GDC 2015 notes on Technical Sound Design

So, Marc Kilborn gave a great talk on basically what a technical sound designer does. This was of particular interest to me as many skills that fall under the purview of techical sound design are ones I need more experience in. The skills Marc outlined as being part of Technical Sound Designer's skill set:

  • Implemention expert
  • Unique scripted solutions
  • can code in high level/scipting languages)
  • can make sounds as well

He went on to compare a technical sound designer as an audio equivalent of gameplay programmer in so far as they should be adept at high level programming, but not low level (that would be the domain of the audio programmer)

The rest of my notes are regarding the scripting part of his talk. They're pretty inane (my notes, not Marc's talk).

  • C# in unity is good example of scripting
  • python is a good place to start for scripting (Lua is good too)

I recently read a comparison between Python and Lua that seemed to indicate that because Lua has a smaller footprint, is faster, uses less memory and was designed intially as a configuration language it is well suited to games in general - I believe it gets used for audio scripting a bunch.

  • Learn to build with unity and unreal
  • disect a game a try to break it
  • Learn to program in C# and PD
  • Device 6

I honestly can't remember if these were my take away's from Marc's talk or things he actually said. I mean, they're all good things to do and definitely on my distressingly long post GDC to do list.

But the most quotable thing that Marc said was a riff on a post audio quote that I've heard attributed to both Spielberg and Lucas - the old "film is 50% sound" chestnut. Marc's take on that was "Game Audio is 50% implentation. Tattoo that on the back of your eyelids"

But let's be realistic. This is audio. When is there time to even close your eyes?

 

GDC 2015, Dialogue 101 notes.

Another talk I wasn't planning on attending. With over 15 years post experience I've done just about everything dialogue related you can think of, and I wasn't sure that even Michael Csurics . Recordist, re-recording, processing, ADR, editing, directing, I've even spent a bit of time on the other side of the glass. Fortunately, I get a perverse kind of pleasure from being proven wrong, probably because it usually means I get a chance to learn something.  

So right out of the gate, Michael starts talking about casting. Whoops, I've never done that before. Climbing off of a high horse is difficult, but fortunately the conference rooms at the Moscone center are relatively spacious, so I'm pretty sure no one noticed.

  • Give the actors sides, not scripts, which should contain, that actors name, character name, sex, age, race, accent, vocal traits, role description, notes, some lines of dialogue, artwork of the character
  • Relevant character info only, don't overwhelm the actor, help them do their best job. Try to keep sides to 1 page, 2 max

The lines don't have to be from the script, but they should be something that an actor can really show their idea for the character with. IMO the lines should also be where that character "lives" in your story (reluctant hero, naive sidekick, bitter but sage veteran) and emotional range. Further to that, I would also add that if you're doing callbacks that might be a good time to have your actors read lines where there characters "turn" or do something against archetype or seemingly out of character. Obviously great writing goes a long way to make such moments believable, but story setups get cut, and if you have an actor that can really sell those moments where a charcter turns a sharp corner in their development or acts in a way that is seemingly inconsistent with their core values, that's worth its weight in gold. 

  • Get your actors to read Onos / onomatopoeia lines. Some hits (tense jaw, tight core) some getting hit (mouth open relaxed, core not engaged) some pain / death screams.

If a character has a lot of systemic dialogue, there's no point in hiring someone who can deliver good Ono's. Its really immersion breaking for players.

  •  ACTRA rates are 890.13 including agent fees - good for 4hr call and is minimum
  • For non union budget 200/hr and a two hr minimum

That non union rate looks high to me, as its the same as ACTRA. My guess would be 100/hr 2hr min is accurate and that I've made a typo in my notes.

  • studio cost 200/hr on avg. includes engineer

This is going to vary by region and city, but that's a good benchmark. In Toronto you should be able to get a good room for about a grand a day, especially if you're booking a bunch of sessions.

  • Don't book a studio without visiting if you can help it, visit first see if the space feels comfortable. 
  • Walk on to the floor and give it a clap, make sure it doesn't sound too live (you can here the clap ringing for a long time or it sounds unnatural) or there's no natural decay, it just sounds like a muted pop

I'll add here, that an especially dead room doesn't really present any technical problems. In fact, a really dead room is great for screams and any loud sounds which excite a "live-r" room and generate a lot of reflections that get picked up on mic. Not having any room reverb makes it easier to "sell" stuff as being outside. However, a really dead room can be  difficult to work in for actors. A dead room sounds and feels unnatural, which is just one more thing taking an actor out of their character, scene and imagined environment. By and large, technical considerations should come second to performance - the whole point of recording VO is to capture a performance, so the technical considerations should always be in service to that performance.

