Re: Small Horizontal Slits/Lines on edges of diagonal images

HandBrake for Windows support
Forum rules
An Activity Log is required for support requests. Please read How-to get an activity log? for details on how and why this should be provided.
Post Reply
match
Enlightened
Posts: 111
Joined: Sun Mar 14, 2010 5:08 pm

Re: Small Horizontal Slits/Lines on edges of diagonal images

Post by match »

Hi. I'm having a problem with seeing jagged horizontal lines, pretty much exclusively on diagonal images. I've seen interlacing artifacts before in my encodes, however it has always been just on small images in the distant part of the picture. And they were lines that went completely through the object. These jagged lines are present even on very large, close up objects in many places of the video. I have never seen these types of lines just on the diagonals before and have previously done hundreds of encodes with Handbrake. I have already encoded two and 1/2 full seasons of this show without a problem, so I'm thinking it isn't the source. My only guess is that it has something to do with the decombing filter. I download nightlies on a fairly regular basis (maybe once a week). Has the decombing filter been changed at all in the last few weeks? I haven't seen anything on the trac-timeline. But if so, is there somewhere where I can download a previous nightly version (perhaps one about 2-4 weeks old)?

If the decombing filter hasn't been changed, or there isn't a way to download previous nightlies, is there something I can do with the decomb parameters? If so, could you please be kind of specific. I've read the documentation more than once, but I am really not sure which parameter(s) to tweak or by how much.
I believe these are the defaults:
mode :1
metric :2
motion-threshold :6
spatial-threshold :9
block-threshold :80
block-width :16
block-height :16

I'm also curious as to what range the numbers go up to? For example, Motion-threshold is at 6...how high can it go up to...100?

Also, If I wind up tweaking the decomb settings beyond their defaults, will that degrade the picture quality (like when interlacing something that doesn't need interlacing)...and/or does it just slow down the encode? I am not too worried about the encoding speed as long as it is not crazily slow. I'd rather it be more effective/stronger (as long as it doesn't mess with picture quality), than to have a speedier encode.

One more thing...I'm going to post this in the request thread, but I thought I'd mention it here too. I would love to see some preset decomb options in the future. Kind of like the interlacing options. Maybe one for default, one for powerful, etc. As I am guessing not too many regular users know when and how to tweak these settings. Just a thought.

Keep up the good work and thank you so much in advance!

Image
Notice the horizontal lines from her neck down to her left shoulder

Code: Select all

# Windows GUI svn3576 2010100701
# Running: Microsoft Windows NT 6.1.7600.0
# CPU: Intel(R) Core(TM) i7 CPU         860  @ 2.80GHz
# Ram: 4055 MB
# Screen: 1280x1024
# Temp Dir: C:\Users\Chornomaz\AppData\Local\Temp\
# Install Dir: C:\Program Files (x86)\Handbrake
# Data Dir: C:\Users\Chornomaz\AppData\Roaming\HandBrake\HandBrake\0.9.4.3606

# CLI Query:  -i "A:\Ripped DVDs\DVD IMAGES\Farscape - Season 4 Disc 5\FARSCAPE_S4_5\VIDEO_TS" -t 1 -c 1-8 -o "A:\Ripped DVDs\Farscape-s04e17.A.Constellation.of.Doubt.mkv"  -f mkv --detelecine --decomb -w 720 --loose-anamorphic  -e x264 -S 600 -2  -r 23.976 -a 1 -E copy:ac3 -6 auto -R Auto -B auto -D 0 --markers="C:\Users\Chornomaz\AppData\Local\Temp\Farscape-s04e17.A.Constellation.of.Doubt-1-chapters.csv" -x ref=9:bframes=9:subq=10:mixed-refs=0:8x8dct=1:trellis=2:b-adapt=2:direct=auto:me=umh:merange=64:no-fast-pskip=1:analyse=all:b-pyramid=1 -v 1
# User Query: False
-------------------------------------------

[18:40:12] hb_init: checking cpu count
[18:40:12] hb_init: starting libhb thread
HandBrake svn3607 (2010102001) - MinGW i386 - http://handbrake.fr
8 CPUs detected
Opening A:\Ripped DVDs\DVD IMAGES\Farscape - Season 4 Disc 5\FARSCAPE_S4_5\VIDEO_TS...
[18:40:12] hb_scan: path=A:\Ripped DVDs\DVD IMAGES\Farscape - Season 4 Disc 5\FARSCAPE_S4_5\VIDEO_TS, title_index=1
libbluray/bdnav/index_parse.c:157: indx_parse(): error opening A:\Ripped DVDs\DVD IMAGES\Farscape - Season 4 Disc 5\FARSCAPE_S4_5\VIDEO_TS/BDMV/index.bdmv
libbluray/bluray.c:960: nav_get_title_list(A:\Ripped DVDs\DVD IMAGES\Farscape - Season 4 Disc 5\FARSCAPE_S4_5\VIDEO_TS) failed (0x195dbb8)
[18:40:12] bd: not a bd - trying as a stream/file instead
libdvdnav: Using dvdnav version 4.1.3
libdvdread: Encrypted DVD support unavailable.
libdvdread: Device (null) inaccessible, CSS authentication not available.
NAME OPEN FAILED
libdvdnav: Unable to find home directorylibdvdnav: DVD disk reports itself with Region mask 0x00000000. Regions: 1 2 3 4 5 6 7 8
libdvdread: Encrypted DVD support unavailable.
libdvdread: Device (null) inaccessible, CSS authentication not available.
[18:40:12] scan: DVD has 14 title(s)
[18:40:12] scan: scanning title 1
[18:40:12] scan: opening IFO for VTS 1
[18:40:12] pgc_id: 1, pgn: 1: pgc: 0x1962690
[18:40:12] scan: vts=1, ttn=1, cells=0->7, blocks=0->832487, 832488 blocks
[18:40:12] scan: duration is 00:45:56 (2756020 ms)
[18:40:12] scan: checking audio 1
[18:40:12] scan: id=80bd, lang=English (AC3), 3cc=eng ext=0
[18:40:12] scan: checking audio 2
[18:40:12] scan: id=81bd, lang=English (AC3), 3cc=eng ext=0
[18:40:12] scan: title 1 has 8 chapters
[18:40:12] scan: chap 1 c=0->0, b=0->25157 (25158), 72175 ms
[18:40:12] scan: chap 2 c=1->1, b=25158->96909 (71752), 234390 ms
[18:40:12] scan: chap 3 c=2->2, b=96910->278962 (182053), 616472 ms
[18:40:12] scan: chap 4 c=3->3, b=278963->451014 (172052), 566375 ms
[18:40:12] scan: chap 5 c=4->4, b=451015->631394 (180380), 642677 ms
[18:40:12] scan: chap 6 c=5->5, b=631395->771990 (140596), 423534 ms
[18:40:12] scan: chap 7 c=6->6, b=771991->832483 (60493), 200217 ms
[18:40:12] scan: chap 8 c=7->7, b=832484->832487 (4), 176 ms
[18:40:12] scan: aspect = 0
[18:40:12] scan: decoding previews for title 1
libdvdnav: DVD disk reports itself with Region mask 0x00000000. Regions: 1 2 3 4 5 6 7 8
[18:40:12] scan: title angle(s) 1
[18:40:12] scan: audio 0x80bd: AC-3, rate=48000Hz, bitrate=448000 English (AC3) (5.1 ch)
[18:40:12] scan: audio 0x81bd: AC-3, rate=48000Hz, bitrate=192000 English (AC3) (2.0 ch)
[18:40:12] scan: 10 previews, 720x480, 29.970 fps, autocrop = 0/0/8/6, aspect 16:9, PAR 32:27
[18:40:12] scan: title (0) job->width:704, job->height:400
[18:40:13] libhb: scan thread found 1 valid title(s)
+ title 1:
  + vts 1, ttn 1, cells 0->7 (832488 blocks)
  + duration: 00:45:56
  + size: 720x480, pixel aspect: 32/27, display aspect: 1.78, 29.970 fps
  + autocrop: 0/0/8/6
  + chapters:
    + 1: cells 0->0, 25158 blocks, duration 00:01:12
    + 2: cells 1->1, 71752 blocks, duration 00:03:54
    + 3: cells 2->2, 182053 blocks, duration 00:10:16
