AVS › AVS Forum › Blu-ray & HD DVD › HDTV Software Media Discussion › Which 3D format do you think the BDA will pick for Blu-ray?
New Posts  All Forums:Forum Nav:

Which 3D format do you think the BDA will pick for Blu-ray? - Page 9

post #241 of 275
Quote:
Originally Posted by bdraw View Post

I was able to get Panasonic's PR department to setup a technical call about 3D and I asked about the 3:2 pull down issue, and after a day they got back to me and said that it hadn't been decided yet. So right now all they will say is that the system can do full 1080p60 per eye, but aren't ready to say how they'll present 1080p24 material.

The other interesting thing that was mentioned was that 3D 1080p60 doesn't take up twice the bits of 2D 1080p60 because the right and left images are so similar that the format compresses well.

But it doesn't decode 1080p60 stereoscopic content though , it only outputs in it, just like current players can output 1080p60 2D but not decode it. If they're not taking about 1080p24, they're probably just talking about 1080i30.
post #242 of 275
Quote:
Originally Posted by 8:13 View Post

Since they are using 120Hz plasma monitors for 24p 3D movies, they will have to be 3:2 Pulldown.

Boy my head hurts after reading through this thread, but I've "drawn" my own experimental frame display diagrams and come to the same conclusion as you, 8:13. There is really no way around some kind of pulldown alteration of a 24p source being shown to two eyes when using a 120Hz TV. For each frame, one eye is going to have to get three flashes while the other gets two, or the display of the frames themselves will have to be staggered as you drew them. Either way I can't imagine it'll be very pretty.

All I know for sure right now is that before I personally take the plunge into 3D, I want to wait for a 480Hz TV, that is guaranteed to be able to accept and display 2D and 3D recorded at 48p and 60p (that's probably inevitable for the future). Of course it will also have to be able to show all combinations of 24p, 30p, 48p, 60i, and 60p in 2D and 3D with no pulldown judder. 480Hz should make that perfectly doable since it's evenly divisible by all combos.

24p 2d = 20 flashes per frame.
24p 3d = 10 flashes per eye per frame.
30p 2d = 16 flashes per frame.
30p 3d = 8 flashes per eye per frame.
48p 2d = 10 flashes per frame.
48p 3d = 5 flashes per eye per frame.
60p 2d = 8 flashes per frame.
60p 3d = 4 flashes per eye per frame.

After doing the math, I definitely argue that 480Hz is a better "grail" than 240Hz.
post #243 of 275
Ok, someone explain...

A 240hz display, using the SIDE-BY-SIDE 3D method would mean 120hz for both the left and right eye.

120hz (for both left and right) = 24p x 5.

Why is anyone thinking 3:2 pulldown is needed? It isn't needed for a true 120hz display displaying 24p 2D content. A true 120hz display can show it natively (x5). Now, a 60hz display had to perform 3:2 pulldown, we all know that.
post #244 of 275
Quote:
Originally Posted by taz291819 View Post

Ok, someone explain...

A 240hz display, using the SIDE-BY-SIDE 3D method would mean 120hz for both the left and right eye.

120hz (for both left and right) = 24p x 5.

Why is anyone thinking 3:2 pulldown is needed? It isn't needed for a true 120hz display displaying 24p 2D content. A true 120hz display can show it natively (x5). Now, a 60hz display had to perform 3:2 pulldown, we all know that.

As far as I know, if you're using the shutter glasses method, you need a 240hz TV to avoid 3:2 pull-down judder with 24p stereoscopic content.

If you're using the polarized glasses & TV method you only need a 120hz TV with 24p stereoscopic content to avoid 3:2 pull-down judder, if both left & right images are shown at the same time.
post #245 of 275
Quote:
Originally Posted by Joe Bloggs View Post

As far as I know, if you're using the shutter glasses method, you need a 240hz TV to avoid 3:2 pull-down judder with 24p stereoscopic content.

If you're using the polarized glasses & TV method you only need a 120hz TV with 24p stereoscopic content to avoid 3:2 pull-down judder, if both left & right images are shown at the same time.

That's exactly how I see it. So I can't fathom why we would need 480Hz displays in order to avoid judder. With 240Hz, you don't need 3:2 pulldown.
post #246 of 275
Quote:
Originally Posted by 8:13 View Post

The 3:2 pulldown I showed a few posts up is to how the panasonic plasma they have in thei tour truck has 3:2 pulldown when displaying 24p in frame sequential 3D.

Gotcha, you were just explaining the 120Hz Plasma that Panny was using. Makes sense for 120Hz.
post #247 of 275
Quote:
Originally Posted by taz291819 View Post

That's exactly how I see it. So I can't fathom why we would need 480Hz displays in order to avoid judder. With 240Hz, you don't need 3:2 pulldown.