  • Places to find actors in order of preference:  Agencies, Casting Directors, Local Theater
  • When choosing your actors,  check availability and type of experience. Confirm that they have availability the week before certification to accomodate last minute changes before gold 
  • Make sure they are right for YOUR game (try to avoid the fan effect / the celebrity effect) 

Scripts!

  • NEVER GIVE AN ACTOR A SPREADSHEET (or anything other than a properly formatted script)
  • Scripts need to include context - previous lines, and the scipt should to be formatted to distinguish between between your actors lines, and the context lines
  • separate scripted events / cinenmatics from systemic dialogue

To that last note, I would add save your Onos for the end of the systemic dialogue, with death screams (and screams in general) coming at the end. I would also add to make sure that you have scripts ready for everyone else on the team at the session. Plan for script for the recordist, her assistant, the actors, yourself and anyone else working "behind the producers desk" (producer, your assitant, the writers, narrative director, voice designer etc). You should also be at least perihperally aware of the workflow (intimately, if you're the engineer, assitant, voice / narritive designer) and be sure that everyone is on the same page for deliverables, naming convention and file formats.

And finally, on directing: 

  • collaborate not dictate, let the actors do what you hired them for and only give a line reading if they're really struggling
  • create a safe space, being in the VO booth with the actor can be a huge help here. Also leaving the talkback open in between takes is a big help
  • have a vision, know the script

I can personally attest to getting better performances when the director is on the floor with the actors and when the actors have someone else to act off of. Also nothing interupts the flow of a session more that when the director doesn't know the intention of the scene - if its not possible to read through all the scripts and have a meeting with writer before the session, the writer needs to be at the session. 

Okay, that's all for today, tomorrow I'll tackle both technical sound design and systems in Sunset overdrive. 

 

GDC 2015 Audio Gigs: How to Find Get and Keep them Coming

So the second talk that I took copious notes for, was from Richard Ludlow of hexanyaudio.com. To be perfectly honest, I had no interest in staying for this talk. None of the amazing gigs I've been lucky enough to get have been through networking, they've been through friends, or more precisesly, friends I've made through work. There isn't any magic formula for "getting" gigs and networking events that purport to do that seem to be selling relationships as a means to an end. No thanks.

So. I'm ready to go soak up some SF sunshine when Richard says (and I'm paraphrasing) "When you're at a GDC party, social or networking event, don't go looking for a gig. Look for someone you might want to be friends with.  Talk to those people. If you hit it off, those are the people who someday, may call you with work". This really resonated with me, and it's how I now approach all potential "networking" events. Not as a chance to find work. But as a chance to find someone that I like, can hang out with, and therefore, might want to work with. 

But Richard's talk wasn't all friendship braclets and holding hands.  Once you have a meeting with a potential client, he had a list of things you should suss out in the initial meeting. He recommended start off with casually asking questions - I would also reccomend doing some googling so some of these aren't straight up questions, but talking points you can bring up.  

• how big is your team and what is the breakdown / roles

• are you doing this full time

• how long you have been around

• what are your past projects

• where are you located

• how is the being funded

IMHO that's a mouthful to ask casually. So do some homework on the client first. RIchard then reccomends getting down to question about the game. 

• what platform are you developing on

• what's the development cycle like?

• what is your timeline for audio and ship date?

• what, if any, is the story

• what's the estimated gameplay length

• how many, levels?

• how many main characters?

•what are the core mechanics

• what are the game's influences

• who's the publisher

The above is the beginning of the nitty gritty - you should definitely be writing it down. Now on to the stuff that it's often so tempting to start with. The audio related questions: 

• what's the musical style

• how many minutes of music

• how many SFX assets?

• Is there narration? Budget for casting and VO and ACTRA?

• Is there music interactivity?

• are you interested in using a  middleware solution or implemenatation servicies?  

• do you want live players?

• do you want to music to be a work for hire or do you want to license the music?

• are their cut scenes & trailers that you want my services for.

And finally the questions everyone hates to ask - budge. 

• is there a general range you're looking to keep audio in

If it seems appropriate (ie they ask you what they should spend on audio) ask them what the budget of the game is.  

Now at this point in my career, I usually get told that I have the gig or I don't. But Richard says this is where you make a bid. It should include a cover letter, propsal, and cost breakdown.

• cover letter - introduce yourself, express excitement for the game, outline the highlight of the proposal, have a brief bio

• creative proposal - get them excited about what you want to do for the game, include detailed ideas about music and sound and specific about implementation, as well as cost details.

