I’ve been continuing to post five minute tips videos about DaVinci Resolve via the Ripple Training YouTube channel; folks have really been liking them, so I’m inclined to continue doing them, and I thought it worth giving everyone a reminder that they’re there. Here are most two most recent videos that have gotten the most eyeballs. Enjoy!
There’s a Tumblr making the rounds called “Shit People Say to Women Directors.” It’s worth reading to see what women in our industry are having to put up with. It’s ridiculous, in this day and age, that anyone can make these sorts of comments with a straight face. I’ve spent my entire career, from film school through my various jobs in post, working with a variety of talented directors who happen to be women, and the notion that gender imposes any kind of limitation on the job is ludicrous.
Put more bluntly, there is no shortcoming I’ve seen ascribed to women directors that I’ve not also seen exhibited by male directors. I know from personal experience, as a director of one one feature and several shorts, that directing is a grueling gig. At the end of the day, it’s preparation, experience, creativity, and character that separate good directors from terrible ones.
I studied theater arts with an emphasis in film production at U.C. Santa Cruz, and of the professors I considered very influential, two were women. Deborah Fort was a visiting film production professor, whose critical eye and ability to articulate the importance of taking responsibility for the images in your frame stick with me on every project I direct. Marcia Taylor was a formidable directing and acting professor with vast experience, whose practical advice on stagecraft, and direct critiques of my various directing exercises drove me to work harder and prepare more rigorously; when she told the class that “every production you undertake as a director will require everything you’ve ever learned,” she wasn’t kidding, and I find this to be true even 25 years later.
If my memory serves me correctly, our film program’s small classes were somewhere around 75% male and 25% female, and I fell into to working alongside many of the women in my class on their projects; the nature of the program was that everyone did a little of everything, so I worked as a student on several woman-directed projects, and they worked with me on mine. Not once did I ever feel that the women were somehow less talented, in-charge, or in any other way less capable. We were all in it together, and good work (and tedious work) was exhibited equally by everyone.
Moving to San Francisco where I started my postproduction career, I encountered many women directors at the Film Arts Foundation and Bay Area Video Coalition, both organizations of which were dedicated to enabling work outside of the mainstream. As an editor, then later as a broadcast designer, I worked for many women clients, directors and producers, on many varied productions, and looking back I find no generalizations worth making that relate to gender.
Moving later to Los Angeles and Manhattan, where I completed the metamorphosis of the post-production part of my career into a colorist, I worked with many more women directors. And transgender directors. And of course male directors. The good ones were good because of preparation, experience, and creativity. Gender, in my experience, played no role in who was great to work with, and whose work I thought was solid.
And not that I need any more examples, but I’m married to a woman who’s an extremely talented director, with whom I’ve worked in both production and post on two shorts. With twin backgrounds in acting and art direction (she’s been the production designer on all of my recent films, as I’ve been the editor and colorist on hers), she comes at the craft from a different skill-set then I do, and I often find myself envious at the comfortable way she’s able to work with her actors.
If you think that men make better directors then women, you’re wrong. And if you’re lucky enough to get a job on a film, as tough as this industry is, and you can find yourself able to insult or demean the woman who’s directing because of her gender, then you should be fired. It’s the 21st century, and well past time to leave this kind of baggage behind us.
If you’re a DaVinci Resolve user who’s about to grade a project shot with any of the Blackmagic Design cameras and recorded using CinemaDNG raw, you should seriously consider upgrading to Resolve 11.2.1 if you haven’t already started your project (if you’ve already started, then don’t upgrade yet, as a matter of principle).
One of the more significant updates is an improvement to CinemaDNG debayering inside of Resolve, which will improve any CinemaDNG raw clip being debayered to Rec. 709 regardless of when it was shot or what Blackmagic Camera it’s from.
If you open the CinemaDNG option of the Camera Raw panel of the Project Settings, you’ll see the following Camera Raw controls. The new features are the “Apply Pre Tone Curve” and “Apply Soft Clip” checkboxes.
The “Apply Pre Tone Curve” checkbox, which is on by default for previously created projects, exists to preserve backward compatibility with the previous method of debayering for projects you’ve already graded (in which your adjustments depend on the original debayered output). However, I’m also told that the Pre Tone Curve setting may look better with CinemaDNG raw files coming from other sources, so if you’re importing .dng media from cameras other then those from Blackmagic Design, you can always try toggling this on to see which type of debayering you prefer.
If you turn this checkbox off, or if you create a brand new project in Resolve 11.2.1 or higher in which this checkbox is off by default, then you’re using a newer, nicer-looking method of debayering for all CinemaDNG media in your project. Here’s a comparison of the old vs. new debayering:
Additionally, there’s also a new “Apply Soft Clip” checkbox (it’s off by default). With this checkbox turned on, you cannot reenable “Apply Pre Tone Curve,” as this is an optional function of the new debayering, that serves to bring high dynamic range parts of the signal (super bright highlights) back into the picture as image detail you can see and adjust, similarly to if you’d used the Highlights control to retrieve this part of the signal. Here’s a comparison of a snowy wide shot using the old debayering, versus the new debayering with Soft Clip turned on.
With the current project I’m working on, shot using the Blackmagic URSA camera, I’ve found the new debayering settings immediately provides an image with denser shadows, more neutral color balance, and richer color; an image that’s easier to grade right off the bat when debayering straight to Rec. 709. Obviously this is just a starting point, but a better starting point reduces the number of steps it takes you to get to the final image you want. This is a welcome improvement that, frankly, makes the raw output of all Blackmagic cameras look even better right from the get go, no matter when the material was shot.
I received a question from Anthony that reflects similar questions I’ve gotten from other folks around the world, paraphrased here:
I was wondering if you could answer a question that I’ve been wrestling with… Is any one software better in terms of QUALITY for color correction? …For instance, if I were to use a curves adjustment and then a three way color corrector theoretically using the exact same settings, would quality degradation be more or less prevalent in after effects, premiere, colorista, resolve, or speedgrade?
