I finally got around to hacking my GH2 so I ran a couple quick sample shots to see if I'd notice any difference. I used the stock firmware, Intraesting 80mbps, and Orion V2 up to 154 Mbps GOP3.
The first part of the test is really just me at my computer late at night filming to see if I can notice any change. I walked around and filmed the different light sources in my living room. Really, a hack shouldn't pull more dynamic range but some people said it did. I don't think so. In fact, the depth of field is too shallow and dynamic to really tell anything. I grabbed some stills to give the watcher a chance to investigate.
It would be much better to go outside and test the hacks on a pair of sticks with some panning movement. I wanted a location with movement and lots of fine details, so the park. You can see trees blowing in the wind, cars driving by, and lots of fine details in the grass or shingles on the gazebo roof. Frankly, watching this on a HD tv uuncompressed, I still don't notice a fantastic difference. Especially considering how much card space the hacked footage uses. Key differences were that certain details were rendered better, but not significantly. Being able to count the grass better doesn't change a film. Furthermore, the swing moving in the GOP 3 footage does retain more detail when looked at frame by frame. But, when playing the footage at normal speed we really don't notice a big difference.
Finally, I tested some other features of the hack and that is at the end.
This is no place for a debate on the GH2 hack, but all in all I can say that the camera is not meant to use such a high bit rate and it isn't an efficient user of high bit rates. So, I would say that high bit rates aren't necessarily useless, but over rated. AVCHD is a very good codec, better than H.264 in a technical way- but it isn't a good editing codec no matter how high the bit rate. I think with a higher bit rate color correction might be an easier task, but I didn't test that and really I haven't seen anybody ever test this. As for GOP, no as noticeable as I thought it would be. I think really the big change comes more in the processor of the camera, and the bit depth, not the Mbps or any other superficial numbers. This kinda reminds me of the old Megapixel count battles and how they really didn't matter. My 4mgpx camera took awesome pictures, better than most 16 megapixel cameras- you just can't print large. So, would I hack. Yes, because I need PAL and NTSC; any other gains just seem negligible.
Also, some have said the red channel get better with Hacks, I don't think this is possible because you are not getting 4:2:2 or 4:4:4 all of sudden, just use 5d to RGB software to improve this.
I will say however, that a higher bit rate makes a big difference during snow or rain- or anything with insane amounts of fast moving detail. Not is this shoot, but later on I shot a rain storm with the hack and non hacked and I noticed a huge difference- I have also seen snow video online that prove this point.
Music by Ocean Disco.