    + 4: cells 3->3, 172052 blocks, duration 00:09:26
    + 5: cells 4->4, 180380 blocks, duration 00:10:43
    + 6: cells 5->5, 140596 blocks, duration 00:07:04
    + 7: cells 6->6, 60493 blocks, duration 00:03:20
    + 8: cells 7->7, 4 blocks, duration 00:00:00
  + audio tracks:
    + 1, English (AC3) (5.1 ch) (iso639-2: eng), 48000Hz, 448000bps
    + 2, English (AC3) (2.0 ch) (iso639-2: eng), 48000Hz, 192000bps
  + subtitle tracks:
    + 1, Closed Captions (iso639-2: eng) (Text)(CC)
Reading chapter markers from file C:\Users\Chornomaz\AppData\Local\Temp\Farscape-s04e17.A.Constellation.of.Doubt-1-chapters.csv
Calculated bitrate: 1374 kbps
[18:40:13] 2 job(s) to process
[18:40:13] starting job
[18:40:13] sync: expecting 82627 video frames
[ac3 @ 0x195f5f0] No channel layout specified. The encoder will guess the layout, but it might be incorrect.
[18:40:13] job configuration:
[18:40:13]  * source
[18:40:13]    + A:\Ripped DVDs\DVD IMAGES\Farscape - Season 4 Disc 5\FARSCAPE_S4_5\VIDEO_TS
[18:40:13]    + title 1, chapter(s) 1 to 8
[18:40:13]  * destination
[18:40:13]    + A:\Ripped DVDs\Farscape-s04e17.A.Constellation.of.Doubt.mkv
[18:40:13]    + container: Matroska (.mkv)
[18:40:13]      + chapter markers
[18:40:13]  * video track
[18:40:13]    + decoder: mpeg2
[18:40:13]      + bitrate 9800 kbps
[18:40:13]    + frame rate: 29.970 fps -> constant 23.976 fps
[18:40:13]    + loose anamorphic
[18:40:13]      + storage dimensions: 720 * 480 -> 720 * 496, crop 0/0/8/6, mod 0
[18:40:13]      + pixel aspect ratio: 863 / 720
[18:40:13]      + display dimensions: 863 * 496
[18:40:13]    + filters
[18:40:13]      + Detelecine (pullup) (default settings)
[18:40:13]      + Decomb (default settings)
[18:40:13]    + encoder: x264
[18:40:13]      + options: ref=9:bframes=9:subq=10:mixed-refs=0:8x8dct=1:trellis=2:b-adapt=2:direct=auto:me=umh:merange=64:no-fast-pskip=1:analyse=all:b-pyramid=1
[18:40:13]      + bitrate: 1374 kbps, pass: 1
[18:40:13]  * audio track 0
[18:40:13]    + decoder: English (AC3) (5.1 ch) (track 1, id 80bd)
[18:40:13]      + bitrate: 448 kbps, samplerate: 48000 Hz
[18:40:13]    + AC3 passthrough
libdvdnav: Using dvdnav version 4.1.3
libdvdread: Encrypted DVD support unavailable.
libdvdread: Device (null) inaccessible, CSS authentication not available.
NAME OPEN FAILED
libdvdnav: Unable to find home directorylibdvdnav: DVD disk reports itself with Region mask 0x00000000. Regions: 1 2 3 4 5 6 7 8
libdvdread: Encrypted DVD support unavailable.
libdvdread: Device (null) inaccessible, CSS authentication not available.
libdvdnav: DVD disk reports itself with Region mask 0x00000000. Regions: 1 2 3 4 5 6 7 8
[18:40:13] yadif thread started for segment 0
[18:40:13] yadif thread started for segment 1
[18:40:13] yadif thread started for segment 2
[18:40:13] yadif thread started for segment 3
[18:40:13] yadif thread started for segment 4
[18:40:13] yadif thread started for segment 5
[18:40:13] yadif thread started for segment 6
[18:40:13] yadif thread started for segment 7
[18:40:13] decomb thread started for segment 0
