<?xml version="1.0"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
	<title>Georgi Marinov - sound editor</title>
	<link>http://www.esem.name/gmarinov/audio/</link>
	<description>Blog of London-based freelance sound editor Georgi Marinov. Reflects on post-production, workflow, client relations and general topics.</description>
	<language>en-gb</language>
	<pubDate>Fri, 18 May 2012 13:48:52 +0100</pubDate>
	<lastBuildDate>Fri, 18 May 2012 13:48:52 +0100</lastBuildDate>
	<generator>Underscore Nano</generator>
	<webMaster>gmarinov@gmail.com (G Marinov)</webMaster>
	<atom:link href="http://www.esem.name/gmarinov/audio/blog/rss/" rel="self" type="application/rss+xml" />
		<item>
		<title><![CDATA[EuCon woes]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/eucon-woes</link>
		<description><![CDATA[

<p>How to restart EuControl without restarting the computer.</p>

<h2>Poor advice</h2>

<p>Over at Pro-Tools Expert there is <a href="http://www.pro-tools-expert.com/home-page/2012/5/15/dealing-with-avid-artist-series-connection-issues.html">a post</a> that supposedly explains how to diagnose connection issues with the Avid Artist Series. They will have you fiddle with DHCP, IP Addresses, and so on.</p>

<p>More often than not it's the goddarn software that is asleep at the wheel. Unsurprisingly, the capable engineers at Euphonix (now Avid) have been pouring their efforts some place else, instead of making sure the software is rock solid. I've had enough encounters with a variety of their products (ranging from the MC Control v1 to System 5), all suffering from the same problem - fragile software. It seems to be in the company's DNA. They should learn from the Delta 2496 - probably the most solid card in the universe.</p>

<p>So very often my Artist Control would be deaf to EuCon, simply because I didn't pray and cross my fingers whilst restarting the computer prior to expecting it to work.</p>

<p>What bullshit. Read my lips: I do not restart.</p>

<p>Things get plugged in and out, in no particular sequence on my Mac. It's how it's always been. So if you do the same (totally reasonable), and have the same problem - "controller can't hear computer and vice versa" - here's how to make Eucon and the Control talk again without restarting OSX. </p>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/eucon.gif" alt="eucon woes" /></p>

<ul>
<li>Start Activity Monitor</li>
<li>Find and kill all three processes: EuControl, EuConDiscovery, and MC_Client. </li>
<li>Start EuControl again.</li>
</ul>

<p>You're welcome.</p>
]]>
		</description>
		<pubDate>Fri, 18 May 2012 13:35:46 +0100</pubDate>
		<guid isPermaLink="false">eb150744e156055905b7768c117b2d49</guid>
	</item>					<item>
		<title><![CDATA[This channel is illegal]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/this-channel-is-illegal</link>
		<description><![CDATA[

<p>Or how we, soundies, sometimes come across as anoraks.</p>

<h2>Snicker</h2>

<p>I was at this indie-style shoot, an indoor thing, lots of guerilla rigs (Panasonic GH-series DSLRs with manual lenses, 5Dmark2s, and SQN+MixPre multichannel sound tracked direct to laptop). I like seeing every last drop of creative juice squeezed out of a piece of equipment. More so, I love seeing people dare to do it. It's nice to recall any initial skepticism as the finished product appears.</p>

<h3>38, 51, 69</h3>

<p>This TV guy (I think) walks up to us and introduces himself, a fellow sound person from nearby, whose radio mics were picking up our radio mics' signals. It's not unusual to bump channels and it makes for a pretty standard conversation. </p>

<p>But then, as he glanced over one of our transmitters, the first thing he said was: "Well, this channel is illegal". </p>

<p>Not sure he had to rub that in my face. I know channels and their legality are <a href="http://www.gbaudio.co.uk/radio.htm">undergoing change</a>. Yet I didn't bother too much fiddling with channels. All I needed to know is we could switch them quickly if we had to. </p>

<p>That so familiar urge to shine with technical knowledge.. It's near uncontrollable. But then came that delicate moment, and I watched him back down a bit, after we kind of failed to establish who needs a license when the kit is on hire. Subtle. And maybe this is how we, sound people, fail at having the warm conversations we could. </p>

<p>Granted, it's much more difficult to listen than it is to talk.</p>
]]>
		</description>
		<pubDate>Sat, 28 Apr 2012 12:08:30 +0100</pubDate>
		<guid isPermaLink="false">96d6577bee27a635cf24c845a5b8a0dc</guid>
	</item>					<item>
		<title><![CDATA[Drive (work that goes unnoticed)]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/workgoesunnoticed-drive</link>
		<description><![CDATA[

<p>Part of the story in Drive (the easy part) can be inferred from this one shot. </p>

<h2>Intense focus</h2>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/drive_supermarket.jpg" alt="still frame from Drive" /></p>

<p>At some 14 minutes in, this beautiful wide-angle shot of the local supermarket in the town where Drive takes place pushes the figures of Irene and her little son - even further away than they are. We're looking from where the main character stands in that very same moment. </p>

