[Michael Mccune]

Thanks, Michael. Actually, from the BMD Resolve Forum, someone points out that this is a feature, not a bug! That is, from the manual, p. 321, “Frame Display Mode,” it seems to speak to this. Here is the helpful reply.

TheBloke

Posts: 1679Joined: Sat Nov 02, 2019 11:49 pmReal Name: Tom Jobbins

Re: Resolve 17 fails to import TIF sequences—-and a fix

Sat Nov 20, 2021 11:36 am

It’s not a bug that TIF doesn’t import as a sequence by default, it’s deliberate. The new Frame Display Mode “Auto” setting will automatically detect sequences from DPX, EXR and DNG files, but is specifically excluded from doing so with JPG, PNG and TIF. I don’t understand why they did it like this, but it’s definitely deliberate.

From the manual (p321):

More info here: viewtopic.php?f=21&t=149318#p798412

—Thanks all, Mike

[Michael Mccune]

Thanks, Glenn. Glad an expert had a quick look at this. And also that you did not seem to recognize this issue otherwise: That is, when a guy has a problem he is rarely the first one if it is a genuine app issue.

I did some of the usual re-install and re-start and, happily, it is much better, perhaps fixed.

In the background, I wonder about how, exactly, Resolve handles alphas, masks and mattes in the Resolve node area. We all know that blending those can be done in a million ways. But since Black Magic supports also Fusion stand-alone Studio and Resolve/Fusion, I am happy to think that, at bottom, they really do have it handled.

Must be very tricky for them, though, to cram all that nuance and subtlety into a user-friendly app.

Thanks again, Glenn.

Mike

[Michael Mccune]

Thanks, Simon. Very glad to know you are still out there. Saw one of your latest videos and also wanted to say that I benefited greatly from your Fusion video on understanding framing as it helped immensely with a project using multiple frames (archive film in Academy, HD, UHD) and all resized for various Ken Burns style pan and zoom. This can drive a person nuts without basic understandings such as you outlined in your video. Thanks.

Still no joy in Fusion Studio–any flavor: 16.0 beta, 16.1.1, version 9, or the most current version. Uninstalled, reinstalled Fusion and even Resolve multiple times; even manual post-removal file deletion and also used Revo Uninstall to find many remaining files. Verified with our admin at our tiny non-profit that there were no domain firewall rules for BMD and had it whitelisted. Still no joy.

BUT did find a provisional workaround: Resolve continues to accept all usual file types. You may recall that he main problem with Fusion is that it will not load, say, 90% of otherwise totally standard files, such as tif or png file images, image sequences, EXR, mov, etc. Often not even a console error message. No apparent pattern among the few files it does accept.

The workaround is to bring the files into Resolve and then, from the Media Pool, bring them into Resolve Fusion. Would prefer Fusion stand-alone but that still does not accept most files. (As mentioned, tried multiple source disks, different depths of the folder hierarchy, used short file names, verified permissions, etc.)

But I understand the the sun is shining all over the world; just not on this Fusion. Waiting for the next revision.

Thanks again, Simon.

Best, Mike

[Michael Mccune]

Haven’t tried to reset the Windows mouse behavior( to not go the default button in a new dialog box).

I, like most people I expect, would like to have Windows (but not Resolve) move the cursor to any new dialog box. Doesn’t serve any purpose that I know of in Resolve.

Seems that Resolve would have a way to avoid this annoyance.

Mike

[Michael Mccune]

Complication: the xmp lines of info, described in the previous post, seem to be place ONLY in the first frame of the sequence. And the last frame of the sequence seems to terminate the frame and–and the frame sequence–with a series of “NUL/FS/NUL” triplets.

The frame sequence will nonetheless import into Resolve without the first or last frames but playback seems to show that the timing is off.

Also, Marc, you suggested to check the Edit page Clip Attributes. It does indeed show a value for fps BUT it appears to be only a drop down selection list with the field initially populated with the timeline fps.

So, as you say, Marc, it seems prudent to set up the frame sequence for the editor’s timeline fps before importing it. That would seem to avoid some very tricky and hard to understand jitter issues, such as I found in playing with a frame sequence imported without an xmp head or a tail frame.

Any thoughts???
Thanks, Marc.
Mike

[Michael Mccune]

Found it!!! This answers the question of how Resolve knows the frame rate of an imported frame sequence. Also shows other helpful info about the frame sequence, such as dates and times and pixel aspect ratio.

