Blackmagic Forum

I hope they do it, I would like to be able to configure this camera as a shoulder mount, and since there is no lanc, maybe some third party might cobble together a lanc to usb adapter. That seems a much more robust solution then a bluetooth to lanc adapter. Especially if you had multiple cameras but only one adapter. The usb-c would be plug and play, whereas you would have to link to the camera.

Jim Giberti wrote:Einstein was very succinct.

I like that.

Well here's succinct, I'm not interested in debating the moral ethics of it anymore. Both sides need to examine their moral ethics. Active hearers are active listeners.

Yes, USB as a controller (even wifi)? But will a SSD for Raw recording on a hub with a controller interface, work seamlessly? Useful to have two USB then?

I'm very interested in syncing multiple cameras together for 3D? Probably more a future upgraded micro question though.

The new Sony A7III claims 15 stops of DR in a small mirrorless body. With that and Sony's amazing low light sensitivity, I feel like if they had better color science and included prores/raw codecs in a blackmagic pocket form factor body, they could build a much better camera than the Pocket 4K. DR claims are always exaggerated, but Daniel Peters (an Ursa Mini Pro shooter) said that it's almost on-par with his UMP which is very impressive for not being a dual-gain sensor.

Wayne Steven wrote:Yes, USB as a controller (even wifi)? But will a SSD for Raw recording on a hub with a controller interface, work seamlessly? Useful to have two USB then?

I'm very interested in syncing multiple cameras together for 3D? Probably more a future upgraded micro question though.

If the usb-c was taken by the hand grip controller it would still be possible to record to SSD through the CFast connector as is often done on the UMP.

Savannah Miller wrote:The new Sony A7III claims 15 stops of DR in a small mirrorless body. With that and Sony's amazing low light sensitivity, I feel like if they had better color science and included prores/raw codecs in a blackmagic pocket form factor body, they could build a much better camera than the Pocket 4K. DR claims are always exaggerated, but Daniel Peters (an Ursa Mini Pro shooter) said that it's almost on-par with his UMP which is very impressive for not being a dual-gain sensor.

I own a Sony A7rII. I love these cameras for run and gun filmmaking but there's a few caveats to the dynamic range you're mentioning. Remember that
1) you're using Slog2 (3 is absolutely not usable), which is a 10 bit curve which gets mapped on a 8 bit signal. That means color data is lost in the end product. This is is most obvious when filming people, as their faces get very plasticky, and overall detail is lost. This means that for the best results you'll be filming in less flat profiles that'll lose you 2-3 stops of dynamic range, putting you right around the Blackmagic.
and
2) The high signal to nose ratio of 8 bit means you'll probably not really get Sony's claimed 15 stops of range (A7SII had 14.5 claimed but most tests put it around 12.5) as shadow detail is quickly lost. This can be partly corrected by over exposing so as to increase shadow detail but in high contrast situations this will clip the highlights as well, so you'll get decreased noise but little to no dynamic range advantage.

That's what I'm saying. If Sony were 10 bit it would be so much better than anything that's out there.

Can the usb c be used to charge the battery which can extend the shooting time? interested to know how much it can extend the shooting time?

They stated the camera can be charged via usb-c, but only while the camera is off unfortunately.

Sean van Berlo wrote:They stated the camera can be charged via usb-c, but only while the camera is off unfortunately.

What's the source of that statement?

Savannah Miller wrote:The new Sony A7III claims 15 stops of DR in a small mirrorless body. With that and Sony's amazing low light sensitivity,

Hope you know, that the only reason for its 'amazing low light sensitivity' is the heavy internal noise reduction? Because if you record the signal onto a Shogun Inferno it is noisy as hell.

Kim Janson wrote:That raise the question, is there any advantage of doing the NR in camera, other than no processor power needed for NR during editing phase.

