Yes. Unfortunately, that is what I am trying to do. And what you suggest is very helpful and an approach I think could work. I remember seeing a demo of letters bouncing around with other objects in a similar fashion that might give us
some insights. I will have to go look for it. Found it.
Having said this, after thinking about your suggestions, maybe a different approach might work as well? What if I could always constrain the text to fit within a known text box size? Then, instead of rendering the strings/text, we just used standard
farseer boxes as graphical objects with physics. Then after the fact overlay the text for each box after the fact based on box position. Similar to what you were suggesting at first, but with a twist. Don't display the graphic boxes at all.
Have them render behind the scenes (if that is even possible), lookup the location for each box and then display the text at that specific location where the underlying farseer boxes are located. This would completely avoid the need to do all of the
display/on screen work and should significantly improve the performance of displaying text because it doesn't have to be rendered as sprites.
This all sounds pretty crazy to me, but if you think about it from an old dumb terminal perspective, you could then just display the text on screen via one swipe/review of the hidden physics screen with everything in the "right" location and do all of the
physics work centrally on a server and with only a few passes to the data. Remotely displaying the text as needed, and at whatever speed the device/network allowed.
Is any of this even possible, or are we delving into the bizzare, stupid and unknown.
Thanks again so much for sticking with me on this thread. It has given me a ton of new insight.