If I weren't whacked out of my head on cold medecine, I probably wouldn't post this. BUT since I am - here we go.

Now that I've shot with one, and played with the footage in post.

Here are my thoughts:

The jello is horrible... HORRIBLE. That's gotta change. They did fix many of my gripes with the latest firmware release in terms of iris control. The Focus peaking is nice. The shadow response is kinda weak. I think it should come with a sticker that says, "NEVER UNDEREXPOSE YOUR SHADOWS WITH THIS CAMERA.....EVER." Meter you scene, make sure the highlights will fit within 11-12 stops or so, and let them blow, cuz you can get them back... but you're not gettin' your shadows back, brutha... no way. BTW a good and accurate False Color function would help with this.

Verdict: If on a tripod or an extremely steady hand held rig, it's a nice camera. It won't get my money, but there are many worse options out there in this price range.

In Post: I'm curious why Resolve doesn't have as comprehensive an interpretation dialog as AE. Cuz the exposure slider alone ain't cuttin' it. I want CONTROL. I haven't updated to the new version of resolve (9.1 i think it is) to see if they've remedied this yet, but I'd like to have independent control over everything stored in the DNG - the highlights, shadows, exposure, contrast, and brightness. Everything else is already there; color temp, etc.

And as a bonus: Just some thoughts and predictions that I took a lot of heat for last year before the release.

http://www.personal-view.com/talks/discussion/comment/63170#Comment_63170

http://www.personal-view.com/talks/discussion/comment/69912#Comment_69912

@GravitateMediaGroup

It's a different issue. Focus peaking tells you if something is in or out of focus.

Having the right FFD, determines if a lens can hit focus at it's furthermost distance (infinity) and then how the witness marks or distance scales on the lens line up for accuracy.

Ideally, a lens will be set to hit infinity when the little mark says it's focused to infinity on the lens and at the correct FFD on the camera, which in this case is 44mm.

As BMD have discovered, stills lenses lenses aren't that accurately calibrated for both their distance markings and their FFD. Faster wide angle lenses tend to show this up the most and the Tokina seems to have the worst tolerance (judging by the number of FFD issues that are associated with this lens on other cameras including Canon)

jb

Why can't focus peaking reduce this issue? Being that it's a feature available.

@nomad - not true, there is a Sigma 10-20 with constant ap. - f3.5. I have it and it is a great lens.

http://www.bhphotovideo.com/c/product/633609-REG/Sigma_202101_10_20mm_f_3_5.html

They wanted cinema lenses that have a hard infinity stop and accurate witness markings for focus pulling to be accurate and found that many EF mount lenses aren't that accurate.

Most of their buyers do not have even single cinema lens (I do not count for such Samyangs with gears).

@robmneilson

i'm glad one brave soul is attempting this. let me know if they are true to their word about a quick turnaround. I've heard a lot of nightmares people have had with sending equipment back to larger manufactures (panasonic, canon), so I'm curious to see how BM does.

Shipped my bmcc back today to be recalibrated, they say its up to a ten day turnaround. I guess ive already waited this long.....

The cameras are all shimmed and calibrated to 44mm as per the reported focal flange depth of EF mount. I use the term "reported" as Canon does not publish the specifications officially so the exact expected tolerances are not known.

When I tried the same collimated lens on some 5Ds, 7Ds and even a C300, the witness marks fall short of the expected distance which indicates to me that there's some fair amount of tolerances built into the Canon cameras to cater for still lenses where the tolerances are not as critical as cine lenses.

To address this, we need to put in the same allowances. We have tested this with some of the cameras from customers reporting this and it works.

If you think you're encountering this issue, please contact your local Blackmagic Design support office and we'll sort you out.

The camera will have to be sent back to be recalibrated

To be short - somehow BM though that they do not need to allow tolerance for ideal lens to shoot past infinity. They also did not make measurements of actual flange distance of enough samples of Canon cameras before issue became known.

After spending 10 weeks helping to color-grade the film [Oh Brother Where Art Thou?], Deakins said he was impressed with the new technology’s potential, but remarked in the October 2000 issue of American Cinematographer that “the process is not a quick fix for bad lighting or poor photography.” (This quote can stay in the present tense, of course, because it will be relevant in perpetuity.)

He was using DataCine scans from Super-35mm. Totally, @eatstoomuchjam , all RAW requires processing. Unprocessed RAW looks like crap no matter the camera. Some of the cutting edge HDR displays may be able to one day show a RAW image in near totality but they're still going to be applying a LUT of some kind.

I have a very basic question: the BMCC claims 13 stops of dynamic range in a 12bit DNG file (raw). How can a 12bit raw file exhibit 13 stops of dynamic range? 12bits mean 2^12 values for the luma information, hence also 12 stops (luma doubles with every stop).

Or is there some confusion regarding the term raw? Raw means for me no processing, just a huge junk of data from the sensor. In this case raw seems to be more, namely 13 stops mapped onto a 12bit "raw" file via some sort of curve.

I understand the mapping of lets say 11 stops of DR mapped via log (or something else) on an 8bit codec. But RAW?

What am I missing?

@BurnetRhoades

Yep, but it is always required to add all test details before this video, as this guys used very strong ND filters to specially show big color shift.