[18:40:13] decomb thread started for segment 1
[18:40:13] decomb thread started for segment 2
[18:40:13] decomb thread started for segment 4
[18:40:13] decomb thread started for segment 6
[18:40:13] encx264: min-keyint: 24, keyint: 240
[18:40:13] encx264: encoding with stored aspect 863/720
x264 [info]: [18:40:13] decomb thread started for segment 7
using SAR=863/720
[18:40:13] decomb thread started for segment 3
[18:40:13] decomb thread started for segment 5
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
[18:40:13] reader: first SCR 146 id 224 DTS 22254
[18:40:13] mpeg2: "Chapter 1" (1) at frame 0 time 3003
x264 [info]: profile High, level 3.1
No accelerated IMDCT transform found
[18:42:08] mpeg2: "Chapter 2" (2) at frame 2164 time 6501495
[18:49:32] mpeg2: "Chapter 3" (3) at frame 9211 time 27663636
[19:07:55] mpeg2: "Chapter 4" (4) at frame 27706 time 83204121
[19:24:02] mpeg2: "Chapter 5" (5) at frame 44695 time 134222088
[19:42:58] mpeg2: "Chapter 6" (6) at frame 63964 time 192086895
[19:55:50] mpeg2: "Chapter 7" (7) at frame 76682 time 230279049
[20:01:15] reader: done. 1 scr changes
[20:01:15] mpeg2: "Chapter 8" (8) at frame 82685 time 248306058
[20:01:15] 2758.956200s: Video -> Film
[20:01:19] work: average encoding speed for job is 17.001905 fps
[20:01:19] sync: got 82686 frames, 82627 expected
[20:01:19] mpeg2 done: 82686 frames
[20:01:19] render: 66148 frames output, 560 dropped and 8 duped for CFR/PFR
[20:01:19] render: lost time: 48002955 (15985 frames)
[20:01:19] render: gained time: 48002955 (63802 frames) (0 not accounted for)
[20:01:19] render: average dropped frame duration: 3003
x264 [info]: frame I:419   Avg QP:18.26  size: 42386  PSNR Mean Y:45.98 U:50.24 V:50.63 Avg:46.97 Global:46.53
x264 [info]: frame P:14227 Avg QP:21.16  size: 15004  PSNR Mean Y:43.72 U:48.31 V:48.73 Avg:44.78 Global:44.25
x264 [info]: frame B:51502 Avg QP:24.24  size:  4618  PSNR Mean Y:42.02 U:47.97 V:48.50 Avg:43.26 Global:42.78
x264 [info]: consecutive B-frames:  0.9%  1.8%  5.0% 30.9% 19.9% 28.3%  6.2%  5.8%  0.9%  0.3%
x264 [info]: mb I  I16..4:  7.4% 81.4% 11.2%
x264 [info]: mb P  I16..4:  2.7% 11.7%  1.5%  P16..4: 62.9% 12.2%  2.3%  0.3%  0.1%    skip: 6.2%
x264 [info]: mb B  I16..4:  0.2%  0.9%  0.1%  B16..8: 43.1%  5.5%  1.3%  direct: 6.6%  skip:42.2%  L0:43.0% L1:47.8% BI: 9.2%
x264 [info]: final ratefactor: 19.69
x264 [info]: 8x8 transform intra:75.7% inter:73.3%
x264 [info]: direct mvs  spatial:100.0% temporal:0.0%
x264 [info]: coded y,uvDC,uvAC intra: 78.4% 65.9% 29.0% inter: 24.1% 17.0% 0.9%
x264 [info]: i16 v,h,dc,p: 27% 35% 10% 28%
x264 [info]: i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 10%  6%  6% 11% 16% 16% 12% 13% 11%
x264 [info]: i4 v,h,dc,ddl,ddr,vr,hd,vl,hu:  9% 31%  3%  7% 11% 12%  9%  9%  9%
x264 [info]: i8c dc,h,v,p: 35% 26% 21% 17%
x264 [info]: Weighted P-Frames: Y:4.9%
x264 [info]: ref P L0: 52.2%  8.2% 15.4%  7.6%  5.5%  4.4%  3.8%  2.6%  0.2%  0.0%
x264 [info]: ref B L0: 86.0%  6.6%  3.2%  1.6%  1.1%  0.9%  0.6%
x264 [info]: ref B L1: 94.4%  5.6%
x264 [info]: SSIM Mean Y:0.9716266 (15.471db)