<p>The soundtrack is an intense but delicate tapestry of warm synths, piano tones, discrete vocals, washing over the quiet punctuation of the supermarket ambience - the beeps of cash registers, cart wheels, a bottle clinging. And in the middle of all this - as Benicio throws a bag of crisps inside the cart - the foley, loud and clear and so very close. There is, in that brief second, a glimpse of how intensely focussed the character of Ryan Gosling is, on the two, behind the otherwise calm face. Also how central to the story these fragile mother and son will become. And then, as he continues strolling between the shelves, and stops for another brief second, overhearing the cart wheels on the other side of the wall of products, the delicate exchange between Irene and Benicio. </p>

<p>The kind of level film sound works on.</p>
]]>
		</description>
		<pubDate>Sun, 01 Apr 2012 17:16:02 +0100</pubDate>
		<guid isPermaLink="false">e22c1d84684410fca143b172c95c8fd5</guid>
	</item>					<item>
		<title><![CDATA[Killing Noise]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/killing-noise</link>
		<description><![CDATA[

<p>I was asked by a video editor to share my ideas on how to salvage sound from a no-plan shoot destroyed by background noise and poor mic placement (read: reverb).</p>

<h2>First..</h2>

<p>It's somewhat difficult to write yet another take on the subject and have it add to what's already out there and findable via Google (here's <a href="http://izotope.fileburst.com/guides/iZotope_RX_Restoration_Guide_v_1.pdf">a hardcore guide to RX</a> for instance). So alongside the techniques, I've listed the context, idea behind them, and how I use them.</p>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/cassettedeck_pm.jpg" alt="photo by Matthew @wblo from flickr.com/photos/taurusaficionado/3576693538" /></p>

<p>Think of cassette tape. It was noisy, had a quarter of today's standard dynamic range, so it required recording hot, and noise reduction was a must. History says recording technology before tape was even worse. The battle with noise - the inherent-limitation-of-the-medium kind of noise - has lasted long before this other kind became a problem:</p>

<h2>"Return of the Sensorial"</h2>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/sd722.jpg" alt="Sound Devices 722 front panel" /></p>

<p>This 7-series above is a <em>very</em> quiet piece of equipment, especially when used with a good microphone. The real problem with its recordings is not system noise, but how to keep background noise out.</p>

<ol>
<li>Call it "aural focus", our brains are incredibly effective at extracting only the content we're interested in, from the mush of sound stimulating our ears every second.</li>
<li>Isolated sounds are rare. As I type this there's an airplane passing overhead, a car passing by, a dog barking in the street, my laptop fan whirrs quietly, the CPU makes a barely audible high-pitched whine, and my fingers strike the keyboard. The typing sound echoes around the room (i.e. sound and space go together). I can hear the reverb but only if I really tell myself to. My brain can tune in to each of these sounds separately.</li>
<li>Recording is not interested in said intelligent processing. A microphone captures sound blindly to the extent of its abilities. This is by design. Things go ugly very quickly when recorded. I.e. with a microphone placed away from where I sit, I couldn't possibly record just my typing.</li>
</ol>

<h2>Sound post noise</h2>

<p>... is usually the two types combined: Equipment noise and background noise from people and spaces.</p>

<ul>
<li>Hiss and other system noise (the one I mentioned first), including poor radio mic reception.</li>
<li>Hum, buzz, and other electrical noise (on location), including bad cabling or noisy lights.</li>
<li>Unwanted sounds from coincident events. (cloth rustling, jewellery, birds, transport, etc).</li>
<li>Constantly present background noise (air conditioning, refrigerator hum)</li>
<li>Unwanted reflected sounds in the space (echo, reverb)</li>
</ul>

<p>To do the heavy lifting of noise reduction, the brain uses information from the other senses: To filter out reflections it's a matter of sizing a room with our eyes. We'd notice a bird more easily if the space didn't smell of trees. Once spatially positioned, the hum of air conditioning, or the buzz of lights, can be easily ignored.</p>

<p>In recorded sound, that information is not present.</p>

<p>And then there's information outside of the particular moment, for instance London has a deep ground level roar on weekdays; sound travels faster when it's cold; a mountain is very quiet but not by a waterfall; people in Hanoi perceive car horns differently.</p>

<h2>Killing echo (and other noise)</h2>

<p>The best noise reduction is sheer planning. "It's done in advance" because of everything I just listed and maybe another hundred reasons including that no recorder captures context, and no mono (or stereo) microphone is as good as a pair of ears on a real human head.</p>

<h2>Noise-reduction after the fact.</h2>

<p>Back to reality, noise reduction is a chore in sound post. Sound editors know there is no silver bullet to it. The separate techniques combine to lift the veil of unintelligibility from dialogue. A little bit of each is what it takes and a bit too much of one is enough to ruin everything.</p>

<h3>Volume automation</h3>

<p>Often overlooked but simply riding the gain fader on a dialogue track can reduce the impact of background noise in the parts where it is most noticeable - when things are supposed to be quiet. Immersed in a dedicated atmos/background track, location sound takes on a new life. </p>

