Blackmagic Pocket Cinema Camera 4K

[BTM_Pix]

1 hour ago, drm said:

By the way, is the communication two-way? I mean, if I select a focus point using the touch screen, are you able to use that information with your focus system to maintain focus?

In this iteration, no.

But it forms part of a more comprehensive focus controller that Andrew will be doing a piece about that will give more context to the reasoning.

6 minutes ago, drm said:

I suspect that he could quickly modify this solution to send the commands to a follow focus motor instead of sending the focus commands to the camera.

Indeed.

[independent]

So far this looks like more of a wireless focus system, without need for a external focus unit. Kudos! 

However, for continuous autofocus, I seemed to have missed it. This looks like a straightforward rack focus between two set points. 

If you’ve in fact developed continuous autofocus, is it happening within the center frame or is tracking a particular focus point (presumably through touchscreen)?

Great work either way.

 

[BTM_Pix]

1 hour ago, independent said:

So far this looks like more of a wireless focus system, without need for a external focus unit. Kudos! 

However, for continuous autofocus, I seemed to have missed it. This looks like a straightforward rack focus between two set points. 

If you’ve in fact developed continuous autofocus, is it happening within the center frame or is tracking a particular focus point (presumably through touchscreen)?

Great work either way.

 

It is continuous but we were just moving between two targets.

If the man had come over to steal the plant it would have tracked him and ditto if the plant had come to eat us.

It tracks nearest point but is driveable.

Take what it is that you see it doing now as a jumping off point for what is possible as more info and nuance will be released over the coming months.

I'll keep it out of this thread now and in the main one so normal Pocket4K discussions can be resumed.

[drm]

Just now, BTM_Pix said:

It is continuous but we were just moving between two targets.

If the man had come over to steal the plant it would have tracked him and ditto if the plant had come to eat us.

It tracks nearest point but is driveable.

Take what it is that you see it doing now as a jumping off point for what is possible as more info and nuance will be released over the coming months.

I'll keep it out of this thread now and in the main one so normal Pocket4K discussions can be resumed.

LOL...one of the main P4K complaints is related to its focus ability (or lack thereof), hence the interest in your type of solution . Although, this could be said of many of the cinema cameras.

To be honest, I am only somewhat satisfied with the push to focus on the P4K. It works some of the time. I use manual focus with focus peaking most of the time but I do use the push to focus on shoots sometimes to get initial focus. Unfortunately, the P4K push to focus also flakes out some of the time. For example, during my last shoot I noticed several times during the day when I used the one touch AF on the P4K that the camera would get confused, then focus in and out repeatedly, then ultimately fail to focus (with the Olympus 12-100 f/4). (it was a run & gun type shoot)

Even though my GH5 & GH5s cameras have continuous autofocus, I never use it. I set them to one touch AF, just like the P4K. I can't stand to see the pulsing while the camera is trying to focus. If I see that Panasonic focus pulsing in one of my shots, I would consider that shot ruined and not use it.

I would love to have a reliable *non-pulsing* continuous autofocus system that I could turn off when I didn't want it.

[BTM_Pix]

This can be set to be the equivalent of back button focus so it can be momentary and only active while you have the controller wheel pressed in.

When released, it detaches and you can use the manual focus wheel or just leave it parked until you engage it again etc.

[Kisaha]

@drm i am planning to use the P4K with the 12-100 Olympus for r n g situations. Can you comment on the combo?

Does the lens's IS help a bit? 

I see you have issues with the AF..

[drm]

4 hours ago, Kisaha said:

@drm i am planning to use the P4K with the 12-100 Olympus for r n g situations. Can you comment on the combo?

Does the lens's IS help a bit? 

I see you have issues with the AF..

Kisaha, I have a lot of m4/3 lenses (maybe 20? and Canon) but use Olympus 12-100 lens a lot. It is great because it is a constant f/4 and has a great range. The lens is also very sharp and well built. The lens IS *does* help and works well.  

I also use the Panasonic 12-35 & 35-100 f/2.8 lenses regularly. I find that I use the Olympus 12-100 lens in run/gun situations more than the Sigma 18-35 because of the better zoom range on the Olympus (plus you get IS).

After reflecting on the issue, I think that the focus issues with the Olympus lens that day were mostly due to the environment in which I was shooting. It was very bright outside and I was running handheld (cage, side handle, SmallHD monitor on top).  If you hold the camera still when focusing, the one touch focus works well. If you move the camera while focusing, you can cause the camera to freak out. I suspect that I didn't do a very good job holding the camera still that day   In the studio and most other times, the one touch focus has worked very well. I can't recall ever having that many focus issues before on a shoot, so, I chalk my problems that day up to user error. It didn't help that I couldn't see the screen very well in direct sunlight.

[thephoenix]

 

[Novim]

A very interesting test of Rick Lang on BM forum (here) : a comparison between Q0 and Q5. One his conclusions: "Q0 versus Q5? Of course Q0 must be better because it’s a lot higher data rate. The design goal of the ‘Q' is a quality image and how could Q5 match Q0 with so much less data? In a word: it’s Blackmagic. Sometimes some elements in either Q5 or Q0 are ‘better’ than the other; it’s not a race in which one runner is in the lead throughout the race. Q5 makes some sacrifices that you can see, but it seems to have some less understood running style and I have to admit in this video, it emerges victorious, not only in fidelity but also edge sharpness."