x264 [info]: PSNR Mean Y:42.413 U:48.053 V:48.561 Avg:43.609 Global:43.076 kb/s:1360.09
[20:01:19] decomb: deinterlaced 709 | blended 465 | unfiltered 65526 | total 66700
[20:01:19] starting job
[20:01:19] sync: expecting 82686 video frames
[ac3 @ 0x3c08a70] No channel layout specified. The encoder will guess the layout, but it might be incorrect.
[20:01:19] job configuration:
[20:01:19]  * source
[20:01:19]    + A:\Ripped DVDs\DVD IMAGES\Farscape - Season 4 Disc 5\FARSCAPE_S4_5\VIDEO_TS
[20:01:19]    + title 1, chapter(s) 1 to 8
[20:01:19]  * destination
[20:01:19]    + A:\Ripped DVDs\Farscape-s04e17.A.Constellation.of.Doubt.mkv
[20:01:19]    + container: Matroska (.mkv)
[20:01:19]      + chapter markers
[20:01:19]  * video track
[20:01:19]    + decoder: mpeg2
[20:01:19]      + bitrate 9800 kbps
[20:01:19]    + frame rate: 29.970 fps -> constant 23.976 fps
[20:01:19]    + loose anamorphic
[20:01:19]      + storage dimensions: 720 * 480 -> 720 * 496, crop 0/0/8/6, mod 0
[20:01:19]      + pixel aspect ratio: 863 / 720
[20:01:19]      + display dimensions: 863 * 496
[20:01:19]    + filters
[20:01:19]      + Detelecine (pullup) (default settings)
[20:01:19]      + Decomb (default settings)
[20:01:19]    + encoder: x264
[20:01:19]      + options: ref=9:bframes=9:subq=10:mixed-refs=0:8x8dct=1:trellis=2:b-adapt=2:direct=auto:me=umh:merange=64:no-fast-pskip=1:analyse=all:b-pyramid=1
[20:01:19]      + bitrate: 1374 kbps, pass: 2
[20:01:19]  * audio track 0
[20:01:19]    + decoder: English (AC3) (5.1 ch) (track 1, id 80bd)
[20:01:19]      + bitrate: 448 kbps, samplerate: 48000 Hz
[20:01:19]    + AC3 passthrough
libdvdnav: Using dvdnav version 4.1.3
libdvdread: Encrypted DVD support unavailable.
libdvdread: Device (null) inaccessible, CSS authentication not available.
NAME OPEN FAILED
libdvdnav: Unable to find home directorylibdvdnav: DVD disk reports itself with Region mask 0x00000000. Regions: 1 2 3 4 5 6 7 8
libdvdread: Encrypted DVD support unavailable.
libdvdread: Device (null) inaccessible, CSS authentication not available.
libdvdnav: DVD disk reports itself with Region mask 0x00000000. Regions: 1 2 3 4 5 6 7 8
[20:01:19] yadif thread started for segment 0
[20:01:19] yadif thread started for segment 1
[20:01:19] yadif thread started for segment 2
[20:01:19] yadif thread started for segment 3
[20:01:19] yadif thread started for segment 4
[20:01:19] yadif thread started for segment 5
[20:01:19] yadif thread started for segment 6
[20:01:19] yadif thread started for segment 7
[20:01:19] decomb thread started for segment 1
[20:01:19] decomb thread started for segment 2
[20:01:19] decomb thread started for segment 3
[20:01:19] decomb thread started for segment 4
[20:01:19] decomb thread started for segment 5
[20:01:19] decomb thread started for segment 6
[20:01:19] encx264: min-keyint: 24, keyint: 240
[20:01:19] encx264: encoding with stored aspect 863/720
x264 [info]: using SAR=863/720
[20:01:19] decomb thread started for segment 0
x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 FastShuffle SSE4.2
[20:01:19] decomb thread started for segment 7