<h3>Gating/Expansion</h3>

<p>A gate or expander module (or plugin) reduces volume when the signal falls below a certain threshold. It is very useful for hiss, and widely used in music, but where higher level of control is required, volume automation wins.</p>

<h3>EQ</h3>

<p>EQ is useful in two ways. The more traditional way is about boosting and cutting specific frequencies (think our ears are most sensitive between 2-4kHz and content above 10kHz often has more noise than useful content so it's safe to gently roll off). </p>

<p>The one less talked about is notch EQ around room modes. It's a method I favour even though it's limited to small spaces. The theory behind it is that every closed space exhibits resonances at specific frequencies. Room modes, as they are called, greatly shape the character of echo in that space. To my ears, EQ-ing those frequencies first does a great deal to reduce perception of reverb in a room.</p>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/empty_room_modes.jpg" alt="modes of a small empty room" /></p>

<p>The picture above is just a sample. Every room differs and so do the numbers. While voice jumps between a lot of frequencies, room modes stay the same. They can quickly be uncovered by sweeping a narrow parametric EQ, and there's usually a mathematical relationship between the resonant frequencies. A few notch cuts can make for another good step towards reducing echo, at some risk of taking away from the voices.</p>

<h3>Multiband dynamics</h3>

<p>Easily my favourite, this is expansion combined with filtering, to form separate bands processed in parallel. It's realtime, and does not destroy sound the way spectral noise reduction does. The idea is the same as in expansion - reduce the signal (in each band) below a certain threshold. But the result is a dynamic EQ curve enveloping the content, applied in realtime, and thus a reduction in noise.</p>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/wavearts_multidynamics.jpg" alt="Wavearts Multidynamics plugin" /></p>

<p>My take is the best interface for this is on the <a href="http://wavearts.com/products/plugins/multidynamics/">WaveArts Multidynamics</a> plugin, otherwise a great underdog (you can see it in action applying a dynamic EQ curve above). Waves C4 does this in a less usable interface, and <a href="http://www.waves.com/content.aspx?id=10907">Waves C6</a> adds two custom bands at the cost of some confusion in the interface. Logic's Multiband Compressor (which also does expansion) can do the job, but hasn't got the surgical timing required for post.</p>

<p>Multiband is difficult to set up (think scene by scene at the least) but very rewarding when it comes to noise reduction. It reduces both hiss and echo while retaining or improving the clarity of speech. It's also easy to overdo.</p>

<h3>De-verb</h3>

<p>More often than not these are one-knob multiband expanders (see <a href="http://spl.info/software/microplugs/de-verb.html">SPL De-verb</a>), but a "true" de-reverb has recently been introduced..</p>

<h3>FFT Noise reduction</h3>

<p>The heavyweight of noise reduction, and not unlike, say, a 1000-band dynamics plugin. Adjusting threshold on that lot is as simple as taking a "noise fingerprint" before processing. It is slow and easy to overdo, but capable hands extract incredible results using tools such as <a href="http://www.izotope.com/products/audio/rx/">iZotope RX</a>, <a href="http://www.waves.com/content.aspx?id=9943">Waves WNS</a>, or <a href="http://www.cedaraudio.com/">CEDAR DNS</a>.</p>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/izotope_rx.png" alt="iZotope RX" /></p>

<p>The trick, especially with iZotope RX advanced is to apply different amounts to different frequency regions. Also, the "no silver bullet" idea still applies, so this is only an element of the larger whole.</p>

<h3>Spectral editing &amp; repair</h3>

<p><em>"Killing birds with iZotope RX"</em> </p>

<p>The above is from a tweet I got one day. It talks about a technique called spectral repair, which is about selecting sound in a brief point in time (and spectrum!) which is then deleted and replaced with surrounding information, not unlike the heal brush in Photoshop. There's spectral editing in Adobe Audition too, and that can be used to filter out whine from lights or extractor fans. When the offending noise fluctuates in frequency, try processing in separate sections.</p>

<h3>De-click, de-crackle, de-buzz, etc.</h3>

<p>I'm listing these last but they are usually applied very early, before anything else has happened to the signal. I think of them as a cleanup phase, before hardcore processing has begun.</p>

<h2>The mix</h2>

<p>When it comes to video, it helps to take the headphones off after all processing has been applied, and let the brain re-process the sound and picture as a whole. Quite often, unwanted noise noticeable on headphones disappears when played on speakers. </p>

<p>Which brings us to the final part - mixing. The techniques above are aimed at cleaning up one element of the mix. It will inevitably be recombined with all other elements - backgrounds, foley, music. Music especially can obscure a lot of reverb and background noise. It won't make it less unwanted, but it does help perceive it as less annoying. The mix is important in noise reduction, just as anything else already mentioned. </p>

