Blackmagic Pocket Cinema Camera 4K

[Kisaha]

5 minutes ago, leslie said:

i guess if you can afford one of those lenses you can afford to have the sensor stack fixed as well, if its an issue for you . Although i dont think its the manufacturers fault for not making exotic, expensive, fast, wide, adapted glass work at 100 % on their camera. I swing the other way, cheap, old glass adapted ?

(Cheap) old glass is the most affected it seems, because they didn't have any sensor stacking back then!

I am sure there will be some tests soon. Can't wait for the results.

[leslie]

Just now, Kisaha said:

(Cheap) old glass is the most affected it seems, because they didn't have any sensor stacking back then!

shouldn't the adapter take that into account ? its not like the sensor stack moves around very much

edit : my understanding is flawed i guess, sorry if i antagonize you. I don't know how bm could get it wrong with the sensor on the  first pocket camera but i remember vaguely that there was some issue back then. i presume that most manufactures build mounts so that they sell more of their own lenses rather than have those dollars go else where. which is a bit of an oddity for bm as they dont make lenses do they ? perhaps we should consider ourselves lucky if we can get adapted lenses to work at a reasonable   level of performance

[Kisaha]

18 minutes ago, leslie said:

shouldn't the adapter take that into account ? its not like the sensor stack moves around very much

That's exactly the point Metabones does with this adapter. Most sensors are close to 2-2.5mm, the m43 are 4-4.5mm so they need special construction.

BM is m43 but unlike the other m43 sensors, their sensors are close to 2.5mm. That changes..well, something.

"1. The difference in sensor-stack thickness between the camera the lens was designed for and the camera actually being used.

2. The maximum aperture of the lens. Wide-aperture lenses are going to be more sensitive than narrow aperture lenses.

3. How telecentric the lens is. (More specifically, how far forward the exit pupil of the lens is.) A lens with the exit pupil far away from the sensor is not affected by the thickness of the sensor stack very much. A lens with the exit pupil very close to the sensor is affected a lot."

So, a combination of those 3 can yield specific undesired results.

The worst case scenario could be an ultra fast wide angle rangefinder lens.

As I said, it can be nothing in practice, in theory definitely is something, so we need an independent test to learn what we are missing, or not, in real life.

For those who want the best performance out of their cameras, these adapters are going to sell, no question about it.

[Yannick Willox]

What does this actually mean for the new PL 10-25 mm native m43 lens ? Does it also suffer, as it has been designed for thicker sensor stack ?

[BTM_Pix]

Finally got around to buying and testing these mini-XLR adapters.

They are the official Blackmagic ones that they sell for the VideoAssist but obviously work fine with the Pocket4K as well.

Decent build quality, very practical length of 40cm and they come in a pack of 2 for £30 so a good value as well.

More to the point they are actually available, as mini-XLR to XLR is generally still a tricky cable to track down.

https://cvp.com/product/blackmagic_bmd-hyperdaxlrmini2

[Kisaha]

24 minutes ago, BTM_Pix said:

Finally got around to buying and testing these mini-XLR adapters.

https://cvp.com/product/blackmagic_bmd-hyperdaxlrmini2

+1

These are the ones I have spoted myself. I have a pair for my Zoom F8, but I want to have backupa anyway.

How do they sound? I remember people complaining for weak signal and/or hiss and stuff. 

Have you tried a phantomed power mic or any kind of mic?

[drm]

6 hours ago, webrunner5 said:

Kind of a bit outdated article, but I think you get the point. Should help with wide angle lenses, which is a pretty big fad in video now. Doesn't look like a earthshaking difference though. But every bit helps.

https://www.lensrentals.com/blog/2015/01/a-thinner-sensor-stack/

Thanks @webrunner5 for this article. This was fascinating reading.

So, here are my big picture conclusions:

1. This is only an issue at really wide apertures (1.4+)
2. The sensor stack on the P4K is 1.5mm thinner than normal M4/3, so if you are using an adapter without optics, the P4K should perform better than other M4/3 cameras because its sensor stack is close to the same thickness as a normal SLR. 
3. The Sigma 18-50 f/1.8 lens is pretty much immune from the problem because it has an exit pupil distance of 150mm. 
4. MOST importantly (and unexpectedly) the P4K may show problems with *native* M4/3 lenses at very wide apertures because those lenses were designed to use a 4mm filter stack but the P4K uses one that is 2.5mm.

