-design fight scenes in film examples raging from Raging Bull, Indiana Jones to Ted and/or kung-fu movies.

TASKS:

• Identifying the potential software to use and learning it (Max MSP, Kontakt, SynthEdit, SynthMaker)
• Researching the bigger than life Hollywood sound and principles behind it
• Learning the plug-in development in chosen programme
• Recording the sounds
• Create the software
• Re-design existing movies using software
• Feedback
• Evaluate

DELIVERABLES:

• The software, the plug-in itself
• Re-scored examples of film using the created software
• Everything else we discussed in class like the search list etc.

DISSEMINATION:

• Plug-in can be tested and played-about-with in a real-time demo
• Demonstration and play-back of clips

Some final notes: I do understand very well that this could all go tits up and that the created plug-in could simply suck. One of my reasons for creating this is so that film-makers wouldn’t have to rely on stock sounds, but then this plug-in could prove a one-trick-pony where it’s good only at creating one type of sound/punch, so the whole idea behind it would kind of become moot. However, I do like the ratio of work split on this project: there is research to be done, software to be learned, but then there are thousands of sounds to be recorded and edited, so I would be getting enough ‘book-time’ and enough studio-time; enough balance to help me stay sane!

Late night Brainstorming Session and a New Idea for Proposal

 I don’t know why but in the last couple days I really struggled to write anything meaningful for my soundscape sound design project idea. I also can’t seem to come up with a valid rationale for an academic project like this and all I can seem to think of is personal development which is just not good enough.

 Anyway, I really can’t sleep tonight. I am really stressed, depressed and generally feeling down because of my lack of progress. Since I couldn’t sleep I decided to a brainstorming session. I know it sounds weird, but I decided that if the Uni is stressing me out so much I need to fight it head-on, so that’s why I decided to have this impromptu brainstorming session. I began by thinking totally outside the box. I thought about sound design but from a different perspective. Normally the way we do sound design is done in a DAW like Pro Tools, and for example layering of sound effects is carefully thought out process. I thought, what could happen if you turn this idea on its head and if I could create a sound design tool/plug-in that could randomise the process to hopefully create some interesting sounds. I thought it could be a good idea to apply this technique to generate a certain type of sound, like a gun-shot, a punch, or a cinematic trailer hit. Below I present some notes that I took down during the brainstorming session:

• The hit generator: Create a VST like software (or plugin) to generate customised ‘bigger-than-life’ Hollywood style punches and hits out of a library of pre-recorded and synthesised sounds and based on a set of user defined parameters.
• Combine elements of real-life recorded sounds (newspaper hits, vegetable abuse, face slaps etc) with synthesised sounds (like bass hits and claps for add low end and crack to the attack)
• [All samples would be phase aligned and matched to remove possible unwanted cancelation]
• Every punch or hit would be different than the other because of randomised sample selection and could range in scale from small hits to big cinematic speaker shattering knock-outs
• That can be used in low budget film productions and or video games, where there are budget limitations in hiring a professional foley stage and to not to rely on stock sounds.
• Demonstrate the use of the created software on a series of clips from existing films, and re-design fight scenes in film examples raging from Raging Bull, Indiana Jones to Ted.

I will proceed to expand and flesh out this idea and send it to Kenny for feedback by the end of the week.

Proposal Meeting with Kenneth. New ideas - Soundscape sound design and Recording guns

 Today, I had my second meeting with tutor Dr Kenneth McAlpine. It was a rather lengthy meeting that I really enjoyed. I prefer one-to-one meetings like this with less time constraints, rather than the ones with the class where you only get 15-20 minutes maximum. Because of this, we managed to discuss pretty much all of my ideas that I had up until this point. I started off with the gun-shot recording idea that James previously wasn’t very fond of. Kenny had said that there is some potential there, but again I had big problems trying to justify the idea and in no way “wanting to photograph myself at a military base recording gun-shots” was going to be a serious rationale for an academic piece of work.

 We ended up talking at length about my film sound idea. The topic of that changed to ‘Hyperrealism in Film Soundtracks’ after the last meeting with James. However we have soon abandoned the idea of ‘hyperrealism’. I really like the way Kenny makes you really think and really questions all of my ideas and makes me think even harder! We backtracked a bit and talked about film sound design in general and last year’s Audio Production Methods module. I brought up the idea of sound modality in film and Kenny started enquiring what about ‘modality’ makes it so interesting to me. My reply was that is a very fascinating new concept that can totally change how a scene in movie plays/sound completely affecting the mood. I gave examples that I discovered while doing last year’s blog. I also said that I really like how dynamic modality and the way sound modality can gradually change depending on the scene/storytelling requirements. It can be very effective when used on ambiences and having dynamic sound design in ambiences can really affect the way a scene plays out and I think that’s another reason why I find ‘modality’ so interesting as it has tremendous sound design potential when used well. Having heard all of this Kenny asked me why I don’t focus on film sound ambiences in particular. I said that this could be a good idea. I like the way we can sound design a place, time, or a particular setting with just sound, and most of the time ‘setting the scene’ with sound is done over a very limited time, like few seconds, when the establishing shot happens. The end result of this discussion was that we established that it could be a good idea for a dissertation and at the end of the meeting the draft question proposal was:

Setting the scene: Soundscape sound design in film.