• cost breakdown - keep this all modular. music can be broken down by themes, arrangements, transitions, stingers. SFX by primary sounds, UI, ambiences as well as number of iterations. Dialogue can be broken down into casting, actors, direction, recording, and line processing. Finally, although Richard didn't address this in the talk, I emailed him about mix and implentation costs. Based on his response you may want to roll mixing and middleware implentation costs into your other assets - asking a client to pay for both the middleware licsense and for you to use it to implement can seem like insult to injury - same goes for the mix. Seeing as when you throw assets over the fence it can ruin the sounds (and your reputation), I'd be willing to try and find a way to absorb mixing and implementation (if the implementaion begins and ends at the middleware) depending on the project. However, if assets are to be implemented directly into the engine, that's definitely something you can charge for, preferrably at an hourly rate.

Richard also reccomened sending two quotes, one as music for hire, one as licensed music. 

After this, my notes get pretty monosyllabic. Here they are, verbatim:

• copyright

• what is being delivered

• delivery milestones

• how much

• developer rights

• composer rights

• credit

This seems to revolve mostly around music, after you've been offered the gig. Sorting out the copyright is very important, obvs. So is having a clear handle on what exactly is being delivered, and how much of it. Also having a rev limit, and some discussion about what contitues a revision, and what constitutes a "change order" - ie an entirely new piece of music with a new style / emotional content / pallet. I'd also personally have an agreement esp with work for hires, about who owns rejected content. Having specific talks about both composer rights and developer rights are crucial.

Finally richard also mentioned that audio style guides are for short conversations, but uh, I didn't include any examples in my notes... He also said that for an in depth coversation, dig deep and ask for six tracks that appeal to the dev for this project. Once you get those tracks, send a couple of tracks back to the devloper based on your discussion, and get some detailed feed back from them. 

I'm sure that Richard probably said something about how to keep clients, or maybe he ran out of time. So I'll leave you with this nugget: you can be brilliant, easy to work with, and meet your deadlines - to keep working you have to be at least two of those. 


 

GDC 2015!

This year I was fortunate enough to go GDC and attend the audio track. I took copious amounts of notes in point form, so I figured I might as well share them here. Some of them may not be entirely comprehensible, hopefully I can figure them out as I go along. Points marked with a "•" are basically what my notes where. Anything with a "-" is me trying to explain my note post-hoc.

The first talk I attended was Sound Design the Bungie Way, by Jay Weinland.

• When you have too many things to assign unique sounds to, try to solve it with UI                           

  - Jay was talking gun sounds here, I think the idea here was that the acutal firing sounds for the guns are so design intensive, giving them different reload and ammo PU sounds can help give them a unique sonic identity

 • Think of sounds as being attached to the player, or detached from the player. The former should be 2D sounds, the latter, 3D sounds

 • Bungie removed weapon reverb tail from shots themselves, and assigned them to envinronment type and surface type. Tails were broken into three basic weapon type (shotguns, assualt rifles and sniper rifles) then broken down by environment, then by surface type

 - reverb talks seemed to use the same source sounds, but were treated with EQ (and additional in game reverb) when they were further away from the player

 • Apparently you don't have to get special permits (beyond gun permits) to use tannerite to create explosions (for recording) 

• Bungie augmented all their gun recordings with a transient layer off the top and a "punch", "snap" and "lowend beef/mech" layer as well

• Enemy guns had two basic variation sets, distant and close. Because the vast majority of gun play happens at the middle distnace they made the decision not to change between distant and close sound at the mid point, to much switching between sounds all the time. Instead they crossfaded into distant sounds further away. For the mid range, they used the close sounds with some HPF and LPF.

• a mix of three global dynamic systems were used for state mixing

*  Ducking - gentle slow release, a few db ducking happens of incoming input level

- presumably this was to prioritize medium/high priority sounds without a lot of volume

•Sidechan - working at about 6db of gain reducion, 

- input dependant, so this makes more room for loud sounds

• Silence Duckers - aggressively kills everything for an instant to leave room for just a huge explosion

- So the way I understand the silence plugin to work is that WWISE just looks at the level it is set at as opposed to analyzing the sound it is at. By making it super short, setting its level to the max and having it play just before your explosions in an explosion event, you can kill all other sounds faster that you can hear them cut out giving room for the explosion to really cut through. 

• Instead of a long continous loop, rocket by's were treated as long one shot sound that played in proximal relationship to the player. If the rockets were fired very close to the player, the audio engine would seek into the middle of the pass by audio file, quickly fading up. 

 - I followed this conceptually, but Jay didn't have time to get into how it was scripted.

• Consider using state mixes (ie scopes affecting pan)

- I litteraly just remembered that "scopes affecting pan" refers to using a separate mixing state for each gun scope and narrowing the pan to help the feeling of focus for looking through the scope.

Okay, with any luck, I'll be able to summarize another talk for y'all tomorrow.

