Capture: Split by Scene
Moderator: Ken Berry
-
lancecarr
- Advisor
- Posts: 1126
- Joined: Sat Dec 11, 2004 6:34 am
- System_Drive: C
- 32bit or 64bit: 64 Bit
- motherboard: eMachines ET1861
- processor: 3.20 gigahertz Intel Core i5 650
- ram: 12GB
- Video Card: ATI Radeon HD 5400 Series
- sound_card: ATI High Definition Audio Device
- Hard_Drive_Capacity: 700GB
- Location: Taipei, Taiwan
- Contact:
No, VS can only capture as one continuous MPEG.
There is another program call HVDsplit (I think) which can do it.
Try here: http://strony.aster.pl/paviko/hdvsplit.htm
There is another program call HVDsplit (I think) which can do it.
Try here: http://strony.aster.pl/paviko/hdvsplit.htm
-
lancecarr
- Advisor
- Posts: 1126
- Joined: Sat Dec 11, 2004 6:34 am
- System_Drive: C
- 32bit or 64bit: 64 Bit
- motherboard: eMachines ET1861
- processor: 3.20 gigahertz Intel Core i5 650
- ram: 12GB
- Video Card: ATI Radeon HD 5400 Series
- sound_card: ATI High Definition Audio Device
- Hard_Drive_Capacity: 700GB
- Location: Taipei, Taiwan
- Contact:
-
jmone
- Posts: 93
- Joined: Sun Jun 17, 2007 12:28 pm
- System_Drive: C
- 32bit or 64bit: 64 Bit
- motherboard: ASUS P8Z77-VPro
- processor: i7 3770K OC
- ram: 16GB
- Video Card: ASUS GTX660 OC
- sound_card: FiiO E10
- Hard_Drive_Capacity: 60TB
Please ask support for this - the more of us that do they may decide to add it (it drives me mad!). The problem with HDVSplit is it brings the files across as TransportStream (TS) not ProgramStream (PS), at least VS11+ does an on the fly conversion to PS.
FYI the response I got from Ulead on the Split by Scene (Timestamp) during HDV Capture was the lame...."Unfortunately, we do not have any available updates regarding it. We're continually developing updates and will forward it as a request although there aren't any guarantees that a plug-in will be released on the next update."
FYI the response I got from Ulead on the Split by Scene (Timestamp) during HDV Capture was the lame...."Unfortunately, we do not have any available updates regarding it. We're continually developing updates and will forward it as a request although there aren't any guarantees that a plug-in will be released on the next update."
-
Trevor Andrew
-
sjj1805
- Posts: 14383
- Joined: Wed Jan 26, 2005 7:20 am
- System_Drive: C
- 32bit or 64bit: 32 Bit
- motherboard: Equium P200-178
- processor: Intel Pentium Dual-Core Processor T2080
- ram: 2 GB
- Video Card: Intel 945 Express
- sound_card: Intel GMA 950
- Hard_Drive_Capacity: 1160 GB
- Location: Birmingham UK
I don't have HD but my understanding of the issue is that video recorded as DV will contain time/date recording information and so when capturing (or to be more accurate transferring) video from the camcorder to the computer, the program will detect changes in the time/date recording information and treat each change as a new scene. This gives the user two choices
a. Split by scene during the transfer - resulting in several small files
b. Transfer as one large continuous file and then later use the split by scene function to generate several thumbnails which are rather like chapter points on a DVD Video.
High Definition would work in the same manner as video captured from an analogue source such as a TV Card. It is in MPEG2 format - or in some cases MPEG4 or even DivX. These do not have time/date recording information and so can only be captured as one large file. Following capture the program can then detect scene changes based upon things like the entire picture and or sound change dramatically - such as when the adverts come on during a TV program.
Like I said, I don't have HD and so this is just my best guess.
Perhaps some users with HD can confirm or negate this theory.
a. Split by scene during the transfer - resulting in several small files
b. Transfer as one large continuous file and then later use the split by scene function to generate several thumbnails which are rather like chapter points on a DVD Video.
High Definition would work in the same manner as video captured from an analogue source such as a TV Card. It is in MPEG2 format - or in some cases MPEG4 or even DivX. These do not have time/date recording information and so can only be captured as one large file. Following capture the program can then detect scene changes based upon things like the entire picture and or sound change dramatically - such as when the adverts come on during a TV program.
Like I said, I don't have HD and so this is just my best guess.
