Panasonic 4K at NAB?

[adkimery]

Yeah, 4K for cinema purposes, for film makers...makes sense, 4k for consumers...marketing hype that will come around but it'll take a long while. Just look at TV, HD has not even become the standard for ALL channels yet....you get some HD channels and some SD...some HD isn't even real HD.

I'd guess that 4k is going to come in via streaming & download services. I don't think we'll see 4k OTA or cable/satellite because of the expense involved in upgrading a lot of back-end gear (a big reason why the transition from analog SD to digital HD was decades in the making). When it's just 1's and 0's over the internet change can happen at a much quicker pace. YouTube, for example, when from offering only sub-SD streams to 4k streams in about 5yrs. I think in 5-10yrs it will be common to have 4k as an option for streaming media (assuming of course you have a fast enough internet connection and a device that can display it).


-Andrew

 

[pulpfiction007]

Good point, and yep...you're right...but seeing the difference on many of these IPADS, tablets, phones, laptops....people will just say yeah, I've got 4k for the buzz of it. I can't count the number of average consumers I've heard say things like: My phone is so awesome, it's xx mega pixels.....now we know that's a very small determining factor in the IQ. but they don't!

 

[mik303]

You're comparing raw 4K to 4:2:0 720p DVCPro 50? Don't you think the raw part muddies the comparison, so that the advantages you cite in color correction, etc., could be because of raw, not 4K?

If you're wanting something better than DVCPro 50, then you are merely agreeing with everyone, 2K or 4K.

I meant good ole' DVCPROHD, my 720/24pN setting on the HVX 200 that I have grown to love over the years. I always thought I was working with a 4:2:2 codec and not 4:2:0.

Also, yes, you make a great point. 4K RAW is not the same as 4K. But 4K resolution is better than 1080p every day of the week, as far as POST PRODUCTION is concerned. Are you seriously making the "4K is useless" argument? That is just not logical. It's like saying 24bit audio is useless because retail Audio CD's are all 16bit or because most music today is distributed online as compressed files that are 10 times smaller than 24bit .wav's or .aif that we work with in the studio.

 

[combatentropy]

Are you seriously making the "4K is useless" argument?

No, just that 4K is not worth the expense, at this time.

If you made two cameras exactly alike --- and I mean exactly alike, no more of these shenanigans comparing 4K RAW to 8-bit 4:2:0 720p --- if you made two cameras exactly alike, except one is 2K and one 4K, in other words, ALL ELSE BEING EQUAL, PEOPLE!, then the 4K camera would:

- have 1/4 the recording time per card
- cost 4 times as much in media
- take 4 times as long to render each effect, and 4 times as long to transcode to another format
- have two stops less dynamic range (since the pixels are one quarter the size, gathering one quarter the light, which is 2 stops)
- have two stops less light sensitivity
- may be a manufacturer's deciding factor between global and rolling shutter, due to the dynamic range and noise going below a certain threshold (coincidentally, global-shutter sensors seem to have 2 stops less DR than their rolling-shutter counterparts)


I'm not saying you can't find a use for 4K. It's just that the things it seems to help with (reframing in post, photoshopping backgrounds) I would use seldom, yet I would be paying the expense in every shot.

 

[adkimery]

If you made two cameras exactly alike --- and I mean exactly alike, no more of these shenanigans comparing 4K RAW to 8-bit 4:2:0 720p --- if you made two cameras exactly alike, except one is 2K and one 4K, in other words, ALL ELSE BEING EQUAL, PEOPLE!, then the 4K camera would:

Couldn't the exact thing be said for SD vs HD (w/a slight exception for HDV which had the same data rate as DV so storage requirements stayed the same)?

Some people will be okay making some compromises now, or in the near future, to have a 4k camera because they see a worthwhile upside and some people will wait. Just like some people jumped on the HD train as soon as it started leaving the station and others waited for it to get more established.

 

[bruceallen]

No, just that 4K is not worth the expense, at this time.