<h2>All, combined</h2>

<p>Again, a little bit of each of these techniques is enough, as together they make more than the sum of all parts. And really 10dB of reduction achieved is a lot. This processing takes time, but is cheaper than bringing the talent in again, or repeating the shoot. (Plus, some directors just won't do ADR). More importantly, this is why good sound recording on location is so important. Don't overlook.</p>

<h2>One more thing</h2>

<p>I've been long interested in this topic, so my BA dissertation was all about noise. Not how to eradicate it, but how we need to leave some of it in, so that things feel natural and the listener doesn't feel uneasy. Some food for thought. </p>
]]>
		</description>
		<pubDate>Sun, 25 Mar 2012 14:34:23 +0100</pubDate>
		<guid isPermaLink="false">b633c41c443c57a0eff6f9f04d90a334</guid>
	</item>					<item>
		<title><![CDATA[Codespeak]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/codespeak</link>
		<description><![CDATA[

<p>What I said, what they heard.</p>

<h2>Catch 22</h2>

<p>One day I got an email from a director about potentially working on their film. Then we spoke on the phone. The film was to be mixed at a big facility. One of my earliest lines: <em>"I'll need to talk to your sound mixer about how they want their tracks organised."</em> The (unspoken) thought behind it: <em>"Time will be limited and material has to be organised in a particular way."</em> The thought behind the first thought: <em>"I hate overly confident talk so I never say 'Sure no problem' and think later."</em></p>

<p>Bad mistake.</p>

<p>That was followed by a month of silence, and then I got turned down on the basis of not already having an established relationship with the sound mixer. </p>

<p>Now that's a Catch 22 for anyone relatively new to the field. But, in hindsight, I guess some of this has to do with how what I said sounded to a director's ears.</p>

<p>Asking about how people down the line want their material is common in the sound post world. But to a director that may have translated to "I have no idea how to organise the mix on your film."</p>

<p>The thing is, we - sound people - go deep into detail such as "do we channel breaths in the main voice stem or group out to an optional extras track". It saves everyone trouble. Hence the questions.</p>
]]>
		</description>
		<pubDate>Wed, 07 Mar 2012 09:15:00 +0000</pubDate>
		<guid isPermaLink="false">eea316504acaabbf4bf508e1676a51b8</guid>
	</item>					<item>
		<title><![CDATA[We loved your sound...]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/we-love-your-work</link>
		<description><![CDATA[

<p>...but we decided to go with someone else.</p>

<h2>.. or why I don't pitch:</h2>

<p>A while ago I received a short invitation to pitch for the sound design of an app. Then I was asked to sign an NDA. Then I received a concise brief listing the sound set and requirements. The brief demanded originality and high quality.</p>

<p>Now, I don't know about other people, but in my world this means high expectations, and puts a stop on talking about budget in a straightforward manner, especially to a seemingly serious client. There was a tight deadline, hardly any space for negotiations or revisions, and I was effectively asked to do the work without knowing whether I'd get the work. </p>

<p>Not a situation to be in. But I went with it. I made an exception to my rule not to do unpaid pitch work. I thought for a few days, then came up with an aesthetic. Then worked on it, made the deadline, at the cost of some fuss and hurry. Wrote detailed notes. Explained why I won't give a straight quote. That's how I do. </p>

<p>Then, fully prepared, I waited for the "no", hoping it would be a "yes". And in itself, a "we loved it but decided to go with someone else" is nothing to write about. But then the app came out. </p>

<p>With a sound set one third of the brief. </p>

<p>At about 10% of the seemingly expected effort.</p>

<p>Talk about over-delivering.</p>

<h2>Two things I don't get:</h2>

<p>Even if a brief asks for significantly less effort, but in an implicit way, I find it near-insulting to the client to assume that is so. And no less insulting to double-check with them. And this brief didn't. So tell me. It's the only way. In hindsight this explains why I got no references (even after insisting). So generally if a client is after just a bunch of sounds, why not make it clear? Makes my life easier too. </p>

<p>If I was turned down on my vague non-quote (there's no way I was turned down on the quality of the near-complete work), then why keep mum about it? I can survive straight talk, and don't disagree with the decision, just how it's communicated. More like how the non-communication of it leads to loss of enthusiasm for everyone involved. </p>
]]>
		</description>
		<pubDate>Sun, 04 Mar 2012 23:05:21 +0000</pubDate>
		<guid isPermaLink="false">88a332ee5d7e04040e7e293f67aba20a</guid>
	</item>					<item>
		<title><![CDATA[Cyberscience and Noise]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/cyberscience-and-noise</link>
		<description><![CDATA[

<p>Digital primitives for noise measurements are starting to appear. </p>

<h2>Noise pollution</h2>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/pixeli.gif" alt="noise" /></p>

<p>One of the unwanted side effects of a busy modern-day metropolis is the noise. Find a quiet place and you can hear the booming rumble of the city; to "unignore" that in order to hear it is then a conscious act of will -- amazing. From afar and/or up above, the city background becomes a low-mid frequency mush of transport, traffic, an endless reverb tail of all the city bustle. Through it, one can sometimes make out, of all things, the monochromatic beeping pulse of a reversing service vehicle -- that and sirens are possibly the only two creating an easily identifiable pattern, regardless of distance. </p>

