@enim: don't hijack other peoples posts, that's just rude; aside from the fact that you got an question, your question has nothing to do with the question of the topic starter. -> open up your own thread; about your question: there's a filter plugin for avisynth with might help -------------------------- @: those blocking artifacts are there the whole time (clearly visible if you boos the brightness a bit). First I would recommend to check out if there wasn't a better version available (the whole thing looks like some broken NTSC->PAL conversion from a VHS source). @: those blocking artifacts are there the whole time (clearly visible if you boos the brightness a bit).

Apr 03, 2014  On an encoder, a deblocking filter alters some of the pixels on the sharp edges that can form between macroblocks on predictive frames, esp. In areas of low contrast and at low bitrates. Think of it as similar to. Increasing the number of reference frames also dramatically increases the encoding time, since motion-estimation search, which is the slowest part of video encoding, has to occur on all possible reference frames in order to find the best match.

First I would recommend to check out if there wasn't a better version available (the whole thing looks like some broken NTSC->PAL conversion from a VHS source). I know, it would have probably been more precise if I addressed those horizontal lines that it creates during flashes in first place.

Those bother me more than the actual blocks. I want to fix those as well if that's possible. There isn't, I wish there was. Build and index file then enable the deblocking feature when opening the source an AviSynth: Mpeg2Source('filename.d2v', CPU=6) That will deblock and remove DCT ringing artifacts from Y,U, and V. The horizontal lines can be reduced with Vinverse(). I did that, it deblocked well.

But Vinverse() didn't help. I don't think this issue is actually to do with combing, is it? Anyway, so far the best result I achieved with this script: quite frankly I'm not sure I can do much better with this source. Code: DirectShowSource('DWT Buenos Aires Human Nature.VOB') ConvertToYV12() Deblock_QED(quant1=60,quant2=60,uv=3) daa() daa() daa() daa() ChromaShift(C=-4) MergeChroma(awarpsharp2(depth=25)) super = MSuper(pel=2, sharp=1) backward_vec2 = MAnalyse(super, isb = true, delta = 2, overlap=4) backward_vec1 = MAnalyse(super, isb = true, delta = 1, overlap=4) forward_vec1 = MAnalyse(super, isb = false, delta = 1, overlap=4) forward_vec2 = MAnalyse(super, isb = false, delta = 2, overlap=4) MDegrain2(super, backward_vec1,forward_vec1,backward_vec2,forward_vec2,thSAD=1200). Deblockers and debanders both attempt to mask damage than to reconstruct it, You can try any number of plugins and plugin combos.

It's unlikely that you'll get rid of 100% of the stripes in the worst of the striped frames, which occur just after a bright flash and before returning to darker backgrounds. Those stripes (and the macroblocks) are the result of over-filtering the original tape, removing fine detail and fine gradation data by removing all the grain.

Nonton film ultraman tiga final odyssey sub indo star. Grain isn't just 'noise', it's also part of perceived detail -- especially in film and analog sources. Removing more grain removes more data. To the previous over filtering, add a bitrate too low for the video's nature.

A target bitrate of 5000 is just too low to maintain clean flat areas and large expanses of dark, bright and shadow, and too low for an action + effect lighting as in a Jackson concert. Then add inexcusably bad deinterlace (half of the original fields were apparently discarded) and you get sawtooth edges, requiring more filtering, more smoothing, and entailing more loss. Look at frame 65 in your vob sample.

Over filtering, bad deinterlace, and low bitrates (and I'm guessing it's been encoded more than once) has literally left 'holes' in the response pattern of clouds and rays of light, rather than gradients and smooth transitions. Field decimation didn't help. Michael's face looks as if he's wearing a silk mask; his costume has lost all detail and texture.

You can smooth all that, but objects and gradients will lose more of their shape. Posterization effects are in the source. After filtering yet again and re-encoding, blocky grunge and more posterization and banding will appear, looking as bad or worse.

I don't think that even plugins as esoteric as Dither() can help very much without destroying lots of what's left. But you might want to spend some time in the dither and GradFuncMod thread. It won't solve the problem entirely, but it will explain much of what happening:. It could lead to more suggestions.