The advantage is, that you can get a good quality compression from that. If they wouldn't denoise then that would result in ugly compression artifacts.
For example: my Canon XF305 has lots of picture controls and you can adjust in-camera noise reduction as you like it - even turn it off. At +6dB the camera has significant visible noise, much too much for its 50MBit/s 4:2:2 MPEG-2 compression. It looks like a blocky mess. So you have to crank up the noise reduction to get a usable recording.

Some denoise techniques also work in the sensor pad circuites themselves, to do a hardware comparison, during sensing.

Temporal noise comparison and image restoration has been my preferred choice for over 20 years. If this is done right, very high quality should be possible with much better compression results, as Robert said.

But remember the stills are just consumer cameras as far as video. They may want you to spend on higher end equipment to get better quality. But BM could do it right here on their cameras.

Sean van Berlo wrote:

Savannah Miller wrote:The new Sony A7III claims 15 stops of DR in a small mirrorless body. With that and Sony's amazing low light sensitivity, I feel like if they had better color science and included prores/raw codecs in a blackmagic pocket form factor body, they could build a much better camera than the Pocket 4K. DR claims are always exaggerated, but Daniel Peters (an Ursa Mini Pro shooter) said that it's almost on-par with his UMP which is very impressive for not being a dual-gain sensor.

I own a Sony A7rII. I love these cameras for run and gun filmmaking but there's a few caveats to the dynamic range you're mentioning. Remember that
1) you're using Slog2 (3 is absolutely not usable), which is a 10 bit curve which gets mapped on a 8 bit signal. That means color data is lost in the end product. This is is most obvious when filming people, as their faces get very plasticky, and overall detail is lost. This means that for the best results you'll be filming in less flat profiles that'll lose you 2-3 stops of dynamic range, putting you right around the Blackmagic.
and
2) The high signal to nose ratio of 8 bit means you'll probably not really get Sony's claimed 15 stops of range (A7SII had 14.5 claimed but most tests put it around 12.5) as shadow detail is quickly lost. This can be partly corrected by over exposing so as to increase shadow detail but in high contrast situations this will clip the highlights as well, so you'll get decreased noise but little to no dynamic range advantage.

3) It has a very bad rolling shutter which is I think the weakest point of the A7

In math and physics, I think the best solution is referred to as “elegant” and that usually implies a beautiful simplicity. The loveliest example I can think of is the expression of the energy (primarily of matter) equals its mass times the square of the speed of light. We all understand the meaning of that equation, but very few could ever have derived that simple expression without previously knowing Einstein’s elegant solution.


Sent from my iPhone using Tapatalk

rick.lang wrote:In math and physics, I think the best solution is referred to as “elegant” and that usually implies a beautiful simplicity. The loveliest example I can think of is the expression of the energy (primarily of matter) equals its mass times the square of the speed of light. We all understand the meaning of that equation, but very few could ever have derived that simple expression without previously knowing Einstein’s elegant solution.


Sent from my iPhone using Tapatalk

HUH?

MarcusWolschon wrote:

Sean van Berlo wrote:They stated the camera can be charged via usb-c, but only while the camera is off unfortunately.

What's the source of that statement?

Interview with Grant at NAB where he said you could charge the camera while it's in your bag. I'm inferring from that, to be fair, that it doesn't charge while recording. The fact that it charges at all via USB-C is on Blackmagic's product page.

Leon Benzakein wrote:

rick.lang wrote:In math and physics, I think the best solution is referred to as “elegant” and that usually implies a beautiful simplicity. The loveliest example I can think of is the expression of the energy (primarily of matter) equals its mass times the square of the speed of light. We all understand the meaning of that equation, but very few could ever have derived that simple expression without previously knowing Einstein’s elegant solution.


Sent from my iPhone using Tapatalk

HUH?

Okay, I shouldn’t have said “we all understand.” My bad.




Sent from my iPhone using Tapatalk

Kim Janson wrote:It explains the relation of mass and energy, it is one thing to use the formula, other to understand it. I do not think anyone understands mass, limited speed of light, gravity, magnetism, time, ...existence. Sure there is formulas explaining how they are related to each other, how a photon hits the sensor and release energy etc.

