I too have a Panasonic, HC-V700, and later today will make clips at all the settings to see how my eyes see them.
Personally, I now always shoot in 60P 28Mbps, as I got tired of interlacing, have gotten used to putting my computer to work overnight while I sleep, and most importantly, I use 60P because YouTube lets me upload massive files.
I just taped a 1.5 hour talk I gave, in 60P, edited it in FCPX, used MpegStreamclip to cut it down to "only" 7GB in 720P, and uploaded to YouTube. It was in a room that was dark and totally concrete, and the highest setting made a huge difference. I can post it if you wish.
I will post later when links to my clips so anyone can see what the difference looks like on my camcorder.
But I am a home user not a professional, and invested in a 2TB Firewire drive, so I can afford to make 228GB 1080P files.
Make sure your tests include movement!
When I used to test the lenses of still and cine cameras forty years ago I used to stick the camera on a tripod and photograph buildings with good sharp outlines and detail as it was easier to see how the lens definition fell off to the edges.
However, digital video cameras are totally different and if you film a test chart or brick wall you will get very misleading results.
Even the lowest bit rate setting will produce the highest quality image, but introduce movement and the lower bit rates will usually show unpleasant pixellated artifacts, sometimes to the point of being unwatchable.
That's why I was somewhat surprised to discover no visible artifacts on the clip of someone digging and a slow camera pan.
I don't think it would have any real impact on the editing as AVCHD files are very small to begin with and if you converted them to ProRes I am sure they would be the same size (nearly a gigabyte per minute) regardless of the original bit rate.
The only real advantage is during shooting when 5mbps would give nearly 4 hours on an 8GB card as opposed to the normal one hour of 17mbps.
Here is a quick test of the five sizes and bit rates.
I really didn't know that three times the bit rate really doesn't result in large file sizes, though the quality is different.
It will take YouTube a bit to process the video.
Thanks Hugh for the test.
As you can see, it's not easy to do and it's impossible to replicate things exactly especially when the camera is on auto.
Watching and judging the results is also difficult because of eye fatigue and memory tricks . . . . maybe they should be kept to about 5 seconds each.
I think that the higher settings gave better results but there was nothing dramatic as you would expect.
I tended to look at the trees and bushes in the background which were in constant motion thanks to the breeze.
In the lowest (HE) setting they seemed slightly blockier with that "painted" look but even that is not conclusive proof because the breeze may have been agitating them slightly more or some other shooting factor may have altered.
I don't think an average viewer would notice any difference.
Regarding file sizes, the higher bit rates obviously produce proportionally larger files in the camera but once converted to ProRes 422 they are all approximately the same.
My 34 second HA file was 78MB in AVCHD and 510MB in ProRes.
The figures for my 33s 20f HE were 31MB and 506MB.
That's an interesting surprise! I assumed that because the bit rate was lower, the recorded info would be less, and the editing would be easier (on the computer!). If the "unstuffed" file is nearly the same, it's not really worth it. One good thing is it might make using the cards economically viable as an archive, like good old DV tape!
Lower bitrate = less info. Now if your are shooting a relatively low detail scene with little movement you may not see a big difference but as soon as you have movement you will see the difference - also when you hand hold the camera you'll generate more smear at lower bit rates. Also shooting at lower light eg at night you'll see much more noise and blockyness.
The lower bitrates are fine for filming speeches and such like but if you want to get the best quality and plan to shoot lots of short little clips then keep it on 60p. Having said that 60p is a right royal pain in the arse to ingest and edit (FCPX won't even see it) but the other modes on the panny TM700 are interlaced so look horrible if you are shooting vertical lines.
The only reasons for using the lower data rates are to be able to shoot more (in an emergency when you are running out of cards) or to economise on the cost of a card.
It is false economy if the quality is noticeably reduced but for talking heads and other low-motion shots it could be useful.
There is no advantage when it comes to editing.