Hello, I'm fairly new to video encoding, I have a few questions if anyone could help me please:
1) Does re-encoding an encoded mkv file lead to further quality loss? "Copy of a copy" kind of thing? If for example I have a 720p file that takes 4 GB, and I re-encode it into about 1.5GB to save space. Would the new file be of less quality than encoding the "original" file directly into 1.5GB?
2) Same question about Audio. In case I re-encode the audio into 320kbps AAC format, from a 640 / 1.5M AC3 / DTS format?
3) Downscaling - Whats better? If for example I have a 1080p file that takes 8+ GB and I want it to be 4GB. Would it be preferred to downscale it first into 720p and re-encode, or leave it at 1080p and re-encode?
4) Downscaling - Is it preferred to downscale a video into "half" of its dimensions for more smooth looking / easier scaling? 1080p into 540p for example? Rather than 720p?
5) Bitrates - Whats the preferred bitrate for various mkv file resolutions? Whats considered "good" quality vs overkill? I've seen 720p files that take 6+ GB and others that take 4GB (considering theyre about the same length ofcourse). Same goes for 1080p files. Some can take 8GB, I've seen others going up to 15GB and more (again, considering same length more or less).
For now, my general 'rule of thumb' I use is - 720p at about 3-5 mbps, 1080p at about twice of that 8-12mbps, and 480p/360p at about 2-3mbps?
6) AAC format - I've read at some forums that this format can cause problems / artifacts. Others claim it can turn a 1.5mbps audio track into 320kbps with barely any quality loss, is that true?
For general information - The program I currently use for conversions is RipBot.
Thank you for your help
+ Reply to Thread
Results 1 to 2 of 2
-
-
Yes.
Encoding "original" to 1.5 GB would be higher quality than twice encoding to 4 GB, then 1.5 GB.
Yes, but double encoding audio is less obvious to most.
Depends on method.
1080p60 -> 1080p30
1080p60 -> 720p60
1080p60 -> 720p30
or are you talking about film rates?
1080p24 -> 720p24
Depends on software used and method (algorithm). But in general halving pixels h or v is less compute intensive than interpolation for progressive video but motion "smoothness" is a different issue.
Question is relative to what? Correct method of measurement is Mbps. GB depends on clip length.
Lets compare to ATSC TV broadcast (forget Blu-Ray)
Original distribution file ~ 35-144 Mbps
Typical ATSC 1280x720p60 MPeg2 compression is 10-14 Mbps
Typical ATSC2 (near future) 1280x720p60 h.264 is 6-8 Mbps.
Youtube 1280x720p30 crushes this down to ~1-2 Mbps (for their "high" quality stuff)
Youtube re-encodes what you send them so your upload file needs to be as high as possible (withing max file size limits).
1080p has ~2.2x pixels vs 720p so should use ~ double bit rate.
320kbps is high quality for AAC or AC3. 224 Kbps is usually enough for 2.0 (320 for 5.1 channels) but many go for less. /QUOTE]Last edited by edDV; 25th Dec 2011 at 00:11.
Recommends: Kiva.org - Loans that change lives.
http://www.kiva.org/about
Similar Threads
-
Combine Mpeg-2 Files (without re-encoding or loss of quality)
By Anonymous642 in forum Newbie / General discussionsReplies: 2Last Post: 26th Oct 2011, 06:58 -
--
By MGadAllah in forum EditingReplies: 33Last Post: 25th Sep 2011, 19:53 -
Downscaling While Capturing Video?
By fmagreed2 in forum Capturing and VCRReplies: 7Last Post: 6th Dec 2010, 17:20 -
encoding .VOB to .FLV with least loss in quality
By altavistasf in forum Video ConversionReplies: 9Last Post: 15th Oct 2010, 17:33 -
Bad quality downscaling...
By Emillister in forum Video ConversionReplies: 33Last Post: 29th Aug 2009, 12:48