Existence though, the biggest wonder of all.

HUH?

Kim Janson wrote:Existence though, the biggest wonder of all.

Sent from my iPhone using Tapatalk

Now there is something I can hang my hat on.

"I think therefore I am."

Which begs the question, does the BMPCC4K really exist or does it not? It cannot think or can it?

We saw it, some of us touched it but where is it now.(JULY?)

Leon Benzakein wrote:

rick.lang wrote:In math and physics, I think the best solution is referred to as “elegant” and that usually implies a beautiful simplicity. The loveliest example I can think of is the expression of the energy (primarily of matter) equals its mass times the square of the speed of light. We all understand the meaning of that equation, but very few could ever have derived that simple expression without previously knowing Einstein’s elegant solution.


Sent from my iPhone using Tapatalk

HUH?

Lol. You walked into that Rick. Kim is right. It appears to be succinct because you know the meaning already. But that is the issue, people interpreting and misinterpreting things from their own understadandings without putting in the hours to understand everything little thing. The equation is actually very abstract and complicated. Most non professionals and those not trained in its meaning would know what it means I mean where are the words light or even energy. You are starting to think about it. Yes, people usually don't even know that e is energy. See, people arrogantly think because they can understand, or not, something, there is nothing wrong with there understanding. But it is hard work to train yourself to understand.

Now, what often do, is succinctly describe a series of things. This means a string of succinct explanations strung together, or to save spsce like the upstart crow, woven together according to their relationships into single sentences (much like the equation actually does).

Now, ask the quantum physics that within years displaced Einstein/St. Augustine, what they thought of his objection to quantum physics. They could probably call it "simplistic".

Now, take a look at this simple one: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4129395/ (Im out and it takes too long to hunt down one with the full text not behind a pay wall ) and tell me how many people can understand a lot of papers more complex than this, and then tell me that just because it isn't easy to understand, the researchers don't know their stuff and that they don't know it better than most other people? There are those that write and those that read what others write. But if they can't read properly in the first place how can they understand properly. Unfortunately, understanding is like film making, it requires a lot of work, and unfortunately it is often just oneself doing the work.

Now, can we get back to the pocket cinema cameea please.

Wayne Steven wrote:

Leon Benzakein wrote:

rick.lang wrote:In math and physics, I think the best solution is referred to as “elegant” and that usually implies a beautiful simplicity. The loveliest example I can think of is the expression of the energy (primarily of matter) equals its mass times the square of the speed of light. We all understand the meaning of that equation, but very few could ever have derived that simple expression without previously knowing Einstein’s elegant solution.


Sent from my iPhone using Tapatalk

HUH?

Lol. You walked into that Rick. Kim is right. It appears to be succinct because you know the meaning already. But that is the issue, people interpreting and misinterpreting things from their own understadandings without putting in the hours to understand everything little thing. The equation is actually very abstract and complicated. Most non professionals and those not trained in its meaning would know what it means I mean where are the words light or even energy. You are starting to think about it. Yes, people usually don't even know that e is energy. See, people arrogantly think because they can understand, or not, something, there is nothing wrong with there understanding. But it is hard work to train yourself to understand.

Now, what often do, is succinctly describe a series of things. This means a string of succinct explanations strung together, or to save spsce like the upstart crow, woven together according to their relationships into single sentences (much like the equation actually does).

Now, ask the quantum physics that within years displaced Einstein/St. Augustine, what they thought of his objection to quantum physics. They could probably call it "simplistic".

Now, take a look at this simple one: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4129395/ (Im out and it takes too long to hunt down one with the full text not behind a pay wall ) and tell me how many people can understand a lot of papers more complex than this, and then tell me that just because it isn't easy to understand, the researchers don't know their stuff and that they don't know it better than most other people? There are those that write and those that read what others write. But if they can't read properly in the first place how can they understand properly. Unfortunately, understanding is like film making, it requires a lot of work, and unfortunately it is often just oneself doing the work.

