Beyond the looks of StartUp: LightBender’s favorite HDR workflow techniques

We were super excited to see that the StartUp series, one of our all-time favorite projects, is now available on Netflix. This has brought up so many great memories that we wanted to share some HDR workflow tips and tricks that enabled us to work faster, smarter, and deliver a great-looking show.

Shooting in high resolutions, VFX-heavy shots, tight schedules, and several different deliveries (including HDR) are some of the most common challenges when creating episodic features that comply with the major OTT’s delivery standards. However, with the right mindset, meticulous preparation, and keeping in mind the following five points, you will avoid some of the unnecessary stress and exceed expectations.

Let’s crack on!

Don’t clip your signal

Let’s start with the shooting. One of the most common and straightforward recommendations for exposure that will work well with SDR and HDR versions is not clipping your signal. Make sure you capture the full latitude of light, because there is nothing that looks worse in HDR than a full blown-out sky! Providing all the information to your colorists will give you way better options when it comes to grading. HDR is about texture, even in the highlights. So the more you protect them, the better. HDR is a process that happens 100% in post, so you need the best possible options from the get-go to avoid getting yourself in a bad corner later.

Besides the typical daytime exterior with a gorgeous sky, or shot with bright colorful features characteristic on HDR demos, many other scenarios can look absolutely beautiful in HDR. Here are a couple of examples:

1. Night shots with practical lights. Darker scenes with light sources create a really nice and colorful contrast, making the scenes feel more detailed and layered.
   

2. Shadowy shots with bright highlights: When you have a subject in the shade (or semi-shade) but still have some sunny spots, those spots really beam with life and light. They will absolutely FEEL like sunlight.
   

First SDR and then HDR

When designing the production and post-production workflow, the final delivery plays a crucial role, especially if you are dealing with high resolutions or different aspect ratios. In the case of StartUp aspect ratio was not an issue (the show is 16:9 throughout) but Sony wanted to ensure we had a full UHD workflow from beginning to end. So from the camera tests, to the conform, color, reviews, and outputs, everything was designed to be UHD as the prime format and all the other online and HD deliverables would be derived from that. In that regard, you have to do your numbers and make sure you allocate enough hard drive space and speed to allow everybody to work comfortably.

In the case of HDR, it depends on what color workflow you are going to follow. You can usually get better results and better collaboration when you start your work with the Rec709 version (without clipping the signal of course!) and then create the HDR after the show look has been approved. This has three advantages:

1. It allows people to work and visualize in a color space that is familiar to them, and it allows for more flexibility when it comes to viewing devices.
   

2. It helps ground the project’s look in a way that DP, director, producer, and studio are used to seeing… I’ve seen a few native HDR sessions where everybody starts second-guessing themselves, and at the end, the project suffers.
   

3. When you start the work on the HDR version, you already have a perfect reference for your Dolby Vision trim pass that everybody loves, so it really simplifies and speeds up the trim process.
   

HDR: Much more than just a brighter and more saturated SDR

Sometimes when people talk about HDR they just think about something brighter and more saturated than SDR, but in reality, the more dramatic change happens in the textures of some shots. Having a wider light and color range creates more differences between shadows and highlights, which in turn contributes to being able to see things more clearly, with more separation. The change doesn’t have to be super-dramatic for the audience to feel it… Some of the latest HDR shows don’t even reach very bright peaks for the most part, but just being able to go past the 100 nit limit of SDR adds so much more.

Think about HDR as a paint bucket. It’s a MUCH bigger bucket than an SDR bucket! But the fact that you have more capacity, doesn’t mean you have to use it all. It’s all about having the freedom to push it when pushing it helps the project aesthetically. You can still have your dark scenes with deep shadows and never go past the 100 nits. It’s all about what works best for the show, the scene, and the shot… And that’s why I feel so strongly about creating the Rec709 pass first. It gives a common ground for everybody to work on, without second-guessing or worrying. And then, when creating the HDR pass, everything becomes much more playful because the look is defined already.

How to save time (and money)?

