Approach: Part 1
A few questions crossing my mind a lot lately are:
- What is my thesis statement as a recordist?
- Why do I pick the things I want to record?
- How do I approach recordings?
- What are the things I want to keep improving on?
What's your approach to recording?I'm curious how different people here think about recording something and what their end goal is. For instance, you find some interesting sound while messing around and you realize this is your one shot to capture it. You have every tool in your arsenal, so what tools do you pick and what do you want to express in your performances. When do you start rolling and when do you cut?
I'll take a stab at answering my own question in a follow up post, but for now I wanted to preserve these really insightful responses for future reference:
Aaron Brown:
[1/2]
Main approach is to start at average listening distance at average listening perspective to get a natural take. That's usually the most useful recording. If I'm recording for specific purposes that can change. For sound design I often get closer for effect and reducing room tone. In studio adding baffling to reduce room tone and noise even further for cleaner files. In the field I may get closer or alter perspective to negate noises.
For file lengths I try to follow a 8:1 rule. If I need 1 minute of ambience I record 8 minutes as after editing I often get a great minute. For props and sources I try to start with the normal set of performances with extras. Funny enough this also seems to follow a 4:1-8:1 rule where some are too similar and others aren't great. Then other performances and perspectives to be safe and have options for happy accidents.
[2/2]
End goal: Make a recording that's as easy to edit as possible that fits in my library as usable source matching or exceeding quality of existing materials. Use markers to mark bad moments to edit or brilliant moments. Be tight with performances so edits have clean gaps, but not too long so its' easy to see segments of sfx and edit them with strip silence. Top and Tail with slate and anything important to note when editing. Streamline the process.
Arnoud Traa:
[1/3]
Great questions!
There are 2 types of recording for me:
Ad-hoc: the weather is really good and i hear something interesting so I grab some of my gear and hope that everything is hooked up and recording in an appropriate timeframe. I press Rec as soon as I can and try to capture it roughly at first (so to have at least something useful). Then i listen/look for the best angle (this depends obv. like @Aaronbrownsound explained very well). I take as many performances as sound interesting or at least 6 minutes to record ambiences (many times longer, but that depends (again!)).
Scheduled Session: i have a really clear idea and goal for a recording session. I think about the gear needed (make a list), the location (contact if necessary) etc and create a rough list of things to record at the minimum (everything from different perspectives etc). I make sure I have enough time to record that list and that the location is accessible for the time i need it. I don't have rules for 'when I role' in this scenario, but have made a plan beforehand, because it depends on just about everything (content/context/goal).
In general:
I try not to be too strict about things, that only makes me rational about everything and I want to be 'in the moment'.
For ambiences though I try to think about how I could use the background sound and what kind of alt. perspectives are useful.
Mastering is a process that is mainly about getting the sound to work for the intended goal, sometimes it's top and tail, get the gain to a good level, nice stereo image (or choose mono) and add metadata. Other times I need to clean things up (wind etc) or have a specific sound in mind that needs to be massaged out of it.
But for my personal library, I generally go for 'this is useable and can still be altered' so i don't do excessive EQ, denoise or dynamic processing. The most important reason for this, I don't want to spent too much time mastering unless it is needed badly (then i maybe just trash the recording). As long as it's not taking a lot of time to edit this in a project, it's fine.
[2/3]
One thing I have been trying to do better (but am hopelessly failing at) is setting a standard for myself regarding mastering object performances. I've been using Radium more and more, but run into issues where sometimes levels are not uniform, gaps are too short (or too long), which results in more time spent tweaking the region settings in Soundminer. I need to get that straight so that process goes quicker. Might even do gentle noise gating to fix some of my older recordings.... but that is a whole other problem
[3/3]
Last thing: i have 3 setups ready to go (at the moment) : hard fx (mono mkh60 or 50), ambiences (dpa 4060 and mkh40/30), 'on the go' sony pcm d50 with usi's. I just bought the sennheiser binaural headset and have been testing/comparing it's sound to the usi's (no conclusions yet). It will be an additional on the go set.
In soundminer i have custom workflows for basic metadata for all these setups (more or less) so I can have technical metadata added and embedded in 2 or 3 clicks. This has made the mastering phase much quicker.
