View Only

Questions about Efficiency

  • 1.  Questions about Efficiency

    Posted 9 days ago
    With some new XPression features and new ideas for how to do things, I have a few instances where I'm questioning if a new process is better/more efficient, or just change for the sake of change, which I want to avoid.
    I'm sure there aren't really perfect answers, and everything is very dependent on a project's specific need. To a certain point, if it works, it works, but I am curious if even down to the software itself, certain things just work better a certain way or if I'm handling something with a sub-standard inefficient process.

    First thing worth mentioning is that for most of our systems, I am currently limited to XPression Prime, so unfortunately, scripting is out of the question.
    Secondly, our productions are mostly based around horse racing, so day to day things like graphics don't really change, just new live data sources being fed through Datalinq Server. Figuring this will give better context to some of my questions below.

    1. Importing images as materials vs Dynamic Materials using (relative) file paths
      • Day to day, our graphic assets don't change, but they are used very dynamically. For example; a single template for a scene showing betting markets is built to display anything from 3 to 24 Runners using Visual Logic and a different graphic asset for each size.
      • Importing the assets makes sense since they don't change, but I would like to make the templates even more dynamic so things like the color scheme (a different bundle of assets) can be adjusted without creating new templates to import and link new materials, and instead just push new paths to the quads (using variables/macros), and I can see that being very easy to do with dynamic materials, as long as it isn't a detriment to performance.
    2. Datalinq blocks feeding Text objects in Visual Logic vs Text Objects using a Datalinq Source in Layout (more Visual Logic or more Objects)
      • Currently, nearly all data is handled in Visual Logic, in most cases so it can be formatted before being shown on air in a Text Object. Things seem to be improving on this front, but the features and options between the two processes are not congruent.
      • Converting to Datalinqs being handled directly in Text Objects will offer the benefit of being more dynamic (able to publish the text object and modify the datalinq as TemplateData), but will also substantially increase the amount of text objects in a scene. Curious to the trade-off.
    3. Many 'static' scenes in the Sequencer vs Few dynamically-driven scenes in Sequencer
      • Currently, a template will have around 12 copies in the sequencer, one for each race in a race card (per track), with a datalinq key determining which race number to pull data for. Either hardcoded (1,2,3,4...) or using a macro like %relid%.
        But, I also have instances where instead of copying the scene out many times, I'll have one instance where the Datalinq key is instead looking to a text object linked to a Dashboard panel parameter; Datalink Key = @raceNum@, making the scene template 'smarter' and keeping the sequencer very clean.
      • Both seem to work just fine, but curious if there is a seen detriment to having lots of scenes in the sequencer. There are plenty of TakeIDs available, but has anybody ever run into any sort of cap? Maybe a factor as well, but we call all of our scenes (and completely run our productions) using Custom Panels in Dashboard, so less TakeIDs makes things simpler on that side, but certainly not enough to totally commit to that direction.

    Appreciate any insight anyone has on these items!
    For us, this stems heavily off of the ability to use Table Presets for static RSS/HTML Datalinq Sources, a huge game changer for us! Quite similar to an idea I brought up a while back, and I'm ecstatic to see it implemented! Being able to use macros to change Datalinq sources inside a template will immensely simplify how we handle updates and bug fixes for our XPression projects. Really looking forward to putting it into action once I make some building decisions based on the questions I listed above.

    Mike DeMarco
    Video Production Engineer
    Las Vegas United States