Everyone knows that I consult for DaVinci, and that Resolve is the software I use for both my client and personal work. That said, I’ve used many apps, have written about several, and as far as I’ve seen, you can do high quality work with the current version of just about every image-processing postproduction application on the market, be it Resolve, Baselight, Rio, Nucoda, SpeedGrade, Scratch, After Effects, Nuke, Symphony, Smoke, Flame, Premiere Pro, Final Cut Pro X, or plugins such as Magic Bullet, Colorista, what have you.
In terms of image quality, one could certainly choose to nitpick the quality and/or feature-richness of a particular operation in one application versus that in another; comparing implementations of curves, or noise reduction features, or qualifiers. But, minutiae aside, I honestly think that when comparing the basic image processing features that comprise the bulk of one’s day to day adjustments, you can get where you need to go in ANY of the above solutions and produce a high-quality image. All modern software that I’ve seen has perfectly good image processing, and each application I mentioned can boast of projects in cinemas, on television, and winning awards in high-profile film festivals; proof that nobody is getting fired because of what they’re using.
That said, not all apps are created equally in terms of the efficiency of working your way through a project as a colorist, and everyone has their own opinion about what software they prefer to use in that regard. I’m not going to take any sides here, because the price/performance/efficiency dial has many positions, and you can do well with any combination of apps and plugins you like providing you have the time and expertise for that solution. Along these lines, I have four metrics you should consider when choosing a grading environment:
- One of the chief differentiating factors between today’s generation of color grading environments is the number of tools available to help you make focused adjustments, work more specifically, and hopefully more quickly. There’s a tradeoff between the time you’ll spend learning to use an expanded set of tools before applying them, but in theory, you’ll get all that time back working more efficiently.
- The best software for you to use also depends on your personal style of working. This often relates to what application you first learned to do color correction with, as everyone tends to favor the methods they originally learned; this usually depended on what features your first grading environment had, and which it lacked. Because of this, users new to a particular app typically spend their first month wishing that app worked more like the last application they used, until they start getting used to the methodologies and features of the new environment.
- The best grading solution may also depend on the type of project you’re working on, with some applications being perfectly fine for short form projects, yet cumbersome for longform.
- Lastly, the postproduction workflow you need to follow may mean that the most feature-rich tool is not actually the best, depending on how the idiosyncrasies of project exchange and reconform relate to the editing and finishing methods you’re using, and your project’s schedule.
In a more concrete example, grading inside of your NLE can ease your workflow in terms of not having to move your edited timeline between apps, and you can reap efficiencies by having your grades live with your edit, thereby avoiding reconforms when last-minute reedits are necessary. On the other hand, dedicated grading applications are made to work quickly, do more to the image, and speed up your grading workday, offering many more dedicated tools for color tweaking then the average NLE. Some grading apps work better when coordinating with specific applications in a regimented workflow, while others are designed to be more flexible in order to accommodate more unpredictable situations.
In making your choice, I’d say that your primary concerns should be efficiency of workflow (in your situation) and flexibility of grading tools. Whatever software you pick, the quality of the result is going to be more up to your ability then the application, pretty much all applications are capable of high-quality processing.
The scene I shot in December, “An Unwanted Job,” is coming along nicely. In an effort to eat my own dog food, I’m doing as much of the postproduction as I can inside of Resolve, from syncing the timecoded dual-source audio and making offline dailies from the original 4K CinemaDNG media, to editing both picture and sound, editing music and doing simple sound design, through adding visual finishing effects and doing the grade. Eventually, I’ll be sending the audio I edit to someone else for sweetening and mixing, probably in ProTools, probably via exporting xml to Final Cut Pro 7, and out to OMF from there, unless I find a more contemporary workflow.
I used the Metadata Editor in the Media page to annotate the Description, Shot, and Scene for each clip, and then I sorted the Media Pool in list view and organized the clips into logical bins for my use. Overall, the new Media Pool columns and bin organization features have made wading through material much, much nicer then in previous versions. For the smoothest overall experience I created a set of transcoded 1080p ProRes LT media with cloned file names and timecode to work with, making it easy for me to reconform to the original 4K once I start with the grading; I have separate bin hierarchies for the camera original and proxy media, and I can switch back and forth using the “Reconform from bin” command (after turning off “Force Conform Enabled” for each clip in the Timeline).
Assembling my timeline went swiftly. As I edited, I definitely found things that could be improved, particularly in the realm of audio, but I was easily able to accomplish what I wanted, and as I got into the details of the edit I’m happy to say that I like the trimming tools in Resolve as much as an end user as I did when demoing them in front of audiences. There are idiosyncrasies in Resolve 11.1.3 such as audio not playing in reverse (you can hear audio playback scrubbing or moving a frame at a time in reverse, just not at speed), or double-clicking to open a clip into the Source Viewer for trimming not working when the Trim tool is selected (you have to choose the selection tool first), but nothing stopped me, and I’ve passed these tidbits along to the Resolve team. One thing I’ll add is that easy access to Roll edits in selection mode make split edits in dialog ridiculously easy to perform.
After cutting a few different versions of the scene, I “soft-locked” my primary edit a few weeks ago. At the moment, I’m going back and forth on the music cues with composer John Rake. I’ve been deliberately having him compose longer pieces of music that I can cut into the timeline in different ways, which has been forcing me to do a bit of music editing in Resolve. Given that audio cutting isn’t necessarily what Resolve is designed to do at the moment, it’s actually been going nearly as well as it ever did back in FCP 7. I often try to avoid locking the edit until the music is finished, as great cues always make me want to push and pull things around. In this case, it’s a great excuse to push Resolve a little out of its comfort zone.
I’ll post a work in progress probably later in February, but for now I’ve started using examples off of my timeline in a couple of free tutorials that are now available on the web.
I’ve done the first in what is to be a series of “Resolve in Under 5 Minutes” videos for Ripple Training (you can subscribe to their YouTube channel), in which I’ll be showing a different technique or tip in every video. This first one covers image stabilization, which is a useful tool whether you’re an editor, colorist, or finishing specialist. For this, I used one of the first shots in the scene, a 20 foot remote-operated jib shot that was being buffeted by wind on a wintery day. Exactly the kind of shot this tool shines at improving.
I also did a webinar for Imagineer Systems and Boris Effects about how to use Mocha and Boris OFX plugins with DaVinci Resolve to do corner pinned match moving, motion-tracked lens flare addition, lens correction and dead pixel/wire removal, and other techniques that you either can’t do or that are otherwise difficult in Resolve by itself.
