Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

RED Workflow Manual v.1.2

brandonk said:
You mean, write a comprehensive manual themselves?
Click to expand...

I'll take this one when I have a free moment. Should happen somewhere around Tuesday, 2027.
But seriously, I wasn't trying to be snide, the things I pointed out are in the "release history"
section for the L&T plugin download.

v0.0.3
Release Date: 2008-07-23

* Requires FinalCut Pro Studio version 6.0.4 or later.
* Requires RED Quicktime Codec 3.0 or later
* All aspect ratios supported
* 4K clips ingest as 2K, all other formats (3K and below) will ingest as native size.
* Supports 4 channel audio.
* Supports RED ONE Firmware Build 16
 
Nice summary of the FCP / CRIMSON workflow. But... no Scratch? This is, hands down, the best way to finish Red originated material at this time. Why leave that out?

Also I disagree with setting your monitor to sRGB colorspace for RedCine. RedCine doesn't provide a monitor out feature, so it's particularly important to be working on a monitor calibrated to the relevant target display device. sRGB has almost no relevancy outside of computers, so unless you're doing web stuff you'd be flying blind.

JT
 
Kidding aside, I seriously wish RED would chime in more on the workflow side of things.

I did 2 posts in "Workflow" called "Basics..." with specific questions that, if answered, would go a long way with users but both posts have died without answers.

Here's one question that sits unanswered:

-When using RED Cine or Alert, if trying to generally brighten your overall image, which is less destructive to the image quality: Exposure, brightness, or ISO/ASA?

I am not one to say "tell me how to do it," but this is something that could probably be explained (ie XYZ degrades the image, so start with ABC first).

There are also full controls in the software that I have a feeling a lot of people have no idea what they do. "OLPF Compensation?" Uh, okay. When should we do "high" versus "low"? "Debayer-Leading Lady?" "DRX?"

Come on, RED! This is the equivalent of putting a button on a camera and not explaining what it does. I'm not new to post-production, so I know these are not typical settings you find in other post-production software. So a little bit of explanation on what each item does would be a great thing.

So...I guess we'll wait on the manual a few years, but maybe a simple FAQ on the software would go a long way in us not feeling like we're completely on our own in figuring some of this stuff out.

Thanks!

Kevin
 
lunacie said:
Nice summary of the FCP / CRIMSON workflow. But... no Scratch? This is, hands down, the best way to finish Red originated material at this time. Why leave that out?

Also I disagree with setting your monitor to sRGB colorspace for RedCine. RedCine doesn't provide a monitor out feature, so it's particularly important to be working on a monitor calibrated to the relevant target display device. sRGB has almost no relevancy outside of computers, so unless you're doing web stuff you'd be flying blind.

JT
Click to expand...

If I had access to a Scratch system, I wouldn't likely have ever started writing this manual. Yea it's the best workflow if you have that hardware in house. Most places do not and can not justify gearing up a post system for one particular camera unless that is all they shoot with. I would hazard to guess that most people are working with FCP when it comes to RED footage.

I will have to review my notes, but Adobe 1998 or Gamma curve settings of 2.2 are what the monitor should be set to since this simulates 709 Gamma space. The Redcine manual (written by RED) suggests this setting. I don't believe I wrote about "sRGB".
 
Kevin Wild said:
-When using RED Cine or Alert, if trying to generally brighten your overall image, which is less destructive to the image quality: Exposure, brightness, or ISO/ASA?
Click to expand...

Adjusting brightness would probably be best, but that depends on just how bright you need to go with the shot. Adjusting ISO will often times add a gain type effect that would not look very good, depending on the circumstances. I tend to use the curve adjustment tool in Redcine to do my 1 lights.

Remember, Redcine is a transcoding tool that is used for making 1 lights, not for doing your final grading. You should still run your transcoded footage through a 3 way color corrector or better to achieve your final results. With that in mind, you may need to do two pass one lights on hard to grade scenes and then composite the transcodes when doing final CC to make the best image possible.
 
Brandon, you mentioned the one I forgot: Curves.

Yes, there are 4 ways you can brighten a shot. That's great...options are good. But I would love to hear from Graham or someone who developed the codec as to which is the least to most destructive of the image. I have done my own tests using all methods and can achieve basically the same look with 4 different methods. But I still have a feeling that there is a "best way" mathematically that will provide the least grain or breaking of the image.

