importing 1080i footage (from GH2) to FCP: interlaced or progressive ProRes 422?

[Shane Ross]

Get ShotPut Pro for the future. It will offload the cards and ensure the full structure is intact.

Why did you shoot 1080i if your intention was for the web?

I’d deinterlace the final…work at as high quality as you can when you start the edit.

Shane
Little Frog Post
Read my blog, Little Frog in High Def

[White Yufeng]

You’d better deinterlace your 1080i files to 1080p. In fact, the quality of 1080 60i is equal to 1080 30p.

———————————————–

Camcorder Video Tips

[Jay Church]

[Shane Ross] “Get ShotPut Pro for the future. It will offload the cards and ensure the full structure is intact.

Why did you shoot 1080i if your intention was for the web?

I’d deinterlace the final…work at as high quality as you can when you start the edit.”

i didn’t realize the camera was set for 1080i, i thought it was set for 30p.

deinterlacing at the end (at export) was my instinct as well. thanks for the confirmation.

[Jay Church]

[white yufeng] “You’d better deinterlace your 1080i files to 1080p. In fact, the quality of 1080 60i is equal to 1080 30p”

the question is when/where in the workflow to do so.

when transcoding it for import, or importing it as interlaced (as suggested above) and then deinterlacing it to 30p at export.

my feeling is the latter. are you of another opinion?

thanks,

_jc

[Shane Ross]

[white yufeng] “You’d better deinterlace your 1080i files to 1080p. In fact, the quality of 1080 60i is equal to 1080 30p.”

Sorry, but that is wrong. When you deinterlace, you effectively throw out half of your resolution. You discard one field of information, and stretch the other to fill in the missing information. 1080i60 de-interlaced will be soft, and half the resolution of 1080p30. 1080p30 fills in both fields with the full picture information.

No, if you shot 1080i and need to deliver 1080p…best to de-interlace when you compress the final. Less resolution loss. Otherwise you toss out half the information….then edit…then toss out more information when you compress the final.

Shane
Little Frog Post
Read my blog, Little Frog in High Def

[Andrew Kimery]

Yeah, 1080i60 is a 1920×540 field every 1/60 of second as opposed to 1080p30 which is one 1920×1080 frame every 1/30th of a second. The date rates are similar but the image quality and motion characteristics are different.

My gut reaction was the same as Shanes (cut in 1080i then deinterlace later) but any GFX you use will most likely get killed if you start in “i” and end up in “p”. So you could either convert to “p” first or cut in “i”, export a high res ProRes master (w/o GFX) in “p”, drop that into a new progressive timeline, add GFX, then export again.

[Jay Church]

ok, i’ve imported transcoded the source as interlaced and plan to deinterlace as i export/compress.

i use the x264 encoder via quicktime conversion in FCP to make my “h.264” web videos.

to anyone that is familiar with that encoder, is it better to check the “deinterlace” option in the “size” section of the quicktime movie settings, or to choose one of the conversion settings (interlaced conversion, progressive conversion) under “chroma subsampling” under the behavior tab in the x264 settings? or a combination of the two?
and which conversion setting would you choose in this instance in the x264 settings?

[Rafael Amador]

Make sure that the video is really interlaced.
When you shoot p30, the files shows up also as i60, but in fact ther are what is called PSF.
In that case you only need to check the clips as Progressive (NONE) and edit as Progressive.
So, get somme pics with a lot of movement on, set the Canvas at 100% and see if they are really 2 fields in the picture.
rafael

http://www.nagavideo.com

[Rafael Amador]

[jay church] “to anyone that is familiar with that encoder, is it better to check the “deinterlace” option in the “size” section of the quicktime movie settings, or to choose one of the conversion settings (interlaced conversion, progressive conversion) under “chroma subsampling” under the behavior tab in the x264 settings? or a combination of the two?”
Is important to avoid the “Deinterlace source video” in the QT GUI (size). The default QT deinterlace is crap.
Deinterlace with whatever other application before to make the xh264 compression.
Then on the x264 options (Behavior) you must chose “Progressive conversion” because your footage is already Progressive. That affects only to the way Chroma is compressed.
rafael

http://www.nagavideo.com

[Jay Church]

[Rafael Amador] “Is important to avoid the “Deinterlace source video” in the QT GUI (size). The default QT deinterlace is crap.
Deinterlace with whatever other application before to make the xh264 compression.
Then on the x264 options (Behavior) you must chose “Progressive conversion” because your footage is already Progressive. That affects only to the way Chroma is compressed.”

how/where would i deinterlace before the quicktime conversion export?

my transcoded footage that i’m using in FCP and my sequence setting are both interlaced.

thanks,

_jc