Now, can we get back to the pocket cinema cameea please.

HUH? X 2

The Pocket4K hopefully will be a good complement to the URSA Mini 4.6K. Both can be used for client shoots, but the Pocket4K will excel where that additional base ISO 3200 will be most useful: (lower light) interior/exterior night. In those shooting situations where the lower dynamic range is not an issue. The Mini 4.6K would be fully rigged and usually on sticks, but the relatively naked Pocket4K could be handheld with a gimbal for greater mobility or mounted to a monopod when only stability is important.

The Pocket4K

On client shoots that would benefit from having two cameras, such as the theatrical videos, as a sole operator, I could lockdown the Mini 4.6K with a 18/25mm wide angle to cover the stage and move about with the Pocket4K and a 32/50/85mm to get my medium/close shots. Wedding videos possibly could benefit in a similar arrangement some of the time.

At this point I would still plan on using my manual full frame APO PL/EF primes on both cameras.




Sent from my iPhone using Tapatalk

Kim

Do you have any footage from using this rig?

It looks like it should work really well.

What is the maximum weight that the gimbal can carry?

rick.lang wrote:The Pocket4K hopefully will be a good complement to the URSA Mini 4.6K. Both can be used for client shoots, but the Pocket4K will excel where that additional base ISO 3200 will be most useful: (lower light) interior/exterior night. In those shooting situations where the lower dynamic range is not an issue. The Mini 4.6K would be fully rigged and usually on sticks, but the relatively naked Pocket4K could be handheld with a gimbal for greater mobility or mounted to a monopod when only stability is important.

The Pocket4K

On client shoots that would benefit from having two cameras, such as the theatrical videos, as a sole operator, I could lockdown the Mini 4.6K with a 18/25mm wide angle to cover the stage and move about with the Pocket4K and a 32/50/85mm to get my medium/close shots. Wedding videos possibly could benefit in a similar arrangement some of the time.

At this point I would still plan on using my manual full frame APO PL/EF primes on both cameras.

Sent from my iPhone using Tapatalk

You are absolutely on the right track.
I have just put together a travel Manfrotto tripod that should work well with the Pocket 4K.
I will have to wait and see if the weight factor of a long lens will be an issue.
The tripod comes with a ballhead but I have replaced it with a Manfrotto befree live fluid head.
One of the legs screws off to be used as a monopod. The head can be transferred to the monopod.
I have found that if I replace the riser column with a 3/8 inch coupler, I can have the camera way down on the ground working as if on a high hat.
I am blown away by the versatility of the tripod.

Can't wait to try it with the BMPCC4K.

Very interesting... hope you post a few pics of your setup as a tripod and a monopod and as a substitute hi-hat. I should look at Manfrotto again.


Sent from my iPhone using Tapatalk

rick.lang wrote:Very interesting... hope you post a few pics of your setup as a tripod and a monopod and as a substitute hi-hat. I should look at Manfrotto again.


Sent from my iPhone using Tapatalk

Hi Rick posted pictures here

viewtopic.php?f=14&t=72832&p=405744#p405744

Much appreciated, Leon.

Did you consider the BeFree Traveler Carbon Fibre Tripod with BeFree Fluid Head?


Sent from my iPhone using Tapatalk

rick.lang wrote:Did you consider the BeFree Traveler Carbon Fibre Tripod with BeFree Fluid Head?

I ordered the Element online thinking that it was the BeFree Traveler Alumium Tripod with ballhead.
When I got it I was disappointed, but when I saw that the leg could be used as a monopod and I could get the head down low by removing the riser I realized what a find I had made.

I think that the Element is better constructed than the BeFree tripod. All parts are metal. Not so on the BeFree.

