@kholi it looks pretty close to BV1's, or you can say on par with BV1? you've done tones of work on the BMCC so you sure know what the BV1 ProRes look like @ it's best.

I think there's plenty of detail with modern glass on the camera, maybe a chart will tell how close it is to BV1, so far it actually looks like even ProRes from BV1's a decent amount more detailed. But, there seems to be less moire in all of these shots.

Ooh shoot John Just posted some more footage they looks amazing Wow!!! the amount of detail is breathtaking.. I think there are 3 more clips on the way.

@bwhitz I don't necessarily disagree, due to GH lens correction, sharpening and the compromises made in its downsampling (which is not as high quality as you'd get in say Photoshop) it's a complex soup.

Also don't forget that the GH2 does heavy in-camera NR which again softens things a bit and removes grain. But grain also gives images a false sense of detail (adding grain to a soft image can do wonders to its perceived sharpness).

And don't forget lens quality/aperture etc. We'd need to compare the same scene, lens and exposure to be sure right?

@bwhitz From your observation do you think the BMPCC Resolution match it's bigger bother BMCC? since you have worked with it. What about RED footage how does it stack-up?

@bwhitz, a Bayer sensor shares the photosites between red, green, blue, with the green getting twice as many samples. Luma is mainly composed of green (as that's what our eyes are most sensitive too), and there are only 1/2 green samples - so you are not getting a full resolution luma image either. It's all interpolated in the debayering step. The interpolation is complex, but the upshot is that you really want 1.3x oversampling to get the full theoretical resolution.

re. GH2 sharpening, if you read my last post properly you know we're on the same page (I shoot with GH2s). Nevertheless the GH2 has access to far more pixels. But it looses some resolution due to lens correction (on native lenses). Hard to compare them as you can't turn the GH2 sharpening off - but if you have to do lens correction on the Pocket in post, you'll loose more res.

@jrd I was just expressing an opinion relative to others' comments about NR workflow requirements. I'll withhold final judgement until I can grade the images myself and assess the properties of the grain. So far the noise in JB's ungraded files doesn't look all that different from my GH3 files with contrast set to -5. Probably not a big deal at all.

@smsjr I have to say I am amazed that you can tell that other ppl's monitors are not calibrated.

re. the GH2 resolving more, it's bound to. The Pocket sensor is 1080p native, but due to the Bayer photosite layout, you don't get the full 1080p resolution. You want approx. 1.3x larger capture resolution to make up for it, that's why the original BMD cam is 2.5k.

The GH2 in the meantime oversamples its 1080p from a much larger res (over 4k), so it has more detail to play with. On the other hand it oversharpens (partly to hide the lens corrections), especially bad once you hack and your image is clearer already, and you can't completely turn it off.

It's all tradeoffs.

A base ISO of 800 should resolve cleanly in daytime shoots unless there is some sort of electrical noise taking place within the internal circuitry. Having to use noise reduction passes as a standard part of the camera's workflow would probably lead me to just return the camera and walk away, so I hope that is not the case with the BMPC.

Amazing! Brawley stated that he had some footage that he really didn't want to put out there because he knew the reaction it would get, but some people were bitching and moaning that it didn't matter and he should upload anything that he had. So he decides to post it and now some of the same people are blah blah blah blah blah...

I'm no colorist, but the footage is more gradeable than folks are letting on.... This took all of 5 minutes in Lightroom (thanks to JuMo for the stills).

The original footage was very flat and neutral, which was nice for grading, but there was little to no highlight recovery possible in this prores file.

Grading was nothing too special: 7-8 minutes messing around in AE with levels and curves. No sharpening and no noise reduction. I might spend some more time tinkering tomorrow in Resolve, but I may just hold out for raw footage or for my own camera's delivery.

On a side note, why is the footage so bad/boring? You're charged with revealing a highly-anticipated camera's first footage to the world and this is what you come up with??? Give us something exciting! Is JB instructed by BM to make the footage as amature and boring as possible? I guess the footage provided us a good test of dynamic range, but if that's the goal, then I'd much rather have a few seconds a raw instead.