 

 

 

File renaming from a spreadsheet

I recently had to come up with a way to rename simple file names to more complicated file names from a spreadsheet on a Mac. It occurred to me that for those of you without Soundminer this technique might be useful for renaming SFX. Sometimes when you purchase a library, the file names are simply unique identifiers (like SD12321.wav), and the descriptive comments are included in an accompanying spreadsheet. By fiddling with the method below, you should be able to replace the original file names with the descriptions, allowing you to use the Finder to search them.     

  • Create a safety copy of your files

  • Make sure that the filenames are in column A and the descriptions are in column B

  • delete the other columns and save the spreadsheet as a .CSV

  • Create a new file in TextEdit and change your format to plain text (Format -> Make Plain Text) and enter the following script:

for details in `cat SPREADSHEET.csv`

do

OLD_NAME=`echo $details | cut -f1 -d,`

NEW_NAME=`echo $details | cut -f2 -d,`

echo "Renaming $OLD_NAME to $NEW_NAME"

mv $OLD_NAME $NEW_NAME

done

  • change SPREADSHEET in line 1 to be the filename of the spreadsheet with your names in it

  • save this text file as rename.sh inside the folder with the files you wish to rename

  • open Terminal and navigate to the folder with the files in it (type cd then drag the folder into the terminal, hit backspace once, then return)

  • type sh rename.sh

  • Profit.

 

NB: don’t forget to delete your safety backup of the exported files

NB: if you comment out the second last line with a hash #mv $OLD_NAME $NEW_NAME terminal will echo all the file changes for you without actually renaming the files.

 

DIY binaural rig

`After purchasing my pair of DPA 4060's a while back I have been planning on building a DIY binaural rig. While I have a method for using the DPA's with my own ears a stealth recording rig, this isn't always the best way to go.

 The inability to monitor is an issue along with noise from the operator (breath sounds, footsteps - etc) and proximity to subject. So while it makes for a great stealth rig, I wanted a way to get past those limitations, wihtout spending a boat load on one of these:

http://tinyurl.com/k2lwm7z

image.jpg


 

 

 

I started by tracing out the ears and cutting to the depth of the silicone mount.

After the ears are traced out, cut along the outline

After the ears are traced out, cut along the outline

Slice the mannequin's ear off

Van Goghing for it.

Van Goghing for it.

Scored columns and rows with my exacto knife

image.jpg


scraped 'em out with a spoon

image.jpg


Grab a nail file and smooth out the insides

 

 

Drill hole in ear for lav cables

image.jpg


There's already a hole in the mannequin head that runs thought it, perfect!

image.jpg


But I'm drilling a hole in the back neck so I can run the cables out there and use the hole at the bottom for a stand mount. 

image.jpg


finished

finished

I elected not to glue the ears in - they're a pretty snug fit and getting the DPA's in and out will be easier that way. 

image.jpg

And then my kids got a hold of her and asked to give her a makeover. I figured people would think I was plenty crazy for carrying around a mannequin head anyhow - at least this way I won't lose it. 

image.jpg

Some Fixit epoxy to fit the mic adapter into the head

image.jpg

It's a 3/8" to 5/8" adapter.

IMG_1037.jpg

I've attached a heavy bead with a twist tie and wrapped around the end of the DPA attaching to the XLR connector to thread it through the head. Even so, it takes some doing to thread the microphone through.

IMG_1038.jpg

Threading the mic into the ear

The mics just poke out a bit

And voila, DIY binaural dummy head.


Hello World.

The awkward first post. I thought about starting this blog in medias res seeing as I am in the middle of my career but introductions and explanations are probably in order. My name is Adam Axbey. I am professionally, in about this order, a sound designer, mixer, recordist, composer, producer, writer and director. The aim of this blog is to focus on my adventures in the first four of my aforementioned fortes but I’m sure a sprinkling of the final three will appear throughout.

Let’s drill down a bit on the goals of this blog. After fifteen years working mostly in post sound and more recently in game audio, I realize that I’m not really a significant participant in the audio community. I’d like to change that. I think that I’m also at an interesting crossroads in my career; as I’m well versed in all thing post audio I hopefully have some helpful tips and tricks to share there (though there’s always lots more to learn). But I’m less well versed in the implementation side of game audio, so I hope the mistakes I make here will garner me some constructive criticism, or at least some creative invective. 

Here’s a link to the last thing I mixed, Cache Craze

And a link to some pretty cool sound design I worked on, Lego Marvel Super Heroes: Maximum Overload

Finally, here’s a web series that I co-produced and did all the audio post for, Last In Space

TL;DR - I’m gonna be doing some cool audio stuff and post about it here, stay tuned if that seems like good times to you.