Luckily, I have a Voigtlander 25mm f/0.95 lens and P4Ks + GH5/GH5s cameras. When I get time, I will compare the P4K vs. the GH5s wide open and look for differences. If there are differences, it will probably be on the edges of the images. I would bet that the differences would be most visible in photos, but obviously the P4K doesn't do very good photos...  

So, Blackmagic designed in a difference (filter stack = 2.5mm instead of 4mm) that would likely make the P4K perform better with adapted lenses, but also likely caused it to perform worse with native M4/3 lenses...weird. 

[webrunner5]

4 hours ago, Yannick Willox said:

What does this actually mean for the new PL 10-25 mm native m43 lens ? Does it also suffer, as it has been designed for thicker sensor stack ?

I would think on the wide end it will be distorted at the edges without the new Metabones on it, Especially wide open, which sad to say is the main reason you would buy it, for the speed. I would imagine BMD does this to get a bit sharper image, which is probably happening in the middle on a "normal lens", but really wide angle lenses take a hit on the edges. I don't think it effects telephoto lenses at all or much at all.

The steeper the angle of light hitting the sensor the worse it is going to be. If you look at old Manual Lenses the wide angle ones tend to have the rear glass nearly at the lens mount. The mid Teles and up the glass is way up in the barrel. That means the closer the glass is to the rear mount the more radical the light bend is.

Now on my Sony FE FF lenses even the wide angle ones were recessed a fair bit in and were more of a rectangle than a round exit on the lens. I think Sony probably has some of the best designed lenses for Mirrorless cameras out. I am not saying the RF, or Z mounts are bad, but I have never used or even seen a lens they use. But I bet the older EF, Nikon F lenses have the same old designs for the old DSLR bodies, even the newer versions of them.

[BTM_Pix]

48 minutes ago, Kisaha said:

How do they sound? I remember people complaining for weak signal and/or hiss and stuff. 

Have you tried a phantomed power mic or any kind of mic?

Well its difficult to say anything definitive about noise and signal loss really as I can't A/B them against anything as these are the only ones I have and the Pocket4K is the only piece of equipment I have that has mini XLR inputs.

Yes, have tried with phantom powered and dynamic mic.

[docmoore]

I use the BM mini to XLR cable with phantom powered boom mic ...

Signal is very strong ... if you adjust levels without hiss and no appreciable noise.

Adequate for the weekly blog that I record ... I no longer use an external recorder.

The camera continues to address a wide variety of needs ... it is very well designed and implemented.

[graphicnatured]

18 hours ago, webrunner5 said:

Sounds to me you could use the old one on the new one if there is that little difference? Maybe it protrudes further into the sensor box though?

I have one, but am afraid to try it.

6 hours ago, docmoore said:

I use the BM mini to XLR cable with phantom powered boom mic ...

Signal is very strong ... if you adjust levels without hiss and no appreciable noise.

Adequate for the weekly blog that I record ... I no longer use an external recorder.

The camera continues to address a wide variety of needs ... it is very well designed and implemented.

I've been happy with the audio through XLR on the P4k as well.

[majoraxis]

On 8/1/2019 at 5:54 AM, drm said:

The Sigma 18-50 f/1.8 lens is pretty much immune from the problem because it has an exit pupil distance of 150mm

That is great information.  Here's link that contains other lens exit pupil distances. It also mentions the Blackmagic 2.5k as having an optical sensor stack of 2.4mm so assuming the Blackmagic Pocket 4k's sensor stack is 2.5mm, an older used metabones speed booster made for the BMCC maybe a cheap alternative with caveats like vignetting and corner sharpness to be considered.

https://www.lensrentals.com/blog/2014/06/sensor-stack-thickness-when-does-it-matter/comment-page-1/

[Kisaha]

50 minutes ago, majoraxis said:

That is great information.  Here's link that contains other lens exit pupil distances. It also mentions the Blackmagic 2.5k as having an optical sensor stack of 2.4mm so assuming the Blackmagic Pocket 4k's sensor stack is 2.5mm, an older used metabones speed booster made for the BMCC maybe a cheap alternative with caveats like vignetting and corner sharpness to be considered.

https://www.lensrentals.com/blog/2014/06/sensor-stack-thickness-when-does-it-matter/comment-page-1/

At 35mm is 70 FYI.