Mark Mangini
Most of my recordings are done close-up. Often as close as possible. This is a function of me working mostly in Narrative cinema where the goal is not verisimilitude but some “heightened” or exaggerated version of it with sound. The Science Fiction genre comes to mind. There is an economy to this approach because I can create an “ambient” version of a close-up recording. The inverse is almost impossible. Of course, if I need ambient recording, capturing something the way the ear hears it, I plot that in my record manifest and capture it. But those sounds aren't as much fun to record. I love the “privileged view” that a hyper cardioid capsule captures when it is right up next to or inside of something. We don't hear that way and our ear misses details in those sounds without the assist from our microphones. I posted a coffee grinder in Freebies that I mic'ed from two inches away. I didn't need the sound of a coffee grinder. I did need the sound of spacecraft ailerons and the morning I was cleaning the grinder, it revealed itself. By placing the mics up close one “de-contextualizes” sound from its source creating a new context that is ripe for manipulation or “re-contextualizing”. It's just harder to do that with sound captured “as we hear it”. The flip side to this is real ambient sound. I love the nuances of sound that seems to have little sonic focus because they are really hard to fake. The recent Quiet Tasks CS was a bonanza for me because it's really hard to fake/edit/mix the sound of someone doing the laundry or cleaning up after supper offstage from whole-cloth or close up recordings. I guess what I'm saying is that I don't have much of a middle-of-the-road approach.I value success over technical proficiency. As a musician, I have many scars from missing recording a great musical idea or inspiration because I got lost in the weeds of setting up equipment. I know I can always do better with my setups. But in this way I always get something “on tape”. Which is more important than missing inspiration or an ephemeral event because I was setting up. This is true mostly for transitory sonic events that I'm afraid I'll miss.But most of my recording is planned and staged. In those cases I am most often thinking two things: what must I capture for the project at hand and how else might I capture/perform it for future utility. This is the librarian in me that has screamed a million times when finding a great kit of sounds in the library…but they forgot to record the (fill in blank).All my files are mastered with the intent of using them as elements in a sound edit at a future date. To make that editing efficient, I master multiple occurrences of a sound as a single file (think gunshots, body falls etc). This also keeps one's region list (Pro Tools) clean and easier to navigate and gives the editor variety to choose from.I'm sure this will be shocking but after I have mastered the files, I delete the source. If I'm happy with what I've done today, that's fine for me for the future. I don't keep my Raw A format Ambisonic files, I don't plan on keeping my source DMS files. I know that's really bad from an archivists point of view.Metadata is always a challenge. I hew between descriptions that are quick and easy to read vs contains the valuable data needed to accurately describe what it is. For me browsing a sound library is a necessary evil that keeps me from the creative work I really would rather be doing. So anything that speeds up my browse is good. File names are a real bugaboo. While I adhere to most of what UCS offers in principal, I deviate from it in practice because I want filenames to be readable quickly in a mix window and give specific info on what a sound is. I don't like contractions and I don't like removing spaces to conserve on space and I don't like filename bloat. I never put my name or the microphones or any other non essential data in file names. That can be found later if necessary but is rarely useful for a sound edit in narrative work and makes for cluttered and unreadable Edit Windows. It also makes for unmanageable and messy region lists. Of course what UCS has done to clean up regions lists in the first place is brilliant. I wished I had had Tim Nielsen as my librarian forty years ago! I wouldn't be spending weekends like this one going through 200,000 files trying to get them to conform to some kind of usable protocol.
Nathan Smith:
Rene Coronado
ok, here’s my approach:basically there are three types of recordings I’m generally capturing -opportunity recordings
specific purpose recordings
library recordingsto me opportunity recordings are things like weather events, building implosions, unique access events like helicopter training, race cars on a track, etc. Basically any thing that I have one-off access to that I won’t likely be able to get again in the future.—For opportunity recordings I bring as many channels and mic types as possible, generating every inch of coverage that I can. Sometimes when things are very short notice (like weather events) what’s possible is still limited, but when I have any amount of prep time its ALL coming with me. I try to have a variety of pickup patterns and capsule types as I’m generating coverage, so its common for me to pack a set of shotguns, some PZMs, some omnis, a dynamic or two, and an ortf pair - and always my D50 and senn binaurals as well. I vary my placements and even aim some channels away from the source to capture reverberation as the source instead. In post I often do a lot of phase alignment and stacking of channels, making decisions and committing to them to generate a whole that is my best attempt at being greater than the sum of its parts. this technique is best exemplified in my Porsche and Helicopter libraries. https://echocollectivefx.com/product/the-porsche-911-gt3, https://vimeo.com/98349421?embedded=true&source=vimeo_logo&owner=9031199—For specific purpose recordings I’m generally going minimal and I’m hyper focused on the coverage that I’ll need in order to get the sound I’m looking for. In this headspace I already have a good sense of what I’m building sonically, and I’m just here to gather source material with which to execute. I’m not generating extra coverage, I’m not too worried about how this will look to anyone else, and I’m not usually bringing a huge kit to the project. I’m just 100% focused on getting the exact ingredients I need and that’s it. An example here would be when I had the Bally’s sports graphics package redesign in front of me and I needed some specific metal elements to feed into sound particles for a looping texture I was building. I got the props and a mono mic, performed the moves directly into the session, manipulated them to tasted, evaluated, threw it all away (lol) and did it again with diff props. These types of recordings often don’t get filed in the library, though I WILL often take the final stems and drop those into the library when a project wraps, so they live on in that way.—For library recordings like anything for a crowdsource, anything for echo | collective, and anything for our internal libraries, I’ll take a more structured approach.first I’ll get (or create) a shot sheet and from that I’ll figure out how I want to mic the thing. Occasionally this will require more than a single stereo setup, but often it will not. From there it’s all about coverage and performance. Most of my mental energy here is spent manipulating whatever it is I’m recording - making sure to get takes that are as varied and interesting as possible. slates and external noises are carefully observed so that post can be as efficient as possible.
Comments
Post a Comment