If you made two cameras exactly alike --- and I mean exactly alike, no more of these shenanigans comparing 4K RAW to 8-bit 4:2:0 720p --- if you made two cameras exactly alike, except one is 2K and one 4K, in other words, ALL ELSE BEING EQUAL, PEOPLE!, then the 4K camera would:

- have 1/4 the recording time per card
- cost 4 times as much in media

Well... if it has 1/4 the recording time per card and costs 4x as much in media then I'd expect the image to be better, since you're quadrupling the data rate. And yes, that means if you scale that high data rate 4L to 2K, the 2K image benefits from the increased data rate too.

I think datarates are pitifully low in cameras (why else is everyone doing those GH2 hacks?) so don't really think that's much of a disadvantage.

Bruce Allen
www.boacinema.com

 

[alpi69]

Difference is: SD was pretty much the whole market. Everyone had it. HD was only the next step.
Nowadays HD has not penetrated the full market yet.
On TV I see lot of Sd still. Even the NFL has a lot of 4:3-SD content. All of the youtube-watching is pretty much done in SD (1024 wide is plenty for 99% of the web users). So 4K (just as 3D) is not the next step but two steps ahead. It is cool for a few big screens and also for commercials with big screenings. But the audience doesn´t care if it is (true) 720 or 4K.

 

[Bassman2003]

The train scenario is what I and seemingly others are trying to avoid. There is not much upside imho to 4k right now. Back when in the psuedo HD era all of the televisions in stores were going HD. There was an upside for the camera shortcomings to be able to fill the screen. 2k RAW would be much more useful.

 

[Babu]

Couldn't the exact thing be said for SD vs HD (w/a slight exception for HDV which had the same data rate as DV so storage requirements stayed the same)?

Some people will be okay making some compromises now, or in the near future, to have a 4k camera because they see a worthwhile upside and some people will wait. Just like some people jumped on the HD train as soon as it started leaving the station and others waited for it to get more established.

It's the same thing all over again. I welcome this new technology. 4k scaled to 1080P should look incredible. 4K still grabs should be of great use especially in my field of work. Either way better lowlight, DR, and sharpness than the AF100 period. I'm quite sure the camcorders will all have in cam downscale as the FS700 and C300 do. You're right Adkimery. It's the SD to HD argument over again but now 1080 to 4K

 

[Tzedekh]

No, just that 4K is not worth the expense, at this time.

If you made two cameras exactly alike --- and I mean exactly alike, no more of these shenanigans comparing 4K RAW to 8-bit 4:2:0 720p --- if you made two cameras exactly alike, except one is 2K and one 4K, in other words, ALL ELSE BEING EQUAL, PEOPLE!, then the 4K camera would:

The storage space and rendering times are admittedly issues, but it's also just as likely that before a manufacturer releases what it considers a functional 4K camera it would optimize its codec to reduce storage and bandwidth burdens, such that they are not four times as bad at 4K as they were at 2K. Likewise, when developing sensors for 4K cameras, it wouldn't simply use the same materials and designs as it did with 2K sensors but merely make the pixels a quarter the size. Ideally, they would be developed to work optimally at a higher density. So yes, if that same technology were used to make 2K sensors, they might have greater sensitivity and 2 stops more dynamic range -- but in all likelihood, the tech would be used only in the new, 4K sensors. In fact, one could assume your objections are factors in why 4K cameras haven't been marketed until now -- the technology was arguably adequate for 2K but not for 4K. Now it's improved such that the sensitivity and dynamic range of the 4K sensors should at least rival those of the 2K. At least that's the hope.

 

[Kholi]

We read the same claims about audience not caring yadda yadda when the HVX was released. Clearly that was wrong. Then it was SD versus HD, now it's HD versus 4/5K.

End of the day, it all comes down to price. If the FS700 cost 5,000.00 USD and was 4K out of the box, a lot of people wouldn't care about how the audience felt about resolution, etc so on and so forth. Just so happens that it'll still be out of the range of a lot of people.

Bruce explained quite well why 4K is important for every walk of narrative content technically (please note the word technically used here, artistic choice and technical don't often hold hands peacefully). Naturally, I sincerely hope Panasonic has a 4K+ Solution this year, as I still feel that they are the only company capable of delivering insane specs at an insane price.

 

[keylight]

ALL ELSE BEING EQUAL, PEOPLE!, then the 4K camera would:

- have 1/4 the recording time per card
- cost 4 times as much in media

These are two sides of the same coin. You pay the price by either getting only 1/4 the recording time OR paying 4X the price to get the SAME amount of recording time. Not both.

