"numerically honest 4k", Sony backhanded slap at Red

[Wayne Morellini]

now i am seriously waiting for the "DRAGON"..........

I suppose I am too, the smaller scarlet is getting too dated. Apple has also proven to be a let down for me as well, small shd screens have been available for years and of use to a number of professions, like digital photography, design, publishing, can even be useful future proofing for consumer shd over the 3-10 year life of the system with nearly 3 and 4k consumer cameras coming out this year, and not even the latest imac series have one. Now it looks like the resolution revision, major formst revidion, touch screen and even 3d is coming next year. So, for the vast majority of low end budget users (below eng network tv work and studio work) this year is not the best year for it).

And then there are the cheap digital cinema cameras, which the scarlet announcement probably nearly destroyed with the early release date and consumer post hd cameras. The low end is virtually lost, not least because of just time. You notice that red has withdrawn from the low end market a few years ago, the time to strike was a few years ago, now commodity post fhd - shd cameras are coming out (due to the extensive time they have been able to do 4k cheaply). 4k is needed to just keep up with consumer, the next stop is 8k, but if they thought 4k was hard then 8k is going to be worse the industry does not have an adequate focusing solution for it, apart from infinite depth of feild fixed focus lens and auto focus (I do for my camera designs).

If only they had listened to people that knew the time line, they could have released a interim lowend 2 or 4k camera while working on and waiting for the scarlet range but some people think their smarter than others with a proven track record of estimating these things.

On the main subject here, good on sony, but is it going to be cheap compared to an engineering camera, or a imax camera? Yes 8k bins well to 4k (well the greens at least) and debayers well (panasonic now use this to produce 2k fhd in its top 4/3rds camera). I think i talked about doing this years ago. A 8k fixed Scarlet Dragonwith 4k motion and 8k stills is more ideal and we shoukd petition for it, with a 4k video and stills base model.

[Wayne Morellini]

Well, film has a random grain pattern on each frame, which is pretty hard to do with a fixed sensor. You could emulate this with a higher resolution sensor, but at that point, you'd be lowering aliasing anyway by going to a higher resolution sensor.

But.... When treated well, as we tend to do, aliasing is not very much of an issue, and you get a nice high resolution. Aliasing is much more a problem with lower resolution systems and ones with poor OLPF choices.

At which point the added complexity outweighs the benefit.

Graeme

actually interestingly this random shifting of the grain raises films perceived and practical resolution far above what it may seem, as your eye and brain integrates the increased resolution revealed by the shifting grains between each frame which reveals different asoects of the sane position not to mention the movement in position of film frames as they come through. i first notice this when pausing a voyager vhs tape in the 1990s, suddenly the perceuved resolution dropped. i actually devised a system to record full hd with it for dv cameras. the random vibrating ground glass 35mm lens adapttors should priduce a simular effect to film. So you can explore this effect via a randomonly vibrating screen on the red cameras with a complenentary pattern, which, using liight enough compression tied to the pattern (a shifting pattern mask) and intellegent enough compression software otherwise to follow the pattern (position feedback). People might think i talk BS but i often have allready been where they wish to go.

[Robin Balas]

In debayering of RGB pixel you don’t have to always pick the same neighbouring green pixel.
We have choice of 4 different green adjacent pixels that could be picked in semi random pattern.

But the best would be to shift photosites on the photomask by ½ of its dimension in random in all dirrections without letting fill in factor to go down.

If you search the patent databases you will find that Canon have filed for a patent for something similar where they shift the color matrix 1 px back and forth, although they didn't mention random movement to avoid aliasing - but I mention this to show that the idea might be a good one as others are on the same track.
It was mentioned on several sites some time ago but I don't have the link where I am now.
MHO.

[Andrew M.]

You see the shifting by whole pixel is not as good as shifting by 31% or 57% etc.
For decoding luma Y you can pick 3 different adjucent RGBs from the Bayer pattern.
Then you get colors just by removing one B (Y-B) or one R (Y-R) pixel from it.
Again you remove the R or B pixel from random side of the green pixel so Y-B and Y-R will have the randomness as the luma Y
You have 16 different combination for each pixel area. You can keep picking shorter and longer steps for decoding. You can also skip first green use blue and next green so it would be like starting the pixel line with 1/3 offset or even 2/3 offset.

Now it gets even better, the next frame you can shift everything in bit different random pattern and so on and so on.
As long as you follow the mathematically random patter you will get perfect micro noise on the edges of the diagonal lines instead of aliasing.

Sure the easiest would be to shift in random the starting lines of photosites from the left side and the top on the photomask, but on this subject there are the sensors people that can say something about it.

The disadvantage of the hardware approach is that each frame has the same grain, don’t know if it will look very filmic. I guess it depends of how big this grain will be. For sure it will replace OLPF and the signal degradations associated with it.
As to the software approach, it will look good but it takes computing power and I think EPIC is following the technological limits here any way.

[Blair S. Paulsen]

Unless human beings change physically and/or actually change their behavior and sit within 2 screen heights of their screen by choice I'm not sure there's a big upside past 4K for residential use.

Once we have the requisite hardware (thank you Moore's Law) I would be happy to acquire and post at 8K and beyond and there's certainly IMAXish venues that could benefit from oodles of res, but well displayed 4K should be plenty for the home. Just sayin'

Cheers - #19

[Graeme Nattress]

I envisage a sort of "wall of vision" where the whole wall in a room is a display of very high resolution, and you sit close for a encompassing field of view that's highly immersive.

Graeme

[Gavin Greenwalt]

Unless human beings change physically and/or actually change their behavior and sit within 2 screen heights of their screen by choice I'm not sure there's a big upside past 4K for residential use.

Once we have the requisite hardware (thank you Moore's Law) I would be happy to acquire and post at 8K and beyond and there's certainly IMAXish venues that could benefit from oodles of res, but well displayed 4K should be plenty for the home. Just sayin'

Cheers - #19

I see no reason to have anything less than an imax equivalent FOV screen in my home if possible. :D

Personally I think it's going to move to retinal projectors without any screens at all and when we do then head tracking will come into account. Sure you'll frame a shot but you could also turn around and look back. In that case you'll need to render 360* which would need to be far higher than 4k.

[Francesco Bonomo]

I envisage a sort of "wall of vision" where the whole wall in a room is a display of very high resolution, and you sit close for a encompassing field of view that's highly immersive.

I think they are already testing things like that in Japan, if I remember correctly (I think I've read it in an article about Sony written by Jon Fauer).
Personally I think most people will have good ol' style tv screens (although at higher resolution) for a long while, but that just me....if I'm wrong, well, welcome to the "Minority Report" kind of world... :-)

[Craig W. Bickerstaff]

Wall Screens are a bad idea for a few reasons, you can't put your center channel speaker behind them for one thing but you would also have to built your house around them.
And what if the screen broke, it's more realistic to have front projectors and a screen to project on.
I think 1080p could be the standard in the home for a long time to come.

[Steven Caesare]

Wall Screens are a bad idea for a few reasons, you can't put your center channel speaker behind them for one thing

Accoustically transparent screens allow home theater and cinema owners to locate speakers behind them all of the time.

-sc