AVS Forum banner

101 - 120 of 696 Posts

·
Registered
Joined
·
15,483 Posts
Hell, half the time studios can't even do 1920x1080p on Blu-ray correctly without screwing something up!


What makes you think 4k (we'll never see real 4k for consumers, however) would be any better?


I'd love to see true 12 bit color encoding, added support for progressive video encoding of 30 fps and 48 fps content, and 21x9 enhanced content support (so wider films can benefit from the extra resolution too)...


More so than 3D...


But how will it get delivered to us? The current internet infrastructure isn't good enough by far for streaming of that quality (crappy 1080p at 3 Megabits/sec or less, and no lossless 24 bit audio ... no thank you!!) and we'd need a new and improved pre-recorded storage medium to handle the extra bandwidth and added storage space. It'll be HD-DVD vs. Blu-ray all over again.


And then what will Universal do? Release 1080p upconverts from DVD masters and pawn it off as 4k with the added "benefit" of heavy DNR and EE?
 

·
Registered
Joined
·
254 Posts

Quote:
Originally Posted by Kris Deering /forum/post/20913999


I've heard there is a push for 4K content for consumers on an upcoming format. Guess we'll just have to see.

Why not just give us the same format as the theaters use? Save a lot of money creating masters for new movies. All they have to do is come up with a medium that can hold that much data in a convenient form factor. A trivial task, really
Although maybe the theater format is too limiting?
 

·
Premium Member
Joined
·
11,987 Posts

Quote:
Originally Posted by Chuck Anstey /forum/post/0



Why not just give us the same format as the theaters use? Save a lot of money creating masters for new movies. All they have to do is come up with a medium that can hold that much data in a convenient form factor. A trivial task, really
Although maybe the theater format is too limiting?

Way too cost prohibitive given the file sizes. Memory would have to come WAY down for that. Especially for 4K content.
 

·
Registered
Joined
·
7,004 Posts

Quote:
Originally Posted by Chuck Anstey /forum/post/20914096


Why not just give us the same format as the theaters use? Save a lot of money creating masters for new movies. All they have to do is come up with a medium that can hold that much data in a convenient form factor. A trivial task, really
Although maybe the theater format is too limiting?

Back in 2009 TDK announced the development of a 10-layer Blu-ray disc that holds 320 GB which is certainly more than adequate for 4K video as long as the maximum data transfer rate off of the disc were also increased. Other optical disc formats, not based on Blu-ray, have been claimed to hold up to a 1 TB. In any case, either an upgraded Blu-ray format or a new optical disc format could be produced that would technically support the requirements for 4K and even 4K-3D video. However, the process to get industry agreement on a single new standard (i.e., no repeat of BD vs. HD-DVD please) and to also convince the movie studios to release movies, would be a several year effort. Also with 99% (my guess) of HDTV's currently being sold having a screen size of 60" or less (and of course none are 4K ready), I'm not really certain there would be much a market for a premium super HD format. Maybe in another 5 to 10 years there will be inexpensive 80" and 90" flat panel 4K displays that would make a consumer 4K disc format more viable in the marketplace.
 

·
Registered
Joined
·
254 Posts

Quote:
Originally Posted by Ron Jones /forum/post/20914207


Back in 2009 TDK announced the development of a 10-layer Blu-ray disc that holds 320 GB which is certainly more than adequate for 4K video as long as the maximum data transfer rate off of the disc were also increased. Other optical disc formats, not based on Blu-ray, have been claimed to hold up to a 1 TB. In any case, either an upgraded Blu-ray format or a new optical disc format could be produced that would technically support the requirements for 4K and even 4K-3D video. However, the process to get industry agreement on a single new standard (i.e., no repeat of BD vs. HD-DVD please) and to also convince the movie studios to release movies, would be a several year effort. Also with 99% (my guess) of HDTV's current being sold having a screen size of 60" or less, I'm not really certain there would be much a market for a premium super HD format.

I wasn't saying it was a good idea. I was responding to Kris' post that a new format may be in the works and wondered why create a new one instead of using the existing one. I mean if all we are going to get is a hyper-compressed version then it won't really be 4K other than it technically extracts out to 4K horizontal pixels but the actual luminance and chroma resolution will be much less. A high quality 4K video takes up what it takes up and squeezing the crap out of it nullifies any advantage of 4K.


Edit - When I am talking about format, I am talking about the actual encoding format not the physical media format so we are probably talking about two different things. I see the media as simply a storage system to hold the number of required bytes for the encoding format. So in that sense "Blu-ray" isn't an encoding format but a media file system.
 

·
Registered
Joined
·
1,003 Posts

Quote:
Originally Posted by darinp2 /forum/post/20912966


That hasn't been my experience. My experience is that it has been much more difficult to find scenes where the higher ANSI CR was visible compared to the higher on/off CR.

That's interesting. Back when we did it, we didn't have to look long for demo material that showed the higher ANSI CR. This was in a dark room too, no perfect batcave, as there were some brown racks in the room.


Quote:
I did with a Samsung A900B I had purchased against a friend's JVC RS20.

What's the ANSI CR of the Samsung? Is it in the >500:1 range?


Quote:
I would be interested in what testing you have done with a good DLP against one of the JVCs in the same room with the same material, as far as models, setup, and results.

All projectors were calibrated and as far as possible matched in brightness. Screen was a 1.0 gain 10' wide screen (16:9, not scope). No a-lenses were used, nor zooming.


We used for comparison: JVC 950, 990, Sim2 Lumis, Mico 50, Sony VW90 and a Barco DP1200 (on a bigger screen). This was when we had all together. Later comparisons were then performed on selected projectors, for example JVC X3 vs. Sim2 Uno and Mico 50.


The results are pretty much as expected. JVC is king of black, but lacked during real world scenes, that's where the VW90 was better and the Sim2s stood out. I think the Barco is in a league of its own and I'd recommend it for everyone who needs the brightness. There are little alternatives when alot of light is required (DPI, maybe Sim2 Teatro). But on/off suffers with these light cannons.



Quote:
Have you ever tried purposely reducing ANSI CR of the images like by putting a white sheet at the back of the room where the viewer shouldn't know whether the sheet is up or not and then seeing whether they can tell whether it is up or not by watching the images off the screen?

Not with a white sheet, but there were two white speaker boxes in the room, which we removed from the room and later moved back in and out of the room for some basic tests. This had an impact on what we saw on the screen of course, but I found the difference going from a JVC to a Sim2 much more dramatic as far as ANSI CR goes.


I'll run another test once the Nero series arrived and I'll try to get my hands on the new dVision 35 from DPI as well. Will put those side-by-side with a Lumis and see how they are. The JVCs have some issues for me, which I can't live with, the Mitsubishi on the other hand is a pretty nice projector. So maybe I'll make that a part of the equation.
 

·
Registered
Joined
·
528 Posts
I am trying to figure out how they achieve 4k x 2k with a single pixel shift in just one direction. They need to generate 300% more pixels, and they only appear to do 100% more. The best I can see is 2 * 1080p, which is definitely not 4k.
 

·
Registered
Joined
·
23,130 Posts
Quote:
Originally Posted by amt
I am trying to figure out how they achieve 4k x 2k with a single pixel shift in just one direction. They need to generate 300% more pixels, and they only appear to do 100% more. The best I can see is 2 * 1080p, which is definitely not 4k.
Just going to throw this out there because I was thinking about it last night. With JVCs wobulation technique, the pixels are overlapped, each shifted pixels covers 4 unshifted pixels. If we think about this a bit like DLP, how you can use different colors sequentially to create additional colors...


Would it not be possible to use the overlap to effectively make 4 different pixels? Let's see if I can ascii art an example, say we've got an unshifted subframe:

Code:
Code:
1 1 | 2 2
 1 1 | 2 2
 - - - - -
 3 3 | 4 4
 3 3 | 4 4
Now lets say the shifted subframe:
Code:
Code:
5 5 | 6 6
 5 5 | 6 6
 - - - - -
 7 7 | 8 8
 7 7 | 8 8
What happens if you combine them? Well it's additive, and overlaps by a quarter acording to the pictures so:

(assume shifted down and to the right)
Code:
Code:
1 1 | 2 2
 1 6 | 7 8
 - - - - -
 3 8 | 910
 3 10|1112
So you see you by just shifting in one direction, even the same frame (unchanged) you create 4 times the number of unique "pixels" as the original frame.


Now the question is how do you use that to make 8 megapixels each exactly the color you want. That I really don't want to think about, but I'd guess through some sort of flashing you could make it work.
 

·
Registered
Joined
·
528 Posts
If you had enough flashing, perhaps it could be done (4x the pixels). However, they seem to indicate that there is only one "flashing" for a 60Hz signal. The other problem, as you indicate, is that I do not think they can get precise control over the new colors/luminance of these "sub" pixels. Honestly, this does not sound that exciting to me at all. Just another gimmick we will look back on and say "why did they do this again?".
 

·
Registered
Joined
·
8,253 Posts
I don't think so. JVC uses something like PWM/dithering to drive their pixels, anyway. They're driven digitally, not analog. So the whole pixels are drawn by "flashing" them on/off very quickly, anyway. I guess that with the right algorithm the final end result could come quite near to true 4K. The cost of this approach is less time per pixel, though, which should increase the dithering noise.
 

·
Registered
Joined
·
79 Posts
Quote:
Originally Posted by Dan Hitchman
Hell, half the time studios can't even do 1920x1080p on Blu-ray correctly without screwing something up!


What makes you think 4k (we'll never see real 4k for consumers, however) would be any better?


I'd love to see true 12 bit color encoding, added support for progressive video encoding of 30 fps and 48 fps content, and 21x9 enhanced content support (so wider films can benefit from the extra resolution too)...


More so than 3D...


But how will it get delivered to us? The current internet infrastructure isn't good enough by far for streaming of that quality (crappy 1080p at 3 Megabits/sec or less, and no lossless 24 bit audio ... no thank you!!) and we'd need a new and improved pre-recorded storage medium to handle the extra bandwidth and added storage space. It'll be HD-DVD vs. Blu-ray all over again.


And then what will Universal do? Release 1080p upconverts from DVD masters and pawn it off as 4k with the added "benefit" of heavy DNR and EE?
Hollywood is always screwing something up. Just look at all the DVD "re-masterings" they have done that looks crappier than my old S-VHS tapes. The same with BD, they try to get us buy the same old movies again as we are hoping to get a better experience, yea right. Now the same train is leaving the station, the 3D train where we will be screwed once more. The 3D stuff I've seen that is just a new post-production-3d-enchanser I could let my projector do for me with for free!!


4k might go the same way, I'm afraid. Sure there will be really good 4k productions, avatar 2 in 4k or lets see what Peter Jackson has in his hat.. But Hollywood and the electronics industry will start their massive campaigns to convince consumers that "more pixels is better" and we will have our own digital camera "megapixel war"..
 

·
Registered
Joined
·
4,940 Posts
Quote:
Originally Posted by Kris Deering
I've heard there is a push for 4K content for consumers on an upcoming format. Guess we'll just have to see.
Redray provides 'visually lossless' 4K at Says Red...
 

·
Registered
Joined
·
254 Posts
Quote:
Originally Posted by madshi
I don't think so. JVC uses something like PWM/dithering to drive their pixels, anyway. They're driven digitally, not analog. So the whole pixels are drawn by "flashing" them on/off very quickly, anyway.
This doesn't make any sense to me but it could be my misunderstanding. I thought the reason for motion blur on panels is that they cannot transition from fully open to fully closed very quickly. Therefore the PWM isn't pulsing the pixel fully open / fully closed like a DMD device does with mirrors but rather the voltage that is trying to drive the panel to be 50% blocked light is modulated between full voltage and zero voltage half the time to generate the equivalent of 50% analog voltage and the pixel is continually half transparent. Can anyone explain the exact workings?
 

