Pages: 1 2 [3] 4 5 6   Go Down

Author Topic: 4K...  (Read 30922 times)

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24308
Re: 4K...
« Reply #40 on: January 19, 2012, 10:05:39 am »

You see? Facts that ignore other facts: those back seats we patronised were way up in the air, a good many degrees above any divine, central, co-axial datum; mark you, the other lines doing the rounds there weren't much better, either, but they worked - twice a week at every change of programm. Viva Hollywood! even if we didn't watch you much.

Rob C

feppe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2906
  • Oh this shows up in here!
    • Harri Jahkola Photography
Re: 4K 36 degree recommended minimum viewing angle, not optimal
« Reply #41 on: January 19, 2012, 01:02:55 pm »

[Edit: what that calculator calls "recommended THX viewing distance" is in fact the recommended _maximum_ for the _back row_ of a cinema, not at all an optimum viewing distance recommendation.]

You are correct. By my viewing being in the sweet spot I meant that the angle-based distance is 2.9 meters, and visual acuity based maximum is 3.2 meters, so my 3.0 meter distance is pretty much spot on. But perhaps most importantly, it's a comfortable distance and the screen looks big.

Quote
Meanwhile, the next big thing for younger viewers is small things. The movie industry might have to start thinking specifically about how to enhance the movie experience on an iPad or jumbo Android phone, and how to make the legally purchased small screen experience better than the pirated alternatives.

While I agree that gadgets are getting smaller, screens necessarily not so. Bendable video displays are already in prototype, and I bet we'll get them in consumer models this or the following year. Next step is foldable displays, which would stow away for transport - think fo a fan or accordion.

Nevertheless, I only see benefit for 4k in cinema setting, and possibly some yet-to-come heads-up display (ie. no physical screen).

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Sony white paper: 4K>2K from less the twice picture height
« Reply #42 on: January 19, 2012, 01:41:30 pm »

Hi,

Yes, but that may also depend on resolution. On the movie side we also have IMAX.

Best regards
Erik


I would also say that most people sit at 3x the distance from the size of their TV or more.

Cheers,
Bernard

Logged
Erik Kaffehr
 

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: 4K 36 degree recommended minimum viewing angle, not optimal
« Reply #43 on: January 19, 2012, 03:07:05 pm »

You are correct. By my viewing being in the sweet spot I meant that the angle-based distance is 2.9 meters, and visual acuity based maximum is 3.2 meters, so my 3.0 meter distance is pretty much spot on.
Yes, but that visual acuity standard is based on 1920x1080 resolution, so if you do someday have 4K video, the same visual acuity standards allow the screen to get bigger, or the chair closer. However ...
But perhaps most importantly, it's a comfortable distance and the screen looks big.
... it might well be that the screen size needed to allow several people to view the screen from closer than 2.3 picture heights (Sony's threshold for benefiting from more than 2K) will not make sense except in a few huge super-luxury home theaters. So in practice, 4K video's benefits may be realized only in cinemas, as you say. (4K display of stills, as Erik seems to be talking about, is another issue.)

While I agree that gadgets are getting smaller, screens necessarily not so. Bendable video displays are already in prototype, and I bet we'll get them in consumer models this or the following year. Next step is foldable displays, which would stow away for transport - think fo a fan or accordion.
Yes; by smaller, I only meant a recent survey confirming a shift in preferences amongst younger people towards viewing on mobile devices, so far smaller than home theater screens. Unless that future foldable screen resembles a bed-sheet!

... I only see benefit for 4k ... possibly some yet-to-come heads-up display (ie. no physical screen).
!
« Last Edit: January 19, 2012, 04:06:49 pm by BJL »
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13985
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Sony white paper: 4K>2K from less the twice picture height
« Reply #44 on: January 19, 2012, 11:17:14 pm »

Hi,

Yes, but that may also depend on resolution. On the movie side we also have IMAX.

Realistically, for most people the distance to the TV is driven by the layout of their rooms, meaning by the size of their rooms.

Cheers,
Bernard

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: 4K...
« Reply #45 on: January 20, 2012, 12:54:15 am »

4k on an iPad may well be silly, but in that case 4k on a 200" projected image would be equally silly if the distance to the viewer was scaled proportionally. That was my point.

Talking about resolution requirements related to screen size without taking into account viewer distance makes no sense.
4K on an iPad is silly and laughable. 1080p on a sub 10" screen isn't much different in image quality to standard color printing. So how 4K will help is beyond comprehension. ...
4k could be equally warranted or not warranted in an iPad or at the cinema. The question is not about screen size, but screen size related to viewing distance. I.e. how many degrees of our field of vision is covered, or how large an angle does one pixel amount to (1 arcminute seems to be the accepted limit).
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
home theater res. needs: maximum screen size, minimum seating distance
« Reply #46 on: January 20, 2012, 09:31:00 am »

... 4k on a 200" projected image would be equally silly if the distance to the viewer was scaled proportionally.

Talking about resolution requirements related to screen size without taking into account viewer distance makes no sense.
Exactly: the two questions to ask are
- A. how big will home theater screens get?
- B. how close to the screen will people be comfortable sitting?