<p>In a huge city like London, the base level of noise never dips too low. But even worse can be the piercing noise events of stock delivery containers thrown inside large trucks; power tools tossed in vans. Those sudden one-shot bursts of hurt can trick a noise meter into thinking all is fine.</p>

<h2>Cyberscience</h2>

<p>It's nice to see people with hacking/making interests take up the issue of pollution in general, but, more importantly, noise too. Check out the <a href="http://cybersciencesummit.org/portfolio/what-can-you-do-with-noise-data-2/">noise data projects</a> of Citizen Cyberscience, and the <a href="http://www.everyaware.eu/">EveryAware project</a>. It's entirely beyond me why <a href="http://www.noisetube.net/">NoiseTube</a>, originally sponsored by Sony, seems to have been abandoned, or they never got to release an iOS app (shame).</p>

<p>Which is why it was funny to see the ending of the recent Hackday in London, where the speakers were having trouble being heard without amplification as a loud band was playing right next door to the small conference room.</p>

<h2>Future</h2>

<p>But one can dream. Hopefully crowd-sourcing noise pollution data will pick up and that can lead to the noise levels getting more attention. After all, it's <a href="http://www.defra.gov.uk/environment/quality/noise/research/">proven research that noise hurts the economy</a>. It only makes sense that future economic growth will require better and better health from us - citizens - as we'll object to being replaced by robots..</p>
]]>
		</description>
		<pubDate>Sun, 19 Feb 2012 16:38:33 +0000</pubDate>
		<guid isPermaLink="false">cca2af670c94ddfadbf33450baa1812f</guid>
	</item>					<item>
		<title><![CDATA[What is my day rate?]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/tell-me-your-dayrate</link>
		<description><![CDATA[

<p>It depends..</p>

<h2>It's not that simple.</h2>

<p>If you're asking me to clean up your 10 minute student project doco, add a cellphone ring, a camera snap, or a car horn or two, then I'd give you one kind of day rate. If you want me to score a video that you edited to temp music from a $220 million budget film, and retain the feel, then I'd ask for a different rate (one that covers the grief). And if you want me to work on your production over a long period of time (bringing a bit of security to my life), then that's a third thing. Then there are tasks that I simply don't take on..</p>

<p>But I will never, ever, tell you my day rate before we've established the kind of project you're working on, the kind of work you expect me to do, the kind of timeframe I will be given, and the kind of deliverables you'd ask for. It simply won't happen, especially with new clients, especially if I'm not given a glimpse of the project, be it a still frame, or <a href="http://esem.name/gmarinov/audio/blog/2012/send-me-the-script">the script</a>. Sorry.</p>
]]>
		</description>
		<pubDate>Sat, 11 Feb 2012 19:03:22 +0000</pubDate>
		<guid isPermaLink="false">ef8c71274c5c89cd80a450d4946563dd</guid>
	</item>					<item>
		<title><![CDATA[The extra mile]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/the-extra-mile</link>
		<description><![CDATA[

<p>Would it be enough?</p>

<h2>Arithmetic</h2>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/milestone.jpg" alt="photo by flickr.com/petereed" /></p>

<p>On every project I do there's an extra mile to go. Often I just want to advance my craft in some way. Not so rarely, it's a budget covering a mile too short. </p>

<p>Cut more money, that's less time (or hands, or tools) to do it. On a half-rate budget, that's an extra mile for every mile we go. On a quarter-rate budget, the budget stays serious, but the joke is on me. I go ten extra miles or not, it won't make much difference. There's never a mile-long project.</p>
]]>
		</description>
		<pubDate>Fri, 03 Feb 2012 13:22:57 +0000</pubDate>
		<guid isPermaLink="false">2eb3c00b5dbba31c8dbda81a1166faf5</guid>
	</item>					<item>
		<title><![CDATA[Send me the script]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/send-me-the-script</link>
		<description><![CDATA[

<p>A picture paints a thousand words. But that's not enough.</p>

<h2>Why</h2>

<p>Here's an excerpt of the shortest short I've ever worked on, "Afterlight" from <a href="http://3dstories.co.uk/">3Dstories</a>:</p>

<pre><code>INT. DARKROOM - DAY

With tongs he places the photographic paper in the stop bath 
tray and his wife appears slowly. He hangs her up with a 
clothes peg and sure enough there is depth to the image. 

She appears beautifully rounded, set into the photographic 
paper, the long sandy beach stretching out behind her.

Staring into the image, a tear wells up and he wipes it from 
his face. He hears the sound of the sea lapping at the shore 
and reaches up to touch her.

