Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Do you 'grade' prior to editing?

Do you 'grade' prior to editing?

  • No adjustments before the edit, grade after edit

    Votes: 34 41.0%
  • Give footage a quick 'look' before edit, edit, go back to original footage and grade

    Votes: 37 44.6%
  • Heavily adjust the footage prior to editing, edit, export this footage out after additional grading

    Votes: 7 8.4%
  • Heavily adjust the footage prior to editing, edit, go back to original footage for grade

    Votes: 5 6.0%

  • Total voters
    83
My professional workflow is Mac based, and to take advantage of the new color science, as well as get smaller "offline" files that can be delivered to editors who work off-site (rather than giving them R3Ds) it involves the following steps:

1) Flip through all the shots in Clipfinder, and do one-light corrections by loading them up in Red Alert! Use Clipfinder to lift & stamp RSX files between shots in the same scene that will all receive the same one-light correction.

2) Use Clipfinder to do a REDline batch render of all footage to ProRes 422 720p Quicktimes at half-res debayer, quarter-res if I have a lot of footage and am in a hurry. Give those to the editor.

3) Get an FCP project back from the editor, export an XML and roundtrip it through Clipfinder, tweaking the one-light if needed through Red Alert!, rendering out the shots in their entirety that are used in the project at full-res debayer, often to 1080p ProRes 422.

4) I then load up the roundtripped XML in FCP, and check the project. If there were a lot of long takes and I'm short on rendering time, I've probably rendered in-out points only, which often has problems when multiple sections of one R3D have been used in an edit, so I do some manual re-conforming.

5) Send to Color and do any secondary correction there. I know that is working with limited source material, but until pigs fly and Color supports the new color science, I find it a worthwhile tradeoff as long as my one-light got exposure and white balance taken care off, and I only need Color for secondaries.

6) Back to FCP to check everything, especially any crops, motions, and filters.

My personal workflow is on a PC, so for the stuff I do just for myself, I just drag and drop the R3Ds into Vegas, edit and grade to my heart's content at any step of the process as the mood strikes me, and export it when I'm done.
 
With Redcine-X it is much easier to get a good one light. In the past I would do basic contrast control and CC. Now I find myself producing images much closer to the final grade than before. Redcine-X is far superior for one-light ,IMO, but grading is still a must.
 
The term of course originates from the film dailies process and later applied to the telecine process and now to transcoding dailies. I take it to mean there is only one setting applied to the whole roll of film, if you will. I think this often meant the printer lights were set at default mid values, 25/25/25.

"Best light" always meant to me that one shot was fairly well graded, and that grade was applied to all the film, so your dailies or your inter-positive were getting a look, but that look was not tweaked from angle to angle.

I have asked for labs to give me a best light, setup to a grey card, as prep for a file based color correction when film is involved.

Thanks for the explanation Brandon.
 
Grade every shot, make sure the client already gets used to my look, after the offline it is usually presented to clients for approval, so I want them to see a very good offline.
I use REDcine-X + Tangent Wave and I am testing 3CP software together with their 3CP colorchart.
After the edit online and grade in Scratch from R3D with pro colorist.
I have never liked one lights, not when shooting film not with RAW.
Everything has to be perfect and the nice thing is that I got the control back from the telecine with RAW.
 
Since Red is a Raw camera EVERYONE is giving their rushes a 'look' for the editorial. They might just not know it or care which one is applied... Since that is the case why not either produce a technical grade to allow editorial to see what is in each shot OR push it all a bit towards where you want to end up..

Please hurry with a version of Monkey extractor that drives Red Rocket...

thanks

Michael L
 
I take it to mean there is only one setting applied to the whole roll of film, if you will. I think this often meant the printer lights were set at default mid values, 25/25/25.
Not exactly. A workprint with printer lights at 25,25,25 is generally called a mid-light print. It will not give you balanced color so it's rather worthless - I don't know if it is used for anything other than tests.

A one-light print can be timed to a gray card at the head of the roll, or to a printer light combination specified by the DP. But there is only one correction per roll. So basically, a one-light is an overall look that you are applying to the footage based on how you choose to expose the negative.

A best-light print is timed scene-to-scene, with multiple corrections per roll when required.
 
I think it is important to make a good "edit" grade. because a lot of people are going to see this footage and judge you on it. your test audiences will look at it, and a lot of people who have no idea of what grading is. having comments on how footage looks - could be dark scene that is not particularly dark, because it is meant to be brought down in the grade. It disturbs what people should be focusing on - the story and performance. Same goes for wooing investers or producers and such...

Being able to easily pre-grade it to a fairly high standard prior to editing - is a major advantage of shooting RED. The DP. can even do it himself and create the Redcine/rocket look presets, which a DIT or what ever, can apply to the clips.

NOW someone (wink wink) develop a xml import for RedcineX. So I can export an xml and import it to redcineX to reprocess only the clips I have used for the edit.
 
How can you grade shots before editing when you don't know how they will inter-cut together?

Then again, people work in mysterious ways...

:001_rolleyes:

Bingo. This is the biggest point exactly.

I have also heard the arguement that the one-light is the best chance for a DP to demonstate their intention with the look of the shot, but it still might not be a valid look depanding on how the shots cut together.

David
 
STOCK FOOTAGE - grade first

STOCK FOOTAGE - grade first

Then there are those of us that not only shoot for a production, but most of my time or spare time is shot shooting STOCK FOOTAGE.

So I'm ALWAYS grading first then storing said footage for stock sales. Even for a production, i almost always grade my footage first, then edit later. But that is also since nothing goes in the garbage...goes right back to stock.
 
