Hi there,
I captured 50 minutes of footage into Mediastudio Pro, edited and then authoured the DVD with Moviefactory. The VIDEO_TS folder size was only 1.64GB. Is this normal for this amount of footage or can i change the settings to capture and edit at maximum quality? Thanks in advance.
Final VIDEO_TS Folder Size?
-
Terry Stetler
- Posts: 973
- Joined: Thu Mar 17, 2005 3:34 pm
- Location: Westland, Michigan USA
Follow the Army's M1-Eyeball test: if it looks good don't mess with it 
Sounds like it's about right for the length of the program, and if you encoded at the default 6000 kbps upping it to 7000-9000 won't matter that much and could even hurt playback on some DVD decks. Upping the quality setting matters more.
Sounds like it's about right for the length of the program, and if you encoded at the default 6000 kbps upping it to 7000-9000 won't matter that much and could even hurt playback on some DVD decks. Upping the quality setting matters more.
Terry Stetler
-
Terry Stetler
- Posts: 973
- Joined: Thu Mar 17, 2005 3:34 pm
- Location: Westland, Michigan USA
File/Create/Video File/Options/Compression
Once in the Insert Video File dialog select MPEG as the filetype then click Options. In the Compressoin tab pick NTSC DVD or PAL DVD and adjust the slider at the top to 90. It'll take longer to render, but should give better results.
Unless you're experiences in things MPEG stay away from the Advanced menu.
Once in the Insert Video File dialog select MPEG as the filetype then click Options. In the Compressoin tab pick NTSC DVD or PAL DVD and adjust the slider at the top to 90. It'll take longer to render, but should give better results.
Unless you're experiences in things MPEG stay away from the Advanced menu.
Terry Stetler
There are many factors that can change the file size and what is desirable. If we start with the source video, at the two extremes, you can have poor analogue footage (e.g., from a VHS tape) and you can have uncompressed video from studio equipment. In the latter case, you want as little video degradation as possible, but a reasonable amount of degradation would be unnoticeable because it would be drowned out by the poor quality of the source.
That having been said, digital degradation is different from poor analogue quality, so you still have to be reasonable.
As a guideline, let's take an average quality digital signal, such as from a mini-DV camcorder and the individual can extrapolate what I say here, up or down according to his own requirements.
OK, let's start with the audio. We have a choice of 3 systems: LPCM, MPEG and AC-3 (Dolby Digital). The first is uncompressed and gives the largest file. The other two are compressed. The AC-3 is not written into PAL standards but all modern PAL players (e.g., <6 years old) accept it without problems. I therefore strongly recommend using this, because it gives good quality for a modest bitrate. Mostly, ordinary 2.0 stereo will be used and 192 kbit/s is more than adequate for good quality music.
For video, it becomes less cut and dried. Theoretically, you can use bitrates up to about 9000 kbit/s: in practice, you can't. Why? Because many players simply will not accept such a high bitrate, especially from a DVD±R/RW disk, whose 1:0 contrast is much lower than a pressed (commercial) disk. Then comes the question of quality. For our DV average, I've done extensive tests and I've found that 6000 kbit/s is good enough to encode the signal with no significant deterioration of the signal quality under most conditions, even for moving subjects. This is because the signal is already significantly compressed in the camcorder. Very occasionally, 7000 kbit/s will give a slight improvement where you have specific problems. 6000 kbit/s will play in all players, that I know of. The deterioration at 5000 kbit/s is just noticeable to a trained eye but becomes noticeable at 4000 kbit/s to most people on fast motion scenes, although not on scenes with little motion.
A DVD±R with a reasonably simple menu, no sub-title tracks and no extra audio tracks, with sound at 192 kbit/s and video at 6000 kbit/s will fit in at about 2.8 Gb/h so, theoretically, will be good for about 90 minutes, but it is not good practice to fill a disk to more than about 4.0 - 4.1 Gb for a number of reasons, especially for mediocre quality blanks (single layer).
Then there is the thorny question of variable and constant bitrates. Commercial pressed disks almost invariably use variable bitrates, so why shouldn't DVD±R's? Well, in the first place, we don't have at our disposal the equipment that Hollywood uses, unless we spend in at least the six figures. We have, as a rule, encoders that allow only one or two passes. These attempt to up the bitrate where there are significant differences between one frame and the next and to lower it where two consecutive frames are very similar. None of them analyse the actual visual content, like is done in Hollywood, which may do 20 or more analytical passes before deciding on the optimum bitrates for a given scene. Coming back to our DV video, which parts of the video have the greatest difference between successive frames? Where there are transitions. So our low-cost encoders up the bitrate most at transitions. Even worse, let's imagine the source file is analogue. This almost always has a poor signal-to-noise ratio, usually of the order of 30 dB (except for direct connections to studio-quality cameras). When viewing the video, this is not usually very visible for a number of reasons, but it does mean that the encoder will interpret the noise as movement in the video and will up the bitrate, especially on scenes with a low key illumination. Your moonlit landscape will therefore whizz through at a high bitrate, while a clip of field events will encode at a lower bitrate for the 100 m sprint. For these reasons, VBR tends to be grossly over-hyped with our resources. This is why I recommend CBR for all DV videos of less than about 1¾ or 2 h duration and all videos from analogue sources, of any duration or content. There may be a small advantage in going to VBR for >2 h videos from DV sources, where there is much rapid movement, but even this remains to be proved.
