Reduce, or eliminate strobing while “zooming” in a slideshow.

[Andy George]

Hi Richard,

It looks to me like the type of artifacts typical with DV Compression. What codec are you rendering to and what is your composition setup? How large are your stills?

-Andy George
Senior Editor
http://www.chiselindustries.com

[Richard Mamary]

Thanks Andy for your reply.

To make sure I will be overly redundant with my info.
Fyi, the exported mov is then being imported into a Apple Pro Res 422, FCP timeline.

There are only four images in this group.
All of them are 300 pixels/inch

1. 3648×2736
2. 2736×3648
3. 3648×2736
4. 3648×2736

First, a screen shot of my AE composition settings.

and one of AE’s output, (Apple Pro Res 422.)

Next, my FCP project settings.

[Andy George]

Looks like you have everything set up properly.

Check out this link from the AE Online help.

https://help.adobe.com/en_US/AfterEffects/9.0/WS3878526689cb91655866c1103906c6dea-7e5ba.html

Specifically the part about subpixel positioning and the guidlines for fixing the problem.

If setting the image to Draft (turning off anti-alliasing and subpixel positioning) reduces the jaggies then this might be what your dealing with.

-Andy George
Senior Editor
http://www.chiselindustries.com

[Richard Mamary]

Thanks Andy for the idea.

I will read up, and try it out.

[Tudor “ted” jelescu]

Also, I usually turn motion blur on for the zooms (or moves) on stills- this increases render times but eliminates the flicker.
There is no need for 300 dpi resolution for video – 72 dpi is just the same.

Tudor “Ted” Jelescu
Senior VFX Artist
Bucharest, Romania
http://www.ennstudio.ro

[Richard Mamary]

“Also, I usually turn motion blur on for the zooms (or moves) on stills- this increases render times but eliminates the flicker.
There is no need for 300 dpi resolution for video – 72 dpi is just the same.”

Thanks Ted, I will experiment with motion blur, when I am having that trouble again, and I will use 72 dpi (ppi), from now on, in these still sequences, which will also save me a lot of file space. But, in the future…

I’m learning here, so I don’t know if it makes any difference, but the stills I am using are 300 ppi, not dpi.

Are ppi, and dpi basically the same (interchangeable?) Just one is used for printing, while the other is used for video?

“I presume this stuff is 720P, and NOT 1080P.” “Another thought: AE may have interpreted the footage improperly.”

Thanks for the all the input Dave. Yes the project is 720p. I checked the stills “interpretation,” and the fields are set to none.

However, I ended up going back in to Photoshop, and masked out the troubling area and repaired it.

I didn’t want to have to do it this way, because I was trying to find a better, more universal method, that might allow me to avoid going back to Photoshop, where I originally prepped the stills. I have had this occur before, and rather than always return to PS, I wanted to see if there was an AE solution. One less step.

Now, I have a bigger problem, and will make a new post.

Thanks again!

[Richard Mamary]

“If this was a sequence of still images, I don’t know why you had to go to FCP before you imported the footage into AE.
I can’t think of a single good reason for it… but I always seek enlightenment.”

Thanks again Dave. All I do is seek enlightenment!
And, neither can I. Because it’s the opposite!

The sequence of stills was prepped in PS, rendered in AE, and then imported into my FCP timeline.
FCP does a lousy job with still sequences.

But then, once in FCP, there was the strobing of the line under the gurney. So I had to go back to AE to re-render it, and wanted to know if there was a solution in AE, (rather than having to go back to PS, as I have had similar, “occurences,” in the past with animated still sequences.)

I suppose in this situation, PS is probably the only reliable solution.

Now, my big issue is this moire pattern, on a horizontally striped shirt on one of the subjects in this doc. I just posted. it

[Richard Mamary]

“I still think you weren’t working with real progressive-scan footage. I think it was shot interlaced. If this stuff started life at 1080, the chances are overwhelming it was interlaced, even if it was shot progressive scan.”

The digital photos were downloaded to my computer, tweaked in PS, than imported into AE, motion added, rendered as Apple Pro Res 1280×720 sq pixels, 29.97, supposedly the same settings as my FCP sequence. Once rendered in AE the slideshow was imported into a 1280×720 29.97 sq pixel project in FCP.

“If if you’re convinced everything is progressive, then you need to start scrutinizing FCP, which is just as big a screwup as anything else in Software-Land. Check the footage in the Browser and then check the timeline> If you don’t have the field order — which FCP incorrectly calls field dominance — se to none, you could have aliasing problems.”

I am not convinced of anything unfortunately.

My understanding was even though I had AVCHD “lite”, 1280×720 30p footage, I was instructed the proper way to import it into FCP was as Apple Pro Res, with a timeline of 29.97?!

So I have been confused about this from the beginning.

I was told that 29.97 was still progressive, (which strained my brain a bit i will admit.)

Here are my FCP sequence settings, followed by the composition settings from AE.

So when I made my AE Composition, I chose the same settings. You can’t see, but I chose “square pixels,” too.

And here is the Composition Sequence Highlighted in AE’s Browser.

And, finally the info on the AE exported slideshow, as seen from FCP’s browser.

The one thing here that strains my brian more than anything is understanding how 29.97 can be 30p.
That one kills me.