She is brought to life, rubbing sun tan lotion into her shoulders.
</code></pre>

<p>On the cut I got, the photos were still rectangular pieces of green paper with X marks on them, so that said "depth" can be added in 3D post.</p>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/afterlight.jpg" alt="afterlight" /></p>

<p>So no matter how it was shot, there's no way I would hear "the sound of the sea lapping at the shore", nor would I know if our character was supposed to hear it, or, in any case, how much emphasis the director wants to put on that sound. </p>

<p>The only way for me to know is the original script, where that trigger of an action drives the storyline directly. Anything else would be my interpretation of the pictures.</p>

<h2>But there's more</h2>

<p>Now I could have him "hear" this and simulate it being in his head. Or I can choose to "worldise" the beach and play a painstakingly put together second of sound that seems to come out of the photo, the rest of the darkroom utterly quiet.</p>

<p>You know which I chose, but that's beyond the point. How the script was originally worded contains information I find more than precious. For the filmmaker (who has read it hundreds of times) it's easy to say <em>"Here's the cut, just watch it. Plus the script changed anyway."</em> But when I am approaching a new project, all fresh and knowing nothing about a film that's yet to be post-produced, that's the kind of detail I am most interested in, and the detail I am most likely to miss.</p>

<p>So send me the script. I don't mind the changes.</p>
]]>
		</description>
		<pubDate>Sat, 28 Jan 2012 23:44:51 +0000</pubDate>
		<guid isPermaLink="false">a895e1d07ad0e965e333a43456a581f2</guid>
	</item>					<item>
		<title><![CDATA[Most sound designers...]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/andyfarnell-on-sounddesigners</link>
		<description><![CDATA[

<p>Quoting Andy Farnell:</p>

<h2>"worldly apprehension"</h2>

<blockquote>
  <p>"Amongst the many good sound designers that I have met, they all turn out to be really quite well rounded smart people. They aren’t purely phenomenal,  see the world not only through an artists gaze, but an informed, worldly apprehension. Sound is about going in to the world, what happens inside and outside things, its about deep knowledge about how things work and what does that mechanism mean to your emotions. I think the role of what a sound designer is becomes somebody who’s language is not about computers but understanding the mechanisms of sounds.[...] Deep knowledge of sound is what a sound designer has, even when they can’t express or vocalise it."</p>
</blockquote>

<p><a href="http://designingsound.org/2012/01/procedural-audio-interview-with-andy-farnell/">I didn't say this</a>. It's what I strive to be.</p>

<p>Farnell makes the case for embracing procedural audio, not only on the basis that it can save time and money, but that it can untie the hands of the sound designers and let them focus creatively on the things that really matter. He also examines how a project's technical resources are split, which is not unlike how a project's financial resources get distributed. <a href="http://designingsound.org/2012/01/procedural-audio-interview-with-andy-farnell/">Thoroughly enjoyable interview</a>.</p>
]]>
		</description>
		<pubDate>Wed, 18 Jan 2012 14:47:17 +0000</pubDate>
		<guid isPermaLink="false">04cef08a6c4b8737c72376e83e60eb78</guid>
	</item>					<item>
		<title><![CDATA[Do you still have the files?]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/doyouhavethefiles</link>
		<description><![CDATA[

<p>On keeping and/or archiving project files:</p>

<h2>Do you by any chance still have a copy of my project?</h2>

<p>No. You asked for the files and I was nice enough to send them over. Or you gave me your project to work on and then I gave it back. Our lo/no pay deal never included archiving your project. If I've packed and sent original session files, I've already gone the extra mile. If I haven't archived, I was under no obligation to do so. And if you want me to dig up those files and do some extra work, ask about my schedule first, <em>then</em> ask about the files.</p>

<p>So no. I did not archive your files, as I don't offer that service. Or maybe I did, as far as archiving my work goes, but that disk space is mine to use however I see fit. It was your say a year ago but if I decide to demolish that area, it's now my say. If you had your files and you didn't look after them, don't expect that I did.</p>

<p>Where I cannot offer you a guarantee, I won't offer you the service. No disrespect but I had to adopt this as a rule. While I may keep your files for a bit after we're done, the answer is: No.</p>
]]>
		</description>
		<pubDate>Fri, 13 Jan 2012 13:05:07 +0000</pubDate>
		<guid isPermaLink="false">0be8c8d2f55ed6a3f06848298c32147f</guid>
	</item>					<item>
		<title><![CDATA[Dictation from the Universe]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/dictationfromtheuniverse</link>
		<description><![CDATA[

<p>Is it right to talk about the decisions we're <em>going to make</em>?</p>

<h2>Hardwired</h2>

<p>From <a href="http://boingboing.net/2012/01/09/ending-an-endless-game-an-int.html">an interview with writer Julian Gough</a>:</p>

<blockquote>
  <p>We’re hardwired to be storytellers, and when we look back on our lives we build them into stories. And the more we find out about the nature of human consciousness, the clearer it is that we are making up stories after the facts a lot of the time, to make sense of decisions that we’ve made at a totally unconscious level: we have to make them into a story in order to navigate our own personal universe.</p>
</blockquote>

<p>In that same interview Gough talks about working to a totally open brief to write the Minecraft (hint: a video game phenomenon) ending, and the <em>unusual</em> way his work resonates with the audience - it leaves them thinking. Ignore the slightly sarcastic tone of the suggestion for a moment, I think one of the best currencies we have at the moment is the extended length of a thought. </p>