It’s time for some tough love.
Since I do both production and post, I’m in a position to observe where efficiencies lie all through the filmmaking process. In speaking with different people out in the industry, and comparing my practices with theirs, I’ve decided there’s a public service announcement I can make to the filmmaking community at large regarding how you can save money. A lot of money.
Rehearse your actors in advance of the shoot, dry run your blocking without running the cameras, and record fewer takes.
I know. You’re thinking “but digital is cheap,” “I don’t want the scene to go stale,” “why not run the camera in case some magic happens,” and “I need to get through those takes to find the scene.”
Bullshit. Digital is not cheap. Hard drives are expensive. Backups are expensive. Archiving camera cards takes time, and having a ton of cards because you want to shoot like a drunken sailor costs in either expendables or rentals. Generating dailies and logging all of your crap takes consumes hours and hours of your editor’s time. Dealing with Terabytes of unused data in the finishing process takes time. Time is money. Time is also frustration, which makes post folks not want to give you a discount. Less media is less hassle, and less hassle makes editors more inclined to maybe cut you a break.
Rehearsing in advance, assuming you have capable actors, will not make your scene go stale; centuries of theater give the lie to this way of thinking. What advance rehearsal will do is let you work through the scene in the abstract, without the necessary distractions of the cinematographer’s queries or your schedule running behind or what scarf should the lead wear or any other of the hundred things you need to weigh in on that have nothing to do with the actor’s actual performance.
In exploring the scene with your actors in rehearsal, I guarantee they’ll find nuances in performance that might otherwise be missed, and you’ll have time to try things out and make adjustments you need for consistency with your overall vision. All of this will let you know where there are moments to emphasize and trouble-spots you’ll need to correct in the performance. Ultimately, you’ll be better prepared for what’s to come. Come shooting day, you and the actors will both know what the goal is, and you’ll be less at risk forgetting about a key moment, gesture, or intention because you were distracted by a lens change. And, in the rush of production, there’ll be more then enough energy to hype up both you and the actors to keep things fresh, since you’re likely in the location for the very first time.
Running the camera while you’re rehearsing your blocking is a colossal waste of storage, unless you’re chasing a sunset or some other time critical element. Whether you’re doing jib, crane, dolly, gimbal stabilizer, or steadycam work, or just panning and tilting a camera on sticks, there is no magic while you and your actors are fumbling your way through the on-set blocking with a camera operator who’s seeing the actor move for the first time. Save your editor the heartbreak of watching four unusable takes and just rehearse the blocking without recording. The actor will be more at ease to experiment, and you’ll have more psychological freedom to change your mind and explore different ways of executing the move with your actor, cinematographer, and operator.
Regarding multiple takes, I am firmly of the belief that if you’ve had the opportunity to discuss and rehearse the scene with your actor in advance (even if it’s in advance of the rest of the crew arriving for a day’s shoot), there should be no reason to ever go through more then 4-5 takes, unless there’s something physically specific you’re needing (I once did eleven takes trying to get a coin to fall on the ground in the right place for a closeup, what are you going to do?). There are plenty of stories of directors pushing an actor through 99 takes to wear them down, which is more strategic cruelty and not so much about trying to get a good read, and that’s fine if you’ve got 100 million dollars. Myself, I’d rather work out what I need in advance through appropriate casting and rehearsal, and then get the scene in three takes and move on to have another setup. Back in my film school days, a production professor told us that shooting twelve takes was probably a waste of time, since you’ll probably be unable to tell the difference between six of the takes later on when you start editing, and I’ve generally found this to be accurate.
So now you’re thinking, “okay, smart-ass, so what’s your shooting ratio?” My most recent project, which was one day’s shoot with over twelve camera setups with some additional alts spread among two locations recording to 4K CinemaDNG raw, ended up totaling about 500 GB of media for 50 clips in all for a generously covered 3 minute scene. Going back to my short, “The Place Where You Live,” four shooting days recording to 4K R3D raw resulted in a total of 487 GB of media. Usually when I mention this, people ask me what went wrong, and what I did about the lost media. Then I say that was all the media.
I started out shooting 16MM film, and being a broke-ass filmmaker I had all too many days of having pages to get through on two 100-ft rolls. Those experiences instilled in me an economy that serves me well even today. And so, I try to be careful about casting. I rehearse in advance whenever possible. I block the scenes with the actors without running the cameras. Then at that point I try to get out of any particular camera setup in 3-5 takes, sometimes altering the framing of any takes I want “for safety” just to squeeze in another push-in. Obviously if an unexpected challenge comes up, I’ll do more takes to get what is needed, but I try really hard to be prepared enough in advance to avoid that.
The result is that I never have to debate whether or not I want to shoot uncompressed RAW based on available storage, my DITs are usually bored, logging the footage is a breeze, and as a result the edit goes quickly. Backing up the project media doesn’t break the bank, and moving the project around for finishing is resultantly easy.
So that’s my advice. Save everyone some hassle and take a page from the theater—rehearse in advance. You’ll get better results dramatically, you’ll get more out of your time during the shoot without going over schedule, and you’ll have a project that should be noticeably cheaper to post.
Anyone who knows me knows I’ve got two main opinions about gear. First, don’t buy gear, rent it. Today’s awesome stuff turns into tomorrow’s laughably obsolete curiosities. Second, I could care less about cameras. I’m a director, not a DP, and as long as the DP likes whatever camera we can afford to use, and the resulting image data is gradable, then I’m happy.
So it’s with some reluctance that I admit I’ve started buying camera accessories to have in reserve for the increasing number of small projects I seem to be directing. Starting with my next directorial excursion, a scene rather then a fully realized movie, designed to create some nicer material to use for my next book, video training, and presentation projects, and in the process work with another great local Minneapolis crew to create something very different from my previous two movies for my director’s reel, which by god I’ll actually put together one of these days.
