vcassel
Veteran
I'm considering two now as well. Want to play with the one first and see how it goes. Being used to the URSA, I can't see how it will disappoint, though.
4k bmpcc
- Thread starter starcentral
- Start date
I'm considering two now as well. Want to play with the one first and see how it goes. Being used to the URSA, I can't see how it will disappoint, though.
this will probably be the m.2 of choice ...for the moment... http://www.samsung.com/semiconductor/minisite/ssd/product/consumer/960evo/
John Brawley
Well-known member
Huge disappointment if true! Any reason behind this? Is it technically possible?(opens a window of hope if they decide later down the line).
John Brawley
Well-known member
I have a feeling there might be a technical reason for it. ProRes 12bit might be harder as it’s encoded than RAW 12 bit. From memory that's why it wasn't on the first pocket.
JB
JB
Last edited:
John....will you be a beta tester ?
morgan_moore
Major Contributor
Im not expert but I beleive 444prores is very innefficient space/processor wise vs raw - it would never be sensible when one is at design limits for heat and processing power - its a nice have on bigger cams but ultimately on a 'pocket' camer you are just grabbing data and heavy lifting happens in post.
It's definitely less efficient space wise. 4K ProRes 444 is huge. It's actually pretty easy on the processor side of things (not including the image processing that goes along with color grading, obviously -- that's dependent entirely on whatever you do in color grade), but requires a lot of disk bandwidth.
Samuel H
Major Contributor
I'm becoming less optimistic on the Sony side of things (I don't like the slow-lens RX100VI, and there's still no sign of a7S III, or any indication that it will have 4k 60fps, internal 10-bit, or improved DR), so I'm starting to lean strongly towards the pocket 4k.
So, some thoughts on what is always the main issue for me: dynamic range. We'll know for sure when people get their hands on the camera and run real-world tests, but I just wanted to set my expectations.
The sensor seems to be the same one as in the GH5s, and for video I think it will be limited to 12-bit mode. Luckily, dpreveiw shows both 14-bit and 12-bit results on their underexposure comparison tool.
I compare that sensor with the two cameras that I own right now: RX100 IV and a6000. I think the "+3EV" shot would be the limit for me on the GH5s in 12-bit mode. In 14-bit mode it has maybe half a stop more range, but I think the sensor can't shoot 14-bit at video frame rates, only 12-bit. Looking at those results, the RX100 IV has about the same DR as the GH5s 12-bit, and the a6000 has about half a stop more.
That's a very good result: I have traveled shooting stills with the a6000 and RX100 IV and video with the RX100 IV and a7S, and the amount of headroom I get on those stills is waaaaaay bigger than on the video files (for which I always use slog2).
In video mode, the RX100 IV has about half a stop less DR than the original a7S, and I think the a6500 and a7 III are half a stop above the original a7S, so I could get an extra stop just by moving to a more recent Sony.
The a6000 in stills mode has between one and 2.5 stops better DR than the RX100 IV in video mode (depending on whether you just care about big shapes, or also textures: RAW makes a huge difference if you care for textures because compression usually destroys textures in the shadows).
Take away the half-a-stop difference that the a6000 seems to have over the GH5s in 12-bit mode, and here's my expectation:
I think the pocket 4k shooting RAW will have half a stop less DR than the a6500/a7III shooting slog3 if you test them with a chart (where only big shapes matter and you don't have to worry about 8-bit banding), but one stop more if you care about textures in the shadows. That's no BM4.6k, but I think I'll be a happy p4k user.
So, some thoughts on what is always the main issue for me: dynamic range. We'll know for sure when people get their hands on the camera and run real-world tests, but I just wanted to set my expectations.
The sensor seems to be the same one as in the GH5s, and for video I think it will be limited to 12-bit mode. Luckily, dpreveiw shows both 14-bit and 12-bit results on their underexposure comparison tool.
I compare that sensor with the two cameras that I own right now: RX100 IV and a6000. I think the "+3EV" shot would be the limit for me on the GH5s in 12-bit mode. In 14-bit mode it has maybe half a stop more range, but I think the sensor can't shoot 14-bit at video frame rates, only 12-bit. Looking at those results, the RX100 IV has about the same DR as the GH5s 12-bit, and the a6000 has about half a stop more.
That's a very good result: I have traveled shooting stills with the a6000 and RX100 IV and video with the RX100 IV and a7S, and the amount of headroom I get on those stills is waaaaaay bigger than on the video files (for which I always use slog2).
In video mode, the RX100 IV has about half a stop less DR than the original a7S, and I think the a6500 and a7 III are half a stop above the original a7S, so I could get an extra stop just by moving to a more recent Sony.
The a6000 in stills mode has between one and 2.5 stops better DR than the RX100 IV in video mode (depending on whether you just care about big shapes, or also textures: RAW makes a huge difference if you care for textures because compression usually destroys textures in the shadows).
Take away the half-a-stop difference that the a6000 seems to have over the GH5s in 12-bit mode, and here's my expectation:
I think the pocket 4k shooting RAW will have half a stop less DR than the a6500/a7III shooting slog3 if you test them with a chart (where only big shapes matter and you don't have to worry about 8-bit banding), but one stop more if you care about textures in the shadows. That's no BM4.6k, but I think I'll be a happy p4k user.
Last edited:
Akos Baranya
Active member
I agree, dynamic range is the main issue for me as well and without real world tests, we are blind. For me the bar is the previous pocket, if the new pocket has a lower dynamic range than the previous one, what is the point?
PaPa
Major Contributor
For me, the only reason I got rid of my bmpcc was because of the horrendous moire and aliasing on certain patterns/fabrics. But I loved the dynamic range and colour science. The colours were just sooo cinematic. (whatever that means). Things just looked right, skin tones, etc.... All I would want is a clean 4k version of the pocket and I'm sold. All other things being equal. Especially at that price point.
cyvideo
Veteran
Like many I'm curious about the new BMPCC 4K. To that end I downloaded some original camera clips that were made available. Overall quite impressed. Not as quiet as I thought it would be but then again I must remember it is a 4/3 sensor. Something I did see in one of the clips though did raise a few question. Wasn't looking for anything in particular but when playing the "Close up at night" clip I'm seeing, and on a 4k monitor it's hard too miss, vertical lines running through certain areas. especially in the out of focus lights, the 'bokeh' behind the model. The lines seem to follow the horizontal camera movement. Question? what is this? There is obviously some dirt and dust on the sensor or in the lens that was used on this clip but the vertical lines appear to be following the movement of the camera. Are they coming off the sensor? I have no idea. I've looked at this clip now on two different systems on two different pieces of software on two different monitors so it is a clip based issue. The clip can be downloaded, along with others from:
https://www.blackmagicdesign.com/products/blackmagicpocketcinemacamera/gallery
The following images are not magnified. They are just various crops from the 4K screen grabs. Anyone venture to think what these lines may be?
Chris Young
https://www.blackmagicdesign.com/products/blackmagicpocketcinemacamera/gallery
The following images are not magnified. They are just various crops from the 4K screen grabs. Anyone venture to think what these lines may be?
Chris Young
AndreeOnline
Veteran
Water/dirt/residue on the windows behind the actress.
And they aren't following the movement of the camera, they are following the movement of the window+frame just behind the model.
And they aren't following the movement of the camera, they are following the movement of the window+frame just behind the model.
Last edited:
morgan_moore
Major Contributor
Dirt/fluff on rear element - as for the line I’d say that is the sensor doing bad things on a very thin looking file?
morgan_moore
Major Contributor
Thin in terms of exposure?