Sounds like your rendering something MP4-like, perhaps?
Rendering in 32-bit makes no difference for what is normally 8-bit footage unless you are doing multiple operations that have very different intermediate output results in 32-bit vs. 8-bit, primarily in the context of compositing.
There is some depth to the use of this, more than can be provided in a short forum post. Good books are available if you want to learn this at a higher level.
The fact is that all broadcast is 8 bit, and while this is just my opinion, I think it's correct; the only benefits to be had by 32 bit are in things like compositing, where you can get a better key because you're working with a little more room to work while keying and then when you render it out ( depending on your output you will probably go back to something that is 8 bit anyway.
32bit is also done in a different gamma for the time being ( I say for the time being in the hopes that the next version of vegas will not still operate this way, but whether that actually happens is another thing ), which can cause some color shifting issues when working with different formats out there.
You're not likely going to gain much, and while some people do have problems with there systems being unable to handle working with video being converted to 32bits per channel floating point, it is not a given that it will crash your system, I've done plenty of work with 32bit that was just fine and my system was completely stable during it.
It will kill your playback (most likely) and so editing in that way, can be a bit of a bear, but overall it's still useful for certain things, just not very many at this point, and it has it's draw backs as well.
My dvd's seem a bit grainy, so I think the camera is having some problems. I am trying to get JVC professional to reply to my service request but I am having some trouble.
Video doesn't have grain, it has noise. More light is the best cure.
If you're encoding at a low bitrate then the noise can become 'sticky' and look kind of like grain.
There are some very visible advantages of going 32bit if you want to do linear light compositing. e.g. cross dissolves will look somewhat more like film dissolves.
The easy way of working is to stay in 8bit. Use the SMLuminance plugin for cross dissolves if you like that look (8bit plugin, but same results more or less as the 32bit way of working). Nest 32bit projects into your 8bit project if you require linear light compositing.
Or, you can avoid the 32bit stuff altogether.
The fact is that all broadcast is 8 bit, and while this is just my opinion
Broadcast is 10bit Y'CbCr for signals going over single-link SDI. It might not make a huge difference since many HD masters are delivered on HDCAM, which is 8bit I believe.