It depends whether James Cameron or other people ever make any movies in 48p stereoscopic and you're using the LCD shutter glasses method
post #248 of 275
Quote:
Originally Posted by Joe Bloggs View Post

It depends whether James Cameron or other people ever make any movies in 48p stereoscopic and you're using the LCD shutter glasses method

I don't think they're going to change everything because of one Producer/Director. Don't get me wrong, if there is no price difference between 240hz and 480hz, may as well go 480hz.
post #249 of 275
Quote:
Originally Posted by Jamie E View Post

Boy my head hurts after reading through this thread, but I've "drawn" my own experimental frame display diagrams and come to the same conclusion as you, 8:13. There is really no way around some kind of pulldown alteration of a 24p source being shown to two eyes when using a 120Hz TV. For each frame, one eye is going to have to get three flashes while the other gets two, or the display of the frames themselves will have to be staggered as you drew them. Either way I can't imagine it'll be very pretty.

All I know for sure right now is that before I personally take the plunge into 3D, I want to wait for a 480Hz TV, that is guaranteed to be able to accept and display 2D and 3D recorded at 48p and 60p (that's probably inevitable for the future). Of course it will also have to be able to show all combinations of 24p, 30p, 48p, 60i, and 60p in 2D and 3D with no pulldown judder. 480Hz should make that perfectly doable since it's evenly divisible by all combos.

24p 2d = 20 flashes per frame.
24p 3d = 10 flashes per eye per frame.
30p 2d = 16 flashes per frame.
30p 3d = 8 flashes per eye per frame.
48p 2d = 10 flashes per frame.
48p 3d = 5 flashes per eye per frame.
60p 2d = 8 flashes per frame.
60p 3d = 4 flashes per eye per frame.

After doing the math, I definitely argue that 480Hz is a better "grail" than 240Hz.

I made some calculations from the 3D thread in this forum I made and here are the results below. This is without 3:2 pulldown.

Section f

Code:
24p........................ = 23.97600 FPS.. = 41.708375 Milliseconds Time between frames
NTSC's....................... 29.97......... = 33.3667 Milliseconds Time between frames
48p........................ = 47.95200 FPS.. = 20.8541875 Milliseconds Time between frames
NTSC's....................... 59.94 FPS..... = 16.68335 Milliseconds Time between frames

Code:
1000 / 23.97600 = 41.708375
1000 / 29.97    = 33.3667
1000 / 47.95200 = 20.8541875
1000 / 59.94    = 16.68335
There's 1000 milliseconds in a second.
MS / FPS = MS: 1000 ms / FPS = Time between frames

"The REAL D system uses triple flash to provide the best motion rendition possible. Use of triple flash puts the refresh rate above the normal flicker fusion threshold for humans, providing smoother motion. The triple flash approach also makes the left and right eye images to appear closer in time, giving significantly less motion induced parallax errors and therefore more comfortable motion rendition."

http://www.edcf.net/edcf_docs/real-d.pdf

Flicker fusion threshold

"If the frame rate falls below the flicker fusion threshold for the given viewing conditions, flicker will be apparent to the observer, and movements of objects on the film will appear jerky. For the purposes of presenting moving images, the human flicker fusion threshold is usually taken as 16 hertz (Hz)."
LINK

MS / HZ = MS: 1000 / 16 = 62.5

• What does this mean?
Code:
1000ms / fps = ms > 1000ms / 16hz =  62.5ms = fps:ms faster than 62.5ms: No flicker
1000ms / fps = ms < 1000ms / 16hz =  62.5ms = fps:ms slower than 62.5ms: flicker
• In Actual practice though, for 3D, the Hz to use is the one Real D uses: "triple flash", as this is standard in digital cinema = 72Hz 2D (71.928fps) per eye.
Code:
1000ms / fps = ms > or = 1000ms / 72hz
1000ms / fps = ms > or = 1000ms / 72hz =  13.8888889ms = fps:ms faster than or equal too 13.8888889ms: if equal too it's decent quality 3D(but can be better).
On a 240Hz monitor, 60p video using 2:2 pulldown making it 120p video
1 000ms / 119.88fps = 8.34167501ms
8.34167501ms is faster than 13.8888889ms so there should be no flicker when using a 240hz monitor and 60p tv in stereoscopic 3D.

This is from post #2.
post #250 of 275
The Panasonic demo at Cedia used the Xpand glasses which can do double flash 96 and triple flash 144 for 24fps or double flash 120 for 30fps content.