Note that this is a light tripod for light cameras and not too fancy pans and tilts, but ideal if you need support for interviews or locked off shots and gentle pans and tilts.
The tripod can take 17.6 lbs

https://www.bhphotovideo.com/c/product/ ... veler.html

The BeFree tripod head can take 8.8 lbs.

https://www.bhphotovideo.com/c/product/ ... video.html

Savannah Miller wrote:The new Sony A7III claims 15 stops of DR in a small mirrorless body. With that and Sony's amazing low light sensitivity, I feel like if they had better color science and included prores/raw codecs in a blackmagic pocket form factor body, they could build a much better camera than the Pocket 4K. DR claims are always exaggerated, but Daniel Peters (an Ursa Mini Pro shooter) said that it's almost on-par with his UMP which is very impressive for not being a dual-gain sensor.

It's pointless to think that this will ever happen.

BMD's whole camera business model is doing something that the big players will never ever do.

When they first launched the BMCC it was the first ever camera (before even Alexa) to do on-board internal uncompressed RAW and ProRes recording. In a camera that cost 3 thousand bucks.

I remember many saying that Canon would come back hard and offer the same, the hacker community was big then too. Everyone thought they would show BMD how to make a camera.

AJA were going to be the ones for a while, they gave up.

GoPro for a while would do it...and nope...

I also remember some pretty disparaging comments from RED users about this "toy" camera that did ProRes years before they did. Having used ProRes recently on RED I can't say it's very well implemented either !

And here we are...6 years later. And we've only just seen Canon sort of do something meaningful with the C200. A kind-of-sort-of-crippled RAWlite.

These guys want to segment their markets

They want you using their own codecs or their own media or both.

They want to define what you can and can't do based on how much you pay them.

BMD just don't think that way.

Panasonic are the only ones (so far) aside from Arri and BMD to do on-board ProRes recording (Varicam LT)

And RED too I guess.

JB

Well said John

There have been a lot of silly suggestions since this camera was released about what Blackmagic should have done, how it should have been a different chip size, different form factor, different battery etc... that I feel it's appropriate to make one of my own.

It seems the level of interest in this camera is unprecedented, and Blackmagic are going to sell a ton of them, and it's going to become even more of a gateway drug into the Blackmagic ecosystem. I really, really, really wish they would slap built-in NDs on (make it an EF mount if they need the flange difference) and replace the HDMI with an SDI. There, just that. Call it the Pocket Pro, and sell it alongside the Pocket, for double the price. I, and I'm sure many others here, would definitely buy that.

John, that does appear to be the way things are currently, but in tech, never say never.

Well, it sure will sell well if they don’t scare the potential customers with the first batch issues which would be classic BMD. Too good specs for the price already although, correct me if I'm wrong, it’s before VAT. Can’t see the point of adding SDI instead of HDMI to a supposedly pocketable camera. It surely was meant to compete with highly popular compact Sony and Panasonic cameras. I’m sure they want a piece of that fat pie, too, that’s they’re extending their Pocket line and are willing to sell the camera for some marginal gain (or even at a loss, given the cost of warranty repairs which, I bet, is their worst pain). But, let's be honest, in that market, low light performance and image stabilization are what people are looking after, and BMD will be able to catch up with the big players in this field no sooner than in a few decades, if ever. Look at GoPro. Innovative companies, as they are, should be very careful with every step forward. As a fan of BMD, I wish them well but I’m afraid they don’t realize what the compact camera users want nowadays. RAW, SDI (and maybe even XLRs) are meant for bigger projects/budgets. Recently, I spoke with a friend who is a wedding videographer, and he mentioned, as a matter of fact, how people become gradually less impressed with the size and complexity of a videographers’s gear (which back in the day would have justified the rate) and how they’re more and more demanding to what they want to see in their wedding movies. So, basically, what they want is more versatile shots and transitions. And because complex movements are big part of it, you absolutely need good IS an AF. So, he recently sold his UM4K and is considering a gh5 or gh5s. I’m telling this not because I’m dissatisfied with my own UM4K but to illustrate how forgiving some customer groups become to “a man with a dslr”.