Thanks.

Kevin
 
Thanks for putting this together- this is VERY helpful :).
 
brandonk said:
If I had access to a Scratch system, I wouldn't likely have ever started writing this manual. Yea it's the best workflow if you have that hardware in house. Most places do not and can not justify gearing up a post system for one particular camera unless that is all they shoot with. ...
Click to expand...

True.

BUt also remember that there are post houses/service bureaus/digital labs in every major market that have put in -or are putting in- a dedicated R3D processing pipeline. In many cases, this includes a redbOXX/Scratch Cine system which can crank out your one-lighted, conformed dailies within hours - or almost certainly the next day.

With 2000+ cameras in the wild, and more coming online every week, it's likely that many "smaller markets" will also soon have somebody offering these services.

I know it can be hard to build in the cost (and charge your client) for out-of-house processing -especially if they are used to HD workflows. But if you think about this in the context of a replacement for 35mm film, it's not that much of a stretch...

I think your document is fantastic. I would love to see something about the "Scratch-enhanced" pipeline options.

Cheers!
 
Awesome document. A few things I might suggest/change/add/whatever:

The H suffix on the proxies is for "half," I believe, but I thought it was "high," too.

The proxies do not have to be in the same folder as the .r3d file to work. We move ours around (I'll explain why in a moment).

We use a Red Alert workflow (or at least have on the one job we've done so far, and plan to for the next).

The shooter we're working with does one lights on location with the client, then gives us a drive with his .r3d files and his .rsx files. We create the M proxies for use in Final Cut. This is where moving stuff comes in. We've come up with a workflow where we can version our color correction (should the client want to see different looks) by moving both the .rsx files and proxies to different folders. Red Alert reads the .r3d file using whatever .rsx is in the folder with it. If there is none, it creates one. The proxies created always playback using the settings as they were when it was created (in other words, if I continue to make changes in Red Alert after I make the proxies, the proxies don't change).

We pull the M proxies of the look we like into FCP, do our offline as usual, and then the old fashioned tediousness begins:
We do paper EDLs of the cut. File name, start frame, end frame. Then, we go into Red Alert to set in/out points and any fine tuning on color correction (or first pass if the shooter doesn't do it for us before offline). Using RedRushes, we can queue all of the clips to render DPX sequences at the desired final resolution, which we import into Smoke and finish as usual.

Not the best workflow by any means, but when you're not finishing in Final Cut, it's going to be messy getting it out from anything I can tell.

We're probably going to adapt to using Crimson now that I've been reading about it, but if anyone has any suggestions, we're wide open to them.
 
Kevin Wild said:
Kidding aside, I seriously wish RED would chime in more on the workflow side of things.

I did 2 posts in "Workflow" called "Basics..." with specific questions that, if answered, would go a long way with users but both posts have died without answers.

Here's one question that sits unanswered:

-When using RED Cine or Alert, if trying to generally brighten your overall image, which is less destructive to the image quality: Exposure, brightness, or ISO/ASA?

I am not one to say "tell me how to do it," but this is something that could probably be explained (ie XYZ degrades the image, so start with ABC first).

There are also full controls in the software that I have a feeling a lot of people have no idea what they do. "OLPF Compensation?" Uh, okay. When should we do "high" versus "low"? "Debayer-Leading Lady?" "DRX?"

Come on, RED! This is the equivalent of putting a button on a camera and not explaining what it does. I'm not new to post-production, so I know these are not typical settings you find in other post-production software. So a little bit of explanation on what each item does would be a great thing.

So...I guess we'll wait on the manual a few years, but maybe a simple FAQ on the software would go a long way in us not feeling like we're completely on our own in figuring some of this stuff out.

Thanks!

Kevin
Click to expand...
In my experience doing one lights with Red footage, I tend to use all of the tools available in Redcine/Red Alert to achieve a base look for the shot. This includes Curves, which I use a lot because it allows me to tailor the low end very nicely and then I'll also play around a bit with exposure. In terms of OLPF, detail, etc, I generally start with them off and if the shot needs it, I'll add some until it works for the shot.

When the exposure is locked in and the focus of the shot is good, often there isn't a whole lot to do. Some basic curve adjustments, getting rid of the noise floor on the low end, etc. When the exposure was not good, it does make for a lot more work to get it right, often using all the tools in the arsenal.