Today was all about prepping the aforementioned camera equipment. DP Michael Handley, AC Brian Suerth (pictured later), and I poured over the three Blackmagic Design cameras we’ll be using, a brand new Ursa for most of the coverage, a Production 4K that we’ll be using with a Ronin gimbal stabilizer to fly around one of the day’s two locations, and a Pocket Cinema Camera for POV shots in tight spaces. I’ve been wondering if I’d find utility in the three different form factors of cameras that Blackmagic now offers, and I now appreciate that it’s really nice to have different sized camera bodies for different tasks, each of which can be prepped simultaneously with different rigs. A far cry from my days of having a single CP 16 film camera to work with.
The Ursa itself is a beefy camera, but resultantly quick to set up thanks to an attached battery mount plate (extra), built-in rails mounts, and a built-in handle; I really, really liked the speed of setup. The editor in me wants as many angles of coverage as I can shoot in a day, and I hate feeling like I’m waiting for a camera to be rebuilt from setup to setup.
Since I usually prefer to have the camera on support rather then handheld, given the type of coverage I tend to go after, the Ursa’s weight (16.5 lbs without lens/etcetera) doesn’t really bother me—we’ll be using a jib, dolly, and slider to move things around when necessary. Of course, that’s easy for the director to say, I don’t have to move the camera from setup to setup, but nobody on the camera crew has had any complaints so far.
Incidentally, the 10″ swing-out screen is pretty awesome. As are the two other touch-screens on either side of the body that switch between UI and monitoring for the assistant. Brian mentioned he still may use his own display for working the focus, but more because it’s the display he’s used to, which makes sense to me.
We toured the menu system, and were pleased to find it pretty straightforward. So, at least so far as setup goes, the Ursa seems to be a no-drama camera. And, pointing the camera at random things at Tasty Lighting’s Acme stage, the EF mount Rokinon cinema lenses I’m using coupled with the Ursa’s Super35 sensor made the kitchen area and random lighting instruments positioned here and there look delightfully dramatic framed in 2.35.
So far so good. I’m looking forward to seeing how it performs on location.
It’s an old saw that if you want to write, you should read as much as you can. Every piece of work you consume adds to your mental inventory of artistic possibility, and it’s as true for image-makers as it is for writers. If you want to create memorable visuals, you owe it to yourself to see what others have done before you.
In my various travels, I always try to squeeze in a trip to a gallery or two; no matter how good a print or digital reproduction is, it will never equal reflected light off an original canvas. The last time I was in Nürnberg, Germany, I squeezed in a trip to the very excellent Germanisches Nationalmuseum, which has what has become one of my favorite paintings, Johann Liss’ 1623 “The Prodigal Son With the Prostitutes.” (Image thanks to TimSpfd at flickr)
It delights me no end that most of the techniques we seek to employ when shooting and grading cinematic images were pioneered and mastered centuries earlier by generation after generation of painters, sculptors, and fabric and glass artists. It’s staggeringly instructive to notice particular techniques made in completely different media that have direct application in your own work.
In the painting above, note the startlingly careful control of multiple planes of highlights. Not only is the overall image brilliantly high-contrast (although the photo diminishes this somewhat, in person the contrast is deep), but the highlights on each plane of subjects fall at progressively diminishing levels relative each figure’s distance from the camera. This control of highlight creates depth by keeping the most brilliant and crisp highlights on the couple nearest the viewer, while the highlights of figures farther away (the servers, couples in the background) are more muted. The one exception is the prodigal son himself, at the center of the room and thus behind the front-most couple; his cuff and sleeve are rendered with brilliant, sparkling highlights equal to those of the foreground characters, thus keeping him at the forefront of the viewer’s attention.
Negative space is utilized naturalistically by leaving the background to fall to shadow, with the exception of a few subtle details to keep the surrounding area from being too flat. Texture is preserved even as the distant wall and floor are kept both dark and desaturated, creating an organic vignette behind and offset from the foreground figures. I’m a huge fan of “intrinsic vignettes,” (for lack of a better term), where a natural pool of shadow or saturation can be used to guide the viewer’s eye, rather then an overlaid oval of black. Any time I can accomplish this by reinforcing and extending a forward-thinking cinematographer’s contrast scheme, I do, because it’s a much more organic way of shaping the visuals.
Similarly, saturation is also used to focus there viewer’s eye. Flesh tone is rendered with the most saturation on the faces and hands of the revelers who are central to the scene. Brilliant splashes of red and gold define the main players of this drama, while background players are rendered with more muted greens, ochres, and grays; even the skin tones of the background players are more muted, doubtless justified by the rendering of smoke lending a haze to the room.
Naturally, you’re only going to get this kind of look through cooperation with the cinematographer and art department (wardrobe and set dressing). This isn’t the kind of look that you as a colorist can simply create without some fundamentals first built into the image (insert joke here about an indie film’s white T-shirt wearing protagonist standing in front of a beige wall).
Still, given a careful shadow ratio with thought given to pools of light and shadow, and given the introduction of selective regions of color into the scene via creative costume and set choices, you can be provided with the opportunity to shape the image creatively using only careful control of contrast and saturation, both in your Primary grade and in targeted Secondary adjustments designed to push specific elements forward or backward in the scene. I also love using Hue and Saturation curves to sculpt saturation in specific ways. While one could easily play some games with selective sharpening or blurring of parts of the image to emulate the painters deliberate ability to over- or under-render the detail of particular individuals, no other tricks should really be needed.
Alas, my web site was hacked while I was away in Florida for the Fort Lauderdale International Film Festival screening of “The Place Where You Live.” In fact, my whole family of web sites were hacked, and a week passed wherein I could do nothing. Happily, I’ve fixed the issue and taken steps to secure the help I’ll need to prevent this from happening again.
From the very beginning of this blog, I’ve resisted placing ads for anything other then the fruits of my own activities (my books, training videos, and soon my films), and the perennial Amazon ad for books I have on my shelf. I figure if you’re here, you’re probably interested in what I’m doing and what I’ve created, but you’re not particularly interested in ads for colleges, cars, or whatever other random adverts might pop up. Despite the increased cost of running this site in a world requiring hardening from hacking, I’m determined to continue to offer this site as a free resource. It’s fun, and I can continue to justify my occasional and eclectic posting habits.
