Joseph Owens
16
COW Coins
16
DaVinci Coins
Forum Replies Created
-
[Glenn Sakatch] ” why did you delete sections of the timeline and not just close up the holes?”
I do this occasionally in the time-honored (or habitually-obsessed) strategy of checkerboarding A and B rolls. For example if I have a melange of acquisition frame rates that I might want to eventually segregate in order to keep or delete either all the 23.98 or 29.97 source media and then re-build an appropriate sequence for either universal (23.98) or domestic broadcast (29.97) and try to keep the media as native as possible. That is — trying to avoid conversions that compromise the available net quality. The worst is trying to get the original 29.97 fps stuff back down to 23.98P. Can’t see why I would want to convert and then re-convert.
However, the attempt to keep the edit segments (if they need to keep a sync relationship) separate is not lost on me. This can be as subtle as the SMPTE-A relationship that 2:3 pulldown will introduce with respect to the timeline reference timecode. Sometimes, it makes things slightly easier to find placement when you need to start re-conforming elements that have gone out for treatment, VFX… blah, blah, blah. (“Uh, I think that clip came from somewhere around the 43-minute mark…”)
Had a breakfast meeting not that long ago trying to sort out how to streamline re-conforms when all (496 or so) VFX clips come back into a feature-length project and they’re all named “Untitled” and start at 00:00:00:00 time code. And there is a producer screening in about 10 minutes… from five minutes ago. (“What! Those things are already a week late!!???!?!”)
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
Duplicate your timeline and delete what you don’t need.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[Ben Dittrich] “does anyone now what the max Gui Resolution for Resolve is, or is there a possibility to enlarge text (& menus)?
“From the Resolve 12 Configuration guide:
“DaVinci Resolve 12 offers support for a wide variety of UI resolutions in single and also dual UI monitor configurations. The application is optimized for the Retina 5K screen (5120×2880) with a resultant resolution of 2560×1440 and scales in a number of ways to even operate on laptops screens with 1440×900 res. If you do not have a Retinal 5K screen the UI layout works extremely well at all resolutions from 3840×2160 to 1920×1200. Lower resolutions are supported with abridged name and icons at the controls.”
I find that 2560 on an ASUS PB278 is acceptable. Also running R12 on a Retina 5K iMac at 5120 is workable at least as far as the UI is concerned. I would not dream of using either of these displays for grade judgement. Been playing with a display calibration system that cannot be named on this forum without triggering the nanny-filter, but at least with a viewer 3DLUT in place, the client confidence goes up.
Some windows in the UI are expandable, but the font and placement are not customizable.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
The real issue with respect to “Broadcast” legal gamut is that no RGB display will give you a true representation of where your luma + chroma values are going to wind up after matrix and baseband composite are computed. There are a few outboard scopes that have built-in functions to simulate the values that will result. The notion that simply limiting RGB values to between 0-100 complies to “legal” gamut is not correct.
Your example and query about the clipping in the blue channel depends on whether those values are truly clipped, or will result in *yellow-blacks* below 0 IRE and whether those will be passed through a composite baseband encoder. Those would be out-of-gamut values.
In reality, there are separate parameters for luminance and chrominance legal gamut boundaries. For Y’, usually -2 to 102 IRE are permissible, and 75 to 100% saturation (while not maybe advisable, anyway) are the outer limits for chromaticity. In no event though are a combination of Y+C permissible beyond -20 and +120 IRE — and this is the combined baseband signal that no software scopes are likely even capable of displaying.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
There are change-source-folder strategies with “loose naming” convention but it does depend on clip names, reel numbers and identical time code. Other than that, the Mark 1 Human Eyeball.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
This would be a question for either Jason Livingstone or Giovanni Galvez of Telestream to answer definitively.
My understanding of .scc is that it supports EIA608/Line21 only and that people still ask for them associated with HD out of ignorance that it is not 708 compliant. Some caption authoring software will allow you to import a scenarist file to generate EIA708-wrapped stand-alone files, such as .cap, .aaf, and the soon-to-be more widely recognized .mcc as the VANC-compatible donor file.
The fact that several NLEs and disk-authoring software allow the import of .scc glosses over the reality that while the text and time-stamp are just fine, all the other metadata isn’t, including the extremely important requirement (being enforced more rigorously now) for custom positioning on-screen. This is not part of the 608-Line21 technical description, but it is a major feature of 708. The other issue is that scc is tied to 29.97 and if you try to make one that is some other frame rate, the authoring program will warn you (or should warn) that you are not EIA-compliant anymore, and if your work is gong to be QC’d for broadcast, either be prepared to justify the export or wonder if the deliverables are being grab-bagged from the internet and they are just adding more” important-sounding stuff” to the document.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[marko milovanovic] “it messes up shots, moving them closer to each other or even putting them one above other..”
Sounds like a mismatch between how the NLE creates the PTRZ factors and how Resolve interprets them, and could be the result of a number of other import options, a consequence of the original media not being the same dimension as the original timeline and a “stretch-to-fill” or similar transform has been applied. Especially if these are non-square, there will be havoc.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
Steal them from some original source and add them as superimposed layers.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
[PHil Side] “the timeline, 1900×1080. How can I maintain the original size?”
Unusual dimension for a timeline.
In the edit browser, there is likely a transform being applied to that element that you can uncheck in the properties box.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.
-
I’m finding that if the source media has been exported in any kind of linearized codec, then the chances are that a reverse LUT is unlikely to help place the media back “in the basket.” Sometimes the “next” LUT, and not the expected one, is less injurious and then an experienced colorist can step in and take it from there.
If you have done your million and a half feet of dailies/rushes as an orientation to the trade, herding pleasant images back into the middle — from both ends of the exposure bracket, is a fairly straightforward exercise, as Marc will attest. Sometimes the machines really aren’t of much assistance, like the auto-approach on an airliner failing right around the decision altitude. Go around or try to stick it: judgement day.
jPo
“I always pass on free advice — its never of any use to me” Oscar Wilde.