By: Jesse Korosi, Director of Workflow
On more and more jobs, we are starting to see the need for file based pixel accurate rack leaders.
As you have probably seen if you work in a camera department and/or post production, rack leaders (or “Framing charts” if you will) are often shot during the camera prep. Here is an example from a show shooting anamorphic:
I am still a huge supporter of shooting these charts as they are very valuable! However, the issue with this leader is that it will never be pixel accurate. For most TV shows or lower budget features, this has not been an issue. However, having the ability to ensure every department is working with the exact same frame, down to the pixel, has really proven to be very valuable for some of our clients!
I guess something to ensure everyone reading this understands before I continue would be, who is it for? Why do we need it in the first place?
Let’s say we are shooting with the Arri Alexa, 1920×1080 and we’re delivering the show at 1920×1080. There is a 99.9% chance you are just framing for the entire sensor. Therefore, the entire area recorded will be what you use in post production…simple! So your framing chart may look something like this:
In this situation it isn’t actually critical; we often don’t need a rack leader considering the direction to each department would be to just use the entire frame.
However, let’s say you are shooting with a RED, or an ARRI at a higher resolution…maybe 4K, maybe 5K, etc., and you are only finishing and delivering the finished job in 1920:1080. Now that you have all of this extra resolution, you could choose to frame for a “safety” area. Therefore, maybe it’s 10% punched in from what you are recording. If by chance editorial needs to stabilize a shot, or slightly reframe, now they have some extra wiggle room with that 10% you gave them! Here is an example for what you would see on your monitor on set:
This would be the entire image you see on your monitor on set:
Within your monitor you would actually have “look around” area. This is area that will NOT be recorded. Not as important to this blog in regards to framing…but something to be mindful of if you have this look around turned on in your monitor!
This is your recorded image:
Now this is the important one. This is what they were actually framing for on set! Notice that it is 10% punched in from the green.
Now aside from wanting the safety area, another reason it’s important to shoot a framing chart when having any customized punch in or re-frame, is because we need to ensure every department is working with the same area!
Now that we have shown those images above, remember that the final recorded image will just look like this (excluding the center cross-hair of course):
Therefore, when working with this recorded image, you will now NOT see where those frame lines were. My reference above was also a simple 10% punch in. Pretty easy to do the math there. Sometimes the re-frame may not be so simple. I have been on a job when there was a punch in for safety, but that punch in was not centered… and then there were separate lines for any 2.40 delivery and separate lines for a 1.78 delivery!
So again, our goal is to ensure that every department will be cropping in the same amount and therefore looking at the same frame.
A few examples of this may be that, once VFX gets their shots, whether that be converted or the RAW, they need to ensure they know where this punch in is as well considering they are ONLY going to work within the “framed for” area.
Once VFX has completed their work, different jobs will have different workflows for how these shots are returned as final images to your finishing facility. Some people may deliver the image now with mattes, matching the original resolution. Therefore, just to show an example of how this may look coming back from that original frame I showed above:
Then again, some jobs may crop that matted area and just deliver the usable frame:
Now we hit our final conform/online.
The facility (often different from each of the others in this chain thus far) now needs to also conform back to the master files, which means they need to know how to crop in and re-frame the image.
Now, to come full circle…even with these leaders, you would have seen from the samples above. The arrows will be stuck on with tape and hopefully be close, but the edge of the arrow is often cut off on the odd one and even the most skilled crew will never get these pixel accurate. So…
The Arri, as an example, creates an XML file when you make your framing choices IN CAMERA. This XML can be given to your lab and they can then take these numbers and make a pixel accurate leader within Photoshop. Ideally, you would have one image for every camera you are shooting with. Here are a few examples:
Those were quite compressed to get up on this blog, but the idea would be they would match the exact resolution of the camera, they have a grid, they show the specific active pixels, the arrows are bang on to the exact pixel, and it’s easy to tell what the intended frame versus recorded area is.
Also, you can take it to the next level and make one for what VFX is going to deliver back to the online so everyone understands exactly what’s happening here as well. Here is an example, if they were then going to scale the above frames down into a 2K deliverable:
So in the end, with this method using a photo-shopped leader, we can ensure that everyone is cropping in to the EXACT same pixel as each other!