Отправлено с моего iPhone используя Tapatalk

John Brawley wrote:And here we are...6 years later. And we've only just seen Canon sort of do something meaningful with the C200. A kind-of-sort-of-crippled RAWlite.

I agree with much of what you're saying, but not this piece here.

I don't think there is anything crippled with Canon raw light. 12bit, Goldilocks-Bitrate, wide gamut raw is a very nice compromise in this day and age. Current compression is around 3-5x or so? I'd welcome an additional version that was 7-10x.

But I wouldn’t for one. I’ve tested all flavours of raw and ProRes and each step does reveal slight changes from the upper limit of raw and ProRes 444 XQ. Raw 3:1 and ProRes 444 are both very close to the best, perhaps that’s my Goldilocks, but after that I’ve seen more noticeable changes. I’ve still used raw 4:1 when I needed longer recording time and no one complains, but I always want to start from the best image I can manage. Definitely wouldn’t use anything more compressed from a BMD camera than raw 4:1.




Sent from my iPhone using Tapatalk

But Prores 444 is less compressed than both Raw 3:1 and 4:1, right?

I'll ad that ProRes HQ is what we produce in day to day for broadcast and web delivery and it's robust and well beyond what's required for those mediums.

Everything else is incremental improvements, but with ProRes, you're already there for 95% of what's being done professionaly.

Steve Yedlin's tests probably haven't disabused any Red fanatics of their resolution fetish, but I wish he'd smash the "raw" idol as well. That 5% may be real some of the time, but the notion that it will make any difference in the moving pictures on what amounts to no-budget productions is maybe unduly optimistic.

Prores was good enough for last year's Best Picture Academy award, so it just might be good enough for wedding videos.

John Paines wrote:Prores was good enough for last year's Best Picture Academy award, so it just might be good enough for wedding videos.

I agree. Also, I guarantee that the customer doesn't notice. A GH5, used properly, will WOW them. I think RAW is something that has taken on a life of it's own in this community. Ever since it's made it's way to the prosumer market there has been an obsession with it. Myself included. It's more for people like us who can actually see the difference (or think we can at least). If you can't tell without a side-by-side comparison, pixel peeping, etc. it's probably ok. I had to put my bmcc raw footage next to my t2i footage and point out differences to my wife a few years ago. To me, she's the average person. They don't care, they're watching the story.

That being said... I can't wait for my Pocket 4k. I think I'm going to shoot everything in either 3:1 or 4:1.

Sean van Berlo wrote:But Prores 444 is less compressed than both Raw 3:1 and 4:1, right?

Nope- it's compressed much more than that, almost 7:1.
ProRes444 XQ is 4.5:1.

You have to understand that RAW bayer data is a monochromatic image, eg:



During debayering we convert (some pixels are interpolated as we have no real data for them) 1x monochromatic data to RGB channels. This causes debayered, final RGB data to be 3x (assuming same bit depth and full sampling) bigger than RAW. We can "loose" some of this "big" data by converting RGB to YUV, where we can sample UV (color info) every 2nd pixels horizontally (so we get 4:2:2) or on both axes, so we get 4:2:0. On top of this we apply compression- ProRes, etc. but even then we still have more data to compress than RAW. Y channel (this is as big as RAW) and at least 2 channels sampled every 2nd pixel in both axes, so this is another 0.5x of RAW data. This means 4:2:0 YUV data is still 1.5x bigger than RAW data (assuming same bit depth).
Compressing and recording RAW is much more efficient than doing the same on debayered data, so RAW recording is good in terms of efficiency (to get big frame size, fps and bit depth).

Jim Giberti wrote:I'll ad that ProRes HQ is what we produce in day to day for broadcast and web delivery and it's robust and well beyond what's required for those mediums.
Everything else is incremental improvements, but with ProRes, you're already there for 95% of what's being done professionaly.

Yes, that is my take also. The new Pocket Camera 4K is looking better all the time.