[20:01:19] reader: first SCR 146 id 224 DTS 22254
[20:01:19] mpeg2: "Chapter 1" (1) at frame 0 time 3003
x264 [info]: profile High, level 3.1
No accelerated IMDCT transform found
[20:03:20] mpeg2: "Chapter 2" (2) at frame 2164 time 6501495
[20:10:06] mpeg2: "Chapter 3" (3) at frame 9211 time 27663636
[20:27:11] mpeg2: "Chapter 4" (4) at frame 27706 time 83204121
[20:42:39] mpeg2: "Chapter 5" (5) at frame 44695 time 134222088
[21:00:32] mpeg2: "Chapter 6" (6) at frame 63964 time 192086895
[21:12:51] mpeg2: "Chapter 7" (7) at frame 76682 time 230279049
[21:17:48] reader: done. 1 scr changes
[21:17:48] mpeg2: "Chapter 8" (8) at frame 82685 time 248306058
[21:17:48] 2758.956200s: Video -> Film
[21:17:51] work: average encoding speed for job is 18.024767 fps
[21:17:51] sync: got 82686 frames, 82686 expected
[21:17:51] mpeg2 done: 82686 frames
[21:17:51] render: 66148 frames output, 560 dropped and 8 duped for CFR/PFR
[21:17:51] render: lost time: 48002955 (15985 frames)
[21:17:51] render: gained time: 48002955 (63802 frames) (0 not accounted for)
[21:17:51] render: average dropped frame duration: 3003
x264 [info]: frame I:419   Avg QP:18.55  size: 40708  PSNR Mean Y:45.76 U:50.11 V:50.51 Avg:46.77 Global:46.39
x264 [info]: frame P:14227 Avg QP:21.14  size: 15087  PSNR Mean Y:43.73 U:48.32 V:48.74 Avg:44.79 Global:44.36
x264 [info]: frame B:51502 Avg QP:24.29  size:  4701  PSNR Mean Y:42.03 U:47.97 V:48.50 Avg:43.27 Global:42.88
x264 [info]: consecutive B-frames:  0.9%  1.8%  5.0% 30.9% 19.9% 28.3%  6.2%  5.8%  0.9%  0.3%
x264 [info]: mb I  I16..4:  7.9% 81.1% 11.0%
x264 [info]: mb P  I16..4:  2.7% 12.0%  1.5%  P16..4: 62.1% 12.2%  2.7%  0.3%  0.1%    skip: 6.2%
x264 [info]: mb B  I16..4:  0.2%  1.0%  0.1%  B16..8: 42.8%  5.6%  1.4%  direct: 6.4%  skip:42.6%  L0:43.0% L1:47.8% BI: 9.3%
x264 [info]: 8x8 transform intra:75.9% inter:73.8%
x264 [info]: direct mvs  spatial:99.0% temporal:1.0%
x264 [info]: coded y,uvDC,uvAC intra: 79.0% 67.0% 29.6% inter: 23.9% 17.1% 0.9%
x264 [info]: i16 v,h,dc,p: 28% 35% 10% 27%
x264 [info]: i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 10%  6%  6% 11% 16% 16% 12% 13% 11%
x264 [info]: i4 v,h,dc,ddl,ddr,vr,hd,vl,hu:  9% 30%  3%  7% 11% 12%  9%  9%  9%
x264 [info]: i8c dc,h,v,p: 35% 26% 22% 17%
x264 [info]: Weighted P-Frames: Y:4.9%
x264 [info]: ref P L0: 52.6%  5.2% 16.5%  8.7%  6.2%  4.7%  3.8%  2.2%  0.1%  0.0%
x264 [info]: ref B L0: 85.6%  7.1%  3.4%  1.5%  1.1%  0.8%  0.5%
x264 [info]: ref B L1: 94.3%  5.7%
x264 [info]: SSIM Mean Y:0.9719909 (15.527db)
x264 [info]: PSNR Mean Y:42.421 U:48.058 V:48.563 Avg:43.617 Global:43.171 kb/s:1373.87
[21:17:51] mux: track 0, 66148 frames, 473783220 bytes, 1373.82 kbps, fifo 16
[21:17:51] mux: track 1, 86216 frames, 154499072 bytes, 448.00 kbps, fifo 128
[21:17:51] decomb: deinterlaced 709 | blended 465 | unfiltered 65526 | total 66700
[21:17:51] libhb: work result = 0
Rip done!
HandBrake has exited.
Last edited by match on Mon Oct 25, 2010 11:19 pm, edited 1 time in total.
hunterk
Bright Spark User
Posts: 179
Joined: Tue Jun 03, 2008 2:27 pm