Perhaps some users with HD can confirm or negate this theory.
Steve,
It's taken me awhile to get a handle on these container formats, which is what we are actually discussing.
The term MPG is a container format, within the container lies the video & audio elementary streams.
Both formats still use elementary streams within the container format. The difference is whether the container format is PS or TS.
TS format is designed to be de-multiplexed on the receiving end.
If it's PS format then the streams are mpeg1 or mpeg2 video. The TS container format can carry more video formats as elementary streams
The broadcasting stations usually broadcast their HD in a TS format container. It's the capturing device that determines how to manipulate and process the video/audio. The broadcasters can put a few different video/audio codecs inside the of TS container, including more than one video or audio track. The TS container is still an mpeg2 stream used to carry the information.
Ulead can do this and make it work, I'm sure eventually they will.
Ulead converts the tape mpeg2 TS to PS format during capture, so does MF6+, PS can edited easier and smart-rendered almost like a file copy. TS can still be smart rendered, but the container format still has to be rebuilt which takes much longer than smart rendering PS format. PS format uses a fixed timebase reference.
HDV tape contains the time & date information, it's all in the capturing software. When using HDVSplit to capture from the middle of the HDV tape playing the raw m2t files back on my media player the time thats' displayed is the actual time that was on the tape. Even though I'm playing that file from the beginning the time counter reports 25:00:00 etc.
DV tape is similar to Transport Stream, if it wasn't then to capture you would have to start from the beginning of the tape, you wouldn't be able to capture from the middle of the dv tape. DV tape is not written on tape as DV-Type1, it's raw data, conversions are performed during capture.
Confusing, we can have a video that's in mpeg2 Transport Stream container format that holds mpeg2 video & dolby audio elementary streams.
Or, we have a video that's in mpeg2 Transport Stream container format that holds avc/h264 video & dolby audio elementary streams.
With the release of VS11+ & MF6+ ulead now sparses the mpg file to determine the container format and format of the elementary streams contained within the TS or PS container.
This is so confusing to explain, reminds me of "Abbott & Costello's" Who's on First..................no, your on second.........
It's taken me awhile to get a handle on these container formats, which is what we are actually discussing.
The term MPG is a container format, within the container lies the video & audio elementary streams.
Both formats still use elementary streams within the container format. The difference is whether the container format is PS or TS.
TS format is designed to be de-multiplexed on the receiving end.
If it's PS format then the streams are mpeg1 or mpeg2 video. The TS container format can carry more video formats as elementary streams
The broadcasting stations usually broadcast their HD in a TS format container. It's the capturing device that determines how to manipulate and process the video/audio. The broadcasters can put a few different video/audio codecs inside the of TS container, including more than one video or audio track. The TS container is still an mpeg2 stream used to carry the information.
Ulead can do this and make it work, I'm sure eventually they will.
Ulead converts the tape mpeg2 TS to PS format during capture, so does MF6+, PS can edited easier and smart-rendered almost like a file copy. TS can still be smart rendered, but the container format still has to be rebuilt which takes much longer than smart rendering PS format. PS format uses a fixed timebase reference.
HDV tape contains the time & date information, it's all in the capturing software. When using HDVSplit to capture from the middle of the HDV tape playing the raw m2t files back on my media player the time thats' displayed is the actual time that was on the tape. Even though I'm playing that file from the beginning the time counter reports 25:00:00 etc.
DV tape is similar to Transport Stream, if it wasn't then to capture you would have to start from the beginning of the tape, you wouldn't be able to capture from the middle of the dv tape. DV tape is not written on tape as DV-Type1, it's raw data, conversions are performed during capture.
Confusing, we can have a video that's in mpeg2 Transport Stream container format that holds mpeg2 video & dolby audio elementary streams.
Or, we have a video that's in mpeg2 Transport Stream container format that holds avc/h264 video & dolby audio elementary streams.
With the release of VS11+ & MF6+ ulead now sparses the mpg file to determine the container format and format of the elementary streams contained within the TS or PS container.
This is so confusing to explain, reminds me of "Abbott & Costello's" Who's on First..................no, your on second.........
Last edited by etech6355 on Sat Feb 09, 2008 5:03 am, edited 1 time in total.
