Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums AJA Video Systems Playback offset frame drift

  • Playback offset frame drift

    Posted by Chris Keppy on May 20, 2005 at 2:35 pm

    At Corus Entertainment we have 7 suites running FCP. FCP 1 – 4 use Cinewave for capture, FCP 5 – 7 use the AJA Io for capture. All captures from SDI and work is in 10 bit. In our test lab we currently have a Decklink Pro.

    When testing for a project we found that the playback offset in FCP1, FCP5 and FCP6 (the only rooms available for the current test) would intermittantly change. This includes rooms with both Cinewave and Aja Io capture devices. With Cinewave we are using a 2 frame offset, in the AJA rooms it is a 7 frame offset. On some of the outputs the edit would happen at the correct frame but the timeline would start to roll 1 frame late giving me a 1 frame freeze of the first frame and delaying the rest of the sequence by a frame.This would happen between 20 and 60% of the time. The edit point was always accurate. It did not matter which machine we output to. We also found in our tests that the Decklink card in the lab did not change. Since ther are so many things different in the lab as compared to the edit suites that we could

    Steve Covello replied 20 years, 11 months ago 4 Members · 3 Replies
  • 3 Replies
  • Aja Sales department

    May 20, 2005 at 4:29 pm

    Hi-

    Have you contacted our support department?

    support@aja.com
    (530) 271-3190

    Thank you,


    AJA Video Systems
    (530) 274-2048 Intl.
    (800) 251-4224 US
    sales@aja.com

  • Kim Rowley

    May 20, 2005 at 4:49 pm

    I read your post with interest and applaude you on your test. I work for a company that currently has 4 M100 edit suites. Just yesterday we placed our order for 2 FCP Studio’s with the lastest G5 and the AJA I/O.(We’ll change the remaining 2 suites shortly) I researched just the exact problem you’re having since frame accuracy is of paramount importance to us in laying off to Digi Beta. I was assured by the AJA tech team that frame accuracy was not an issue. Could you take the time and trouble to update this forum as to how everything works out for you? Thanks a million.

    Kim
    Editor
    CSC Video

  • Steve Covello

    May 24, 2005 at 1:53 pm

    I had the same problem (double frame, output offset) insert editing a single clip in the middle of a program with the Kona HD working in 23.98 1080p mode back in 2003. At the time, I found the following advice:

    >>>The only way to make frame accurate insert edits on Kona HD is to go back to the 2.3b7 BM drivers from over a year ago.
    If you have to make inserts with your current setup, try resetting the Device Control and you will be able to make 1 edit that doesn’t repeat the first frame. (lousy work around)
    Ever since the 4 series of drivers came out I’ve lost the ability to do frame accurate inserts or assembles in 1080P 23.976 to D5 or HDCam. I have told both BM and AJA about this problem, the exact same problem exsists on the Decklink HD.
    Maybe when AJA comes out with thier own drivers this will be fixed.<<< I have found that Panther Cache Cleaner has cleared up a few output problems (incorrect channel patching on Kona2), though it seems recent drivers have permanently fixed this problem. Also, do you edit to tape from the Canvas or Viewer? i always output from the Viewer and never get that "incompatible compression" error, and who knows what else I am avoiding. I also seem to recall back in 2003 that if I tried to output consecutive masters with IO with the same Edit to Tape window (without closing the window then re-opening it to do the second tape), I would get a terribly inaccurate output on the second tape -- unlike my old Avid where you could output a million masters with the same Outut setup. This problem seems to be solved on K2, but I haven't checked on my IO for same problem. I have done several tests similar to yours and only have issues in HD. IO and Kona2 in SD all seem to be right on when outputting an entire program. I will test out an insert edit and report back. Steve Covello double wide post

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy