q matrix question, grainy in dark scenes

[muu]

does anyone use the cymat 1.1 q matrix for 1 dvd9 -> 1 dvd5?

is cymat 1.2 matrix better quality if decimating bitrate from 8500avg dvd9 to 5600avg on dvd5

i understand undot avs filter increases compressability, does it degrade quality noticably?

my 5pass 5600avg test encode looks grainy in dark scenes, should i set quantizer characteristics - flat priority above 30? i dont think i want to decrease contouring, i want to decrease mosquitoing so a lower flat priority? is mosquitoing the proper term for 'grainy'?

does a vbr bias of 30 seem appropriate for this source? i am not entirely clear on how to determine a proper vbr bias even after reading the avbr comparisons, could someone suggest a vbr bias that would decrease the 'grainy' effect or perhaps a q matrix more apt at dark scenes?

i appreciate any help

[freelock7]

No, CYMAT works better in low bitrate than CCE original matrix (VBR:2500).
Test it with QCCE_CYBEREDITION-from Amnon82, a really very good tool !
:)
About quantizer caracteristics, try between 10-20 to avoid mosquito noise.
Look at this thread:
http://forum.doom9.org/showthread.php?s=&threadid=82904
Concerning noise in the video, it's another problem that you have to resolve through denoiser filters (undot, ...).