<h2>Intent</h2>

<p>Many a time I am asked to provide an advance "write-up" of how I see a project and the kind of work I am going to do on it. I see this as summarising the work beforehand, talking about the work, before I have actually done the work. </p>

<p>There are two things wrong with this: </p>

<ol>
<li>Talking about the work instead of doing the work. </li>
<li>Describing the result before it actually exists.</li>
</ol>

<p><a href="http://sethgodin.typepad.com/seths_blog/2012/01/out-on-a-limb.html">Creativity is about exploring ideas, a high number of which fail</a>, says important man Seth Godin. <a href="http://sethgodin.typepad.com/seths_blog/2011/02/art-is-what-we-call.html">And it's risky</a>. So by providing an early roadmap we lock the exploration process, taking a huge chunk of the creative magic away, by not allowing things outside of that roadmap to occur.</p>

<h2>Risk</h2>

<p>Embracing risk is not easy. That's not how companies run and that's not how I usually get paid for my work, especially the music side of it. But let's entertain the thought for a minute. </p>

<p>Three times in 2011, due to time constraints, I was trusted blindly. I did not have to explain anything in advance, and subsequently the work did not need to be <em>corrected</em> by someone. The clients were happy and I will hold these projects very close to my heart for a long time. On three other occasions, the explanations came first, then the work, then the clients  kept asking for revisions. Awkward. Expensive.</p>

<p>That's a lot of coincidence. </p>

<p>With both types of people I get my projects from, I have either a) worked with them already, or b) been recommended by someone. That's the beauty of word-of-mouth. </p>

<p>Film directors are happy with just sound. They never ask me to fill three roles at once, or I just turn down everything that's not "sound post". They give me their script and leave me alone without asking for a roadmap. I get trust from them more often.</p>

<p>Creative directors will usually want music+sound. (Under pressure from corporate types) they want the roadmap first thing, a quote to go with it, and the assurance that the work will be fresh and original, usually without having much completed work to serve as an anchor. </p>

<p>Sure, we can do that, no problem. But the combined risk includes an uncertain roadmap, an incorrect schedule, and work of questionable originality (decisions under pressure are usually risk-free). In simple words one can't be sure what will be done, when it will be done, and whether it would be special. Add to that the risk of reversing the natural flow of things Gough talks about above - by asking for explanation of actions not yet committed. </p>

<p>Surely that amount of risk is just the same as any open brief. Definitely higher than that of a brief on the back of some work already completed. And let's not forget word of mouth. The reason we met in the first place was most likely work done to a more liberal rule.</p>
]]>
		</description>
		<pubDate>Tue, 10 Jan 2012 16:44:10 +0000</pubDate>
		<guid isPermaLink="false">a12a05982707ee7d7d30a0e198e4864d</guid>
	</item>					<item>
		<title><![CDATA[Work that goes unnoticed]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/workgoesunnoticed</link>
		<description><![CDATA[

<p>Start of the series in which I will be pointing out the detail that we sound editors go into, in our work, but which rarely goes noticed.</p>

<h2>The Wire</h2>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/0107-thewire.jpg" alt="The Wire still frame" /></p>

<p>The still frame above is from around the 24th minute mark in the first episode of "The Wire" Season 3 (Supervising Sound Editor: Jennifer Ralston). Look again and you can just about make out the edges of two huge faces to the far left and right in the frame. The camera looks beyond that table to focus on the characters. </p>

<p>A subtle start to councilman Tommy Carcetti's run for mayor in this scene, but for me the most interesting detail (plot aside) was in the pauses of the chat: you can actually hear the blurry faces on each side of the shot, having a conversation in the far stereo end, close and clear, just very quiet.</p>

<p>All in the name of realism.</p>
]]>
		</description>
		<pubDate>Sat, 07 Jan 2012 00:06:00 +0000</pubDate>
		<guid isPermaLink="false">97c83b5a2d78577587fb50ae8bbe8060</guid>
	</item>					<item>
		<title><![CDATA[2011 in review]]></title>
		<link>http://www.esem.name/gmarinov/audio/blog/2012/2011inreview</link>
		<description><![CDATA[

<p>It has been an amazing, if a tad difficult, year. I feel very lucky (and somewhat proud).</p>

<h2>The work</h2>

<p>At the start of 2011 I decided not to say "no" to any work. Whatever it was, I had to find a way to do it, preferably avoiding working for free. For the most part this happened, and the year saw anything from three-day projects all the way to the end of a three-year one. All word-of-mouth (my credo), it was also quite diverse, ranging from TV station branding, through sound-art, to iOS interactivity; as well as the usual - short films, sound for CG/animation.</p>

<p>Excluding some under NDA, here are a few projects that deserved listing:</p>

<h3>№1 most challenging - "The Silverstone Chase" (INK)</h3>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/thechase.jpg" alt="The Silverstone Chase" /></p>