For someone used to decent cinema 3-D the UP and Grand Canyon sequences definitely lacked Movie Theater presentation quality. There was an obvious cadence jitter going on that completely destroyed the 24fps originated segment of the presentation. On the other hand the Olympic skiing footage shot at 30 or 60fps was breathtaking and perfect. Perhaps the plasma was turned down a bit on brightness I doubt that it was 4.5 ansi lumens. Seating on the second row I did see some crosstalk.
The guy behind me did not let me take pictures so I sneaked these from the hip. To show viewing angle and content shown.




I asked what would be done about 24 fps content and they said that they were in committee and the Hollywood representatives were pushing for 144 hz triple flash accommodation within the format.
post #251 of 275
Thread Starter 
Quote:
Originally Posted by CINERAMAX View Post

I asked what would be done about 24 fps content and they said that they were in committee and the Hollywood representatives were pushing for 144 hz triple flash accommodation within the format.

They might have been referring to the Panasonic 3D displays since from what I have read the 3D Blu-ray player would send the 1080p24 3D video signal and it would be up to the display on how it is shown.
post #252 of 275
Richard that would be fantastic, how sure are you of that? The 103" plasma was definitely doing some 2/3 pullup to 120hz on UP and Grand Canyon.
post #253 of 275
Thread Starter 
Quote:
Originally Posted by CINERAMAX View Post

Richard that would be fantastic, how sure are you of that?

Well in terms of the 3D Blu-ray player I have seen that shown in a Panasonic document, Keith has mentioned the 3D formats supported by Blu-ray, and based on the mandatory 3D formats for HDMI 1.4. As for whether Panasonic is thinking of adding 144 Hz support to their 3D displays that depends on whether what they said was even partly accurate.
post #254 of 275
Panasonic announces UK 3D Blu-ray roadshow, full tour dates revealed

http://www.homecinemachoice.com/blog...ealed+02+10+09
post #255 of 275
Panasonic's 3-D HDTV Among CEATEC Highlights

Quote:


Panasonic's prototype 50-inch 1080p 3-D plasma display, which uses active shutter glasses that enable the viewing of theater-quality images, is set to reach the market next year. No pricing information was offered, but during a press seminar on the technology, Panasonic executives said it would be tagged for a consumer audience. Possible screen sizes have not been decided upon as yet.

The company noted that its new high-speed 3-D drive technology, which Panasonic claims enables rapid illumination of pixels while maintaining brightness, uses crosstalk reduction technology to minimize double-images that occur when left- and right-eye images are alternately displayed.

Panasonic highlighted its work with Hollywood studious to develop its FullHD 3-D technology, which it said works best with plasma displays.

The company noted that unlike analog TV or HDTV, broadcasters will not play a key role in introducing the new format. A 3-D-capable Blu-ray Disc drive, also set for next year and also without a suggested retail as yet, will play the premier role in introducing 3-D.

http://www.twice.com/article/356914-...Highlights.php
post #256 of 275
here is a picture of the 3D BD player along with the shutter glasses that Panasonic is using for their 3D demos:

post #257 of 275
Quote:
Originally Posted by Lee Stewart View Post

here is a picture of the 3D BD player along with the shutter glasses that Panasonic is using for their 3D demos:


The glasses in this article look much cooler...
http://www.electronichouse.com/artic..._dominance/D1/

But the red ones look like they could take a beating, great for families with small children.
post #258 of 275
They look very much like the ones Xpand Cinema uses:

http://www.xpandcinema.com/products/glasses/
post #259 of 275
Quote:
Originally Posted by Lee Stewart View Post

here is a picture of the 3D BD player along with the shutter glasses that Panasonic is using for their 3D demos:


You really should really credit or link the source of images you get off the internet if you are going to repost them.

http://www.engadgethd.com/2009/09/11...ience-eyes-on/
post #260 of 275
Quote:
Originally Posted by Lee Stewart View Post

They look very much like the ones Xpand Cinema uses:

http://www.xpandcinema.com/products/glasses/

That is because they are XpanD glasses.
http://www.xpandcinema.com/news/23/
post #261 of 275
Is the max bitrate for 3D going to be exactly the same as on current 2D players? For compability with current players I suppose it has to (ie. spin the disc at the same speed), but for the same quality isn't 3D FULL HD going to need more max bitrate than current players?
post #262 of 275
Possibility;
3D BD in 2D playback on current machines extracting 24fps @ 1x speed.?
3D BD in 3D playback on future 3D machines extracts 48fps @ 2x speed.?

"Somewhat similar" to the DVD redlaser 1080p HD format; 3x speed ROM, from standard DVD discs for playback on HDDVD machines that had "varispeed" drives.
post #263 of 275
Quote:
Originally Posted by Joe Bloggs View Post

Is the max bitrate for 3D going to be exactly the same as on current 2D players? For compability with current players I suppose it has to (ie. spin the disc at the same speed), but for the same quality isn't 3D FULL HD going to need more max bitrate than current players?