If
(minimum comfortable seating distance)/(maximum home theater screen height) is about 2.5 or more,
or in other words
(minimum comfortable seating distance) > (maximum home theater screen diagonal)
then current "2K" or 1080p is enough.

And that criterion comes from Sony's 4K propaganda, so I am not being unfair to 4K.

With most modern ceilings 8 or 9 feet high, and the bottom of the screen needing to be about 3 feet or more above the floor to avoid an uncomfortable downward looking view (IMAX makes some people sea-sick for related reasons) I get maximum screen heights of about 5 feet, and so the question is how many people would want their eyes within 12.5 feet of the screen, meaning the front of the sofa within 10 feet, the front of the "popcorn and soda table" within 7 feet ... (Serious questions: I am mostly ignorant of home theater layout ... I do not even own a TV, unless streaming on a computer counts.)
« Last Edit: January 20, 2012, 09:37:08 am by BJL »
Logged

feppe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2906
  • Oh this shows up in here!
    • Harri Jahkola Photography
Re: 4K 36 degree recommended minimum viewing angle, not optimal
« Reply #47 on: January 20, 2012, 06:21:39 pm »

Realistically, for most people the distance to the TV is driven by the layout of their rooms, meaning by the size of their rooms.

And that's my main argument in saying 4k adoption won't be driven by home consumers. Even though I'm sure we can get affordable 100+ inch LCD/LED/OLED/AMOLED/whatever displays within just a few years at 1080p or better resolution, we won't use the entire screen for viewing a movie due to physical limitations of viewing environment (living room sizes and people's move towards handheld devices). The screens of the future might be wall-sized, but it will display 40-80 inch video in a typical setting unless it's for mood (think digital fireplaces).

!

I wasn't pulling HUDs of displays of virtually any size projected directly to your eye out of my hat. We already live in a scifi world, and it's only getting more so.

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: 4K 36 degree recommended minimum viewing angle, not optimal
« Reply #48 on: January 20, 2012, 06:31:25 pm »

I wasn't pulling HUDs of displays of virtually any size projected directly to your eye out of my hat. We already live in a scifi world, and it's only getting more so.
Agreed: my very brief comment was meant to complement you for literally "thinking outside the box".
Logged

Alan Goldhammer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4344
    • A Goldhammer Photography
Re: 4K...
« Reply #49 on: January 21, 2012, 12:11:55 pm »

I would rather spend the money on a new camera than a home theater set up! :D
Logged

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24308
Re: 4K...
« Reply #50 on: January 21, 2012, 12:22:31 pm »

I would rather spend the money on a new camera than a home theater set up! :D


I couldn't agree more!

Rob C

feppe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2906
  • Oh this shows up in here!
    • Harri Jahkola Photography
Re: 4K...
« Reply #51 on: January 22, 2012, 10:27:29 am »

I would rather spend the money on a new camera than a home theater set up! :D

Not really a surprise on a photography forum. Go to Audio Video forum and you'll get a different response.

Also, it's a false dichotomy: one can forgo one generation of camera upgrades and get a decent home theater setup. It's not like everyone needs to upgrade their camera every generation. It was just ten years ago when "generation" actually mean generation, not 2-3 years. My digital camera is just as good as I bought it almost two years ago.

Farmer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2848
Re: 4K...
« Reply #52 on: January 22, 2012, 05:27:35 pm »

And recent threads in other forums would suggest an aweful lot of people would put software upgrades a very distant third to either of those things!

Even pros are distracted by shiny toys ;p
Logged
Phil Brown

Graeme Nattress

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 584
    • http://www.nattress.com
Re: Sony white paper: 4K>2K from less the twice picture height
« Reply #53 on: January 22, 2012, 05:53:18 pm »

As to video cameras though: Sony's calculations are for full RGB at every pixel, so for Bayer CFA cameras like Red's, matching full 2K projection needs 3K to 4K recording, so dazzling even the front row customers might push even a bit beyond 4K projection, and maybe to 6K or 8K Bayer CFA capture.

Any good camera should use some optical low pass filtering to avoid aliasing artifacts, and that's the case for Bayer CFAs, 3chippers, foveons etc. Most of the resolution loss in a Bayer CFA is down to the OLPF rather than the Bayer pattern itself and our visual system is not really tuned to see chroma resolution. So if you had a hypothetical 3 chip 4k video camera, you'd still only measure around 80% of that in linear resolution terms - or you'd have excessive aliasing.

With the Epic at 5k I can still see some good contrast at 4k, and negligible aliasing. Even with older RED One footage at 4k it's pretty clear that there's benefits over 1080p.

The main thing to think about is angle of view. The higher the resolution the closer you can sit, and not see pixels. Just think how the wide field of view of IMAX makes for a very immersive experience.

Graeme
Logged

Alan Goldhammer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4344
    • A Goldhammer Photography
Re: Sony white paper: 4K>2K from less the twice picture height
« Reply #54 on: January 22, 2012, 06:04:58 pm »


The main thing to think about is angle of view. The higher the resolution the closer you can sit, and not see pixels. Just think how the wide field of view of IMAX makes for a very immersive experience.

