ChibiBoi
30th April 2013, 03:10
Hi everyone,
I'm having a few problems with my VHS rips, because you see the video footage was hard telecined from 23.976 to 29.970 with blended fields, and AnimeIVTC at mode=2 works great at getting rid of them, but it seems that it makes the hard subtitles really wiggly. The wiggles don't seem to be apparently when I use QTGMC at default plus srestore to 23.976.
However, the reason I don't use QTGMC with srestore is because it seems less effective at removing the blending and also causes some jerky footage. QTGMC takes a lot more processing power than AnimeIVTC and also I've read QTGMC shouldn't be used on anime.
Does anyone have any ideas on how to fix the blended fields as well as keep the subtitles looking good?
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
animeivtc(mode=2)
fastdegrain()
lanczos4resize(640,480)
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test clip.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
animeivtc(mode=2,bbob=4)
lanczos4resize(640,480)
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
qtgmc(fpsdivisor=2)
srestore(frate=23.976)
lanczos4resize(640,480)
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
animeivtc(mode=2)
qtgmc(inputtype=1)
lanczos4resize(640,480)
Here are two encoded samples and the raw .avi captured footage:
capture http://www.mediafire.com/?dvnjozw7npck27n
animeivtc http://www.mediafire.com/?12b9kmdyh83bm9k
animeivtc bbob=4 http://www.mediafire.com/?h577ow7zhat2uq5
qtgmc+srestore http://www.mediafire.com/?78moq7cdbblc9zz
animeivtc + qtgmc http://www.mediafire.com/?smo5csw4dusup2z
EDIT:
I tested out bbob=4 setting in AnimeIVTC which utilizes TGMC and this seems to make the subtitles really stable, like QTGMC.srestore, but since it's built into AnimeIVTC's functions, it seems to prevent jerkiness in the footage. However, one downside to both of these is that more blended frames are left in the video, and also the bob makes the subtitles look funky at times during high motion.
I guess now it comes down to choosing the lesser of the two evils: less stable video and subtitles vs more blended frames and slower encoding time. Does anyone have any opinions on this?
EDIT:
I think I've finally got the best of both worlds! Since AnimeIVTC(mode=2) at its default settings got rid of blended frames much better than QTGMC, but QTGMC stabilizes the video better, I decided to use both filters since QTGMC offers a progressive input parameter which gets rid of deinterlace shimmering.
I'm having a few problems with my VHS rips, because you see the video footage was hard telecined from 23.976 to 29.970 with blended fields, and AnimeIVTC at mode=2 works great at getting rid of them, but it seems that it makes the hard subtitles really wiggly. The wiggles don't seem to be apparently when I use QTGMC at default plus srestore to 23.976.
However, the reason I don't use QTGMC with srestore is because it seems less effective at removing the blending and also causes some jerky footage. QTGMC takes a lot more processing power than AnimeIVTC and also I've read QTGMC shouldn't be used on anime.
Does anyone have any ideas on how to fix the blended fields as well as keep the subtitles looking good?
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
animeivtc(mode=2)
fastdegrain()
lanczos4resize(640,480)
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test clip.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
animeivtc(mode=2,bbob=4)
lanczos4resize(640,480)
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
qtgmc(fpsdivisor=2)
srestore(frate=23.976)
lanczos4resize(640,480)
setmtmode(2)
AVISource("C:\Users\Public\Recorded TV\test.avi", audio=false).AssumeFPS(30000,1001)
assumetff()
converttoyv12(interlaced=true)
animeivtc(mode=2)
qtgmc(inputtype=1)
lanczos4resize(640,480)
Here are two encoded samples and the raw .avi captured footage:
capture http://www.mediafire.com/?dvnjozw7npck27n
animeivtc http://www.mediafire.com/?12b9kmdyh83bm9k
animeivtc bbob=4 http://www.mediafire.com/?h577ow7zhat2uq5
qtgmc+srestore http://www.mediafire.com/?78moq7cdbblc9zz
animeivtc + qtgmc http://www.mediafire.com/?smo5csw4dusup2z
EDIT:
I tested out bbob=4 setting in AnimeIVTC which utilizes TGMC and this seems to make the subtitles really stable, like QTGMC.srestore, but since it's built into AnimeIVTC's functions, it seems to prevent jerkiness in the footage. However, one downside to both of these is that more blended frames are left in the video, and also the bob makes the subtitles look funky at times during high motion.
I guess now it comes down to choosing the lesser of the two evils: less stable video and subtitles vs more blended frames and slower encoding time. Does anyone have any opinions on this?
EDIT:
I think I've finally got the best of both worlds! Since AnimeIVTC(mode=2) at its default settings got rid of blended frames much better than QTGMC, but QTGMC stabilizes the video better, I decided to use both filters since QTGMC offers a progressive input parameter which gets rid of deinterlace shimmering.