Cinema Camera 3:1 Raw?

Hey, this is quite a simple question, but does the Cinema Camera have 3:1 RAW?

The Blackmagic Cinema Camera 2.5K does, indeed, have visually-lossless compressed Adobe CinemaDNG raw with a ratio of 3:1.

The Blackmagic Cinema Camera 2.5K does, indeed, have visually-lossless compressed Adobe CinemaDNG raw with a ratio of 3:1.

I think you might got something confused here. The BMCC 2.5K has mathematically lossless RAW with a compression of about 1.5:1
(It's visually lossless as well, because it's exactly the same image as uncompressed after decompression)
The 3:1 RAW ist only available in the URSA and URSA Mini right now.

I've got a feeling, being that it will be on the Micro. It will trickle down to the 2.5K and Pocket.
I mean it sounds really good alongside a version 3.0 of the firmware, doesn't it?

The Blackmagic Cinema Camera 2.5K does, indeed, have visually-lossless compressed Adobe CinemaDNG raw with a ratio of 3:1.

I think you might got something confused here. The BMCC 2.5K has mathematically lossless RAW with a compression of about 1.5:1
(It's visually lossless as well, because it's exactly the same image as uncompressed after decompression)
The 3:1 RAW ist only available in the URSA and URSA Mini right now.

Huh. I could'a sworn...but I looked back through the webs and I can find no mention of 3:1 on BMCC so I believe you are absolutely correct. I must be getting my wires crossed being that I've had Ursa Mini on my mind for so long. At least I know the compression ratio of my BMCC now! Thanks! I do hope 3:1 trickles down to BMCC some day, though.

A lot of the confusion is likely due to the unfortunate promotion of the rather meaningless buzzword "visually lossless" by manufacturers, according to the esteemed John Brawley, when it in fact it just leads to users getting befuddled--its gotten so bad that I saw a bunch of folks on a RED forum insisting that R3D 3:1 was (mathematically) lossless which is quite a silly claim indeed. If we could all agree to refer to lossless as lossless, and anything else as something else as he suggests, then this sort of confusion can in all likelihood be avoided. One can only hope...

Info on the BMCC compressed (but visually lossless) RAW update a while back: viewtopic.php?f=2&t=33575

Info on the BMCC compressed (but visually lossless) RAW update a while back: http://forum.blackmagicdesign.com/viewt ... =2&t=33575

Remember - the BMCC is not "visually lossless" - it is Lossless.

Visually lossless is just another way of saying Lossy. It just sounds better than saying "yes its lossy, but you probably won't be able to see the difference"

Lossless means that when unpacked at the other end - the data is - bit for bit - identical to an uncompressed image.

Visually lossless is just another way of saying Lossy. It just sounds better than saying "yes its lossy, but you probably won't be able to see the difference"

When properly designed, "visually lossless" can indeed be indistinguishable from the original. There is a catch though. In the scientific community, "visually lossless" has always been meant for final images, with compression parameters calculated under specific viewing conditions with fairly strict statistical psychophysical vision models in mind.

Enter post-process into the equation and all this goes south. You never know how the image will be abused in post, hence you can't really claim "visually lossless" for post-oriented compression.


And related to the OP's question, there might be something cool coming soon in an update to a certain program linked in my signature.

Visually lossless is just another way of saying Lossy. It just sounds better than saying "yes its lossy, but you probably won't be able to see the difference"

When properly designed, "visually lossless" can indeed be indistinguishable from the original. There is a catch though. In the scientific community, "visually lossless" has always been meant for final images, with compression parameters calculated under specific viewing conditions with fairly strict statistical psychophysical vision models in mind.

Enter post-process into the equation and all this goes south. You never know how the image will be abused in post, hence you can't really claim "visually lossless" for post-oriented compression.


And related to the OP's question, there might be something cool coming soon in an update to a certain program linked in my signature.

Hence the word "probably"

Side by side, I can clearly see a difference in the shadows of an image, it also affects the noise characteristics too.