Re: Jagged Horizontal Lines on diagonal images

Post by hunterk »

I may be mistaken, but I think those jaggies are the result of a deinterlaced frame, similar to what's going on in this pic from the decomb documentation:
Image
You can see a number of the diagonal lines (e.g., the table edge, kid next to Bender's mouth, etc.) exhibit similar behavior.

EDIT: I believe this is the same problem, so their solutions are probably applicable here too:
http://forum.handbrake.fr/viewtopic.php?f=11&t=18090
match
Enlightened
Posts: 111
Joined: Sun Mar 14, 2010 5:08 pm

Re: Small Horizontal Slits/Lines on edges of diagonal images

Post by match »

Yes, I am seeing similar types of lines as in the picture you posted. However, I read the decombing guide and it seems that this picture example was showing what happens when the deinterlacing (not decomb) filter is left on all of the time and tries to deinterlace a progressive frame...as quoted here: "The downside of deinterlacing is that it reduces picture quality. The filtering process can lose up to half the frame's horizontal lines, throwing away detail. It's not necessary for progressive video, which isn't interlaced. "

Also quoted from the guide: "So it's not good to run a deinterlacer all the time. If you don't need to reduce picture quality, why do it? Not to mention, it can really slow down an encode."

Then it goes on to talk about the decombing filter: "You can leave this on all the time and never have to worry about interlacing. It will handle everything for you."

So if my thinking is right, the decombing filter should NOT produce these horizontal lines (on diagonal images) since it should not try to deinterlace a progressive picture as the example picture shows.

As for the other thread you pointed me to, I could be wrong, but it seems like it may be a different problem. It looks like the entirety of the strings on the guitar look interlaced (not just lines on the border of images like my problem). I've seen this happen when images are very small. Perhaps that is happening because the strings are so thin and it doesn't see it as something it needs to deinterlace. Of course I could be wrong.
In any event, somebody on that thread mentioned turning off detelecine since it is a pal version. However, mine is ntsc, so not sure if that same approach would help.

They also mentioned turning off decomb and using deinterlace instead. But wouldn't that mess up the video if parts of it were not interlaced? I don't know how to tell if the entire video is interlaced. Even if it is, I read that it's quicker to use the decomb since it won't always try to interlace still/low movement shots and usually provides better quality results. I'd much rather use the decomb filter which has worked brilliantly for me in the past, but seems to be behaving differently on this and some other encodes I have been recently doing. If anyone else has any other comments or tips on custom decomb options (or any other suggestions) I'd certainly welcome and appreciate it.

[EDIT:] Out of curiosity, I tried downloading the latest stable 0.94 and tried to encode this episode again to see if the decomb filter acted any differently. It came up with the same horizontal slits on the edges of the diagonal images. So I'm thinking maybe it is more this particular source than the decombing filter being modified. Perhaps the dvd producers changed something in the way they created the video (since it has been happening on the more recent encodes (latter episodes). So anyway, I'm still open to suggestions. Would trying to use the full deinterlacing filter help or hurt? Is there any way to tell if the entire video is interlaced? Or better...is there something I could change in the decombing filter custom field? - Thanx
hunterk
Bright Spark User
Posts: 179
Joined: Tue Jun 03, 2008 2:27 pm

Re: Re: Small Horizontal Slits/Lines on edges of diagonal im

Post by hunterk »

You're correct that decomb shouldn't be deinterlacing a frame that doesn't need it, but it could be a false positive in its detection algorithm. If you would like to mess with the parameters, you can find the documentation in the decomb.c source code here (the explanations are right there near the beginning):
http://trac.handbrake.fr/browser/trunk/ ... c?rev=2956

I personally would recommend trying it without detelecine enabled and see if that helps at all.

Also, have you looked at the same offending frames in the source video to see how they look, i.e. are they interlaced frames or what?
Post Reply