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 :)

4k workflow - dpx, edl?

Måns Winberg

Member
Joined
Jun 1, 2009
Messages
6
Reaction score
0
Points
0
Can I create an edit decision list in final cut pro and use in redcine or red alert to export dpx-files? Do I need Assimilate Scratch for this? Can I include audio recorded with an external recorder? I want to be able to deliver dpx files in 4k RGB to a post facility for grading and film-out (like it says in red.com/cameras/workflow/2k or 4k di finish).
 
Can I create an edit decision list in final cut pro and use in redcine or red alert to export dpx-files? Do I need Assimilate Scratch for this? Can I include audio recorded with an external recorder? I want to be able to deliver dpx files in 4k RGB to a post facility for grading and film-out (like it says in red.com/cameras/workflow/2k or 4k di finish).

You'll need one of these- Monkey Extract, Crimson, JamFuse or any other tools, that seems appear every day. Scratch would work as well...
 
You'll need one of these- Monkey Extract, Crimson, JamFuse or any other tools, that seems appear every day. Scratch would work as well...

You'll also need a lot of money if you really plan to use 4K.
 
It just seems strange that you can't use Redcine or Red Alert for this, or any other Red software, in a straight-forward workflow. It will probably be 2K in the end, but right now it's a short film so it might not cost a fortune in 4k. I'm not the one who's paying, luckily enough, I just want to be able to tell the producers what their options are.
 
By far the best way to do this (I think) is in a SCRATCH suite with a good colorist. Here's how the process goes....

1) export the edl out of final cut (CMX 3600, all video on track 1)
2) export a low-res quicktime file of the project out of final cut
3) Load the edl and reference QT into SCRATCH
4) SCRATCH will instantly match back to the R3D files
5) Play down the R3D assembled program and compare to reference cut for accuracy.
6) Color grade each scene in REAL TIME
7) Export DPX frames to whatever specs you require for film-out
8) To double-check, re-import the DPX into scratch and compare to reference cut.

There are several benefits to this workflow.

1) Your color grade will be far better/detailed and take way less time
2) You'll be able to VERIFY your DPX sequence
3) You'll be working with an experienced colorist who has done this many times before.

Best of all, this shouldn't be expensive. Yes, it will be more than "free". But it will be the cheapest insurance ever...and I guarantee you'll be happier with the final result.

As a SCRATCH owner, I'm definitely biased...but I wouldn't have invested in the system if it didn't make a night-and-day difference.

IMHO.....
 
Another workflow, less expensive (Scratch costs piles of money)

1. Edit your material in FCP, using M proxies
2. Save XML in FCP
3. Load XML in Crimson. Generate another XML to be loaded in Redcine
4. Load the new XML in Redcine. Color Grade in 4K (if you like)
5. Render your output in Redcine (quicktime or dpx sequence)

If its necessary,
6. Return to FCP via Crimson, with Roundtrip process
 
By far the best way to do this (I think) is in a SCRATCH suite with a good colorist. Here's how the process goes....

1) export the edl out of final cut (CMX 3600, all video on track 1)
2) export a low-res quicktime file of the project out of final cut
3) Load the edl and reference QT into SCRATCH
4) SCRATCH will instantly match back to the R3D files
5) Play down the R3D assembled program and compare to reference cut for accuracy.
6) Color grade each scene in REAL TIME
7) Export DPX frames to whatever specs you require for film-out
8) To double-check, re-import the DPX into scratch and compare to reference cut.

There are several benefits to this workflow.

1) Your color grade will be far better/detailed and take way less time
2) You'll be able to VERIFY your DPX sequence
3) You'll be working with an experienced colorist who has done this many times before.

Best of all, this shouldn't be expensive. Yes, it will be more than "free". But it will be the cheapest insurance ever...and I guarantee you'll be happier with the final result.

As a SCRATCH owner, I'm definitely biased...but I wouldn't have invested in the system if it didn't make a night-and-day difference.

