Joakim Ziegler
Forum Replies Created
-
If you need just sharpen, you can sharpen in Resolve. That often helps slightly soft shots to at least match better (although you’re not really making them any less soft, of course).
If you really need to try to get something that’s out of focus into focus, there’s software that attempts to do that, although it’s fairly specialized, and my experience with it has been definitely mixed. Nuke X/Furnace has F_DeBlur, which is one of the more basic versions of that kind of thing, and it works to an extent, both for motion blur and out of focus. There are also others.
—
Joakim Ziegler – Postproduction Supervisor -
We’ve also noticed this. It’s even more noticable on our Linux system, which has 16 cores (plus hyperthreading), but it’s barely any faster than a 6-core i7 machine. The same 6-core i7 does 16-18 fps JP2K encoding in 2k with DoReMi CineAsset.
We reported this to BMD during the private beta, and they said they were aware of it. It’s particularly disappointing for us Linux users, since we don’t have the option to use EasyDCP.
—
Joakim Ziegler – Postproduction Supervisor -
Yes, this is a bug. We reported it to BMD while Resolve 10 was still in private beta, and I assume they’ll fix it.
—
Joakim Ziegler – Postproduction Supervisor -
Converting them to another color space will probably clip them. You should just convert with no LUT.
—
Joakim Ziegler – Postproduction Supervisor -
While the Arri LUTs are a good baseline, you’re most likely monitoring in rec.709 on set, so if your final grade is to P3, which it often is, it’s not going to match anyway (an exact match between P3 and rec.709 is basically impossible, it’s a matter of getting something that feels and seems the same).
So, it all depends on your workflow. For most high-end DI I’ve been involved with, with Alexa as the source, especially in the last year or two, we have not used the Arri LUTs, preferring instead to work by hand, or in some cases with a film emulation LUT, even though we’re not going to 35mm. This applies both to things we’ve done in-house and at other facilities, like EFILM. But it’s a matter of preference, of course.
—
Joakim Ziegler – Postproduction Supervisor -
LUTs are not inherently any bit depth, they do not quantize your material. However, things like the Alexa LogC to Rec.709 LUT do clip your material, so you will irrevocably lose range.
In general, it’s often not necessary to use the LUT, you will get better results just grading with custom curves directly on the LogC material, and you have much more control.
—
Joakim Ziegler – Postproduction Supervisor -
Again, if your calibration software is offering “gamma 2.2” and “gamma 2.4” as an alternative to “rec.709”, then they’re referring to the gamma curve, and you should use 2.2 or 2.4. Many calibration systems sadly don’t understand that the rec.709 curve is not suitable for anything but encoding and should not be given even as an option for monitor calibration.
—
Joakim Ziegler – Postproduction Supervisor -
While there may be several problems here, you’re getting a few things confused. Rec.709’s “overall gamma of 1.9” is in the encoding curve. A problem with rec.709 has been that the spec only includes the encoding curve (what the camera uses to translate linear light into the spec’s color space). This curve is not relevant for monitors. You’ll want to calibrate your monitor somewhere between Gamma 2.2 and 2.4 to display video correctly. When you say “sRGB Gamma 2.2”, I assume you’re referring to sRGB primaries and a Gamma 2.2 curve, which is correct, since sRGB and rec.709 primaries are identical (but sRGB has a different curve).
Try calibrating to a pure Gamma 2.2 or 2.4 curve, and see what happens.
—
Joakim Ziegler – Postproduction Supervisor -
Your setup isn’t really well suited for what you’re trying to do, as debayering the Red files to DPX before you start is very far from ideal.
But what I’d do in this case would be to debayer to DPX at 2048 wide and wh in RedLogFilm and RedColor3, work with that using curves, no LUT unless you definitely want a distinct film look (with the difficulties that implies for grading), make it work in rec.709, and then experiment with different LUTs to convert that rec.709 to DCI P3/DCI X’Y’Z’ afterwards. The advantage of this is that the files are fairly light weight.
If the client wants a 35mm film out, just do it from the same render, using a LUT to take you from P3 to the native film space. Yes, it’s slightly lower resolution, but no one will see the difference.
Or, as an alternative, if you’ve got the disk space/speed and GPU capacity, debayer to 3k (3072×1278) or 4k (4096×1716) and work from that, that’ll give you sufficient resolution for anything you might want to do.
I would not bother to work anamorphic in Resolve, it just makes your files twice as big for a very theoretical quality improvement that’s only relevant for film out. 3k would be roughly the same bandwidth, and would give you freedom to reframe with little quality loss, and be relevant for both film out and DCP.
—
Joakim Ziegler – Postproduction Supervisor -
It’s not a horrible idea, although I would first of all decide what deliverable is the most important for the project, 35mm or DCP. This also affects things like how you would work with LUTing, etc. In films we work on, the answer to this question today is DCP in 95% of cases.
Also, consider that you can change your project resolution later. If you’re working with 5k material, and always use the highest debayer quality, you can change resolution and render directly to whatever output resolution you need. Another option would be to work in something like 3k square pixel.
However, you should be aware that there are some bugs in Resolve when working with anamorphic, particularly related to tracking, as I remember (windows would track all wrong), but it’s been a while since I did a project in anamorphic, so these may have been fixed.
—
Joakim Ziegler – Postproduction Supervisor