 

[drm]

3 hours ago, Novim said:

A very interesting test of Rick Lang on BM forum (here) : a comparison between Q0 and Q5. One his conclusions: "Q0 versus Q5? Of course Q0 must be better because it’s a lot higher data rate. The design goal of the ‘Q' is a quality image and how could Q5 match Q0 with so much less data? In a word: it’s Blackmagic. Sometimes some elements in either Q5 or Q0 are ‘better’ than the other; it’s not a race in which one runner is in the lead throughout the race. Q5 makes some sacrifices that you can see, but it seems to have some less understood running style and I have to admit in this video, it emerges victorious, not only in fidelity but also edge sharpness."

 

I recently did a two-day commercial shoot using Q5 exclusively. There were extreme variations in locations and lighting, but I am very pleased with the results. I haven't seen a single thing where I felt that I needed to use Q0. I am personally surprised that Q5 is as good as it appears.

[MeanRevert]

Personally, I think there should be something like Q3 since I've found the low-end of Q5 is too low on some static shots.  It should also help the noise on the shadows thing.

[Jim Giberti]

Have used Q5 exclusively on a dozen or so commercial projects to date. Did extensive testing of all flavors of Braw. Q5 is amazing and over 40 minutes on a simple Sandisk SD card.

[drm]

Consider my mind blown... I will have to do some testing with my cameras now.

Native ISO is really 400 and 4000, and it clips at lower levels as the ISO drops below the default ISOs. ISO 100 = 83% clipping point.

https://www.youtube.com/embed/hV7ixU2AmLE

Gerald Undone: Exposure Tips for the BMPCC4K & Why I Don’t Use ProRes

Note: How to I embed a YouTube video?

[kye]

Just paste the URL into the message window, hit enter/return and after a few seconds it should expand out to embed the video  

[drm]

Gerald Undone: Exposure Tips for the BMPCC4K & Why I Don’t Use ProRes

So, according to Gerald, there are several things with the P4K that I didn't know about that are very important:

1. Exposure with ProRes on the P4K can be problematic. To fix this you should use RAW, not ProRes, or be *very* careful when setting your exposure.

2. The maximum clipping point *changes* as you change the ISO, if you reduce your ISO from the "native" ISO levels (400, ~3200). At ISO 100, it clips at 83%! I am familiar with the chart showing the steps above and below middle gray, but I didn't realize they were just clipping the top at lower ISOs.

3. Highlight recovery seems to work differently than I had thought and should apparently not be used much of the time. Highlight recovery appears to cause problems with the image when you use it at ISOs >= native ISOs. I had heard that the highlight recovery used the data from other channels when a color had clipped to rebuild that clipped channel.

If you have a P4K you should watch this video! I am really curious to hear other's thoughts on this material. 

 

Thanks @kye for telling me how to embed the video!  

[MeanRevert]

Weren't you guys bagging on this dude's earlier P4K video where he blew out the highlights? 

[drm]

1 hour ago, MeanRevert said:

Weren't you guys bagging on this dude's earlier P4K video where he blew out the highlights? 

I view Gerald Undone as more of a techy geek vlogger that talks about camera stuff than a filmmaker. But, I think that you are right that people were commenting about the visuals in one of his videos a few weeks back. 

The photo side of my brain is used to getting the ISO as low as possible to reduce noise. I shoot exclusively BRAW on the P4K, so this doesn't apply to me, but I would not have thought to avoid going below ISO 400 (or 3200) to avoid nuking the highlights. This means that NDs are pretty much necessary. 

UPDATE: Captain Hook added some clarifications to Gerald's YouTube video post, so make sure you read Captain Hook's comments as well.  Some of the clipping may be a Premier issue instead of clipping in ProRes. Hook says that Highlight Recovery (Highlight Rebuild) works as I thought. So, much of the earlier video may be a false alarm 

[kye]

7 hours ago, MeanRevert said:

Weren't you guys bagging on this dude's earlier P4K video where he blew out the highlights? 

Maybe that's what prompted the discussion about exposure..  We all make mistakes, it's about learning from them that is what is important  

[JeremyDulac]

Highlight recovery needs to be used at 400 iso still when ettr'ing, it seems super whites get clamped. Not sure if maybe Gerald's understanding is off, but you definitely need to use it at 400 iso in my testing.

 

These images are original, then lowered exposure and then lowered exposure with highlight recovery on.

[drm]

10 hours ago, JeremyDulac said:

Highlight recovery needs to be used at 400 iso still when ettr'ing, it seems super whites get clamped. Not sure if maybe Gerald's understanding is off, but you definitely need to use it at 400 iso in my testing.

 

These images are original, then lowered exposure and then lowered exposure with highlight recovery on.

Thanks for sharing. I have had blown out skies like that recovered when using Highlight Recovery as well. Captain Hook said that Highlight Recovery should really have been called "Highlight Rebuild" because it rebuilds the highlights from the other channels when one of the channels is clipped. He also said that the recovered area is only black/white because they don't know what color to make the recovered areas. So really, we should avoid using Highlight Recovery if we are interested in getting the most accurate image. In your footage, I would much rather have some detail in the sky rather than having it blown out.

At ISO 400, at least the clipping point is a full 100%. At ISOs below 400, the clipping point is lower. As you increase the ISO up to 1000, the area above middle gray increases. So, your sky shot with ISO 1000 should have more detail than the sky shot at ISO 400. I don't think that I have a sky shot handy. Have you tried 400 vs. 1000 on this scene to see if there are noticeable differences?