I have edited variable framerate videos in Sony Vegas. Now I want to export these to a lossless format.
Considering I can only export to 25fps or 30 fps, would it even make sense to render to "lossless" YUV in Sony Vegas? Isnt there by default a quality loss because the frames are converted from variable to 25?
Try StreamFab Downloader and download from Netflix, Amazon, Youtube! Or Try DVDFab and copy Blu-rays! or rip iTunes movies!
+ Reply to Thread
Results 1 to 24 of 24
Thread
-
Last edited by Abas-Avara; 19th Jan 2020 at 10:46.
The flag once raised will never fall! -
Are you sure that the source videos are have a variable framerate ? What do you intend to do with the edited / rendered footage ? If you want to further process it with another utility, or compress it with an external encoder, then it certainly makes sense as it removes one generation loss to the whole workflow. Anyway, from what I could understand, most editors automatically convert the imported footage to RGB straight away, and unless the editor has a “smart render” feature and the editing work only involved straight cuts, the rendering can never be lossless relative to the source videos.
-
vfr was and is a dopey way to save a minuscule amount of file size. vfr would never be used for any serious video work. that said "lossless" encoding is only visually lossless. you can't go back the other way and get back the original file. but anything is better than vfr. if you are going to edit it more than once, sony yuv is a good option. if you are only going to edit the vfr once and export the final video format than it's an unneeded step.
--
"a lot of people are better dead" - prisoner KSC2-303 -
If you are already "editing" your footage or have edited, in anything other than a simple, on-phone app, you will have ALREADY converted your vfr footage to cfr, whether you realize it or not. And it is NOT a lossless conversion process, though you could still store the result as best as possible by using a lossless storage format.
ScottLast edited by Cornucopia; 20th Jan 2020 at 21:18.
-
Not sure what happens in Sony Vegas but... for general reference:
Could you please name a few? So I can avoid them when I someday somehow had the interest to try some... (never tried any serious one before, no joking)
While reducing redundancy does help efficiency... (not only in the file size)
The reason for VFR is to optimize the handling of the "consecutive duplicates", which is actually a smart move.
The real problem is... the world seemed to have got accustomed to CFR.
Who knows?..
So... you mean lossless encoding in Sony Vegas is a completely joke?
* The term is very confusing... Follow-Up
Anyway, avoiding things you are unfamiliar with is a wise move.
Though sometimes you'll have no choice... but to face them.
Related: http://trac.ffmpeg.org/ticket/8482
Not strictly true...Last edited by gdgsdg123; 22nd Jan 2020 at 01:06.
-
lossless avi of any type - huffyuv, utcodec, lagarith, etc., are all a one way street. once you encode to them you can't re-encode back to what any file format was and get the exact same file back.
that said if you are editing in vegas there is no better option than converting all sources file to sony yuv and using it on the timeline.--
"a lot of people are better dead" - prisoner KSC2-303 -
What lossless format are you using?
An image sequence for the render FORMAT?
Anyways, in Vegas 17, I'm only seeing avc, hvec, magix intermediate, mpeg-1/2, p2, avc/mvc, mxf, etc.
All of these appear to me to be compressed formats.
...
If you pick one, like Magix Intermediate, click CUSTOMIZE TEMPLATE, you can pick a few more standard frame rates, but nothing variable or custom.
...
Anyways, there's always a loss when converting from/to lossy formats, eg mpeg 2 to mpeg 4. But if you're going from lossy to lossless, no loss in information.
If you're going from one frame rate to another, as long as the maximum frame rate of the original is the same as the constant frame rate of the output, you shouldn't be losing any frames either.
Eg. If the vfr for one section is a max of 25fps and you output 25fps cfr, no frames lost. 1 fps in vfr simply gets duplicated to 25 frames for that second.
As for color depth and bits, if your source is 4:2:2 10-bit, you'd obviously have to output to at least 4:2:2 10-bit lossless or better. If you encode to lossless 4:2:0 8-bit, you'll lose quality. -
so you're a newb vegas user. render as/video for windows. choose an output then configure to what you want.
[Attachment 51614 - Click to enlarge]--
"a lot of people are better dead" - prisoner KSC2-303 -
Thanks!
Completely forgot about AVI and MOV after Quicktime for Windows was terminated years ago and PCs were fast enough to edit MP4 etc natively without proxies.
Guess yuv and uncompressed avi are among the few accessible lossless codecs now that quicktime is gone (thus no support for other codecs it used to support). -
--
"a lot of people are better dead" - prisoner KSC2-303 -
Guess yuv and uncompressed avi are among the few accessible lossless codecs now that quicktime is gone (thus no support for other codecs it used to support).
Also for op, YUV Sony that was uncompressed I think, not lossless as the term "lossless" might be understood like "compressed without loss".
Also that Prorez export 422 (under Magix intermediate) , so called almost lossless is not bad also , to use to keep size of video in sane numbers. -
all the "lossless" codecs are only visually lossless. there's no way to re-encode a lossless avi back to exactly the same file it started as.
