Split HD clips from HV20 capture bug in VS11+

[fredb]

I just found a bug in my setup, I don't know if it is VS11+ or my Canon HV20.
I am transferring videos shot in HD, but since I don't have an HD burner I am letting the camcorder downscale to DV and editing the traditional way to create SD DVDs. I use the option in VS11+ to split by scene. The problem is that the captured clips all have one extra frame left at the end from the next scene. It takes A LOT of manual work to remove all those last frames.

Has anyone experienced the problem with other HD camcorders, or with the Canon HV20?

By the way I tried to capture in HD from the HV20 (MPEG) and let VS11 convert to DV, but the quality is not as good as straight DV from the camcorder.

Fred

[MrA]

By the way I tried to capture in HD from the HV20 (MPEG) and let VS11 convert to DV, but the quality is not as good as straight DV from the camcorder.

Fred

Once it is mpeg format, there is no need to convert it again to DV. Smart render will handle it. Leave it at Mpeg. (mpeg-2) You will always lose quality cross converting. Once it is at mpeg-2, that is it, thats the target for dvd. When you edit in mpeg-2, (adding transitions, etc.) that will be re-rendered, but, being a transition, you will lose some quality, BUT, so what, it's a transition, who is to notice, except a very bored computer geek. (if your target audience is computer geeks, charge them more..)

I have a very, very bad habit of making the mistake of making the video file a Mpeg-1 file. If they would remove that option from the render, I'd be a very happy person.

To sum it up, if your final target is a DVD, make mpeg-2 files. Use those files in your DVD creation process.

[etech6355]

MrAmigo2121,
I think he means he's capturing HDV mpeg2 (25MBS) and converting the hd-mpeg2 to DV format (rendering)for editing standard video format.

Not sure about the black frames, that's been reported with many of the HDV cams, not just the HV20.

[jmone]

Hi fredB - I've got a HV20 and after trying a few different workflows settled on using VS11+ for Capture, Edit, and Rendering:
1) Capture from the DV20 in HDV (you get a 1440x1080i Program Stream File). Unfortunatly "Split by Scene" is not supported during HDV Catpure but you can run "Split by Scene" after catpure but it is slow and not as good IMHO.
2) Edit away in the HDV format
3) Output either in HDV files (for playback on a PC) or the following for HDV to DVD Disk Template:
24 bits, 720 x 576, 25 fps, Upper Field First, (DVD-PAL), 16:9
Video data rate: Variable (Max. 8000 kbps)
Audio data rate: 384 kbps
MPEG audio layer 2, 48 KHz, Stereo

The upside is you keep everything in the native HDV format till you choose the final output format to render to. The downside for me was the loss of "Split by Scene" during capture and you get one big file - but at least there is no final "Black Frame" issue. This issue has also been noted by some using HDVSplit but I've never seen it.

Thanks
Nathan

[etech6355]

jmone,
Under preferences (F6 Hotkey) make sure <edit tab> resampling quality = Best.
<General Tab> "Notify when inserting first video into timeline" = Yes.

After you capture using VS11+ then VS11+ should ask you if you want to make the project settings match the video. Always answer "Yes".
You may find if VS has matched the project settings to your source video that going to "Share -> "Same As Project Settings" or the "Mpeg Optimizer" -> show details and "Accept" will smartrender the project faster.

The HDV templates can be tricky, they can change, and the templates depend on the source video along with the project settings. You can double check the settings by when you choose the HDV -> template_X in the bottom left hand corner will be the settings that will be used to render the mpeg file to. In some of these templates and depending on the project source video & project settings these settings can change. I believe the HD templates wokr in a similar method. Just thought I'd mention this.
The trial version has expired on my machine, can't remember the exact settings but matching the project settings and "share -> same as project settings" has always been a reliable method in VS.

[fredb]

By the way I tried to capture in HD from the HV20 (MPEG) and let VS11 convert to DV, but the quality is not as good as straight DV from the camcorder.

