Hi all, I did this test for a comparison of dynamic range and have a margin of error within which I can retrieve details of overexposed or underexposed between GH2 with Flowmotion and HVX200 DVCPRO HD with whom I worked very hard and know enough well its limitations. I state that the GH2 was in "Cinema", which by all and recognized as the worst way for the DR. In the color correction I simply lowered the overall brightness with "YUV Curve" and I am impressed by the ability to retrieve color and detail with the DVCPro HD (note scales) in contrast with the AVCHD GH2 Hack Flowmotion. My question is: if this situation improves with other hacks or quality of 4:2:2 is still unreachable for us?
Or CM night? Flowmotion came out 6 months ago. I don't understand using one of the worst settings (Cinema) for color sat for this test.
@oneday, @rockroadpix I'm doing a short film with 30 children of 10 years old and I needed a hack to give me stability and ease of use. After using it I realized that "Cinema" is the worst setup and testing made it just to understand the limit.
@ proaudio4. The tragedy is that the HVX was not for a while and after I used this simple test I can not use it!
@ Rambo. In fact, the HVX, HMC 150 and a HV30 I use for live performances.
@ jclmedia HVX is with scene files 1, nd 1/8, 1080 25P, 5600K. The GH2, cinema, 0-0-0-0, 14-42 panasonic about 17mm, wb sun, hbr Flowmotion about 31 mbps.
What caught my attention and I wanted to highlight the test publishing, is the ability to recall details and color of the DVCPRO HD codec with 4,2,2, etc., that with the avchd, even with hacks (Flowmotion), not you can get. This is a problem of 4,2,0 8-bit?
With Driftwood hack, or whatever, you can achieve better results?
I think many of us shoot for jobs where you are the only element of the "troupe" and some errors may occur but having a file with a margin recovery makes you feel more comfortable.
As soon as I can make a new test with a different hack will post
It looks like you're new here. If you want to get involved, click one of these buttons!