Working with dephalo_alpha or filter to de-halo

...edited out...since script has been shown already...

yes, i am trying to juice out the details. if you look at the samples i have provided...the close-up shots of the people do not really have halos around their edges...but the distant shots of the doctor bent over a patient...you can see his whole character is aglow with the halo. :( if that's what you call mosquito noise...then i'm sorry...but i'm trying to figure out a good solution for my problem what is your suggestion? would love any input...if you feel that there are other filters i can look at...what might those be? thanks for your help in advance.

Could you post some unfiltered, unaltered, lossless png's of the same shots as in your fist post, along with detailing your encoding settings. [edit]after looking at the clip, I beleive what you are talking about is reffered to as "banding", which can be caused by compression, powerful denoising or powerfuld sharpening. In this case it looks like it was caused by compression

Didee was right...it's mosquito noise that I thought was haloing. that is a SS of a guy bending down over a patient...see his back...and all that...mosquito noise. i had thought that was haloing from the getgo. that link above is for original dvd9 material. encoding settings... megui 2-pass automatic quant i/p/b min 2 max 15 quant type is mpeg 4cd rip 2800MB with DTS audio 1.11GB set bitrate with those settings ~1175 if there is more you need...let me know. yes, encoding via Xvid 1.1.2...with of course DTS audio.

@Eu Man: in addition to wht mp4 mentioned sample from the original source from you are making will also be good. @mp4: I was thinking EuMan is using pretty low sigma values for fft3d, so not sure if compression would play a role in this. more of I think the chain of sharperners with high strength on already a sharp can be the cause.

also the script above in post #3 is what i am using...but NOW, i will NOT be needing the dehalo line which made the mosquito noise worse.