Re: Nikon Acknowleges Banding, Sort Of

[Arthur Entlich <artistic-1@xxxxxxx>]

If the problems with banding in the Nikon LS-8000 are anything like 
other banding issues I have deal with, (the HP S10 comes to mind) the 
causes varied, because the symptoms varied.  In other words, people used 
similar terms (banding) to describe numerous different visual defects.

I don't know what it is that Bob has seen which he is describing as 
banding.  It could well be that there are potentially numerous forms of 
abnormalities that are being referred to as banding.  In the case of the 
HP-S10, there were five different "defects" which were being referred to 
as "banding".

1) Evenly distributed darker lines in dark areas of slides, of a high 
frequency, but which were always straight across the whole slide going 
side to side in the narrow direction.

2) An almost concentric ring banding which some people described as a 
phantom "fingerprint" across their image

3) A greenish matrix which was mainly visible in darker areas.

   4) A pattern, that was often referred to as diagonal banding which 
only on the closest observation could be herringbone patterned

5) A high frequency banded "interference" which was only visible in 
color images and which appeared to be mis-converged or misregistered of 
red/green.

Each of these had unique causes, and some of which are still not resolved.

  Number 1 is still not resolved but was corrected in the newer models 
and the extent varied considerably between units from very slight to 
rather substantially obvious.  The result of the surveys from 
respondents indicated certain manufacturing runs were more likely to be 
problematic.

Number 2 was rare, but was determined to be splatter of the IR LED on 
some scanners causing it to be picked up by the CCD, due to the LED 
being misaligned.  This IR LED and the sensor were used by the scanner 
to determine if their was film in the gate.

Number 3 appeared on a number of scanners, including one I temporarily 
owned, and might have been a poorly manufactured CCD or miscalibration 
of the output levels.

Number 4 was caused by either bad powers supplies, or EM interference 
from other devices which were too close, and was usually resolved by a 
change of the "wall wart" or repositioning the scanner or other high EM 
emitting equipment or wiring.

Number 5, might have been due to slippage of the film or a software 
problem.  The film is advanced on the HP S10 via rubber rollers.

All of these image defects were reported at one time or another as 
"banding" by users, yet they did not all have the same description, nor 
were they rectified by the same repairs or fixes.

So, without knowing what Bob was experiencing, I cannot rule him and his 
experiences out so quickly.

And, just to clarify, the non-calibration of several of the Triline CCD 
sensor lines is not my theory, it is a theory provided by people I trust 
and respect, for at least one type of "banding" on the LS-8000 and until 
someone can provide a better one, it seems plausible and reasonable to me.

Art

rafe.bustin@verizon.net wrote:

> On 12 Apr 2002 at 2:18, Arthur Entlich wrote:
> 
> 
>>Be careful Bob - Rafe, Dickbo and Moreno will not only tell you its all 
>>in your head, but that you should be taken away in a straight jacket for 
>>making such public pronouncements.
>>
> 
> 
> If you gave it a moment's rational thought, Art, you'd see 
> that Bob Zwarick's observation flies in the face of the 
> CCD-calibration theory that you pose as fact.
> 
> 
> rafe b.
> 
> 
> 
> Bob Zwarick wrote:
> 
> 
>>I've moved mine away from the computer and monitor, shielded the cable and
>>found improvement.
>>


-
Turn off HTML mail features. Keep quoted material short. Use accurate
subject lines. http://www.leben.com/lists for list instructions.