Normal deblocking filters will only help with the first iteration of encoding, if you reencode blocking content again the will not really help unless you crank the strength up. (machine learning based deblocking like might help better)
Have you tried something like MCTemporalDenoise with a high preset, SpotLess iwth tr=3 and or SMDeGrain?
i.e.:
![[Image: grafik.png]](https://i.ibb.co/Cp85qtJ6/grafik.png)
to get things starting?
The problem with such content is to rescue some sharpness.
Cu Selur
Ps.: there seems to have been a mix up with pc vs tv luma scaling in the past of this clip, so using something like
might also be a good idea.
Ps.: conventional filtered: script, clip stabilizing is a pain with that intro,...
Have you tried something like MCTemporalDenoise with a high preset, SpotLess iwth tr=3 and or SMDeGrain?
i.e.:
# Spot removal using SpotLess
clip = SpotLess.SpotLess(clip=clip, radT=3, rec=True, pel=1, smoother="zsmooth")
# applying deblocking using DeBlock QED
clip = deblock.Deblock_QED(clip, quant1=60, quant2=60, aOff2=4, bOff2=4)
# removing grain using SMDegrain
clip = smdegrain.SMDegrain(input=clip, tr=3, thSAD=600, thSADC=256, interlaced=False, opencl=True, device=-1)
![[Image: grafik.png]](https://i.ibb.co/Cp85qtJ6/grafik.png)
to get things starting?
The problem with such content is to rescue some sharpness.

Cu Selur
Ps.: there seems to have been a mix up with pc vs tv luma scaling in the past of this clip, so using something like
clip = core.std.Levels(clip=clip, min_in=36, max_in=212, min_out=16, max_out=235)
Ps.: conventional filtered: script, clip stabilizing is a pain with that intro,...
----
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.
Dev versions are in the 'experimental'-folder of my GoogleDrive, which is linked on the download page.