If you like the content and feel motivated to support my efforts with filthy lucre, consider picking up a copy of one of my books or training videos (through Ripple Training). You’ll support my extra-curricular activities, and learn something to boot! And if you’ve already done so, then thank you! I very much appreciate it.
I’ve been having an enormous amount of fun flying out to some of the film festivals we’ve gotten into, and meeting other truly independent filmmakers who are tilting at the windmills of this crazy industry of ours. While I’m generally just happy to have the opportunity to screen my work in front of actual human beings, I’m thrilled to announce that “The Place Where You Live” won distinctive achievement awards for Editing, Visual Effects, and Production Design at Des Moines Iowa’s Wild Rose International Film Festival.
It was my second festival in two weeks, and while audience response continues to be really positive, this kind of additional recognition is icing on the cake, and a nice bit of validation for all the folks who’ve worked so hard to make this impossibly ambitious project come to life. This is especially true as the Wild Rose International had an impressively strong lineup of films to show, including a pleasingly high number from other Minnesota filmmakers (shout-out to the Twin Cities!).
Huge congratulations to everyone who pitched in with me on the VFX, most of whom were tragically left off of the certificate (for reasons solely of length). However, I can amend that here by giving a huge shout out to Brian Mulligan, Aaron Vasquez, Joel Osis, Christopher Benitah, B.J. West, Brian Olson, Patrick Burke, and Marc-André Ferguson for his organizational talents putting the original Smoke-based VFX crew together. I couldn’t have done this without this incredibly talented team of artists, and if I were you I’d hire all of them.
I’m also excited that Kaylynn Raschke’s contribution as Production Designer has been recognized. An industry veteran stylist with over two decades of experience in commercial styling for print and broadcast, set dressing, and wardrobe, she went above and beyond with limited resources to design and assemble a fantastic pair of environments for the movie, each of which had a lot going on. Production design is all too often neglected in smaller productions, and I’m proud to say that’s not a problem we had.
Lastly, I’m very pleased to be recognized for the editing of this piece. While I’ve not had my shingle out professionally as an editor for clients since the late nineties, I’ve continued practicing the craft through cutting my own work, that of my wife Kaylynn (a filmmaker in her own right), and in doing the occasional bit of surgery on projects requiring finishing in addition to grading. It’s nice to see that I’ve still got it.
More festival screenings are coming up; go on over to whereyoulivethemovie.com for the latest listings and updates!
Given how much I’ve written about color over the years, from time to time I’m asked in casual conversation for my opinion on the photoshopping of models on magazine covers. This is an interesting topic as it has significant overlap with the process of color correction for cinema and video. It’s easy to find examples of the abuse of Photoshop, shaving pixels off of arms and thighs, blurring complexion until the face is a plastic mask, etcetera. And these examples may lead one to think that the digital manipulation of the people is an inappropriate or even morally questionable step, given how it skews the representation of models, creates unrealistic body-image norms, and generally gives readers and viewers an entirely wrong idea about how people look. The following two grades (using a clip that accompanies my Color Correction Handbook) show what I’m talking about.
This is a worthy topic of discussion, because I believe all those things to be true. However, in an increasingly raw-shooting world of media production, it’s also true that just about every image coming out of a digital camera needs to be adjusted to look its best. There are few cameras that put out a perfectly usable image right off of the memory card, and even in those cases it takes a talented DP to produce nice images. Even so, raw or log-encoded images typically require some decision-making in terms of how you want to produce a human-viewable image, and in cinema and broadcast, those decisions are going to be subjectively made by colorists, who are hired because they have (supposedly) the good taste to make the image look great, and the good sense to not do anything stupid.
And so, knowing on the one hand that all digital images given to me will benefit from some adjustment, and on the other hand realizing that overcorrection will encourage bad habits in an industry that is all-too-often guilty of encouraging unrealistic ideals, I’ve developed a simple rule that I try to adhere to when grading performers in a project.
Don’t make any correction to the color, contrast, or texture of someone’s complexion that couldn’t have been done by a makeup artist doing a naturalistic job.
This is the main rule I live by in the grading suite, and that’s the rationale I use with clients who want to push me to do more. Usually this explanation suffices. Sometimes it doesn’t and the client who’s paying my bills pushes me to go farther anyway, but I’m lucky in that being the exception rather then the rule.
That said, there are some adjustments that I do feel more comfortable making more aggressively. Pimples are obviously temporary, and no actor alive would keep me from removing such a blemish if they’re unlucky enough to have one appear during a shoot, which is easily done using a small window, some tracking, and a bit of blur (or using digital paint if you’ve got a tool that lets you clone one small area’s pixels onto another region).
In another example, actors who tend to freckle can sometimes end up looking craggy in grades where you’re deliberately upping the contrast for dramatic effect. In these instances, reducing the intensity of the freckles in the skin tone of the grade using a bit of blur, noise reduction, or other techniques (such as pulling Resolve’s Midtone Detail parameter to a negative value) actually brings those actors closer to how they ordinarily look in life.
Likewise with bags under the eyes due to exhaustingly long hours; I typically don’t annihilate them, but they’re easy enough to minimize using a couple of windows, some motion tracking, and a combination of slight contrast adjustments and very slight blur. I did this very thing for my short, “The Place Where You Live,” in one of the opening shots. It was a long day’s shoot, and the makeup artist wasn’t aware just how close in we’d be pushing into her face since we were squeezing a couple of shots into the schedule in a hurry.
A slightly tougher call is on reality TV shows where you’ve got people on-camera with blotchy complexions. It’s one thing to adhere to the realistic portrayal of people, but consider that you probably have someone who wasn’t expecting to be on camera, who may not have the best (or any) hair and makeup provided to them that day, who are now being scrutinized in a closeup by an unforgiving lens and an even less forgiving camera format. Blotchy complexions (typically uneven patches of slightly redder hues in the blush areas of the face) are easily minimized using Hue vs. Hue curves to push the blotchy areas closer to the hue of the rest of the face, or by qualifying the face, desaturating it slightly, and then using the color balance control to push the subject’s skin towards the ordinary hue of their face. In these cases, it’s never a good idea to eliminate the blotchy complexion, that will look unnatural. Instead, you want to minimize the problem complexion such that it’s still there, but its importance is reduced within the framing of the image.