My number one advice would be to start conforming as early as possible to maximize the time available for color. I understand this can be a difficult choice when you are working with certain platforms since re-conform and recovering color from one version to another might be challenging, but it really pays off in the end! In our case, we use SGO Mistika Ultima which, in my opinion, has the best conform tools and editorial flexibility available. We often design workflows on the premise that the show “will never lock” so we can both start color soon and give editorial as much time as possible to finesse the cut. This allows us to start discussing looks, spot potential issues, and start reviews earlier than anything else.

Another critical aspect that can save you a substantial amount of time is VFX handling, which leads us to our next point.

Take control of VFX!

If the show has VFXs, we export the plates and test the whole workflow to and from the VFX vendor as early as possible to spot potential issues. It’s important for the plates to be exported by the Color house since that’s the best way to guarantee color, format, and framing accuracy. Once exported, we deal with the shots in the color grading like everything else, so we can get ahead in color even if we don’t have the VFX shot back from the vendor. If all the color workflow is designed properly, once we get the VFX render, the color will match exactly. The shot might need some minor adjustment depending on how big the VFX has changed the nature of the shot, but overall the time spent in tweaking will be minimal.

The first part of the process, once we were getting the shots back, was to test the alpha channel. We would bring the shots into a timeline and push the color in several directions inside and outside the mask, to ensure it was not breaking, losing quality, or artifacting. If the test was successful, the shots were marked as approved by us and placed on the timeline. If the mask had any issue, we would kick it back. That way we made sure we wouldn’t find the problem while we were in the middle of the color session and with little time to react.

To sum it up!

If I had to condense all the recommendations to work in a show like StartUp in just a few points, these would be:

• Start work as early as possible.
  

• Test everything as thoroughly as possible.
  

• Talk with the deliverables department as early as possible to design the right workflow.
  

• Work from bigger format to smaller.
  

• Make sure you keep all the information and latitude throughout every part of the process, from acquisition to delivery, to avoid pinning yourself in a corner or losing quality.
  

• Set the look of the show as early as possible so other decisions can be built on top of it.
  

• Work intimately with other departments to reach the common goal of an amazing-looking show!

Five tips and tricks when grading an indie feature film

Many people believe that working on blockbuster projects is much more difficult than on smaller ones, but anyone who has ever been involved in both worlds knows it is really not like that. A $200 million movie usually has no shortage of lights, sets look amazing, the cameras and lenses are probably the best that can be, and you normally have a huge VFX team to fix anything that couldn’t be fixed on set. That makes color grading way simpler. Almost anything you do looks pretty and is more about subtleties, logistics, and many color versions and reviews.

On the other hand, smaller projects might have some serious light challenges, or they might not have been able to get the right lenses or enough cameras. It is also true that since they normally don’t have economic access to top-notch VFX to fix most problems, a colorist has to spend more time, be more creative, and have the “helping hat” on all the time to fix some of the issues on the suite. However, it can be super rewarding to get something that was very challenging and take it to the point where nobody thinks about the budget and can just enjoy the story, the scenery and not get distracted by all the issues you fixed.

Think about this: Once the movie is out in the world, nobody is going to think “oh, that style sucks, and those shots don’t color match probably because they didn’t have the budget, but I’m sure the colorist did his best”. It’s your name and your work out there, so make sure it represents you! If the time or budget really doesn’t allow you to do a good job (I’ve been there too) just don’t put your name on it.

One of the most important resources to have on an Indie movie is time. With time, you can pick it up whenever you are not crazy busy, and really try some cool stuff… That’s the kind of opportunity we had when we worked on the SyFy movie Atomica, directed by Dagen Merrill, and DPd by Timothy Burton in 2017. The movie didn’t have a set deadline and needed a lot of love for it to be what they had dreamed it to be. So after a few meetings with the team, we jumped to the challenge and took care of VFX, Color Grading, and Mastering.

1. Take control

In some cases, you will find yourself working along with seasoned professionals and robust workflows, but sometimes you might happen to be one of the most (if not THE most) experienced people in the post team. You should talk with all the members of the other teams to have a clear picture of what’s going on and where things stand. Remember: It is better to test everything, even if it feels redundant, rather than finding the problem too late in the process. Try to become a hub of information and give technical advice to whoever might need it, but do it humbly and respectfully. Nobody likes a cocky know-it-all regardless of how many big credits you have.