Here is the metadata from a frame, in this case using Media Encoder to create sequences with known frame rates. We encoded the original DPX into a 24.976 fps frame sequence, 24 fps and 29.97 fps.

Using a text editor (Notepad++) and a frame from the 23.976 fps encoding and then searching for ‘23.976,’ we find:
Line 128: xmpDM:videoFrameRate=”23.976024″

Also works for other example frames, such as 24 fps. Lines below are 123 to 133. (I expect that these xmp lines and data appear at the same line number location within every frame.)
xmpMM:DocumentID=”ab2fb591-9aa3-aa7e-7264-d2ac000000a1″
xmpMM:OriginalDocumentID=”xmp.did:c157f6e3-4966-f243-b874-f1159a4a8448″
xmp:MetadataDate=”2019-08-30T21:05:36-07:00″
xmp:ModifyDate=”2019-08-30T21:05:36-07:00″
xmp:CreateDate=”2019-08-30T21:04:47-07:00″
xmpDM:videoFrameRate=”24.000000″
xmpDM:videoFieldOrder=”Progressive”
xmpDM:videoPixelAspectRatio=”1/1″
xmpDM:startTimeScale=”24″
xmpDM:startTimeSampleSize=”1″
dc:format=”TIFF”>
And I also see the helpful info about pixel aspect ratio, here clearly indicating it is 1:1. (I had a problem recently in Resolve that seemed related to pixel aspect ratio but could only guess and deduce what the original pixel aspect ratio was. But here it is!!! Plain as day.
———-
Note, from Wikipedia about xmp:
Extensible Metadata Platform – Wikipedia
https://en.wikipedia.org › wiki › Extensible_Metadata_Platform

The Extensible Metadata Platform (XMP) is an ISO standard, originally created by Adobe Systems Inc., for the creation, processing and interchange of standardized and custom metadata for digital documents and data sets. XMP standardizes a data model, a serialization format and core properties … for embedding XMP information into popular image, video and document file …

Mike

[Michael Mccune]

Thanks, Joseph. I like your expression, “… should not be ambiguous across the application.”

Well, it does indeed seem to be, as you put it so well, “… ambiguous across the application.”

That aspect ratio mismatch or ambiguity is exactly the problem. As mentioned in my original post, changes to scaling–using any of the several methods you mentioned, either on the edit page or on the color page–do indeed make the changes as expected. I have tried them all.

Changing the scaling in any of these ways is not the problem. The problem (as far as I understand the situation now) is that an adjustment also makes the corresponding adjustment in the other page, color or edit: thus the aspect ratio mismatch is not fixed.

Of course, this scaling change propagation ordinarily is exactly what one expects and wants, assuming that the viewing panes on the two pages match.

And they should match. But they don’t.

I must be missing something but don’t know what that is.

Thanks anyway.
Mike

[Michael Mccune]

George; yes, think you are right that Super Scale is applied on the output render.

And the scaling or size change is only visible during the Deliver page render in the view window.

I see in the Deliver page view window that the image jumps up in size when the render starts. And then the image in the Deliver page view window collapses back to the expected size as soon as the render is complete.

Thanks again, George.
Mike

[Michael Mccune]

Thanks, George. That may well be the answer: that is, the Super Scale properties of the clip are not changed or reflected in the editor (nor in the Color page) and the Super Scale attribute is not even indicated in the Delivery page. But Super Scale, when set in the Clip Attributes of the Edit page Edit Index, is nevertheless incorporated into the rendered output.

Sounds simple enough.

But the 2600+ page manual might have mentioned what you just clarified.

Thanks again, George.
Mike

[Michael Mccune]

As you have no doubt found, with the Creator update MS now limits display options to 25% increments in Display settings (via the context menu for the Desktop).

Other than that, the only other method seems to be to change the display via the Properties for the application. That is, the context menu for, say, Resolve or Fusion.

Right click on the Desktop icon/Properties/Compatibility/Change high DPI settings.

You will find a couple of less-than-obvious options (e.g., “Use the DPI that’s set for my main display when … ,” and, “High DPI scaling override …”).

This will surely require multiple log off/on cycles to experiment with various guesses for a workable setting. With this approach your Display settings page will then show that you are using a “Custom” display setting.

I agree that some Resolve and Fusion dialog text can be microscopic and making those legible may result in too-large text elsewhere.

One clue to watch for is the startup splash screen for Resolve or Fusion: the smaller text used there is a clue as to the eventual text size when it completes its startup.

Good luck!
Mike