[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: re: Cintel Responses / Noise Levels

Dear All

Dave Fenton posted;

>>My point was not to distract though from us all needing to improve the
>>noise characteristics of Telecine, but to say that the Ursa Gold and the
>>Data Cine are actually performing in this respect to a similar level.

The problem with the above statement is that it ignores an essential
difference between the machines, ie the CRT / PEC, which produces
a noise signature that is not easily concealed by any noise reduction 

Well exposed negative is fine but changes in exposure bring a rapid
increase in PEC noise as, also, will a wide exposure latitude in any one
scene. This is made worse by relative grain size and the fact that with
negative transfer the black clamped noise is reversed into the whites.

The addition of third party devices, such as Innovations new PEC bases,
will no doubt help their performance, but the PEC can only be pushed
as a function of electrical gain and this has a direct trade off with noise.

The Spirit handles exposure changes by altering the light output
of a blue rich Xenon lamp and thereby will not increase noise by an 
increase in electrical  "gain".  I am sure that it was this almost  total
absence of  system noise that Dave Fentons  "people in the know
at IBC" were confusing for noise reduction. It may indeed be fair to 
say that it is partly this mix of PEC noise and film grain that give
the Ursa, and all Cintel machines, their  "look", and , whilst the Quadra
may have been criticized for looking  "flat", the unique imaging head of the
Spirit gives a very vibrant picture with only the film grain present.

PEC noise is also a reason why many people, myself included, are
not convinced about the validity of the new Freeze feature for the Gold.
When you run the PEC noise will become apparent, and a normal
level of noise reduction, say 2 to 3db, will not hide it. This may well
be unimportant whilst the exposure is fine but I am not sure how
Clients will react when the clean image that you have just pulled
from their  " two stops down but still vital"  shot,  fizzes like mad 
when you hit the run button. In addition the clean isolation that you
made for a layer of contained primary or secondary correction will 
also suffer.

Regardless of any controversy over split screens it is not correct to 
imply  that the Gold and Spirit are,  "performing in this respect to a
similar level",  when discussing noise. The technologies are different
and PEC noise is as inherent to the Gold as PAL / NTSC crawl is to 
encoded recordings.

Seamus O'Kane
VTR Ltd London