But Raw has its advantages, depending on your work flow. How I would love a way to process a Raw f
I’ll as a monochrome and keep the total resolution.
Cheers

Hmn, yeah. Compare the raw ratios with a Raven. Some of that Red gear gets so high, I question it.

Re-edit:

It's like this visually lossless stuff. Visually lossless would include near lossless which 3:1 might be, and 4:1 is likely. But say on a codec like cineform non raw. You could do 10:1. That likely would have been regarded as in the range of visually lossless, but that doesn't mean 4:1 visually lossless is not way better. I would be reluctant to use anything above 6:1 on that sort of codec (different codecs have different ratios at which they hit these things) for something good. The more good, the closer to 2:1-3:1 I would want (which should be lossless it very close). To do visually lossless, one thing they do is they fudge the accuracy of pixels in a way that is hard to notice outrightly, to make them more compressible. But as you do this the contrast detail between pixels goes down and it gets more smudgy look. As John would put it, your picture IQ goes down. So, how they choose to degrade and limit this degregation will preserve more subconscious picture appeal. So, you might buy a video camera, and that video camera maybe set up for the skin of people in the market it was made for, to render people's skin better. That is one of the things you want to preserve. Their is a hierarchy of how people see to preserve. The shape of things, details, facial structure and details. These sorts of things. You want a certain level of look on skin, cloths around skin, or longer shots cloths, bushes and shapes of leaves blowing around, and preserving more detail where it's loss will be noticed more. So, on the face you might want 4:1-2:1, having that around the face 4:1-2:1, but another thing in a scene you might make them compressible 10:1, others things between 4:1-10:1. Now, you average compression from 4:1(extreme closeup to stop the face it dipping in quality too much)-6:1, some saving here and there, to give more preservation elsewhere. There is a variety of ways a manufacturer can set things up to preserve more important quality areas.

Que.

I find it best to view the difference in terms of how it feels to them. People can subconsciously appreciate things they don't normally perceive on screen. Therefore visually lossless might not have the appeal of lossless (how you do visually lossless can help retain appeal). Pixel peeping shows you the issues of the image on a grander scale even though individually such things may normally be unoticed. So, pixel peeping a little issue occasionally may not harm too much, but a little issue throughout the image is likely going to degrade it. Ask you wife, which looks better, then why (trick question, because people respond more to higher contrast than resolution, but you don't go sit in theatres for hours watching contrasty images to relax. You say to your wife then, if she picks the lower quality contrasty image, ahh but they don't do relaxing films that way. If you put this on a really big screen and sit down the theatre, it will look more ugly, like if you looked at it really closely here).

Wayne Steven wrote:Que.

I find it best to view the difference in terms of how it feels to them. People can subconsciously appreciate things they don't normally perceive on screen. Therefore visually lossless might not have the appeal of lossless (how you do visually lossless can help retain appeal). Pixel peeping shows you the issues of the image on a grander scale even though individually such things may normally be unoticed. So, pixel peeping a little issue occasionally may not harm too much, but a little issue throughout the image is likely going to degrade it. Ask you wife, which looks better, then why (trick question, because people respond more to higher contrast than resolution, but you don't go sit in theatres for hours watching contrasty images to relax. You say to your wife then, if she picks the lower quality contrasty image, ahh but they don't do relaxing films that way. If you put this on a really big screen and sit down the theatre, it will look more ugly, like if you looked at it really closely here).

agreed. I do think there is a feel that comes across in higher quality, natural looking images. You kind of fall into the image. Which is why I’m so excited about the pocket 4k!


Sent from my iPhone using Tapatalk

Andrew Kolakowski wrote:

Sean van Berlo wrote:But Prores 444 is less compressed than both Raw 3:1 and 4:1, right?

Nope- it's compressed much more than that, almost 7:1.
ProRes444 XQ is 4.5:1.