There are some good resources for working with Red footage. FXPHD.com comes to mind.
 
Steve, thanks for replying, but it didn't really get at the meat of what I was asking.

Like you, I also "use all the tools." But I don't know from a very "under the hood" type of understanding, which tool degrades the image more than another. I was hoping Graham or someone would chime in...

So...what does "OLPF compensation" do? DRX? "Leading Lady"??

Kevin
 
I'll let you have some of it from the mouth's of Jim and Graeme, at least in terms of OLPF.

http://www.reduser.net/forum/showthread.php?t=6981

I believe Graeme has also gone into some detail about DRX, which can extend your dynamic range a bit by using info from non-clipped channels to rebuild the clipped channel.
 
afaurote said:
The shooter we're working with does one lights on location with the client, then gives us a drive with his .r3d files and his .rsx files. We create the M proxies for use in Final Cut. This is where moving stuff comes in. We've come up with a workflow where we can version our color correction (should the client want to see different looks) by moving both the .rsx files and proxies to different folders. Red Alert reads the .r3d file using whatever .rsx is in the folder with it. If there is none, it creates one. The proxies created always playback using the settings as they were when it was created (in other words, if I continue to make changes in Red Alert after I make the proxies, the proxies don't change).

We pull the M proxies of the look we like into FCP, do our offline as usual, and then the old fashioned tediousness begins:
We do paper EDLs of the cut. File name, start frame, end frame. Then, we go into Red Alert to set in/out points and any fine tuning on color correction (or first pass if the shooter doesn't do it for us before offline). Using RedRushes, we can queue all of the clips to render DPX sequences at the desired final resolution, which we import into Smoke and finish as usual.
Click to expand...

Creative use of proxies for sure! I would imagine that Smoke can import XML data, if so Crimson can help. Thanks for your input.
 
Thanks, Steve. That does provide some good info. Now why we have to dig through posts to find it is frustrating. It was also dated and several times mentioned "in future releases this will be better." So...wondering if the "better" already happened. :-)

Kevin
 
Kevin Wild said:
Thanks, Steve. That does provide some good info. Now why we have to dig through posts to find it is frustrating. It was also dated and several times mentioned "in future releases this will be better." So...wondering if the "better" already happened. :-)

Kevin
Click to expand...

The principles they are talking about are pretty much the same. They are just trying to optimize them.
 
Yup...most of the post you mentioned is more about the camera. They mentioned a "tool" to check back focus. Not sure if this was ever followed through on.

Some of this is frustrating...I know RED is busy doing stuff for tomorrow, but I think it would go a long way in answering some things for "today."

I just did a search on "DRX" and our wonderful search engine turned up NOTHING...nada.

I'm not a newbie at post and there is still a lot to know and learn about the RED post tools. I just wish they would help us along on that path...some documentation would surely help.
 
Kevin Wild said:
Yup...most of the post you mentioned is more about the camera. They mentioned a "tool" to check back focus. Not sure if this was ever followed through on.

Some of this is frustrating...I know RED is busy doing stuff for tomorrow, but I think it would go a long way in answering some things for "today."

I just did a search on "DRX" and our wonderful search engine turned up NOTHING...nada.

I'm not a newbie at post and there is still a lot to know and learn about the RED post tools. I just wish they would help us along on that path...some documentation would surely help.
Click to expand...

http://www.reduser.net/forum/showthread.php?t=15646
http://provideocoalition.com/index.php/aadams/story/red_highlight_clipping_solved/
http://forums.creativecow.net/readpost/242/1259

That will only cost you a dollar. :)

It will be nice to have all of the info gathered into one logical place eventually, but right now a lot of things are subject to change, including firmware, so it's a moving target. But like I said, the general principles are the same. The camera does no sharpening during acquisition. It's applied in post. The amount of sharpening is dependent on a number of factors. This takes some experimenting and sometimes comes down to taste and deliverables.
 
Hmm... The link I setup to Creativecow is being ambushed somehow. I saw this in someone else's posts earlier too. In edit mode, the correct link is there but it displays with the one that doesn't work.

Anybody know why that's happening?
 
I can't even type Creat.....Cow. It puts in that weird link.
 
You must log in or register to reply here.
Back
Top