Graeme
A lot of years ago when Ben Hur was released (1959 or so) it was showing at one theater in San Diego and you needed to reserve seats ahead of time.  I was 12 at the time and went with my friend up the street.  We were in the second row and it was a really big screen (can't remember the technology used only that it was several years before Cinerama) and that darn chariot race really immersed us!!!
Logged

Graeme Nattress

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 584
    • http://www.nattress.com
Re: 4K...
« Reply #55 on: January 22, 2012, 06:13:11 pm »

I bet it did immerse! The vision I have of 4k+ is that it fills a wall, you sit close, and it's like "home" IMAX, with stunning levels of detail and a large field of view. You could do the same with a smaller display, but our eyes are more comfortable focusing at a larger distance necessitating a larger screen.

I find the old viewing distance recommendations for TVs and screens to produce too small a field of view. I like a good sized field of view, but as IMAX DP's know, you've got to really shoot with the larger screen in mind. But when you get it right, it's superb in a way that a "normal" sized field of view just can't compete with.

Graeme
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
5K Bayer about enough for 4K full RBG with a light OLPF?
« Reply #56 on: January 22, 2012, 06:39:19 pm »

...
With the Epic at 5k I can still see some good contrast at 4k, and negligible aliasing. Even with older RED One footage at 4k it's pretty clear that there's benefits over 1080p.
Graeme, I do not have solid data, but I can believe that my numbers were pessimistic and that 5K Bayer is about a match for 4K full RGB recording. Experiments with Foveon X3 sensors suggests a higher ratio (like twice as many Bayer CFA photosite locations as X3 locations) but that could be judgements based on the "aliasing enhanced crispness" caused by the Foveon sensors' unmitigated aliasing, rather than a fair resolution comparison. And I have no doubt that 4K Bayer CFA exceeds full resolution 1920x1080 HD, despite the "count the red and blue pixels only" argument of some X3 partisans.

Anyway, I am sure the Red is ready to up the stakes to 6K or whatever, if that is where the video resolution race goes.
Logged

dreed

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1719
Re: 5K Bayer about enough for 4K full RBG with a light OLPF?
« Reply #57 on: January 22, 2012, 06:50:06 pm »

Graeme, I do not have solid data, but I can believe that my numbers were pessimistic and that 5K Bayer is about a match for 4K full RGB recording. Experiments with Foveon X3 sensors suggests a higher ratio (like twice as many Bayer CFA photosite locations as X3 locations) but that could be judgements based on the "aliasing enhanced crispness" caused by the Foveon sensors' unmitigated aliasing, rather than a fair resolution comparison. And I have no doubt that 4K Bayer CFA exceeds full resolution 1920x1080 HD, despite the "count the red and blue pixels only" argument of some X3 partisans.

Anyway, I am sure the Red is ready to up the stakes to 6K or whatever, if that is where the video resolution race goes.

Canon's approach with the C300 is to not use Bayer at all. There is one RGGB for each pixel in the 1920x1080 output. Thus rather than a 2K with Bayer, the C300 is an 8.3MP sensor producing 2K video.

If Canon continues to follow this approach with the "C" DSLR, then we would expect to see a 33.2MP sensor delivering 4K video.

But just maybe that is a bit too much to ask for.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: 5K Bayer about enough for 4K full RBG with a light OLPF?
« Reply #58 on: January 22, 2012, 07:00:43 pm »

Canon's approach with the C300 ... is one RGGB for each pixel in the 1920x1080 output. Thus ... the C300 is an 8.3MP sensor producing 2K video.
It is interesting for us armchair camera designers to have Canon try yet another CFA and processing system, but I am for now puzzled by that approach of using a roughly 4000 pixel wide array of photosites to produce 1920x1080 RGB output, since I agree with Graeme that using this number of pixels with Bayer CFA array can out-resolve 1920x1080 (except with subject matter that is all red and blue, no green!) Maybe Canon seeks a win on noise by on-chip binning of the GG pair in each RGGB?
Logged

Graeme Nattress

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 584
    • http://www.nattress.com
Re: 4K...
« Reply #59 on: January 22, 2012, 07:53:39 pm »

Well, Canon C300 does use a Bayer CFA - they just don't demosaic it traditionally. They read each of the repeating blocks of 4 photosites - two greens a red and a blue as one. By summing the greens they lower aliasing a bit.

Out of interest I programmed such a demosaic to compare against others. Such a "average green" method is slightly superior to the ignoring of one of the greens entirely, but not by too much. A more complex demosaic to half resolution using all the data with a downsampling filter produces a much sharper result at the expense of some aliasing, but by far and a way, the nicest is a full resolution traditional demosaic, showing lower aliasing by far, better resolution and contrast on edges, and lower chroma moire.

Of course, the method Canon uses works oh-so-much better than their previous line-skipping cameras, but it's no replacement for a full demosaic.

Graeme
Logged
Pages: 1 2 [3] 4 5 6   Go Up