You have to understand that RAW bayer data is a monochromatic image, eg:



During debayering we convert (some pixels are interpolated as we have no real data for them) 1x monochromatic data to RGB channels. This causes debayered, final RGB data to be 3x (assuming same bit depth and full sampling) bigger than RAW. We can "loose" some of this "big" data by converting RGB to YUV, where we can sample UV (color info) every 2nd pixels horizontally (so we get 4:2:2) or on both axes, so we get 4:2:0. On top of this we apply compression- ProRes, etc. but even then we still have more data to compress than RAW. Y channel (this is as big as RAW) and at least 2 channels sampled every 2nd pixel in both axes, so this is another 0.5x of RAW data. This means 4:2:0 YUV data is still 1.5x bigger than RAW data (assuming same bit depth).
Compressing and recording RAW is much more efficient than doing the same on debayered data, so RAW recording is good in terms of efficiency (to get big frame size, fps and both depth).

This was extremely interesting, thanks!

Wayne Steven wrote:Hmn, yeah. Compare the raw ratios with a Raven. Some of that Red gear gets so high, I question it.

Re-edit:

It's like this visually lossless stuff. Visually lossless would include near lossless which 3:1 might be, and 4:1 is likely. But say on a codec like cineform non raw. You could do 10:1. That likely would have been regarded as in the range of visually lossless, but that doesn't mean 4:1 visually lossless is not way better. I would be reluctant to use anything above 6:1 on that sort of codec (different codecs have different ratios at which they hit these things) for something good. The more good, the closer to 2:1-3:1 I would want (which should be lossless it very close). To do visually lossless, one thing they do is they fudge the accuracy of pixels in a way that is hard to notice outrightly, to make them more compressible. But as you do this the contrast detail between pixels goes down and it gets more smudgy look. As John would put it, your picture IQ goes down. So, how they choose to degrade and limit this degregation will preserve more subconscious picture appeal. So, you might buy a video camera, and that video camera maybe set up for the skin of people in the market it was made for, to render people's skin better. That is one of the things you want to preserve. Their is a hierarchy of how people see to preserve. The shape of things, details, facial structure and details. These sorts of things. You want a certain level of look on skin, cloths around skin, or longer shots cloths, bushes and shapes of leaves blowing around, and preserving more detail where it's loss will be noticed more. So, on the face you might want 4:1-2:1, having that around the face 4:1-2:1, but another thing in a scene you might make them compressible 10:1, others things between 4:1-10:1. Now, you average compression from 4:1(extreme closeup to stop the face it dipping in quality too much)-6:1, some saving here and there, to give more preservation elsewhere. There is a variety of ways a manufacturer can set things up to preserve more important quality areas.

I don't know if there is any study, but I assume you can treat RAW image the same as final debayered. If anything it may compress "better" as we know its nature (Bayer pattern, so we could optimise codec). Not 100% sure. Lower compression ration comes mainly from fact that RAW is treated as best possible recording, so we want it to be high quality. 3:1 is definitely really high quality.
RED has many quality levels (although they use I assume more efficient JPEG2000) and lot of people record 5:1, 7:1 etc. There is nothing what stops you record RAW as h264. How compressing RAW effects final debayered quality (compared to compressing final video) can be measured, so it's not that big mystery if one wishes to evaluate it.

Sean van Berlo wrote:

MarcusWolschon wrote:

Sean van Berlo wrote:They stated the camera can be charged via usb-c, but only while the camera is off unfortunately.

What's the source of that statement?

Interview with Grant at NAB where he said you could charge the camera while it's in your bag. I'm inferring from that, to be fair, that it doesn't charge while recording. The fact that it charges at all via USB-C is on Blackmagic's product page.

There was another video in which a BMD representative claimed that you can power the camera using USB C, It can be deduced that you can charge the batteries while it is powered on. How much extended time you get is something we have to wait for.

As I understood it, you can charge by USB-C when camera is off but not charge the camera internal battery nor run the camera powered by USB-C.

Que Thompson wrote:
agreed. I do think there is a feel that comes across in higher quality, natural looking images. You kind of fall into the image. Which is why I’m so excited about the pocket 4k!

+1