[Jorn Bergmans]

I’ve actually been using GIMP a lot on my linux machine and while some of the tools don’t work very intuitively (especially if you’re used to Photoshop), for ‘conforming’ graphics files it’s been performing fine.

Another option is Krita; https://krita.org/
It’s focused on and tailored more to creative digital painters etc, but can do all the things you’d need it to do for creating title cards and whatnot, and is very scriptable / automatable. (which is amazing as a data manager, since it’ll allow me to wrap multi layered EXR files into photoshop documents to send off to client)

[Jorn Bergmans]

I will echo the information provided in this thread already, of exporting to another (intermedia / mastering) codec, and creating the deliverable from that master file.

However, rendering an h.264 encoded video in resolutions above 4096 pixels wide is not possible.
h.264 has a set of levels restricted to certain resolutions and datarates. See Wikipedia; https://en.wikipedia.org/wiki/H.264/MPEG-4_AVC#Levels

If you need to deliver in such a wide resolution, you will have to break up your composition into several parts and export those to h.264, or ask the people doing the video mapping/playback if you can deliver in a codec which supports that resolution; such as ProRes (LT), Apple Animation, or HAP.

[Jorn Bergmans]

Is this the Fnord VP9 plugin? Afaik there is no official vp9 support for AME in the latest CC version.
Third party plugins aren’t supported, and as such I don’t think there’s much help available online. It also makes it hard to say why this would crash – it might be a hardware trigger, anything software related, encoding related, or the way the plugin hooks into the current CC version of AME. (I believe it was developed for CS6)
(the last push on the project’s github was 8 months ago)

I don’t know of any encoder that handles vp9 natively of the ‘big ones’. Only ffmpeg has an encoder that is maintained.
Therefor, I tend to do vp8 and vp9 using ffmpeg (or any of the GUIs if that is more convenient, particularly VLC for (batch)converting should be using the latest libraries for the vp9 codec, and has cross-platform versions)

To enable this, compile ffmpeg with –enable-libvpx and it’ll give you the vp8 and vp9 options
ffmpeg version 3.4.1 Copyright (c) 2000-2017 the FFmpeg developers
built with Apple LLVM version 8.0.0 (clang-800.0.42.1)
configuration: --prefix=/usr/local/Cellar/ffmpeg/3.4.1 --enable-libvorbis --enable-libvpx
DEV.L. vp8 On2 VP8 (decoders: vp8 libvpx ) (encoders: libvpx )
DEV.L. vp9 Google VP9 (decoders: vp9 libvpx-vp9 ) (encoders: libvpx-vp9 )

I built on mac with Homebrew, using the –enable-libvpx, –enable-libvorbis and –enable-libtheora switches.
Hope this helps.

[Jorn Bergmans]

the (AME) timecode effect just does not have that option.
You can do it in Premiere, by making an adjustment layer, adding the Video -> Timecode effect and changing the opacity of the adjustment layer.

[Jorn Bergmans]

so, the _001 in the RED file links to only a part of that RED file, which break up into parts after they hit the 3,7GB mark.
They do this because RED cards are formatted in the FAT32 format which will only allow files to be as big as 2^32 bytes which is around 4GB.

However, either Resolve or Premiere will ‘stitch’ those files together by linking to the ‘header’ file of A005_C005_051521 and one of the programs seems to be doing that differently than the other.
The mismatch here is that the XML points to the exact _001 file, while the other program things it should import the header file, doesn’t understand what the _001 part is, and doesn’t import any footage.
I’m not sure what it says in your XML – does that refer to the _001 file or to the header file A05C05?

What you could do is import the footage separately, then import the XML and link the offline timeline to the footage that you imported earler. This way, you can first check if the chunks are handled correctly by the application importing the footage.

Alternatively (and this is what I would always advise when working with RED RAW files) you can transcode the .R3D files in RedCineX to ‘proxy’ files (ProRes HQ, or DNxHR) and work with the resulting mov files in Premiere. Since RedCineX globs those _001 – _00x chunks together into single files, the resulting .mov will not have the _00x suffix and will be read the same across both of your editing / color grading programmes.
For color grading, you can relink to the R3D files easily by only having the R3D in your media pool as you import the XML that Premiere generated.