I doubt this will be a problem, but you might have a point. I don't think any current movies come anywhere close to hitting the 40Mbps limit. So there should be plenty of room for 3D, which only uses about 30% more bits than 2D since only the difference of the second frame is stored.
post #264 of 275
Quote:
Originally Posted by bdraw View Post

I doubt this will be a problem, but you might have a point. I don't think any current movies come anywhere close to hitting the 40Mbps limit. So there should be plenty of room for 3D, which only uses about 30% more bits than 2D since only the difference of the second frame is stored.

Actually - it's 50%

Quote:


The second channel stores only the data different from channel one, holding the increase in data volume to about 1.5 times

http://techon.nikkeibp.co.jp/article...081030/160508/
post #265 of 275
Quote:
Originally Posted by bdraw View Post

I doubt this will be a problem, but you might have a point. I don't think any current movies come anywhere close to hitting the 40Mbps limit. So there should be plenty of room for 3D, which only uses about 30% more bits than 2D since only the difference of the second frame is stored.

Look at the video specification thread
http://www.avsforum.com/avs-vb/showt...155731&page=19

http://www.avsforum.com/avs-vb/showthread.php?t=1155731

The maximum video bitrates go over 40 (see the 1 second window graph).
Many titles have average video bitrates over 20 (one I saw is at avg 40).

I'm sure earlier in the thread people said that encoding a second stream (for the difference between the left & the right) would take an average of 50% more bitrate, but that the peak video bitrate needed would be double.

So to maintain the same quality surely you'd need on average 50% more bitrate, but also allow the maximum video bitrate to go up to something like 80 mbit/sec (double the current max video bitrate).
post #266 of 275
Quote:
Originally Posted by Joe Bloggs View Post

Look at the video specification thread
http://www.avsforum.com/avs-vb/showt...155731&page=19

http://www.avsforum.com/avs-vb/showthread.php?t=1155731

The maximum video bitrates go over 40 (see the 1 second window graph).
Many titles have average video bitrates over 20 (one I saw is at avg 40).

I'm sure earlier in the thread people said that encoding a second stream (for the difference between the left & the right) would take an average of 50% more bitrate, but that the peak video bitrate needed would be double.

So to maintain the same quality surely you'd need on average 50% more bitrate, but also allow the maximum video bitrate to go up to something like 80 mbit/sec (double the current max video bitrate).

Standard Technology

Quote:


And, in fact, the technologies proposed by Panasonic for 3D imagery storage, transfer, etc, all utilize existing standard technology. Image encoding uses the two-channel encoding function implemented in Moving Picture Coding Experts Group Phase 4 Advanced Video Coding (MPEG-4 AVC) H.264. The second channel stores only the data different from channel one, holding the increase in data volume to about 1.5 times. The HDMI standard is used to transfer data from the player to the television, with left- and right-eye images alternated in single-field (single-frame) units. All we have to do is define a flag to identify image data, equipment and other elements supporting 3D imagery. We really don't need any other major changes, explained Hiroshi Miyai, director, High Quality AV Development Center of Panasonic.

from my link above
post #267 of 275
So they admit that 3D will require 50% more data on average, they don't say whether or not it will require 100% more max data (peak data rate), and they're not going to make any changes to increase the data rates, therefore picture quality for 3D releases, including the 2D version which will just be stream 1, will be lower on average than current titles.
post #268 of 275
Quote:
Originally Posted by Joe Bloggs View Post

So they admit that 3D will require 50% more data on average, they don't say whether or not it will require 100% more max data (peak data rate), and they're not going to make any changes to increase the data rates, therefore picture quality for 3D releases, including the 2D version which will just be stream 1, will be lower on average than current titles.

That addmission has been well known for quite some time.

And the PQ of the 2D extraction will not be crippled at all. it will be 1080x24P at the normal bite rate used for BD.
post #269 of 275
Quote:
Originally Posted by Lee Stewart View Post

That addmission has been well known for quite some time.

And the PQ of the 2D extraction will not be crippled at all. it will be 1080x24P at the normal bite rate used for BD.

Plus, the encoders get better with time. I'd bet a film encoded a year and a half ago could have a substantially lower bitrate with the same PQ today. Based on the numbers, a film with a 25Mbp/s average bitrate in 3D could fit on a single BD50. Put extras and the "Digital-Copy" on a separate disc.
post #270 of 275
Quote:
Originally Posted by Lee Stewart View Post

Actually - it's 50%

http://techon.nikkeibp.co.jp/article...081030/160508/

Great article, thanks.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: HDTV Software Media Discussion
AVS › AVS Forum › Blu-ray & HD DVD › HDTV Software Media Discussion › Which 3D format do you think the BDA will pick for Blu-ray?