Pixelated FCP Render of linked .motn file using LiveFont

Question

I have a clip that I've used the "Send To" feature to open in Motion. While there, I added two bits of text using LiveFonts (blueprint and timer).
It looks crisp as can be in Motion. I save it, Apple-Tab back into FCP and get the momentary "Media Offline" while it updates.
In the canvas, the frames look every bit as good and sharp as in Motion. I then render the sequence and the results are hideous... but only on the "timer" font. I had the same results using the "flip counter" LiveFont. The "blueprint" font, however, looks fine before and after the render.
The same fonts rendered in LiveType look gorgeous. I can overlay them in Motion just fine. I can use them directly in Motion and export a lossless quicktime (Animation format) and they look fine.
If I use Send To, though, to take advantage of the convenient workflow, I'm required to render in FCP and the results are hideous.
Does anyone know if this is a common problem or if there is some kind of workaround?
Here is a link to a sample JPEG showing the canvas window and the rendered output in FCP:
http://www.halfpress.com/images/font-problems.jpg
Keep in mind that these are two qualities are visible within FCP side by side using the same data. I choose a frame in my crappy rendered output, hit f to match-frame and am presented with pristine on the left and hideous on the right... but again, only for one of the fonts.
Appreciate any help!
- Aaron

Answer 1

Thanks. Actually, that seemed to do it... or very close. I also tried the Animation codec and it appears to be an even more precise match to the Motion source than the uncompressed 8-bit.
I had ruled that out initially because exports I did from Motion itself to the DV format looked semi-cruddy, but nowhere NEAR as bad as what FCP was rendering. That made me start thinking FCP was internally hosing the source when it renders.
I will say that FCP is rendering the drop shadows differently from the Motion source than Motion is previewing and rendering internally.
So is this procedure of running the finished sequence through a new timeline with a better codec pretty standard?
Another thing I had tried before posting my original question was exporting my working DV sequence with Compressor to a better higher quality format to see if it improved. I was doing this to see if only the editing process looked bad and maybe it would internally re-render through compressor to a higher quality. It seems it just renders from the nasty DV source even when going through compressor.
Now assume I need to drop it in a new sequence within FCP using Animation or Uncompressed 8-bit, render, and THEN export to compressor?
Trying to ascertain if this is a hack/workaround or legitimately "the way things should be".
Thanks!
- aaron