DV Res to 8 Bit Res graphics problems

Question

Hi,
Our current project was shot and edited in DVCAM and will be output to Beta SP for delivery. Our budget did not allow us to shoot with Beta SP cameras and we can't deliver in DVCAM, hence the different formats. Rendering up the DVCAM clips suffers some loss of clarity and depth but is ok given that we have to make the tradeoff - the graphics however look terrible - we know why but what I want to know is...
We completed the DV sequence and included Livetype project files (for our text based graphics) that were set to a canvas size 720x486 because I assumed that we might as well do the graphics at the final resolution and render down for the edit (DV NTSC). When we rendered everything up to 8 Bit for the final output, the graphics looked terrible.
What I'm assuming happened is that the down rendered (to DV for the edit) livetype graphics were then rendered up to 8 Bit - essentially rendering up a low res rendered file. What I expected to happen was, since the graphics were already at full resolution, that the earlier render down to DV would be replaced by the full res file - I guess I was expecting FCP to recognize that the graphics were already at the resolution they needed to be and simply leave them as they were created, at the full resolution.
So we had to go through one by one and replace the graphics files with the original full res files to bring them back the way they should look.
So a few questions:
1. Is there an easier way to do this apart from leaving the graphics until after the render up to 8 Bit. (We want to have the final show including graphics in DV format for the sound recording session)
2. I tried to do a media reconnect with the original graphics files thinking that would be a quick way to replace all the affected graphics, but it didn't work, nothing happened.
3. Is this a bug in FCP version 4.5? Shouldn't FCP look at the canvas size of the livetype project files to know what needs to render as it does with media clips?
Thanks for any input or suggestions.

Answer 1

Thanks for the info:
Doing dubs to/from digibeta simply isn't an option as we are under seriously low budget constraints which is why we had a mix of beta SP cameras and DVCAM cameras...
We digitized the footage using a decklink extreme card (8 bit setting for Beta SP) and then NTSC firewire for the DVCAM footage (using a DSR-11 deck). Since we are delivering on Beta SP, it seems appropriate to finish the edit in that format (8 bit). Anyway, the DVCAM footage definitely looses sharpness when it is rendered to 8 bit but it is a tradeoff that we are stuck with for the moment. The livetype graphics as described in the original post suffer horrendously unless we wait until the end (after all the media has been rendered to an 8 bit sequence), and create them at that time.
Does this make the situation more clear? Does anyone have any solutions or suggestions? Thanks.