(really, Premiere is too kind in allowing editors to use raw footage, and shouldn’t allow it, in my opinion)

[Jorn Bergmans]

There is an option in Premiere called ‘Play In to Out’. There’s also an option to play around in-out with a two second preroll/postroll.

The default shortcut for Premiere keybindings is ALT+K and play around is SHIFT+SPACE
The default shortcut for FCP7 keys is SHIFT+\ and play around is SHIFT+SPACE

These will always play the currently active window, no matter if that is source, program, ref or timeline. Anything that can contain in- and outpoints will have a play in to out option.

[Jorn Bergmans]

My advice to you would be to always create a MASTER file of the finished film, as ProRes (HQ, or 4444), and make your deliverables from that Master.
That way, even if all the footage gets lost, you will still have a relatively lossless version of the film at hand to create your web/mp4, DVD, BluRay, whatever else versions from.

It is common and good practice. It will also help reduce export times if you need to make multiple versions for delivery by not having to render from timeline each time.

Personally, I always make a TITLED MASTER (for the main language, usually English), and a CLEAN MASTER (no titles, no subtitles, no end credits, i.e. ‘no localized elements’) in case the film needs to be titled/captioned in a different language than the original.

Hope this helps!

[Jorn Bergmans]

What footage are you applying the lumetri color on? What color space and bit depth is this footage using?

AE is still a bit awkward with the interpretation of color spaces. It will display the correct colors (log, r.2020, whatever you have in your file) but it will handle the gamma curve of a lot of codecs as being r.709. That might cause an issue when you’re working with light and adjustments.
Another issue might be that your AE project’s color bit depth is different from Premiere (which is capped at 10, and tries really hard to interpret everything as 8bit color). If your AE project is on 16-bit or 32-bit float it’ll handle lighting effects differently, and will cause your gamma to shift. (color should be ok-ish, but of course these things are intertwined, at least visually)

The list goes on, and without specifics on codec / settings, post workflow etc all I can tell you is that the Adobe suite isn’t very accurate when it comes to handling color.

To make sure you don’t get shifts halfway through your process, you could try adding the lumetri effects as the last step in your post process, or consider switching to a color correction suite such as Resolve (which does not have a very steep learning curve)

[Jorn Bergmans]

When I see your output specs, I can’t help but wonder if the stutter you are seeing is an encoding problem, or a playback problem.
What sort of stutter are you experiencing?

If there’s a stutter in the video / encoding process, it should be visible also if you skip through the video frame-by-frame in your NLE of choice. Is that the case? Because if so, we’re looking at a blending / interpolation problem.

However, whta I’m thinking is that this might be a playback issue rather than an encoding problem.
What hardware and software are you using to play this UHD HEVC video?

[Jorn Bergmans]

A slate usually includes some credits, and some technical specs. It can be just a text block on a black background with a big number counting down the seconds. Generally, a 10-second slate will count down from 59:59:50:00 to 59:59:57:00 (I work in a PAL environment, 25fps, non-drop), then have 3 seconds of black, then the first frame of video at 00:00:00:00 exactly.
The few seconds of black are a failsafe from back in the day, in case the tech spliced at the incorrect frame.
A black flash isn’t as bad as flashing a slate to the audience. ????

With the exception of a countdown clock (in seconds) and generally speaking a circle in the background, you can include whatever the broadcast station / production company / post house asks of you, there is no set of rules to what it needs or can not contain.

A good example of info would be something like this:
https://fcpxpert.files.wordpress.com/2015/10/template.png

Some people like doing it a little more fancy:
https://www.madpanic.tv/tips/files/VTCountDownClock_Tutorial.html#.WZGuQXcjGA0

I prefer just having a big number count down and drawing up a circle in the background. That stems from the days of tape and anamorphic video, so the next tech can check if his pixels are being displayed in the right aspect ratio. But, whatever works!