avisynth: Temporal Seemed to work better than Convolution?
 

I have just started John Q and the test file looked really great. The first few scenes had a bit of noise, but it just seemed to get better as it went along. Here is the script I used

LoadPlugin("C:\Program Files\DVD2SVCD\MPEG2Dec\MPEG2DEC.dll")
LoadPlugin("C:\Convolution3d.dll")
mpeg2source("C:\JOHN_Q\VIDEO_TS\johnq.d2v")
BilinearResize(352,192,30,0,660,480)
#Convolution3d (1,0,0,7,7,3,0)
TemporalSmoother(2,2)
AddBorders(0,24,0,24)
#IL = Framecount / 100 # interval length in frames.
#SL = round(Framerate) # sample length in frames.
#SelectRangeEvery(IL,SL)


Temporal Seemed to work better than Convolution so I commented that out. LBr template cq at 35 with max bit at 2300 min at 128

I hope more people will post their script and setting suggestions for their successful movies.

Filters before Resize
 

You will get better compression and better qaulity if you put your filters before the resize. One draw back is that the encoding time will increase.

Thanks sooo much NY, I have been trying to figure that out for some time now. I hope it improves my John Q, becuase when I finally finished the enocded, I had excellent quality in some places and then some major noise in others. Hopefully you suggestion will iron these problems out, of course I'll find out in another 20 hours :)