 

[Danielvilliers]

The storage space and rendering times are admittedly issues, but it's also just as likely that before a manufacturer releases what it considers a functional 4K camera it would optimize its codec to reduce storage and bandwidth burdens, such that they are not four times as bad at 4K as they were at 2K. Likewise, when developing sensors for 4K cameras, it wouldn't simply use the same materials and designs as it did with 2K sensors but merely make the pixels a quarter the size. Ideally, they would be developed to work optimally at a higher density. So yes, if that same technology were used to make 2K sensors, they might have greater sensitivity and 2 stops more dynamic range -- but in all likelihood, the tech would be used only in the new, 4K sensors. In fact, one could assume your objections are factors in why 4K cameras haven't been marketed until now -- the technology was arguably adequate for 2K but not for 4K. Now it's improved such that the sensitivity and dynamic range of the 4K sensors should at least rival those of the 2K. At least that's the hope.

I think no one is against 4k, who would not want a F65 for $ 5000. let be realistic it is just not feasible today, but an F3 quality 13 stop DR, clean Iso 1600 to 6400, 10 bit 422 at least would be much more feasible by Panasonic for a sun $ 8k to even $ 6k. What we are fearing is exactly what you are saying, a 4k camera where the manufacturer will reduce the proportional data rate, which mean more compression lower quality. It is not as if by magic they are going to get better DR and low noise by having 4 times more pixel. Perhaps in the state of the art $ 80 000 F65, but for us mere mortal there will be a lot of compromise just for 4k.

 

[cowpunk52]

Even when the HVX200 was released on Dec 29, 2005, high definition television had been broadcast in the US for nearly a decade. Now, due purely to increased pixel information, we're seeing camera specs grossly outpace our infrastructure capability for transmit. 4K as acquisition does have its merits for post-production work, but I'd rather see research and development go into a number of other features before simply cramming more resolution into already stressed low bitrate codecs. When broadcasters start asking for 4K deliverables for broadcast is when I'll start considering 4K acquisition. HD and 2K acquisition are just now entering a mature stage (excellent codec quality with relatively simple workflow) as of last couple of years - though broadcasters are still having a lot of trouble with compression.

Plans have just begun to take shape to begin to explore the possibility of 4k delivery by satellite. Nothing yet is in the works via cable & air, AFAIK. 4K, 8K, & UltraHD will happen - but we're looking at quite a few years before the technology hits mainstream. Five at the most optimistic minimum, if customers really start clamoring for it, but more likely I predict it will be 10-15 years before we see it become anything more than novelty.

I submit this announcement from 3/18/12
"According to a report from Advanced Television, DirecTV is rolling out plans regarding upgrades that will eventually send out an Ultra HDTV signal. While a 1080p high definition signal offers 1080 horizontal lines of vertical resolution, DirecTV will be offering services for both “4000-line and 8000-line services” according to the company. In order to enable the transition, DirecTV will continue moving away from Ku-band satellites within the next four to five years and change over to Ka-band to take advantage of higher bandwidth. While Ka-band technology is far more likely to become garbled and unusable for consumers during a extremely rainy day due to increased cloud cover when compared to the Ku-band, DirecTV has already had success in broadcasting 1080p signals over Ka-band to customers in North America."

 

[Danielvilliers]

We read the same claims about audience not caring yadda yadda when the HVX was released. Clearly that was wrong. Then it was SD versus HD, now it's HD versus 4/5K.

End of the day, it all comes down to price. If the FS700 cost 5,000.00 USD and was 4K out of the box, a lot of people wouldn't care about how the audience felt about resolution, etc so on and so forth. Just so happens that it'll still be out of the range of a lot of people.

Bruce explained quite well why 4K is important for every walk of narrative content technically (please note the word technically used here, artistic choice and technical don't often hold hands peacefully). Naturally, I sincerely hope Panasonic has a 4K+ Solution this year, as I still feel that they are the only company capable of delivering insane specs at an insane price.

For me we are in the law of diminishing return. Moving from SD to Hd was a huge difference. But the move from HD to 4k will be much less. I am all for better acquisition as a professional, but not at the expense of DR, low light. Until we get at least 422 and 10 bit output at good datarate.

