DV sequence to H.264 blurry

[David Johnson]

If I understand correctly what you’re asking, converting DV footage to H264 can’t improve the blurriness since transcoding from any codec to another can’t regain image data thrown away by prior compression. DV (especially DV25, vs. DV50) is notoriously soft and nothing can be done about that short of re-digitizing the footage in a better codec. Perhaps HDV since that’s what the source is, although HDV is really just a variation of DV). You could also digitize to a better editing codec like ProRes (H264 is a delivery codec, rather than an editing codec). You already mentioned you can’t re-digitize so perhaps try a slight sharpening filter, but be careful not to sharpen too much.

[Mark Laslo]

Thank you for your reply,

The problem I’m running into is that the footage looks fine in FCP but when it gets converted to h.264 I get artifacing and distortion especially around the cross dissolves. I understand that I can’t make the footage any sharper, but I was hoping that I could change a sequence setting to atleast get the graphics to be sharper and maybe prevent this distortion from happening during the conversion process.

Sorry if i wasn’t clear in my original post – I think i gave too much detail.

Thanks,

Mark

[Bret Williams]

First, DV isn’t notoriously soft. It’s a very good 5:1 compression scheme where most of the compression is in the color info, making itva lesser choice for bluescreen work. It’s not the best format for super tiny graphics either but most of the time DV should look comparable to betacam. In some aspects better, in some worse. I’m tired of people immediately blaming DV. It’s nit usually the culprit in most people’s complaints. It take a pretty well trained eye to pick out any softness or artifacts.

That aside, if the footage looks fine in FCP, the the DV is fine and the problem is in your compression settings. What are they? Data rate, frame rate, 2 pass, cbr or vbr, etc?

[Mark Laslo]

Bret thanks for your response.

Here is the settings I’m using in compressor

File Extension: mov
Estimated size: 322.56 MB/hour of source
Audio Encoder
AAC, Stereo (L R), 48.000 kHz
Video Encoder
Format: QT
Width: (100% of source)
Height: (100% of source)
Pixel aspect ratio: Default
Crop: None
Padding: None
Frame rate: 29.97
Frame Controls: Automatically selected: Off
Codec Type: H.264
Multi-pass: On, frame reorder: On
Pixel depth: 24
Spatial quality: 75
Min. Spatial quality: 25
Temporal quality: 50
Min. temporal quality: 25
Average data rate: 0.717 (Mbps)
Maximum data rate: 0.717 (Mbps)
Hinted for QuickTime
streaming server

I’ve been using a similar version except it’s been converting my 720p prores to 640×360 for our website and we’ve had no issues so I’m confused why I would be getting all this distortion when converting the DV file.

There are some issues coming out of FCP with the graphics – we have a semi-transparent logo in the corner that gets slightly distorted which is why I was wondering if it would make sense (or if it was even possible) to convert to ProRes so at least the graphics would be clean. My biggest concern however is making sure the video doesn’t get increasingly blurry or gain artifacts during the transcode.

Thanks,

Mark