-
jmone
- Posts: 93
- Joined: Sun Jun 17, 2007 12:28 pm
- System_Drive: C
- 32bit or 64bit: 64 Bit
- motherboard: ASUS P8Z77-VPro
- processor: i7 3770K OC
- ram: 16GB
- Video Card: ASUS GTX660 OC
- sound_card: FiiO E10
- Hard_Drive_Capacity: 60TB
Yup - "Spilt by Scene" in DV works on creating seperate files for each change in the the Time/Date Stamp and I WHISH Ulead would do the same with HDV.
HDV also has a Time/Date Stamp but either:
1) VS11's plug in does not read it,
2) they do not use it, or
3) there is some issue with the informatio being lost in the TS to PS conversion that is done.
A freeware utility called HDVSPLIT does do exactly this but leaves you with TS format files & I prefer the PS format that VS11 saves the capture in on your HDD. So my choice is still:
1) Capture with VS11 and work with one big PS file or
2) Capture with HDVSPLIT and work with lots of TS files then transcode back to PS.
Nathan
HDV also has a Time/Date Stamp but either:
1) VS11's plug in does not read it,
2) they do not use it, or
3) there is some issue with the informatio being lost in the TS to PS conversion that is done.
A freeware utility called HDVSPLIT does do exactly this but leaves you with TS format files & I prefer the PS format that VS11 saves the capture in on your HDD. So my choice is still:
1) Capture with VS11 and work with one big PS file or
2) Capture with HDVSPLIT and work with lots of TS files then transcode back to PS.
Nathan
Yes, if your using smart rendering for HDV video the ulead software does this in PS format 10 times faster than in the TS format.
VS still is smart-rendering the video & audio, the extra/longer time of smartrendering TS format is rebuilding the TS stream.
VS10 had a nice conversion, if you loaded a TS file into VS10+ the program would ask you do you want to convert it to PS format. VS10+ converted the TS to PS very fast, similar to a file copy.
VS still is smart-rendering the video & audio, the extra/longer time of smartrendering TS format is rebuilding the TS stream.
VS10 had a nice conversion, if you loaded a TS file into VS10+ the program would ask you do you want to convert it to PS format. VS10+ converted the TS to PS very fast, similar to a file copy.
This is my #1 complaint with all the Ulead video software (and most other HDV-capable editing.) If this worked, I would still be using MSPro instead of VS11+.
I WANT SCENE-DETECTION of HDV Program Stream FILES.
If HDVsplit - a tiny free-ware program- is capable of this, what the heck is so hard that nobody does it?
Scene detection is the #1 key basic operation for most editing (that has scene cuts)- it's the basic unit of edit operations! How can this not work?
HDV *DOES* have the timestamp info, in the original stream. Why nobody bothers to read it is beyond me.
If it is in the TS and not PS stream, then fine- work around it by editing TS natively much faster please.
Or, still work with PS MPG2, but just record "markers" during the capture conversion process- eg an UPD file, like during rendering MPGs- so that the scene split command can later detect/use that information later.
Or it could split the PS MPG2 files into separate files at each scene break, although I'd prefer both options (physical and virtual splitting) since sometimes it's easier to handle a few large, virtually split files than 100s of tiny short physically-split files.
Whatever method, it's better than all the hoops and extra time/conversion/disk wasted with the extra steps!!
This is EXTREMELY FRUSTRATING AND TIME CONSUMING and kills the HD format. I have been waiting over 1year for anyone to fix this right (or even understand the need.) Corel, please fix this.
I WANT SCENE-DETECTION of HDV Program Stream FILES.
If HDVsplit - a tiny free-ware program- is capable of this, what the heck is so hard that nobody does it?
Scene detection is the #1 key basic operation for most editing (that has scene cuts)- it's the basic unit of edit operations! How can this not work?
HDV *DOES* have the timestamp info, in the original stream. Why nobody bothers to read it is beyond me.
If it is in the TS and not PS stream, then fine- work around it by editing TS natively much faster please.
Or, still work with PS MPG2, but just record "markers" during the capture conversion process- eg an UPD file, like during rendering MPGs- so that the scene split command can later detect/use that information later.
Or it could split the PS MPG2 files into separate files at each scene break, although I'd prefer both options (physical and virtual splitting) since sometimes it's easier to handle a few large, virtually split files than 100s of tiny short physically-split files.
Whatever method, it's better than all the hoops and extra time/conversion/disk wasted with the extra steps!!
This is EXTREMELY FRUSTRATING AND TIME CONSUMING and kills the HD format. I have been waiting over 1year for anyone to fix this right (or even understand the need.) Corel, please fix this.