2. Set limits and expectations

Working on an indy project is sometimes a little bit like doing a job for a friend… You want to help your friend out, but you don’t want to be his slave forever, and also you don’t want to end up fighting with him either! In this case, it helps if you have the conversation about the amount and type of work as early as possible. Don’t rely on good intentions and “we’ll figure it out later”. Assign a maximum number of hours (even if you want to go over out of the kindness of your heart) talk about what happens when you go OVER those hours, and what happens when they ask you to do something that was not talked about before. Try not to have any loose ends.

On the payment side, always get paid in stages. I usually do: 25% before you start work, 50% when color grading is approved, and 25% before sending the deliverables. That way you are covered in case anything happens, and if they pull the old “What’s the matter? Don’t you trust me?” just blame your accountant because he got tired of dealing with unpaid invoices from the past.

3. Have fun and try something new

It might sound like a cliché, but if you are working on Indie, you should try to have as much fun as you can and make an effort to try new things and push the limits… Yes, I get it. The pay is probably less than ideal (and sometimes is absolute crap). Still, if you have already decided to take the job, you will have a much better experience if you approach it positively. You might get something else out of it besides money: nice images for promotion, a good relationship with producers, better knowledge of new techniques, figuring out an issue that you have come across multiple times… But of course, it will all depend on your relationship with the client and how much time they give you to do your work.

Sometimes it is difficult to try new workflows that need a lot of testing, or technologies that might have bugs on big productions, but usually you have more time when you are doing indy and, if something goes wrong, you can always stir the ship in the other direction. I always try to test something new when I do indy. That’s how I did my first ACES project, my first HDR project, my first animation project, and many other things...

4.- Saving the day

Besides color, do you like playing around with VFXs? Do you use After Effects? Do you like design? This could be an opportunity to let loose another creative side of you. As long as you are OK with spending the time, and learning something new in the process, you can have a good opportunity to design some amazing credits, or doing some artful split-screens to fix that terrible thing that happened on set… The team will be incredibly grateful and you will get to test yet another skill for the field!

5.- Build bridges

A friend of mine told me once:

“Most directors do short films because they can’t do features… But cinematographers do short films for fun!”

In other words, try to make your sessions and working with you an absolute delight, because you never know where your next gig is going to come from! I can trace 85% of all the work I have ever done here to ONE event. From there, people introduced me to people, and those people to other people, but it was that single event that skyrocketed my career. So if they are taking advantage of you, they are not clear on their directions, and everything is 3 times more work than what you talked about; just breathe… Collect your thoughts, and write a very polite and professional email to the producer(s) explaining the costs of the actual work needed. The more “surgeon-like” you approach it, the better. Passion and feelings are sometimes difficult to deal with (trust me, I’m Spaniard. I got two servings!) but still, you never know what bridge you will be burning… So build bridges FOR YOU instead of destroying them.

Star Wars - Episode VII: Advanced Workflows

Non-destructive workflows for color and stereo on multi-studio and multi-platform environments.

I can’t think of a better way to inaugurate the LightBender Blog than remembering “Star Wars: The Force Awakens” on the 5th anniversary of its release. It was one of the most memorable projects I’ve ever worked on, both personally (I’m a massive geek at heart) and on a professional level. In this article, I would love to talk a bit about the unique workflows we used through Post-Production.

Before Star Wars, I had the opportunity to work at Bad Robot on Star Trek: Into Darkness. That feature defined a lot of the workflow that we later used on Star Wars. The idea was to buy editorial as much time as possible by taking care of most of the finishing tasks locally at Bad Robot instead of externally at the color facility. Traditionally, editors work on the movie, and eventually a cut gets approved. Then you get the elusive “picture lock” where no more changes are made and you can focus on all the other finishing tasks like conform, color, sound, etc… Both for Star Trek and Star Wars we went as down to the wire as you can be, and I think that helped editorial, VFXs, and JJ Abrams create more solid, cohesive, and polished films. Here are some of the things we took over internally:

• VFX Reviews: All High-res VFX reviews, even with shots within the cut, were run and performed at the Bad Robot Theater, not an external vendor.
  