--
"a lot of people are better dead" - prisoner KSC2-303 -
bit for bit wise. quality wise they are fine.
--
"a lot of people are better dead" - prisoner KSC2-303 -
@aedipuss
It would certainly be a welcome feature for lossless video codecs, if only for testing purposes, as it's much easier to verify that two files are exactly identical (all checksums will match) than to test if they are “visually” identical.
@gdgsdg123
* Quality wise only or bit wise? -
audio wav files are easy. they are like all i-frame video.
unfortunately video uses i, p, b, and now slices(parts) of i, p and b frames. which means a video frame can reference a part of a past frame or future frame to create the new frame. once encoded to lossless avi there's no way to recreate bit for bit any source as it's now a changed source.--
"a lot of people are better dead" - prisoner KSC2-303 -
The lossless codecs are literally lossless if you use them correctly. What comes out of the decoder is exactly the same uncompressed video (bit for bit) as the uncompressed video that went into the encoder.
Of course, if you transcode a YV12 AVC video to, say Lagarith YV12, you will first be decompressing the AVC frames (resulting in maybe a 20 fold increase in file size), then compressing those uncompressed frames with Lagarith. Lagarith may get a 2:1 compression ratio but that still leaves you with a file that's 10x larger than the original AVC video. Later, when those Lagarith frames are decompressed, the output will be exactly the same uncompressed YV12 frames that were given to Lagarith. You'll be back to a file that's 20x larger than the original AVC video. If you recompress those frames with an AVC (or HEVC, or VC1, or whatever lossy) encoder you will not be able to restore the original AVC video's quality and size. Any time you compress video with a lossy codec you will get a loss of quality. The best you can do is use high enough a bitrate to keep the losses minimal.
Typically you want to use lossless intermediates when your editor doesn't work with a particular source format. Or when you are going to use multiple programs sequentially -- you don't want to degrade your video with a lossy encoding by each of the intermediate programs. And of course, you want to capture with a lossless codec if you want to avoid creating compression artifacts before you start filtering or editing. -
...I wonder why would you even care about this, as long as the input and output remain quality wise identical. (unless for testing purposes)
More likely something to do with the muxers * rather than the codecs...
Container information (like metadata, which you can even write an article in...) presents, that does influence the file's content.
* What responsible for fitting the encoded data into the container.
Also FYI, SSIM usage reference: http://trac.ffmpeg.org/ticket/7825
No. It's possible to make the output bit exact by using a deterministic encoder and instructing the muxer not to write random container information.
Jargon Explanation
Before encoding, the input has to be decoded first... Which means: if the inputs are quality wise identical (even bit wise different), they are effectively identical to the encoder.
Which usually ends in identical output (from the encoder) unless the encoder is non-deterministic (which is not so common).Last edited by gdgsdg123; 22nd Jan 2020 at 00:22.
-
Thanks for the replies. The VFR is on average between 25 and 30 fps (concerns old videos recorded with Nokia N95). Whats the best fps to convert to? 25 fps (PAL), 29,97 fps or 30 fps (NTSC)? I will use the Sony YUV codec
The flag once raised will never fall! -
what's your intended final video going to be? pal or ntsc? pal is normally 25fps and ntsc 30fps. i'd say go with what you want to end up with.
--
"a lot of people are better dead" - prisoner KSC2-303 -
Its PAL since I live in Europe. However I wanted to choose the fps that does the best job / preserves as much of the original as possible. I guess I will do PAL if there is no difference
The flag once raised will never fall! -
30fps. Standard cell phone master frame rate.
Maintain the original frame rate of the original all the way through editing in whatever video program you use, then output to whatever final frame rate you decide upon (most match the original).
For tv playback, the selection of 25fps pal might be needed if your playback equipment can't handle 30fps pr 29.97ntsc. You should test first. Keeping the output of the final video at the same original frame rate maintains the best picture quality and details.
Similar Threads
-
Which setting is Variable Bitrate in Handbrake?
By safarian24 in forum Newbie / General discussionsReplies: 2Last Post: 7th Apr 2019, 07:34 -
mediaInfo says using AAC [instead of AC3] bitrate becomes alwais "variable"
By marcorocchini in forum Newbie / General discussionsReplies: 13Last Post: 18th Feb 2019, 02:58 -
"Lossless" NVENC bitrate vs x265
By savvyguy in forum Newbie / General discussionsReplies: 0Last Post: 18th Aug 2018, 23:28 -
Is there an Avisynth Lossless Video Codec that uses lower bitrate for 720p?
By VideoFanatic in forum Video ConversionReplies: 7Last Post: 17th Apr 2018, 20:44 -
Should I start over again? (Video=Variable) (Audio=Variable Bitrate)
By strawberryshortcake in forum Newbie / General discussionsReplies: 5Last Post: 13th Feb 2018, 08:16