I looked at that and was thinking the contrary, until I saw that the d5100 was the more contrasty one. I think it is much more about the film mode. What surprise me is that it seems to have more detail than the gh1/2. I think a hacked gh2 would give much better detail into the shadows.

@danyyyel

I don't know man... if that happened I think that Jim may get 'really' mad... You don't want to see him mad! Last time that happened he sold his last company and started a new one!
(Just imagine - GH2 'Red' style! mmmmm!)

It was a hacked GH2, but the hack is just making better looking noise, it's not changing the DR.
We counted a stop ( or a half ) wherever some detail stuck out of the mud, or all cameras would have counted less. There is noise in an Alexa too – just looking nicer. The hack is shifting the acceptable level of noise on the GH2 ( a very subjective thing ) not the technical level.

Thanks @nomad, very interesting account of your test. If and when your test get online or in press make us know to see the conclusion. Yes I do understand the problem of a log output more so with a 8 bit output. The idea was to get a good flat camera film mode to preserve the best dynamic range, it does not need to be as flat as the log ones. A last question would be if you use a hacked gh2 and does it make a difference on a chart, but I understand if you can't answer.

@nomad could you tell us what was the result from the others you tested against the gh2. I have seen a lot of test that showed different result using different methods or charts. If you refer to Dpreview system most if not all the dslr score in the 7 to 10 stops in the jpeg dynamic range test (I think it is the jpeg the we are talking here as it is more related to the video side). While on other test from provideocoalition or the zacuto test, cameras like the 5d are showing more like 10.5 to 11.2 stops.

In the end I am a little confuse about that because it is difficult to get a good idea how the gh2 stack against the other cameras. For me dynamic range is as important as DOF for the film look and even more important than sharpness in that regard. We know that the gh2 is better than all the DSLR and many camcorders in terms of sharpness, what we don't know is in terms of Dynamic range. The hack itself has improved the DR, because now we have shadow detail which is not being smeared by compression. You can better expose for the highlight because you will be able to boost the mid-tones and shadows in post to a certain level, which was not possible before as it would immediately show macro-block in the shadows.

Now an idynamic mode hack would be very interesting. If the processing is done before compression on the raw data, then we would get very close to the sensor full dynamic range. What I found interesting in the threat on idynamic is that the gradients on the wall with the cat was much better. I hope that @Vitaliy tries to hack it, so that we can get an always on version. It would surely alleviate a bit, the need for a flat cinestyle/slog camera film mode.

@Vitaliy_Kiselev Out of curiousity, do you know if Chris is making any headway in his film modes research, and whether there are things people could do to help him?

For dynamic range you can just use LED panel and few ND filters on it :-)

@DouglasHorn
For daylight, I just use Smooth and more often than not, "sunlight" white balance. I pre-process most footage to normalize exposure and color balance, and I'd rather start with known camera settings than a variety of white balance settings. For nighttime incandescent shooting, I always use Nostalgic @ 2700K with M-4.

LOL!!!

3. Nostaligic also has a pretty bad color shit to yellow :-)

@querty - "Moreover, this kind of thing would probably be better done in post, no?"

Not in this case. If the ambient light you're shooting with is primarily reddish yellow, any attempt to balance it closer to white will reduce the camera's signal-to-noise ratio. And if a warm tone is what you're looking to produce, it's better to get as close as you can to that balance in-camera, in order to minimize the amount of color grading you'll need to do in post.

The question I have is what was the color temperature of the practical you used to illuminate your test shots?

@bitcrusher :D beautiful images!

@sb1 can you post a link for the dpreview you are talking about please?