Imagine I want to encode a file to mpeg that is (approx) 480 x 360 (width x height) for use on a dvd.
I can scale it DOWN to 320 x 240 (half D1) and a bitrate of 2300 and get an okay quality picture on my set-top dvd player.
or
I can scale it UP to 640 x 480 and use bitrate of 2300
which would have a better quality?
I know that is quite low bitrate for full size mpeg 2 but I want to keep same file size.
BTW these numbers are approximate-and I haven't accounted for interlacing or rectangular pixels.
+ Reply to Thread
Results 1 to 6 of 6
-
-
If you want to make a DVD, you have to have it at 720*480 resolution (does 640*480 work, anyone?). 320x240 will definitely not work. Up to you to determine the best bitrate for your content and needs.
I like systems, their application excepted. (George Sand, translated from French), "J'aime beaucoup les systèmes, le cas d'application excepté." -
Here is what's legal
NTSC (NTSC Film)
Video:
Up to 9.8 Mbit/sec MPEG2 or up to 1.856 MBit/sec MPEG1 video
720 x 480 pixels MPEG2
704 x 480 pixels MPEG2
352 x 480 pixels MPEG2 (Same as the CVD Standard)
352 x 240 pixels MPEG2
352 x 240 pixels MPEG1 (Same as the VCD Standard)
29,97 fps (frames/second)
23,976 fps with 2:3 pulldown (NTSC Film, this is only supported by MPEG2 video)
Something will scale your video. What will it be ?
The settop player or your processing ? -
interesting question.
I personally would go with 720x480, just so it is 100% within DVD spec. Although 352x240 would work and work well...
I have experimented with this a few times in the past few years, and always end up upscaling instead of downscaling... Try it out with a 10 second test clip and see what ya get.
Let us know... i would be interested in seeing how it works out. -
Thanks you guys, as I said the numbers weren't corrrect.
This is a question about theory, not symantics.
the question boils down to this...
Using the SAME BITRATE.. and intermediate size...
scale down during encoding, and let the set top player double it..
or
scale up during encoding, (creating new pixels by interpolation of some description) and have the set top player play it at 100%
Zerosix, yes, I'll do some tests and report. -
So I did the test.
the original avi resolution was something like 480x360.
I used ffmpegX to create:
1. m2v (720x480) @ 2300 kbps average and mp2
and
2. m2v (352x240) @ 2300 kbps average and mp2.
#1 was about 10% larger file size.
I built the dvd with 2 tracks in dvdsp 1.5.
Track 1 image quality was slightly better to my eye - not much - more noticable in text.
Zerosix, I agree that the 'in-spec' solution is better. Especially since dvdsp 2 won't accept the smaller ones!
I would still like to know the theory behind this experiment.
Similar Threads
-
Books on subtitling - Practice and Theory
By devilcoelhodog in forum SubtitleReplies: 0Last Post: 11th Mar 2012, 12:33 -
Deinterlacing / Interlacing Theory and Application
By joejoe04 in forum Newbie / General discussionsReplies: 10Last Post: 1st Jan 2012, 23:01 -
UTorrent Theory
By Marco33 in forum ComputerReplies: 24Last Post: 5th Jan 2011, 19:12 -
Virus theory?
By ahhaa in forum ComputerReplies: 9Last Post: 3rd Jan 2008, 15:09 -
Simple question: how to compress 4GB mpeg2 to 1GB mpeg2
By jimdagys in forum Newbie / General discussionsReplies: 29Last Post: 3rd Jan 2008, 08:45