IMHO.....
OK, here we go again. What does this mean "Your color grade will be far better/detailed"? Better than what and how? And when you say "You'll be able to VERIFY your DPX sequence" do you mean, that there is a known problem with r3d to DPX render in Scratch? Is Scratch known to drop frames while rendering? Is that why you need to verify the render with the reference QT? Or you just don't trust Scratch rendering:)
 
Another workflow, less expensive (Scratch costs piles of money)

1. Edit your material in FCP, using M proxies
2. Save XML in FCP
3. Load XML in Crimson. Generate another XML to be loaded in Redcine
4. Load the new XML in Redcine. Color Grade in 4K (if you like)
5. Render your output in Redcine (quicktime or dpx sequence)

If its necessary,
6. Return to FCP via Crimson, with Roundtrip process

JamFuse is $49 and ClipFinder is free. Can't do much cheaper than that:)
 
OK, here we go again. What does this mean "Your color grade will be far better/detailed"? Better than what and how? And when you say "You'll be able to VERIFY your DPX sequence" do you mean, that there is a known problem with r3d to DPX render in Scratch? Is Scratch known to drop frames while rendering? Is that why you need to verify the render with the reference QT? Or you just don't trust Scratch rendering:)

When you do a final output from any system, do you check it?

I was an online editor for years, and no matter what system I was on, I *always* watched every frame of a layback to tape.

SCRATCH doesn't drop frame while rendering, and there are no problems with R3D to DPX rendering. I work for ASSIMILATE... but if I went and did freelance work with SCRATCH (I don't) I would re-import my sequence and verify it.

I trust the app. I just don't always trust myself - especially at 2AM when a deadline is looming and I've been up for 24 hours!

Doveryai, no proveryai. ; )

Best,

Lucas
-----
ASSIMILATE, inc.
LA, CA, USA
 
When you do a final output from any system, do you check it?

I was an online editor for years, and no matter what system I was on, I *always* watched every frame of a layback to tape.

SCRATCH doesn't drop frame while rendering, and there are no problems with R3D to DPX rendering. I work for ASSIMILATE... but if I went and did freelance work with SCRATCH (I don't) I would re-import my sequence and verify it.

I trust the app. I just don't always trust myself - especially at 2AM when a deadline is looming and I've been up for 24 hours!

Doveryai, no proveryai. ; )

Best,

Lucas
-----
ASSIMILATE, inc.
LA, CA, USA
Hi Luki.
Impressive knowledge of Russian and Reagan era history. I can see you're paying attention to my posts, as I had mentioned my russian roots only once (actually ukrainian):)
Empirical knowledge and experience is very important. If one feels, that it is necessary to constantly check everything twice, that's fine. But never in all of my years of working with Baselight, Davinci or Lustre did I ever had a problem with frame accuracy of the render. Conform is a different story altogether... Yes, the confidence check is a necessity and I too wouldn't trust myself at any time of day or night. But, if you suggest that Scratch render needs to be checked for frame accuracy, then users should do it. I'm not arguing.
My main point for the original post was this statement:
"Your color grade will be far better/detailed".
Care to comment on it?
 
Hi Luki.
Impressive knowledge of Russian and Reagan era history. I can see you're paying attention to my posts, as I had mentioned my russian roots only once (actually ukrainian):)
Empirical knowledge and experience is very important. If one feels, that it is necessary to constantly check everything twice, that's fine. But never in all of my years of working with Baselight, Davinci or Lustre did I ever had a problem with frame accuracy of the render. Conform is a different story altogether... Yes, the confidence check is a necessity and I too wouldn't trust myself at any time of day or night. But, if you suggest that Scratch render needs to be checked for frame accuracy, then users should do it. I'm not arguing.
My main point for the original post was this statement:
"Your color grade will be far better/detailed".
Care to comment on it?

No, SCRATCH does not need to be checked for frame accuracy in renders. Users do. : )

Color Grade being far better/detailed. Need to know in comparison to what. I think Ian might be talking about REDCINE? If so, then it's just that REDCINE provides basic primaries and nothing else. SCRATCH provides much more. That's the only thing I can think of without clarification.

Lucas
 
No, SCRATCH does not need to be checked for frame accuracy in renders. Users do. : )

Color Grade being far better/detailed. Need to know in comparison to what. I think Ian might be talking about REDCINE? If so, then it's just that REDCINE provides basic primaries and nothing else. SCRATCH provides much more. That's the only thing I can think of without clarification.