[zerocool22]

Quick question: does the pocket 4K battery charge, when the camera is off and is plugged into a wall socket? 

(as I read in the manual you can charge it over USB when the camera is off, but does not seem to mention through wall socket)

[BTM_Pix]

6 hours ago, zerocool22 said:

Quick question: does the pocket 4K battery charge, when the camera is off and is plugged into a wall socket? 

(as I read in the manual you can charge it over USB when the camera is off, but does not seem to mention through wall socket)

Yes

[nathlas]

 

[drm]

10 minutes ago, nathlas said:

 

Thanks for sharing! Given the difference in the sensor stack on the P4K, I am surprised that there isn't more of a difference. I would have thought removing 1.5mm of glass from the stack would have made more of a difference.

[leslie]

9 hours ago, drm said:

Thanks for sharing! Given the difference in the sensor stack on the P4K, I am surprised that there isn't more of a difference. I would have thought removing 1.5mm of glass from the stack would have made more of a difference.

i grabbed this from comments under the video

Optical comparison between Metabones' new speed booster designed specifically for BMPCC4K and general micro 4/3 cameras. XL 0.64 versions. In the last 20 seconds of this video I demonstrate that the focal plane of the new speed booster seems to have a larger angle of concave curvature than the m4/3 speed booster. The central focal point of the new booster is farther from camera than the edge focal points compared to the m4/3 booster. Physically the two boosters are near identical with the exception of the rear element of the BMPCC4K booster protruding farther (hence its incompatibility with most other m4/3 mount cameras).

 i wonder if it protrudes more to compensate for the reportedly thinner sensor stack the p4k has ?

Personally i'm not seeing a huge amount of difference.There is some difference but not the huge jump i would have expected or perhaps was hoping for. I have to  save my coin  for abit regardless so it will be interesting to see what other peoples conclusions are but i'm still thinking a second hand 0.64 is is better value for me anyway at this point.

[Kisaha]

11 hours ago, drm said:

Thanks for sharing! Given the difference in the sensor stack on the P4K, I am surprised that there isn't more of a difference. I would have thought removing 1.5mm of glass from the stack would have made more of a difference.

The one Sigma 35mm I found, is the F1.4 DG HSM and infinity is 78.8 and close distance the exit pupil is 81.9.

The Milvus 35mm is 66 for comparison, so the difference can be more substantial here.

It is a lens to lens difference, if you have an AF-S Nikkor 70-200mm f/2.8E FL ED VR @ all, where the exit pupil is 146.3 then you do not care much, if at all, but if you have wider primes, or Zeis, or legacy film lenses that can go as low as 50, then maybe you care.

There was a talk of weird examples on the internet about P4K and Metabones, we saw a video above that the film maker stated that Metabones don't match well the P4K, depending the eye of the creator and viewers, but also the lens used, there were complaints, or not.

We spend thousands of ?for the best lens/camera combo, and then you put a 20€ vND in front of your lens? Some do.

Everyone has various limits, some want the best of the best for their P4K, some are fine with lesser - and much cheaper, solutions.

Now, if you are buying a Metabones and you own just the P4K, then you shop the special adapter, the issue is for people owning multiple m43 cameras.

I will be waiting for a few more tests and I would like to see the difference in performance with native lenses also.

[drm]

@Kisaha The difference in sensor stack size between the P4K and other M4/3 cameras should be most visible with native lenses at very wide apertures. I did some quick testing a few days back with my Voigtlander 25mm f/0.95 lens. I didn't notice a drop in image quality between my GH5s and my P4K, even using that lens wide open. At some point I am going to try and set up a test scene and look at this issue in more detail. There may be differences with that lens between the two platforms, but my initial look didn't reveal anything that stood out.

3 hours ago, leslie said:

i wonder if it protrudes more to compensate for the reportedly thinner sensor stack the p4k has ?

Yes. I am sure that is the cause. I also think that Kisaha is correct. That particular lens in that video didn't show a big difference because the exit pupil is about 80. A shorter exit pupil will likely show different results.

You just *know* that in the future you are going to hear horror stories about someone putting the new Metabones adapter on a normal M4/3 camera and messing up or scratching the sensor stack.