Fred

Once it is mpeg format, there is no need to convert it again to DV. Smart render will handle it. Leave it at Mpeg. (mpeg-2) You will always lose quality cross converting. Once it is at mpeg-2, that is it, thats the target for dvd. When you edit in mpeg-2, (adding transitions, etc.) that will be re-rendered, but, being a transition, you will lose some quality, BUT, so what, it's a transition, who is to notice, except a very bored computer geek. (if your target audience is computer geeks, charge them more..)

I have a very, very bad habit of making the mistake of making the video file a Mpeg-1 file. If they would remove that option from the render, I'd be a very happy person.

To sum it up, if your final target is a DVD, make mpeg-2 files. Use those files in your DVD creation process.

Thanks for the help, but I need to convert the HDV source to DVD. The compression is MPEG2 in both cases, but obviously HD won't work on a standard DVD so the frame size needs to be changed. Hence re-rendering is necessary and SmartRender won't help here.

[fredb]

Nathan,

Thanks for the detailed description of your workflow. It is unfortunate that "Split by scene" does not work as advertized. I am glad to read that I am not the only one
Hopefully Ulead will find a way to fix it in a service pack. I got an answer from Coel Tech support, here it is in its entirety:
< START>
Response (Francis) - 07/30/2007 12:43 AM Dear Fred,

Thank you for contacting Corel Customer Support Services. You should click on Scan to make sure that the scenes are cut properly.
<END>

Wow! Now that's is why I buy my software and don't use pirated copies, I get great tech support Well at least they were quick to respond this time...

I find that the video rendered using this workflow is not as good as what is generated by the cam itself in SDV output mode. I'll have to check my encoding settings, maybe I had the field order/frame set wrong. I am in NTSC land, but apart from the frame size all should be the same as what you use in PAL. With DV I know the field order can be different between PAL and NTSC, but with HDV I believe they are identical.

What is interesting too is that as I look at my segmented videos, and ask VS to split them by scene again, VS only finds one segment. So the timecode seems to be right, but the video data got shifted by a number of frames. By the way I don't get black frames added, I really get a few video frames from the preceeding scene added to the current one.

Frederic

[sjj1805]

Are you sure that extra frame is actually on the video and not just the thumbnails on the timeline? By this I mean does it appear on the preview screen. I have noticed with several video editing programs that on the timeline you often get a thumbnail of what the clip was joined to - but does not display in the preview window.

In my opinion this was not a bug but in fact a tremendous help when moving clips around - you can see which clip it was previously joined to.

[fredb]

I am sure there are extra frames from the adjacent clips, usually 3 extra frames that I have to remove. This has nothing to do with the thumbnail.

I just finished a project with 160+ scenes, and had to trim all 160 scenes manually. Not fun

[sjj1805]

This will help explain what I am illustrating. I have taken these screen shots from MediaStudio pro because it is easier to show what I mean on a staggered timeline.

Zoom level = 1 second.


Zoom level = 1 frame.


Like I said I like it this way because if needs be I can see which clip it was joined to if I decide to move the clips around.

[jmone]

fredb - regarding the Quality of DV and HDV. FYI both DV and HDV are recorded to tape at max rate of 25mbs. To get the higher 1440x1080 resolution into the same data rate one of the trade offs were the change of compression methods from DV-AVI to MPEG-2.

1) If you were only ever going to want to view your footage in SD, I'd shoot and edit in DV then transcode to your prefered DVD Specs (eg 480i MPEG-2 for NTSC) as part of your DVD creation process.

2) If you think you will want to view your footage in HD at some point then I'd shoot and edit in HDV the transcode directly to your prefered DVD Specs. Unless this is more that a hobby I would not bother with using an "intermediate" codec for editing (many will argue you get better quality due to the problem of long GOP editing) - also I've not heard of DV being used at all for this purpose.

Thanks
Nathan

More info at http://en.wikipedia.org/wiki/HDV