Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. |
26th February 2011, 04:32 | #1 | Link | |
Registered User
Join Date: Nov 2009
Posts: 2,352
|
FFV1: ffms [error] x264
Im having problems decoding a FFV1 source in x264 r.1867 through ffms:
Quote:
FFV1 15mb: http://www.sendspace.com/file/oaptgp
__________________
i7-4790K@Stock::GTX 1070] AviSynth+ filters and mods on GitHub + Discussion thread |
|
26th February 2011, 13:55 | #4 | Link | |
Registered User
Join Date: Nov 2009
Posts: 2,352
|
I made some tests with the default x264-r.1867 x32 build from x264.nl and the 1867 JEEB patch:
Source same as post#1 but a trim of 696 frames: lavf: 7:10seg ffmpg 7:15seg avs: 7:20seg avs: Quote:
PSNR of ffmpg and avs were 52.1831dB and lavf 52.1614 My eye told me me ffmpg was slightly the most accurate. Depending where I look at lavf also looks better, in any case there's more speed gain than quality loss, so I will stick to lavf for now.
__________________
i7-4790K@Stock::GTX 1070] AviSynth+ filters and mods on GitHub + Discussion thread Last edited by Dogway; 26th February 2011 at 14:15. |
|
26th February 2011, 16:11 | #5 | Link |
Mr. Sandman
Join Date: Sep 2003
Location: Haddonfield, IL
Posts: 11,768
|
why 1867? get a newer revision.
__________________
MPEG-4 ASP Custom Matrices: EQM V1(old), EQM AutoGK Sharpmatrix (aka EQM V2), EQM V3HR (updated 01/10/2004), EQM V3LR, EQM V3ULR (updated 04/02/2005), EQM V3UHR (updated 17/12/2004) and EQM V3EHR (updated 05/10/2004) Info about my ASP matrices. MPEG-4 AVC Custom Matrices: EQM AVC-HR Info about my AVC matrices My x264 builds. Mooo!!! |
26th February 2011, 16:15 | #6 | Link |
Registered User
Join Date: Nov 2009
Posts: 2,352
|
yes, well, that's what I had been using for testings lately, its only for consistency. For the last encoding I use to use the most recent build.
__________________
i7-4790K@Stock::GTX 1070] AviSynth+ filters and mods on GitHub + Discussion thread |
26th February 2011, 17:24 | #7 | Link | |
Registered User
Join Date: Apr 2002
Location: Germany
Posts: 5,389
|
PSNR of [what] against [what] ?
Quote:
__________________
- We´re at the beginning of the end of mankind´s childhood - My little flickr gallery. (Yes indeed, I do have hobbies other than digital video!) |
|
26th February 2011, 17:46 | #8 | Link |
Registered User
Join Date: Nov 2009
Posts: 2,352
|
lossless FFV1 against encoded file through lavf,ffmpg, and avs ffms2 decoders. Fidelity to the original lossless. Im assuming that encoding a file twice (with exact same settings) would give me a 1:1 similarity. Tell me if Im wrong.
For ssim and psnr I load them with ffms2 in avisynth.
__________________
i7-4790K@Stock::GTX 1070] AviSynth+ filters and mods on GitHub + Discussion thread |
26th February 2011, 19:02 | #9 | Link |
Derek Prestegard IRL
Join Date: Nov 2003
Location: Los Angeles
Posts: 5,988
|
Lossless is lossless. You can encode the file as many times as you want, and if decoded properly will be mathematically identical to the original, uncompressed source.
If this is not the case, something in your workflow is not lossless Derek
__________________
These are all my personal statements, not those of my employer :) |
26th February 2011, 19:45 | #10 | Link |
Registered User
Join Date: Nov 2009
Posts: 2,352
|
Maybe I lacked to say I lossy encoded to h.264 through lavf,ffmpg, and avs. But everything else is exactly the same.
Code:
FFV1--->lossy mp4 through --->lavf---> comparison: ffms2 decoded ffmpg avs
__________________
i7-4790K@Stock::GTX 1070] AviSynth+ filters and mods on GitHub + Discussion thread Last edited by Dogway; 26th February 2011 at 19:51. |
26th February 2011, 20:05 | #11 | Link |
Registered User
Join Date: May 2006
Posts: 957
|
So you want to test the same h264 decoder three times (assuming you use lavc through avisynth)?
__________________
x264 log explained || x264 deblocking how-to preset -> tune -> user set options -> fast first pass -> profile -> level Doom10 - Of course it's better, it's one more. |
26th February 2011, 20:12 | #12 | Link |
Derek Prestegard IRL
Join Date: Nov 2003
Location: Los Angeles
Posts: 5,988
|
Ahh, yes. This makes sense now
You should get identical (or VERY close to identical) results regardless of how you get the video into x264, provided all other settings are matched. Your metric scores could be hugely thrown off by an extra or missing frame. I'm guessing this is part of the reason you're doing this test? Derek
__________________
These are all my personal statements, not those of my employer :) |
26th February 2011, 21:07 | #13 | Link |
Registered User
Join Date: Nov 2009
Posts: 2,352
|
so x264 doesn't produce identical encodes runned 2 times?
My test was only for decoding accurateness, but as you say (and I recall reading), once decoded is accurate quality wise. So now I don't know... Maybe there is a mismatch between r.1867 and r.1867 JEEB builds. But avs, and lavfc decoded encodes are also different, and I used the exact same build. Must precise they are identical in the upper half of the image only.
__________________
i7-4790K@Stock::GTX 1070] AviSynth+ filters and mods on GitHub + Discussion thread Last edited by Dogway; 27th February 2011 at 18:04. |
28th February 2011, 16:39 | #14 | Link |
Registered User
Join Date: Nov 2009
Posts: 2,352
|
just wanted to say that the guilty for not producing 2 identical encodes was gradfun2db. probably addgrainc.
__________________
i7-4790K@Stock::GTX 1070] AviSynth+ filters and mods on GitHub + Discussion thread Last edited by Dogway; 28th February 2011 at 16:55. |
Tags |
error, ffms, ffv1, x264 |
Thread Tools | Search this Thread |
Display Modes | |
|
|