<p>Created for Santander/McLaren F1 Team, I was happy the incredible people at INK brought me in to fill a triple role: on-set sound recordist, post-production sound effects editor, and music composer. Every detail on this was fascinating, but my favourite is a fond memory of casually sitting on some stairs in the sunny town of Ruse, Bulgaria, shooting the finished mix files up in the cloud. That's how I prefer to work: Clear brief and assignment, lots of trust, then let me do it whenever and from wherever. <a href="http://www.youtube.com/watch?v=Fha3VIh4PhM&amp;hd=1">Watch on Youtube</a></p>

<h3>Most fun to do - Ogilvy Lab "Soundshower"</h3>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/soundshoweri.jpg" alt="Ogilvy Soundshower" /></p>

<p>What do you do if you're given an ultrasonic super-directional loudspeaker and access to a heavyweight advertising agency's marble offices? Spray that space with the sound of birds, crickets, and all sorts of other romantic, non-business creatures and things, not found in the city soundscape, of course. </p>

<h3>Unexpectedly fruitful - "The Future by Airbus" (Clay Interactive)</h3>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/airbus.jpg" alt="Airbus Concept Plane" /></p>

<p>Beats me why ad agencies fail to budget sound in, but they sometimes do, and then it's down to the smaller studios to sort it out, and when they get in touch, I'm having to ask that we think of a way to make this happen. Clay not only listened, they also passed the message on, came up with the budget, and allowed me to contribute to an iOS app they did for Airbus.</p>

<h3>Top remote project - "Afterlight" 3D (directed by James Uren)</h3>

<p>Never underestimate the power of a good script, to get everybody through, without lengthy face-to-face discussions, talking about the work, instead of doing the work. Maybe the fact that there were sound notes, and that James understands the sound angle perfectly, helped a good deal. Plus he was patient. </p>

<h3>Most demanding: "Silverstone: The Wing" (INK)</h3>

<p><img width="100%" src="http://www.esem.name/gmarinov/audio/blog/2012/silverstone.jpg" alt="Silverstone - the Wing" /></p>

<p>Much as I can sympathise with clients referencing scores from $200M+ films, and expecting to pay in very few digits, it's a great shock when a visual editor temps with a similar scale score, then I get video masterfully cut to truly special music, which I have to replace, <em>after</em> I've quoted for both music and sound design. Temp love at its worst. <a href="http://vimeo.com/20839400">Watch on Vimeo</a></p>

<h2>Looking back</h2>

<p>The year also saw me put on a project manager hat once. That fence is not a spot to my liking as usually one gets stuff thrown at them by both client and team. But that same spot allows one to figure out vital things about budgeting and basically man up in no time when it comes to pitching, organising, and doing the work. </p>

<p>Finally, late in 2011, I had a project fail halfway through. It was a complex affair and blame could get thrown around a lot, but it made me realise how unprepared (but also how unwilling) I am to fail a client. The last time anything similar happened I made a conscious decision to solidify and straighten my knowledge, and never let anything similar happen. But these things do happen; to my surprise the creative studio were nice to me, and I learned that sometimes it's out of my hands (though I'll never stop believing it wasn't).</p>

<p>In hindsight, when everything is much much clearer, I could have avoided a few of these things by just listening to my gut more closely, saying "no" on time, and not saying "yes" to so many things. But as I said at the beginning, saying "no" was last resort throughout the year. </p>

<p>I think not saying "no" made things great, but only so much. It also allowed me to get a good feel for when I should be saying "no". Hopefully I can now fulfil my next year's resolution: <strong>Focus</strong></p>

<h2>Looking forward</h2>

<p>I am not only optimistic for 2012. I'm certain it will be the best yet. 90% of making this happen is avoiding the mistakes of 2011 (and 2010, 2009, and so many other). The other 10% is luck and meeting people. True, the year has been successful, but going forward I'll still shake things up a bit:</p>

<ul>
<li>Take the first few months easy. Sounds crazy given the economy, right? Not really. How do you focus without slowing down? I want to shake off December's stress, and allow my mind to think clearly.</li>
<li>Follow my instincts more. Avoid things like working for free, meeting crazy deadlines or silly demands. Seriously, so often my gut knows better than I do, it's embarrassing. </li>
<li>Put a fresh showreel together. I generally take it as a good sign when someone hasn't got a fresh reel (they've been busy), it has been far too long. </li>
<li>Get into interactive projects. I know how to program all sorts of things. <em>And</em> I've learned to distinguish between actually being creative and merely coding.</li>
<li>Record more sounds.</li>
<li>Run a short-form blog. (Class this text as huge now.) Not an easy decision, as I thought I have nothing to add to the dozen(s) of sound blogs out there but have gradually changed my mind. I'll make sure I don't repeat anything already said.</li>
</ul>
]]>
		</description>
		<pubDate>Fri, 06 Jan 2012 22:38:17 +0000</pubDate>
		<guid isPermaLink="false">b9a86f7a589735424b650d72107124aa</guid>
	</item>				
</channel>
</rss>