• Stereo Reviews: Similar to VFXs, since the movie was being converted for stereo-3D by StereoD, all the shots were updated, conformed, reviewed, and adjusted if needed internally. All in final resolution with no proxies.
  

• Internal color passes: Every time we had an internal screener for producers or the studio, either of some reels or the whole movie, I was doing that temp color grading internally at Bad Robot. That allowed for great freedom to discuss looks and spot potential issues way ahead of time, as well as being able to watch a more cohesive version of the movie.
  

• Final Conform (both 2D and 3D). We did full conform of the movie from film scans and VFX finals, both on 2D and 3D, and on different aspect ratios (2.39 scope, 1.85 digital IMAX, and 1.66 Film IMAX); this involved adjusting pan & scan on all the exterior shots, since the viewing area was changing and we needed to keep the focus on the action. These conformed reels were then sent as DPX sequences to final color grading at Co3 along with an EDL so they could cut the events and color-trace if necessary.

I want to clarify: This was not about taking away work from a vendor for the sake of it, but about creating a flexible and collaborative environment where we could optimize time and resources to help the movie be the best it could be. Making a movie is a tremendous collaborative effort between different people, departments, and companies, and we really wanted to make sure everybody could have the most time and resources to do the best on their specific task. To achieve this is very important to take into account the words on the sub-title of this article:

• Non-Destructive Workflows: The movie was alive. Changing constantly, updating constantly… Whatever the workflow used, it had to be flexible and reversible because everything could change at any moment, and if you pin yourself on a corner where a change creates a cascade of time-consuming events then you are at risk of losing a deadline if said change happens too close to the finish line. For this, it’s imperative to start designing workflows from Pre-Production and keep testing them and tweaking them as production evolves! A rigid workflow is not helpful at all.
  

• For Color and Stereo: It’s one thing to have to deal with a movie with thousands of Visual Effects, all the versions, and the final color grading of it all, but if that movie is also going through a 2D-to-3D conversion process, then it’s a game in its own league!! You have to define a very solid versioning and naming system, and have a conform system that allows you to move quickly between conforms. Everything has to work like clockwork because, eventually, something will fail… and at that point, you will have to be able to spot it quickly and either fix it or kick it back to be fixed ASAP. It is essential to keep very fluent communication with VFX and Stereo editorial teams and have ways to quickly version-check to make sure everybody is on the same page, and nothing falls through the cracks! (or through the crackers, as a friend of mine used to say)
  

• Multi-Studio and Multi-Platform Environments: You need to know who is going to be involved, what locations, what software, understand the limitations of it all, and figure out both workarounds and communication between all these (Once again, Pre-production and communication are key!!!) even the simplest thing, like one of the locations not having adequate internet bandwidth, or differences in file systems (Windows vs. Mac vs. Linux) can become a huge issue when you are on a tight deadline! In the case of Star Wars, we were lucky to be relatively close to most of the vendors involved, and had a quite seizable internet connection, so depending on the asset we would use physical hard drives with a runner going down the street, or services like Aspera to send and receive image sequences quickly and securely. As software goes, all internal conform, review, and temp color was done using SGO Mistika Ultima at Bad Robot. Final color grading happened at Co3 and they use Resolve, so we needed to make sure both systems were able to find common grounds. We did it by primordially relying on DPX sequences, EDLs, CDLs, SDLs, and some scripting to make sure everything was seamlessly transferred.

Here is an overall image of the Episode VII workflow taking into account all these principles:

Most of these processes we had to develop as we were working on the project. Nowadays, five years after release, a lot of these obstacles have been smoothed out with more powerful tools: A better understanding of CDLs, ACES, and other set-to-post tools, most software can accurately interchange AAFs or XMLs, and you even have initiatives like OTIO to facilitate scripting of timeline interchange between apps, and online tools like Frame.IO for accurate remote reviews.