And that, to me, is the important thing for both colorists and directors to keep in mind; the best correction in my view is one that minimizes, rather then eliminates, perceived issues with a performer’s on-screen complexion. Barring some dramatic requirement (a sick or tired character, horror movies), you want performers to look as they normally would on their best, most well-rested and relaxed day. Even (especially) if the requirement is for an idealized grade, you don’t want them to look like a blurry-skinned sparkle-zombie.
At the end of the day, I think it’s important to integrate a more realistic portrayal of people, that doesn’t give the audience unrealistic expectations of what folks look like, with the need to do quality grading that puts the performer or subject’s best face forward. A clear image of a performer or interview subject combines careful lighting, hair and makeup, and sensible choices of lens and shooting format. Should any one of those be deficient, the Colorist should be prepared to step in and make a compensating adjustment, but that adjustment should ideally go no farther then was would have been realistic on the set.
I was fortunate enough to be hired to consult with Jennifer Mendes, the colorist at Lisbon post production facility Loudness Films. Comprised of four principals (Branko Neskov doing sound, Pedro Ribeiro doing editing, Jennifer Mendes doing color, and Nuno Oliveira taking care of the business side of things) this two-story facility has fantastic mixing, recording, and grading rooms designed by Joules Newell (Newell Acoustic Engineering) whom I also happened to meet as he was working on the addition of new audio editing rooms.
The mixing and recording rooms are absolutely top notch, Dolby certified, and palatial to an indie filmmaker like me. Audio editing and mixing veteran Branko Neskov played excerpts from a few different projects for me, and the room sounded fantastic (the large-screen projection wasn’t too shabby either). They’ve also got a secondary recording stage large enough for a band, and fully equipped with foley pits and props for cinema work.
The accompanying secondary mixing room is also a great audio suite in its own right.
Of course, I was there to work with Jennifer Mendes in the color suite, which is fully equipped with 2K Barco projection, a Doremi digital cinema server, a Sony OLED secondary display, and DaVinci Resolve on OS X driven by a set of Tangent Element panels and connected to the facility-wide SAN.
I particularly love the glass-walled machine room just outside the grading suite. Not only does it get the projector out of the room and keep the gear cool, but it looks fantastic, and shows off the equipment nicely to clients.
Jennifer and I spent three days going over all her questions about workflow, Resolve operational details, and grading strategy. We used some of the projects she’s been working on as example footage, and judging from her work she’s got a great eye; I was happy to compare methods and share what I could. Overall, I was very impressed with the quality and variety of the work I saw being done in Lisbon for both cinema and broadcast.
The folks at Loudness Films definitely have a sweet gig; Lisbon is a fantastic town, and the low overhead they’re able to maintain in Portugal makes them a compelling choice for cost-conscious filmmakers interested in working with them remotely to finish a project, or even for post artists flying in to rent the facilities to use themselves while enjoying Portugal’s unbelievable cuisine between sessions (and seriously, the food here is fantastic).
Being able to see how filmmaking and post are done in other parts of the world is one of the things I love about traveling, and from what I’ve seen in Lisbon I hope to be back someday.
Great site – which I have just discovered. I am a technician at a UK university and we have recently made the move to shooting on Blackmagic cameras and using Resolve. You seem to be one of the few people going into depth about editing in resolve 11 – and I wondered if I could ask some advice. Is it now feasible to work completely in resolve 11? I am writing a new workflow and even though we also teach Avid and Final Cut – I thought maybe now is the time to actually teach editing and grading in the one package. Is this covered in your tutorials? creating and editing with proxies all with da vinci?
To answer your last question first, my brand new “Editing in Resolve 11” title from Ripple Training is completely focused on how to edit in DaVinci Resolve, walking you through how to bring media into Resolve, organize it for editing, and cut and trim it into an edited program complete with transitions, composites, and other effects. There are a few lessons included that cover grading for editors, which are designed to give an introduction to those tools for folks that don’t know grading, but the overwhelming majority of the videos are all about the various editing, effects, and audio tools available in Resolve’s Edit page, and how they’re designed to be used together.
Now to answer your previous question. Yes, I consider it completely feasible to edit a project from scratch inside of Resolve 11. Obviously I’m biased since I helped design the feature set, but I’ve been using the editing tools as long as they’ve existed, and have cut a few very short projects with them, and I’m very happy cutting in Resolve.
Of course, the cool thing about Resolve is that it also has extensive support for importing and exporting XML, AAF, and EDL project exchange files between just about every NLE currently in use, so you can mix and match NLEs with your Resolve workflow in any way you want. But, if you want to take advantage of Resolve’s ability to let you cut away in the Edit page and then, with the single click of a button, start grading in the Color page, going back and forth as you please cutting and grading the same timeline within the same application, you’ve got a nice editing environment with which to do so.
Furthermore, Resolve 11 editing is based on an editor-friendly source-record style paradigm, with strong track management in the timeline that makes it easy to segue from craft editing into finishing. You’ve even got the ability to customize the name of each track. Bottom line, editors from other environments won’t have to relearn everything to start cutting in Resolve, and beginners will find a nice, clean UI that I consider to be very approachable.
However, in the spirit of complete honesty, there are a few caveats you should be aware of.
- There’s no multicam editing. If you require multicam, I recommend using FCP X’s wonderful multicam tools, and importing the result into Resolve via XML for finishing (works like a charm).
- Resolve 11’s current audio tools are a bit sparse. On the plus side, Resolve does have keyframable clip level overlays, multi-channel 16-channel adaptive timeline tracks, individual channel muting in source clips (via the Clip Attributes command), multi-channel waveform views in the Viewer and timeline, and a track/clip level mixer with assignable channel routing for both digital delivery and tape output, and crossfades. On the minus side, there are no audio filters, audio mixing cannot currently be automated at the track level, and there’s currently no way to export AAF to ProTools directly. However, you can export XML to FCP 7 and then export an OMF from there to ProTools (I’ve done it and it works).
- Media management in Resolve doesn’t work the way it does in other NLEs. That’s not to say Resolve doesn’t do media management, in fact it has a wealth of media management features, but they’re accessible in different ways, and they require some reading of the manual to get a handle on if you’re used to other applications.
