Compressor settings for 1080i@60 going to DVD, getting jaggies

[First Last]

Pretty sure this has been covered…

Switch the field dominance in your final cut pro sequence settings to None and export a self contained Quicktime (something like, Pro Res 422 1080p30) then run that quicktime through compressor using your 90min DVD Compressor settings. I think the key here is changing your field dominance to none and exporting something at 30p and not 60i.

Go fly fishing.

[Bret Williams]

I’ve had the same issues and no one has had an answer. All the DVDs I’ve made from HD footage OTHER than native XDCam sequences have been great. You’re suggesting exporting native interlaced as progressive (no deinterlacing involved) and creating a progressive DVD as the solution? If someone purposely shot and edited 1080i, they would want 480i DVD, no?

[Michael Gissing]

Try exporting a ProRes 422 quicktime, bring that into Compressor. Don’t forget frame controls to get best resizing from HD to SD

[Chris Borjis]

[Michael Gissing] “Don’t forget frame controls to get best resizing from HD to SD”

That’s what almost everyone does forget.

and the reason they look good out of Vegas is that it uses bicubic resizing when it exports footage, something final cut is sorely lacking.

[Tamaki Ueda]

Hi David,

Here’s my recipe when I edit 1080 60i and author SD DVD with FCP, Compressor and DVD Studio Pro.

1. Edit in Apple ProRes422
2. Export 1. as self-contained QT
3. Compressor: Encode 2. as Apple ProRes 422 720×480 Anamorphic (Resize: Better, Field Order: BOTTOM FIRST)
4. Compressor (again) : encode 3. as mpeg2
5. Author!

Please test this, I hope this works for you 🙂

Tamaki

[David Shirey]

Thanks for that detailed process, it’s great to know what is needed for best results. Unfortunately if we had to actually render out 3 separate files before making a DVD, the process would be so time-consuming it would be in our best interests to chuck the mac out a window and just build another PC for a third of the price. Our FCP guy is trying several methods posted here right now, so I’ll have him try it your way also just to see what yields the best results, thanks!

[Craig Alan]

https://www.kenstone.net/fcp_homepage/hdv_to_sd_dvd.html

OSX 10.5.8; MacBookPro4,1 Intel Core 2 Duo 2.5 GHz
; Camcorders: Sony Z7U, Canon HV30/40, Sony vx2000/PD170; FCP certified; write professionally for a variety of media; teach video production in L.A.

[Carl Rohumaa]

Interesting. Are you totally sure that your media and your sequence are the same? If they are not then the final playback will not be able to de-interlace so you will get jaggies. Jaggies are where you no longer “see” the interlacing lines as they are embedded in the video forever. Check your media in the motion tab and make sure there is no scaling or under distort no aspect ratio setting. The aspect should be 0.
If they are not right you will have to make a correct sequence first and then cut and paste to it. The resultant video will be distorted so make a clip right (get rid of any scaling and aspect distortion) and cut and paste the motion and distort attributes to the other clips. This should be foolproof.
You should be able to go straight to compressor with no rendering or other export. In fact some delete all their render files before export to force a “clean” export especially with complex edits.
If this does not work then there is a serious bug to file in FCP. Don’t make your mpeg from a reference QT export as Compressor uses your edits in the timeline as compression markers treating each clip individually.
FCP can be “clever” at making your media look OK in the timeline when it is really distorting it to fit. Obviously all clips that have had frame resize or speed changes etc will not de-interlace either so in the sequence these clips need to be de-interlaced.
Feel free to shoot me down in flames if this is wrong folks or if you have a better insight, or something else to say about it.
One day there will be no interlacing!?

Video Tech

[David Shirey]

Well we did have all the sequence settings match the exact type of footage we were using. Of all the methods we tried, exporting a 1440x1080i prores quicktime file out of FCP then putting that into compressor seemed to work best. It still left smaller jaggies on some still scenes but it’s way better than what we were getting. A 1080p file had absolutely no jaggies on scenes that were still, but created the worst warbled interlacing effects on anything with motion so obviously that’s no way to go. Exporting 720×480 prores files from FCP then converting those to mpeg2 had awful results. Due to the extra time involved our ultimate solution is to just get another PC system, but it looks like we’re going to keep this around so we can tweak projects that freelancers edit for us, so I appreciate everyone helping us fix this.

[First Last]

I think there’s a lot that wasn’t covered and acknowledged in your workflow. I can vouch for exporting a progressive video from a sequence that defaults to i60 if you shot progressive… the whole 30p in a 60i container and what-not.

If you can give us some more details (it was shot at, captured at, etc.) we might be able to give you more solid advice.

Go fly fishing.