Finally, what about shorter videos? I still apply the 6000 kbit/s CBR rule, even though the disk is not filled. The reliabilty of the results will be better assured. Higher bitrates may cause more glitches. Some DVDs I've produced have lasted only 10 minutes but have been encoded at 6000 kbit/s, even though they are only 1/8 full.
Some people here may dispute what I say, gained from experience, on the principle that they believe "bigger is better", but my views are supported by others. As an illustration, I recently looked at four DVDs lasting from 12 to 70 minutes each and distributed by the Cyprus government on DVD-R disks (quantities insufficient to justify pressing). These were produced by three different European professional houses of renown using obviously professional equipment and were of superb quality (at least for the most part). All four were encoded at 6000 or 6500 kbit/s CBR.
To answer your question, 1.64 Gb is light for 50 Min. You obviously used a lower bitrate than I'd consider ideal. 6000 + 192 kbit/s would require about 2.33 Gb + a few 100 Mb for any menus.
Sorry for the long answer, but I feel it is justified for others, as well as yourself.
That having been said, digital degradation is different from poor analogue quality, so you still have to be reasonable.
As a guideline, let's take an average quality digital signal, such as from a mini-DV camcorder and the individual can extrapolate what I say here, up or down according to his own requirements.
OK, let's start with the audio. We have a choice of 3 systems: LPCM, MPEG and AC-3 (Dolby Digital). The first is uncompressed and gives the largest file. The other two are compressed. The AC-3 is not written into PAL standards but all modern PAL players (e.g., <6 years old) accept it without problems. I therefore strongly recommend using this, because it gives good quality for a modest bitrate. Mostly, ordinary 2.0 stereo will be used and 192 kbit/s is more than adequate for good quality music.
For video, it becomes less cut and dried. Theoretically, you can use bitrates up to about 9000 kbit/s: in practice, you can't. Why? Because many players simply will not accept such a high bitrate, especially from a DVD±R/RW disk, whose 1:0 contrast is much lower than a pressed (commercial) disk. Then comes the question of quality. For our DV average, I've done extensive tests and I've found that 6000 kbit/s is good enough to encode the signal with no significant deterioration of the signal quality under most conditions, even for moving subjects. This is because the signal is already significantly compressed in the camcorder. Very occasionally, 7000 kbit/s will give a slight improvement where you have specific problems. 6000 kbit/s will play in all players, that I know of. The deterioration at 5000 kbit/s is just noticeable to a trained eye but becomes noticeable at 4000 kbit/s to most people on fast motion scenes, although not on scenes with little motion.
A DVD±R with a reasonably simple menu, no sub-title tracks and no extra audio tracks, with sound at 192 kbit/s and video at 6000 kbit/s will fit in at about 2.8 Gb/h so, theoretically, will be good for about 90 minutes, but it is not good practice to fill a disk to more than about 4.0 - 4.1 Gb for a number of reasons, especially for mediocre quality blanks (single layer).
Then there is the thorny question of variable and constant bitrates. Commercial pressed disks almost invariably use variable bitrates, so why shouldn't DVD±R's? Well, in the first place, we don't have at our disposal the equipment that Hollywood uses, unless we spend in at least the six figures. We have, as a rule, encoders that allow only one or two passes. These attempt to up the bitrate where there are significant differences between one frame and the next and to lower it where two consecutive frames are very similar. None of them analyse the actual visual content, like is done in Hollywood, which may do 20 or more analytical passes before deciding on the optimum bitrates for a given scene. Coming back to our DV video, which parts of the video have the greatest difference between successive frames? Where there are transitions. So our low-cost encoders up the bitrate most at transitions. Even worse, let's imagine the source file is analogue. This almost always has a poor signal-to-noise ratio, usually of the order of 30 dB (except for direct connections to studio-quality cameras). When viewing the video, this is not usually very visible for a number of reasons, but it does mean that the encoder will interpret the noise as movement in the video and will up the bitrate, especially on scenes with a low key illumination. Your moonlit landscape will therefore whizz through at a high bitrate, while a clip of field events will encode at a lower bitrate for the 100 m sprint. For these reasons, VBR tends to be grossly over-hyped with our resources. This is why I recommend CBR for all DV videos of less than about 1¾ or 2 h duration and all videos from analogue sources, of any duration or content. There may be a small advantage in going to VBR for >2 h videos from DV sources, where there is much rapid movement, but even this remains to be proved.
Finally, what about shorter videos? I still apply the 6000 kbit/s CBR rule, even though the disk is not filled. The reliabilty of the results will be better assured. Higher bitrates may cause more glitches. Some DVDs I've produced have lasted only 10 minutes but have been encoded at 6000 kbit/s, even though they are only 1/8 full.
Some people here may dispute what I say, gained from experience, on the principle that they believe "bigger is better", but my views are supported by others. As an illustration, I recently looked at four DVDs lasting from 12 to 70 minutes each and distributed by the Cyprus government on DVD-R disks (quantities insufficient to justify pressing). These were produced by three different European professional houses of renown using obviously professional equipment and were of superb quality (at least for the most part). All four were encoded at 6000 or 6500 kbit/s CBR.
To answer your question, 1.64 Gb is light for 50 Min. You obviously used a lower bitrate than I'd consider ideal. 6000 + 192 kbit/s would require about 2.33 Gb + a few 100 Mb for any menus.
Sorry for the long answer, but I feel it is justified for others, as well as yourself.
[b][i][color=red]Devil[/color][/i][/b]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]