·
Registered
Joined
·
5,658 Posts
Quote:
Originally Posted by Chuck Anstey
This doesn't make any sense to me but it could be my misunderstanding. I thought the reason for motion blur on panels is that they cannot transition from fully open to fully closed very quickly. Therefore the PWM isn't pulsing the pixel fully open / fully closed like a DMD device does with mirrors but rather the voltage that is trying to drive the panel to be 50% blocked light is modulated between full voltage and zero voltage half the time to generate the equivalent of 50% analog voltage and the pixel is continually half transparent. Can anyone explain the exact workings?
I'll just paste this from a post I made last November. Hope it helps.


The DILA device does not do anything with light intensity. All it does is rotate the polarization by an amount that is proportional to the intended intensity (brightness) of the pixel. I'll make up some numbers to illustrate this. Light from the bulb passes thru a polarizing filter. Let's say that it's orientation is 0 degrees. That polarized light is then passed on to the DILA panel where is passes thru liquid crystal. It reflects off the pixel electrode and passes back thru the liquid crystal. During the round trip thru the liquid crystal light is still polarized but it gets rotated by an amount which is proportional to the [intended] intensity of that pixel. IOW, light exiting a DILA pixel is still polarized but its orientation is now dependent on signal intensity. In essence, very little light has been lost. No image can be observed at this point. The "only thing" the DILA device did was vary the orientation of the polarized light on a pixel by pixel level -- it does not change the intensity of the light. Light from some pixels may still be oriented at 0 degrees while light from other pixels may be orientated at 90 degrees, (or anywhere in between). As mentioned, up to this point light intensity hasn't changed (only it's orientation) so the trick now is to turn this "altered orientation" into an image. The image is finally created by passing the light thru another linear polarizer. Light that is orientated in the same plane as the linear polarizer passes thru with "minimal" light loss. Light that orientated 90 degrees to the linear polarizer passes thru with lots of loss.


I'll add this...


With a D-ILA device, each Pixel is driven with a PWM waveform. The duty cycle of that waveform is determined by the signal intensity. That in turn varies the voltage across the LC material which in turn varies the amount of polarization rotation. The DC bias across the pixel must be zero though so the average value of the PWM waveform must be 0 VDC.
 

·
Registered
Joined
·
254 Posts
Quote:
Originally Posted by Geof
I'll add this...


With a D-ILA device, each Pixel is driven with a PWM waveform. The duty cycle of that waveform is determined by the signal intensity. That in turn varies the voltage across the LC material which in turn varies the amount of polarization rotation. The DC bias across the pixel must be zero though so the average value of the PWM waveform must be 0 VDC.
So my general understanding was correct. The PWM is on the input side to result in a desired rotation angle. JVC in theory could have used 0V to be degrees, 1V to be 90 degrees and 0.5V to be 45 degrees. Instead they PWM a 50% duty cycle to rotate 45 degrees. It does not constantly switch between 0 and 90 degrees half the time to simulate 45 degrees so there is no light flashing.
 

·
Registered
Joined
·
23,130 Posts
Quote:
Originally Posted by studiox_swe
4k might go the same way, I'm afraid.
While I understand what you're saying, I'd rather give them a better "pallet" to paint with for those who do do it well than to limit everybody. 4K (or higher bit depths, or any other improvement) won't make a bad transfer worse, but it will allow good ones to be better.


So to me it makes no sense to say "who cares, they'll screw it up anyway".


-edit,


Well Alan just posted this in the CEDIA pictures thread:



Looks intriguing.
 
101 - 120 of 696 Posts
Top