- Given Resolve’s continued emphasis on top-quality, 32-bit floating point precision in all of its processing, even the editing tools benefit from the highest performance GPU you can give them. In particular, if you’re planning a classroom full of iMacs, getting the top-of-the-line GPU option is the best way to go (there’s an updated configuration guide if you want more information available at the Blackmagic Design support site).
Keep in mind that this is only DaVinci’s second year of adding serious editing tools, so there are bound to be small features here and there that you may find missing if you’re used to other NLEs. However, the team worked hard to put together as complete a set of editing tools as 24 months of arduous work has allowed, and there has been a lot of thought put into the current set of features to make sure the tools are robust and work together elegantly.
All of the basics are there including full JKL transport controls, absolute and relative timecode navigation and trimming, source-timeline viewer ganging, three-point editing, insert/overwrite/replace/place-on-top/fit-to-fill edits, a fantastic and complete set of trim tools, timeline and clip markers with multiple colors and notes with optional marker rippling, multi-clip selection with select all clips forward and backward commands, compound clip creation and editing, multi-take clip management in the timeline, per-clip transform and compositing controls, linear and variable speed effects with optical flow processing, keyframable effects with an in-timeline curve editor, paste attributes, some really nice media organization tools in the Media Pool, a filterable Edit Index that you can use to list all your markers, offline clips, through-edits, etc., and a great “Smart Cache” system for automatic render caching of processor-intensive effects. Obviously there’s much, much more to the Edit page then I can describe here, but these highlights should give you a good idea of how much there is to be found.
Furthermore, every function has been designed to work well using either the mouse, or via extensive keyboard shortcuts. An excellent example of this is the simple ability to add transitions. Using the mouse you can right-click on an edit and choose one of four different timings of the current standard transition. Or, if you select an edit by pressing the V key, you can use the U key to choose which side of the edit is selected (incoming, outgoing, or center), and then press Command-T to add the standard transition at the incoming, outgoing, or center of the edit, whichever is selected.
And of course Resolve’s extensive format support, support of mixed frame-rates/frame sizes/codecs within a single timeline, and extensive project import/export support for multiple XML, AAF, and EDL workflows makes it easy to use Resolve in an incredibly wide variety of workflows, including converting project exchange formats and exporting to just about any media format you’d want to.
Oh, and of course you can switch from editing to using Resolve’s incredibly deep grading environment with the single click of a button. No reconform needed, as you’re working on the exact same timeline in both pages.
While my current title for Ripple Training focuses on Editing,
they’re working on my already-recorded “Grading in Resolve 11” title which is a completely updated grading title. It’ll probably be available in a month or so there’s a separate title on Grading that’s now available.
With all this said, I’d absolutely recommend downloading the Lite version (for free) and checking it out for yourself. The best way to get a feel for Resolve’s editing is to get your hands on it. I think you’ll like what you find. And don’t forget the newly updated Resolve 11 User Manual, written by yours truly, that comes along with the app (it’s installed in the same folder as the Resolve 11 application). The Edit chapters are totally revised, and worth a look if you want to understand how everything works.
I completed my science fiction short, “The Place Where You Live,” about a month ago. I would have spent more time crowing about it, but I almost immediately launched into the final work I had to do for the pending Resolve 11 release. However, since then I’ve been doing all those other things you need to do once you finish a film; entering festivals, creating press materials, writing blurbs, building a web site, making postcards and business cards, budgeting for screeners and deliverables, choosing a shirt to wear to the premiere, etcetera, etcetera.
The truth is, you’re never done, but you do manage to finish some things along the way, and today I completed the brand new home page of the film, located at www.whereyoulivethemovie.com (which was as close as I could get to the title with an available domain).
There’s lots of information about the movie, including a fantastic trailer edited by Monica Daniel (she of Shitting Sparkles). I’ll cheat and let you watch it here if you promise to check out the web site later…
There’s also (drum roll please) news about THE FIRST TWO FILM FESTIVALS THAT HAVE SELECTED TPWYL! Upper caps because I’m thrilled to have good news to share so soon. The Fort Lauterdale International Film Festival in Florida has selected us for their 2014 festival in November. And the South Dakota Film Festival has selected us for their upcoming September 25th-28th screenings in Aberdeen, South Dakota.
I’m delighted to be included in both, and hopeful that I’ll have more news along these lines to come, as I’ve been told that each additional laurel wreath of film festival acceptance brings a fairy back to life (zombie fairies being an as yet untapped corner of two genres, no less).
The current plan is to run TPWYL through the gauntlet of whichever festivals will have us through 2014 and part of 2015. Until that time, I can’t post it freely on the web as many festivals have prohibitions against that. However, once festival play concludes, I’ll be posting the movie for one and all to watch and enjoy. Until then, keep an eye on this and the movie page, as well as my twitter feed, for news of any possible big-screen experiences coming to a film festival near you.
I’ve had a lot of fun working with DaVinci this year, and version 11 a big new release that expands editing, improves grading, and makes nearly every workflow better. While I’ve been nose-to-the-grindstone finishing my film, “The Place Where You Live” for the last two weeks (I finished the last of the VFX yesterday), I’ve continued to keep pace with the DaVinci development team as they’ve been putting the final touches on today’s giant new release of the DaVinci Resolve 11 public beta.
If you’re a current Resolve user, or curious about what Resolve can do for you, there are public betas for both the full dongle-protected version, and the free Lite version. And I need to point out that nearly every feature I talk about in this article is available for free in the lite version. Both can be obtained at a brand new support page: http://www.blackmagicdesign.com/support/family/10.
This page also includes some videos showing what’s new, with specific looks at editing and grading in Resolve. However, if you were to ask me about my favorite new features, I would tell you to check out the following…
Editing, Editing, and Editing
One of the main themes of Resolve 11 is vastly expanded editing tools; you now have a video editor living directly alongside your grading environment, in which you can cut from scratch and immediately switch to grading with a single mouse-click. Or, if you’re like me, you can go back and forth between cutting and grading continuously, making grading tweaks to scenes right in the middle of your edit, creating quick matches when insert shots don’t look right, or creating that day-for-night look you need to make a particular scene work. I’ve actually cut a couple of short projects with these tools, and I think you’ll find the Resolve editing experience surprisingly robust given this is only the second year the team’s been working on it.
