Tom Brooks
Forum Replies Created
-
I find that the pulsing can consist of very tiny fluctuations in the color of the pixels and that one monitor will show them and another will not. It could be as simple as a difference in monitor brightness and contrast calibration between the two systems.
-
I suspect what you are seeing is the characteristic pulse caused by the 15 frame GOP structure of MPEG-2 encoded with the standard presets in Compressor or DVD-SP. I can’t explain it technically for you and there is not always a cure. For the detailed moving background you have, I’d try to maximize the data rate as much as possible, pushing up close to 7Mbps. Too high a rate can cause compatibility problems for your burned disks, however. You can try modifying one of the CBR presets to a higher data rate and use that instead of a VBR preset. I also find that the pulsing I’m referring to shows up more on some monitors than others (particularly LCDs).
The Compression Techniques forum on the Cow might yield some answers, too. Another MPEG-2 encoder might work better than Compressor (Bitvice, Main Concept, et al).
-
There’s a lot on this forum and others on the topic of stills in Final Cut. Search the tutorials on the Cow and also over on Kenstone.net.
If your output is primarily Web, I’d make the sequence progressive and deinterlace the few video clips in it. Final Cut has a poor deinterlace filter but there are better ones available from third-party sources (some for free). Just open your sequence settings, and change the compressor to ProRes 422 HQ or Uncompressed 8 or 10 bit (if your drives and CPU can handle it). Set the Field Dominance to None. This should solve interlace issue on the stills. Again, if you have video clips, they should now be deinterlaced using a filter. My suggestion here is to match the sequence to the majority of your source material and to the final output.
Try to get away from the term DPI for video. That term is aimed at print. If you do the appropriate math, you see that the intent of using 300 DPI is simply to increase the pixel dimensions of the image. Video is pixels. Always, always look at the pixel dimensions of your stills. To pan and zoom they should be around twice the dimension of your sequence.
Truth be told, I have never had tremendous luck with JPEGs in FCP. I change to TIFF as early in the process as possible. Something about JPEG compression seems to make jaggy edges worse as they are scaled and moved in Final Cut. The progressive sequence should eliminate many of the problems caused by hi-res images in an interlaced project.
-
Your posts are very nicely illustrated. The solution was simple, but it’s one that is often overlooked by editors new to Final Cut.
Final Cut Pro 6.0.5, Mac OS-X 10.5.5, Quicktime 7.5.5, Adobe Prod Prem CS4, G5 Quad 2.5, Kona-LHe V6, 8.5GB RAM, Nvidia GeForce 7800-GT 256MB, G-RAID 2x1TB FW800, 6TB RAID-5 (Enhance E8-ML, Highpoint 2322), Panasonic HVX-200P P2. Also MBP 17″ Core 2 Duo 2.5, 4GB, GeForce 8600M GT 512MB.
-
A search of posts should turn up many answers to this.
https://forums.creativecow.net/readpost/8/883079 -
All I’m sayin’ is that when it’s all you got, some EQ can help. It’s not the reverb you’re getting rid of at all, it’s just the unpleasant effect of room resonance. If a room rings like a bell at 300Hz, you can sometimes make a big difference by using a parametric eq at 300Hz to reduce the effect of that resonance on your recording.
-
Equalization is usually the key. The hollowness is caused by certain frequencies of sound being emphasized by the acoustic properties of the room. Usually, reflections in the room will increase certain frequencies and decrease others. The first thing is find the frequency that is overly loud and reduce that. Then, attempt to judge what is missing from the audio. Finally, a little noise reduction can help. I would work in Soundtrack Pro.
Final Cut Pro 6.0.5, Mac OS-X 10.5.5, Quicktime 7.5.5, Adobe Prod Prem CS4, G5 Quad 2.5, Kona-LHe V6, 8.5GB RAM, Nvidia GeForce 7800-GT 256MB, G-RAID 2x1TB FW800, 6TB RAID-5 (Enhance E8-ML, Highpoint 2322), Panasonic HVX-200P P2. Also MBP 17″ Core 2 Duo 2.5, 4GB, GeForce 8600M GT 512MB.
-
Try this: Use Advanced Format Conversion. Use the DV50 preset — not the DV50 Anamorphic. Go to the Geometry tab in the setting. Make sure Dimensions (encoded pixels) is set to 720×480 601/DV. Set Output Image Inset (padding) to 16:9 1.78:1. In the Preview window, click the button in the far upper right. You should now see a letterboxed 16:9 image on the right side of the source/preview split. This is your output preview. Does it look correct? Set an in and an out a few seconds apart and submit to convert a short test. Should be correct.
-Tom -
Assuming your sequence is indeed 1080i50 I don’t see much wrong with your Final Cut workflow. The pans are really fast for a Web clip. The codec will have to work really hard to reproduce those pans. Are you seeing the problem on your edit monitor, or just in the final compressed result?
-
Animation codec out of AE would be OK. Your mistake was in changing your sequence to match the AE clip. You should have answered “no” to the “change sequence settings dialog.” Then your AE clip would need to be rendered, but your sequence would remain in a proper editing codec (DV in this case).
Do you have a Kona or Blackmagic card? If so, you can use the UC or better yet ProRess 422 HQ sequence setting (and render AE in ProRes too) and output from that.