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

Re: flex-files

In a message dated  12/8/96 7:20,  skot   soil at slip.net  wrote:

> Without receiving much support from the manufacturers we've managed to 
> develop a variety of approaches to flex file management and generation 
> none of which I'm completely pleased with.   <deletia>
> Replying to me directly or throwing it out to be devoured 
> by the group would be appreciated.

Let me take the risk.

> We're equiped with Bosch Quadras, Evertz 4025 / 5150, Evertz Keylogger 
> software and a Beta series ASTEC.

As far as Keycode timing accuracy is concerned, our experience shows 
that Bosch Quadras are trickier to handle than Cintels, because
of the dampening roller situated between the keycode reader head
(be it Aaton, CP or Evertz) and the scanning gate.  This spring 
loaded roller induces a variable length - hence variable delay - between 
the actually read keycode and the scanned film image. No electronic means 
can compensate this erratic mechanical delay.  
If film tension - insured by the winding torque motors - varies during 
transfer, the damping roller changes position, and your initial 
phase and delay settings are no longer valid: you end up 
with a timing error of one (35mm) to two frames (16mm).
(This never happens on a Cintel because whatever the film tension, 
the distance between the Keycode reader and the gate is constant).

That is why we kept telling telecine engineers (KL manual section k.3.2) 
to block the FDL 60/90/Quadra 'elastic' roller; but they 
were reluctant to do it, fearing to introduce steadiness 
problems: "Why did BTS put this roller in if it is of no use?"
Recently we had a Keylink installed on the most recent SPIRIT and 
discovered a BTS-made blocking/unblocking device underneath the 
damping roller!  We thus got confirmation of our say: 
the film path of all BTS machines (Spirit included) was set mid 80s at 
a time FDL60s were mainly used for release-print transfers; since splices 
generate bumps that show on master tapes as vertical jumps, the use of a
damper was necessary - Keycode didn't exist anyway.
Negative transfer is another story; nobody cares about splice 
bumps (if any*) because they occur at the roll head, unusable anyway.
And since absolute Keycode accuracy is becoming a must for the transfer 
of dailies, the blocked damper had to come in.

BTS engineers are 100% positive: blocking the damper DOES NOT jeopardize 
the Spirit steadiness.  What is good for the Spirit at 1000+ lines should 
be good at 525. BTS will make this blocking device available for 
FDLs & Quadras too.

> Thanks and remember to blink when your shift is over.

My shift has been a bit longwinded but I hope it will help you finish 
yours earlier! 


* the 'Hammann Guillotine Splicer' now totally eliminates physical splice 
Jean-Pierre Beauviala          jpb at aaton.com                 www.aaton.com