You need to check out the powerful trimming tools, including a fantastic implementation of dynamic JKL trimming (make a selection and hold the Command key down while using JKL), and the ability to disable tracks from rippling using the Auto Select controls. If you tried editing with version 10, rest assured that version 11 adds most of what you may have found lacking, including timecode entry for navigation and trimming, more JKL transport functionality, better keyframing and a new curve editor, improved copy/paste and option-drag to duplicate clips, Description/Comments/Keywords columns in the Media Pool, bin organization for timelines, trim start/end features and numerous trim functions, four-up trim viewer displays for slip and slide, key shortcuts for just about everything (including moving clips up and down among tracks) and editable key shortcuts, many improvements to the process of adding and modifying transitions, a new film style transition, an adjustable audio crossfade, new 16-channel capable adaptive audio tracks, a clip mixing mode in the Audio Mixer, title formatting improvements, compatibility with OFX transitions such as those in GenArts Sapphire, a find next/previous gap function, flag/marker/through-edit/offline filtering in the Edit Index, editable notes and colors for markers, a Paste Attributes function, new Compound Clip editing, and much, much more. Coupled with the multiple edit types, compositing and transform features, three-point-editing, speed effects, and draggable trimming that Resolve already had, these additions add up to a very nice experience. In fact, Editing is so big it’s now covered in chapters 6, 7, and 8 of the User Manual. And before you ask, no, there’s no multi-cam editing (I would point out that Final Cut Pro X has great multi-cam along with great import into Resolve).
And to reiterate, every single editing feature is available in Resolve Lite, for free, on both Windows and the Mac (there is no Resolve Lite for Linux).
Great New Grading Tools
There are many new grading tools, including the new Color Match palette for automatically grading a clip based on a color chart included in the shot, a new Sat vs. Sat curve that lets you precisely adjust the saturation of pixels in the image based on their level saturation in the picture, new LAB colorspace conversion within a node, vastly improved matte adjustment parameters in the HSL Qualifier palette, terrific new Highlights and Shadows parameters in both RAW and Color Match palettes for easily retrieving highlight and shadow detail in high dynamic range media, Color Boost and Midtone Detail parameters for creating adjustments similar to vibrance and definition, an Opacity setting for windows, improved automatic Color Matching tools for Stereo 3D media, updated LUTs, the ability to create multiple PowerGrade albums, Wipe, Split-Screen, and Highlight buttons at the top of the Viewer, new automatic Broadcast Safe settings, and UI improvements too numerous to get into here. Color grading has now been split into two chapters, 11 and 12, of the User Manual.
A New Take On an Old Tool, Groups
Also for colorists, the all new Group Grading features makes grading with groups easier and more intuitive than before. If you’ve avoided using groups in the past because they were too complicated to manage, give them another try in version 11. Creating a group enables two new modes in the Node Editor, Pre-Clip Group and Post-Clip Group, which can be used for creating node trees prior to and after the Clip node tree, both of which are automatically synced among each clip in the group. The Clip node remains separate from the group, allowing you to make individual per-clip adjustments. This way, you’ve got an easy way of creating one set of node trees that will ripple among the clips in the group, and a separate node tree that doesn’t. This feature let me remove a whole page of explanation from the manual because it’s so straightforward to use. Group grading is covered at the end of Chapter 13.
A New Render Cache
Whether you’re a colorist or an editor, all new Render Cache functionality lets you either manually or automatically (if you choose the Smart setting) cache source clip formats that won’t play in real time, cache Edit page timeline effects that are render intensive, and cache Color page nodes that are render intensive. Caching is done automatically and quickly, sneaking in cache processing whenever you pause working. Colorists can also turn on caching for a specific node in the Node Editor, which forces all image processing up to that node to cache, while leaving all downstream nodes live for editing. The format you cache to is user selectable (in the General Options of the Project Settings) and you can choose from among a wide range of video formats. Also, while exporting from the Delivery page, you have the option of choosing to either output the cached media, or force a re-render. Caching in Resolve is now a big topic, and full information can be found starting on page 97 of the User Manual. Not mentioned (yet) is the ability to delete your render cache, found at the bottom of the Playback menu.
Collaborative workflow (only available with the full version of Resolve) is a huge new feature that allows multi-workstation shops, both large and small, to have multiple Resolve users working on the same timeline at the same time. Setting up a shared Resolve project database to do this is relatively simple (there are complete instructions in Chapter 17 of the User Manual), and once you do so, an editor, a colorist, and some assistants can work together on the same timeline, at the same time, giving you yet another tool to manage those ridiculous client deadlines. Even if you’re a tiny boutique post house with two people, an editor and a colorist, you can set this up to use among your two workstations for the cost of only two licenses of Resolve.
A new clone tool in the Media page makes it easy to duplicate camera card media, volumes, or even individual folders, to one or more destinations, complete with checksum reports written to the destination.
There are even new features for delivery, including a new UI separating the output options into Basic, Intermediate, and Advanced sets of controls depending on how much customization you require, new H.264 one-pass encoding with user-adjustable data rate throttling and AAC audio encoding that produces fast and quality H.264 files, MXF OP1A encoding, IMF encoding for owners of easyDCP, and the ability to output clips of mixed resolutions at their original frame sizes when outputting individual source clips. All this and more is covered in Chapter 14.
Get It Now
These are just the highlights, there’s much, much more to this release than I can easily summarize here. It’s all covered in the beta version of the newly updated User Manual that accompanies the disk installation (the User Manual is now automatically copied to the application folder that’s now installed). The User Manual has been significantly reorganized, and as you can imagine there’s a lot more information in the editing chapters than there used to be. So, download the software, skim the User Manual, and give it a whirl. Integration between editing and grading has never been tighter, and while I’m obviously biased since I work with the DaVinci design team, I think you’re going to really like what you see.
And yes, as you can imagine, I’m hard at work on the updated version of my training videos for version 11, through Ripple Training. This year will be a total overhaul, which is a colossal undertaking, but well worth it. Stay tuned on my twitter feed (@hurkman) if you want to be the first to hear about it.