 

[Kholi]

For me we are in the law of diminishing return. Moving from SD to Hd was a huge difference. But the move from HD to 4k will be much less. I am all for better acquisition as a professional, but not at the expense of DR, low light. Until we get at least 422 and 10 bit output at good datarate.

I can see this. Everyone's got to make a choice for themselves. As Bruce explained, though, once you move the content from 4K to 1080/2K, you've got your 4:2:2/4:4:4 and more. It really isn't about finishing 4K/5K, it's about delivering better HD. When 4K is a delivery standard, 8K will be the way you get to a robust 4K.

 

[Garion]

Another thing to remember is it's differant country to country as well. Here in the UK most of our Television broadcasts are SD, we have some HD channels but most are SD. Our countrys Internet is also not as good
as others so streaming 4K would be high on impossible for most of us as well. If your in a big city or near to a telephone exchange youll probably be ok but a lot of people only get 2-3 Mbit, myself included. heck i cant even stream 720p vimeo footage

i think you'd be surprised at how many people here in the Uk still do not have an HD TV or really care to yet. If 4K came tomorrow... how on earth would anyone watch it? wont work here in Uk for a hellava while yet.

I agree it will be great for people exporting to 35mm film or doing VFX work but apart from giving some people a "mine is bigger than yours feeling" id rather have a better 1080p camera for now please.

 

[combatentropy]

Couldn't the exact thing be said for SD vs HD

No, because HD is near the limit of human vision at traditional theatrical viewing distances.

You can't simply say that every transition is just like all the past transitions, especially when the number of transitions to study in history is so few. How many major transitions in motion-picture history have there been?

From 2K to 4K will not be as perceptible as from analog SD to digital HD. Using that logic, you might as well as say that the transition from 2,000,000K to 4,000,000K would be just like from SD to HD, just another transition that the Luddites should embrace. I'm making such an exaggerated example (2,000,000K!) to make a point. There comes a point of diminishing returns. The only thing to argue over is at what point. Studies of the limit of human vision and audience perception, I argue, say that the Transition in question was at 2K. All further increases in resolution will of course yield some fractional benefit, especially if you consider infinite reframability as highly desirable. I don't. Therefore, all increases beyond 2K are of little to no consequence to me, certainly not worth the cost in dynamic range, bandwidth, etc., and I would prefer that all advances in technology be poured into improving the many other aspects of imagery that I mentioned.

You can have your 4K camera. To me, it is a special-use camera, for rental, like super-high-frame-rate cameras.

 

[Tzedekh]

I think no one is against 4k, who would not want a F65 for $ 5000. let be realistic it is just not feasible today, but an F3 quality 13 stop DR, clean Iso 1600 to 6400, 10 bit 422 at least would be much more feasible by Panasonic for a sun $ 8k to even $ 6k. What we are fearing is exactly what you are saying, a 4k camera where the manufacturer will reduce the proportional data rate, which mean more compression lower quality.

But it's not necessarily the case that increased compression yields a lower-quality image. Some codecs are more efficient than others.

It is not as if by magic they are going to get better DR and low noise by having 4 times more pixel.

I'd be surprised if Panasonic's 4K sensor were exactly the same as the AF100's, only with more densely packed pixels. Actually, it may have fewer pixels, just not binned to do 1,920 x 1,080. Even so the binned "virtual" pixel would still be a little larger. But Panasonic is obviously aware that the AF100's sensitivity and dynamic range are lower than the F3's (and even the FS100's) and going to 4K with the same technology would only compound the problem. When they announced the Scarlet 3K (RIP), Red claimed that its MX sensor, even with photosites whose diameter measured only about 3/5 that of the photosites on Red One's Mysterium sensor, outperformed the Red One with respect to both sensitivity and dynamic range. Let's hope the same can be said about Panasonic's 4K offering(s).

 

[Tzedekh]

You can have your 4K camera. To me, it is a special-use camera, for rental, like super-high-frame-rate cameras.

But there will, at some point (maybe within 5 years), be a transition to generally ubiquitous 4K theatrical exhibition. Then 4K will not be a special-use thing. It will be the norm. And if it can be made practical well before that point, there would be little reason not to use it and a very important reason -- future-proofing -- to use it.