Bingo. This is the biggest point exactly.

I have also heard the arguement that the one-light is the best chance for a DP to demonstate their intention with the look of the shot, but it still might not be a valid look depanding on how the shots cut together.

David

To play devils advocate to your point, performing a one light has been around long before RED. Usually when timing a one light each scene has a "look" the DP intends the scene to look like and the colorist does quick adjustments to keep the scenes consistent and following the DP's vision.
Now if in the editorial process they begin intercutting scenes or mix matching etc than no the scenes won't match 100% but usually the overall look will be consitent enough that it won't be overly distracting. That's part of timing dailies is keeping an overall consistency ie black/white levels etc.
Everyone knows some scenes will changes drastically and
some just slightly in the end but I think it's more distracting for the director to be watching a movie that doesn't resemble near his/her final intent without the one light applied. Just sayin.
 
So Nigel, what's this new app. is all about?
 
The primary purpose to a one-light is not necessarily to demonstrate an intended look or give everyone something pretty to look at. It's a way to normalize content, check exposure and prep it for proper grading and finishing later. A one-light within the RED tools is analogous to developing your film negative and scanning it.

With the RAW workflow, when all the stars align and post tools properly support it, a one-light is really not necessary, other than to normalize your footage if you desire.

However, with the current state of RED workflow, a one-light has become an important consideration again. The most popular workflows, Final Cut Studio and Adobe CS4, are now two generations behind in support for REDCODE specs and RED color science. Also factor in the support for RED Rocket, or lack thereof within these applications. Within these workflows now, it makes more sense to transcode out of the RED tools. A one-light CC pass is always a good approach when transcoding from RAW image.

When conforming and finishing within tools such as on a Pablo 4K or within SCRATCH, there is much better support for RED RAW at this point, so any preliminary CC can just be tossed away. Apple Color also allows for this, but as mentioned above it is two generations behind in RED color science and currently does not support Build 21 4.5K. Final Cut Studio also down-scales REDCODE 4K to 2K and gives the user little control over several aspects of the debayer process. However, FCS3 now supports 4K resolutions and workflows. A one-light and export to 4K ProRes 4444 with a full debayer produces excellent quality and any mid range or better Mac Pro with decent storage performance can online with the 4K ProRes files. Yes, I said online with 4K! This is not the most ideal workflow for film-out, but if I were going out to 35mm film, I would not be finishing out of FCS / Color.

So all things considered, a lot depends on the production, intended delivery, budget, workflow available, etc..
 
The primary purpose to a one-light is not necessarily to demonstrate an intended look or give everyone something pretty to look at. It's a way to normalize content, check exposure and prep it for proper grading and finishing later. A one-light within the RED tools is analogous to developing your film negative and scanning it.

With the RAW workflow, when all the stars align and post tools properly support it, a one-light is really not necessary, other than to normalize your footage if you desire.

However, with the current state of RED workflow, a one-light has become an important consideration again. The most popular workflows, Final Cut Studio and Adobe CS4, are now two generations behind in support for REDCODE specs and RED color science. Also factor in the support for RED Rocket™™, or lack thereof within these applications. Within these workflows now, it makes more sense to transcode out of the RED tools. A one-light CC pass is always a good approach when transcoding from RAW image.

When conforming and finishing within tools such as on a Pablo 4K or within SCRATCH, there is much better support for RED RAW at this point, so any preliminary CC can just be tossed away. Apple Color also allows for this, but as mentioned above it is two generations behind in RED color science and currently does not support Build 21 4.5K. Final Cut Studio also down-scales REDCODE 4K to 2K and gives the user little control over several aspects of the debayer process. However, FCS3 now supports 4K resolutions and workflows. A one-light and export to 4K ProRes 4444 with a full debayer produces excellent quality and any mid range or better Mac Pro with decent storage performance can online with the 4K ProRes files. Yes, I said online with 4K! This is not the most ideal workflow for film-out, but if I were going out to 35mm film, I would not be finishing out of FCS / Color.

So all things considered, a lot depends on the production, intended delivery, budget, workflow available, etc..

I'm in total agreement with you Jeff and that's exactly how I'm working these days with great results.
 
Sounds great. Looks like I want to talk about distribution negotiation :).
 
To play devils advocate to your point, performing a one light has been around long before RED. Usually when timing a one light each scene has a "look" the DP intends the scene to look like and the colorist does quick adjustments to keep the scenes consistent and following the DP's vision.
Now if in the editorial process they begin intercutting scenes or mix matching etc than no the scenes won't match 100% but usually the overall look will be consitent enough that it won't be overly distracting. That's part of timing dailies is keeping an overall consistency ie black/white levels etc.
Everyone knows some scenes will changes drastically and
some just slightly in the end but I think it's more distracting for the director to be watching a movie that doesn't resemble near his/her final intent without the one light applied. Just sayin.

Mike,

I don't disagree, i think a one light should be applied. It helps a lot, but I don't think shots should be "graded" before creating daillies.

A quick pass to even things out is needed, but not detailed grading work.

Also. I do have a film background so I know that timing daillies had been around MUCH longer than RED. Not sure where you got the impression I am miss informed in this area.

David
 
it's an app that will change the face of digital cinematography completely. I'm serious about that. At some point soon I'll ask for some alpha testers :)

Very interesting...

Sounds like something similar to what Ian Bloom has been working on...
 
it's an app that will change the face of digital cinematography completely. I'm serious about that. At some point soon I'll ask for some alpha testers :)
I'm a respected alpha tester... :001_rolleyes:
 
Back
Top