Lucas

Thanks Luki. I think you're correct as always. I have to admit, that at times it may seem, that I sound a little biased against Scratch. Nothing could be farther from the truth. I'd stated numerous times, that Scratch is a world class color grading solution, with some shortcomings, just like any other systems. What I usually object to, is the blanket notion, that somehow one color grading solution is better than another. I appreciate, that there is always an element of bias present in these discussion. Personally, I'm impartial to Baselight and Lustre, but it doesn't mean that they're best in every aspect of color grading process. I recognize those shortcomings and just try to work around them. So, let me make a small list of the various distinguishing features of more popular color grading solutions, as I see it:
1. Scratch. Best RED workflow implementation and most cost effective
2. Baselight. Best interface and control panel
3. Resolve. Best support.
4. Lustre. Best interoperability across entire Autodesk product line.
5. Film Master. Best noise and grain management.
Obviously there are some products that are left out from my short list, but it's only because I have no personal experience with them.
 
1. Scratch. Best RED workflow implementation and most cost effective
2. Baselight. Best interface and control panel
3. Resolve. Best support.
4. Lustre. Best interoperability across entire Autodesk product line.
5. Film Master. Best noise and grain management.
Obviously there are some products that are left out from my short list, but it's only because I have no personal experience with them.

Nice post Jake - couple of comments...

2. How do you think Blackboard compares to the Quantel NEO panel, and the new panels for Resolve?

3. Support? As in from the manufacturer? How do you judge this? (not questioning you - just curious)

4. Well, yeah. Do you expect a non-Autodesk tool to have better Autodesk support than an Autodesk tool? :)

5. Have you seen the new Cinnafilm toolset? It is mindblowing... I think that the DV stuff is very good, and I would have called it the best in the market... until I saw the Cinnafilm tools at NAB.

Lucas
 
Nice post Jake - couple of comments...

2. How do you think Blackboard compares to the Quantel NEO panel, and the new panels for Resolve?

3. Support? As in from the manufacturer? How do you judge this? (not questioning you - just curious)

4. Well, yeah. Do you expect a non-Autodesk tool to have better Autodesk support than an Autodesk tool? :)

5. Have you seen the new Cinnafilm toolset? It is mindblowing... I think that the DV stuff is very good, and I would have called it the best in the market... until I saw the Cinnafilm tools at NAB.

Lucas
Hi Luki.
2. On Neo panel I'm just going to repeat something that my friend had said and I wholeheartedly agree. He mentioned, that Quantel went penis envy with the move from the smallest to the biggest panel out there. It's also the most colorful and brightest panel, which can't be very good for a colorist vision. But it is sure is big and beautiful...
I have seen the new Resolve panel. I forgot it's name. Lately, it seems, that DaVinci got back it's mojo and they finally got some good products in the pipeline. Their booth at NAB was actually quite alive, unlike, say, Pogle. I couldn't get much info on the panel as I ran out of time. My biggest question was about the need for the fourth trackball. And their shape tracker appear to be just as good as Lustre's.
3. DaVinci manufacturer support always been no more than an hour away, be it on the phone or, if need it, an actual physical presence, 24/7. Granted, my experience has always been as LA user with the support contract. Places not near DaVinci field offices obviously don't get the same instant treatment.
4. I guess, I did stated the obvious, but it is still really impressive implementation of timeline sharing. They had to bring all of their high end apps into compliance and that itself is quite an achievement.
5. No, I didn't see it at NAB, but I have seen the retiming demo somewhere after the announcement on the web and it is truly impressive. I had never seen it done before in real time. I believe PF Clean does something similar to the Relativity suite in terms of texture control and with the latest version 4 they also had started using GPU processing. Baselight can use PF Clean engine in the background for noise and grain management, but of coarse, it is nowhere near the tight integration of the DV.
 
Would you elaborate please. What is "a lot" of money?

thanks

Storage
Computers
Infrastructure
Projection

For uncompressed 4K, around US$100K for a decent computer with enough storage to handle the data rates and hold a feature's worth of 4K.

Sony SXRD w/10K Lumens bulb - about US$120K.

Theater to accurately show 4K - several hundred thousand.

Software to deal with 4K - around 100K, give or take.

Remember, there is no such thing as a "pure" RED project. Not unless you have a project without any visual effects, titles, inserts, or other formats.

Lucas
-----
ASSIMILATE, inc.
LA, CA, USA
 
Back
Top