or Connect
AVS › AVS Forum › Blu-ray & HD DVD › HDTV Software Media Discussion › Industry Insiders Q&A Thread: only Questions to insiders please
New Posts  All Forums:Forum Nav:

Industry Insiders Q&A Thread: only Questions to insiders please - Page 60  

post #1771 of 4623
Thank you very much for your fast reply!

Quote:
Originally Posted by Tom McMahon

"The word" is totally inaccurate. Think about it: decoding a progressive frame and outputting is easier than decoding and trying to output while fiddling with interlace nonsense. Doing all the 3:2 stuff is actually *MORE* complex and potentially more memory bandwidth intensive that just doing the basic 24P thing for film content.
I'm aware of that. So I'm very puzzled by the fact that Samsung's BD-P1000 player included an extra deinterlacing/inverse telecine Faroudja Genesis FLI8638-LF chip in order to "deinterlace" the 1080i/60 output of your chip while per your information they could have used the possible progressiv output capabilities in the first place. Why would they have done that??? This seems to be the *MOST* complex solution ever -> going 24p-> 60i-> 60p with two different chips involved. I've read a several times in independent press articles that the reason for that is the inability of the BCM7411 chip to provide 1080p output by itself (be it 24Hz or 60Hz). Perhaps you should do something against this misinformation as it shows your product in a bad light.



Quote:
Originally Posted by Tom McMahon
I should add a few more things:

Depending on which blue-laser Standard you favor, there may be differences in how the video gets processed after the video decoders but before the HDMI connector. While the main program film content is *always* progressive (to answer another question on this list), the PiP channel may in fact be interlaced (unfortunately). This creates a situation where the player must comingle progressive and interlaced streams (not to mention presentation graphics etc.).

There is no easy way out of that box - once you introduce interlace, with it's spatial and temporal screwups, you're dead meat. And if you want to output progressive after that, you *must* deinterlace. Harsh reality.
About this "harsh reality": We agree that decoding and outputting the 1080p/24 stream at its native rate is the best choice. Let's say an upcoming Toshiba HD-DVD player is supporting 1080p/24 output with movie content. What happens when I activate the PIP/IME feature??? In order to maintain 24p output the IME stream has to be deinterlaced and 5.97 fps have to be dropped. Is this situation handled by the HD-DVD spec or is it up to the individual implementation of the CE company?


And please answer the most important question: Can you confirm that Toshiba can potentially provide a simple firmware upgrade for its BCM7411 equipped HD-A(X)1 in order to activate/utilize the 1080p/24 output capability of your chip??? Or asked the other way around - Is it from your point of view and considering BCM7411's design technically possible to do so?

Thank you very much!
post #1772 of 4623
Quote:
Originally Posted by TheLion
Thank you very much for your fast reply!
...So I'm very puzzled by the fact that puzzled by the fact that Samsung's BD-P1000 player....Why would they have done that???

And please answer the most important question: Can you confirm that Toshiba

Thank you very much!
I cannot speak for Samsung or Toshiba. Whatever they want we will supply. Its all just sand and bits.
post #1773 of 4623
Quote:
Originally Posted by Tom McMahon
I cannot speak for Samsung or Toshiba. Whatever they want we will supply. Its all just sand and bits.
... and greenbacks :D

Still, I think this means the Toshiba should be able to provide 1080P/24 using the HD-A1.

- Rich
post #1774 of 4623
...
post #1775 of 4623
Quote:
Originally Posted by Tom McMahon
* All new video cameras are progressive in the sense that they capture a whole frame at once in their silicon sensors, even if some of them serially scan out the charges in their photon-receptor cells. They still sample "reality" all at once in a 2D array (Craig Todd's so-called "splat scan"). In many cases the fact that the information is sent out in odd/even rows of pixel samples is irrelevant. It is reconstructed into a progressive frame at the other end before you need to know about it.
I'm not 100% how to understand this. Sports is still recorded in 1080i60, isn't it? Are you saying that soon sports will be recorded in 1080p24 or 1080p30 or 1080p60?
post #1776 of 4623
Quote:
Originally Posted by RichB
... and greenbacks - Rich
Without pointing any fingers at any one CE company or format, you are right. I mean, at the end of the day, SoC companies like Broadcom make product to meet our customers' demands. We all produce a range of silicon and software stacks ("reference designs") that support either or both formats.

On a case by case basis, we may provide just the silicon, the motherboard design, parts thereof, some of the software, or all of the software (sometimes through partners).

We may produce just the video decoding component (as in the Toshiba and some yet-to-be-announced designs) or more comprehensive solutions as you see (and will see) in other players.

We need to support both BD-J and iHD implmentation issues running native on our chip.

All of the aobve speaks to the issue of why hardcore silicon makes sense going forward. While general-purpose processors can achieve some of the above some of the time (sometimes with GPU coprocessors), *NONE* of them can achieve all of the above all of the time at competetitive price points (or power dissipation).

Sometimes a bit of sand makes all the difference in the world....
post #1777 of 4623
Quote:
Originally Posted by madshi
I'm not 100% how to understand this. Sports is still recorded in 1080i60, isn't it? Are you saying that soon sports will be recorded in 1080p24 or 1080p30 or 1080p60?
Film is always progressive, - done deal. Film is a hugh percentage of what we see in movies and also on TV for episodic televison still.

ABC and Fox (60P) live TV and other network origination (sometimes 50P elsewhere) are all progressive now.

I omit the names of the luddites.
post #1778 of 4623
Quote:
Originally Posted by Tom McMahon
Film is always progressive, - done deal. Film is a hugh percentage of what we see in movies and also on TV for episodic televison still.

ABC and Fox (60P) live TV and other network origination (sometimes 50P elsewhere) are all progressive now.

I omit the names of the luddites.
Just to clarify, ABC and Fox transmit 720HP/60. The other do 1080i.

- Rich
post #1779 of 4623
Quote:
Originally Posted by Cjplay
The HDMI supports multi-channel to the receiver according to Sapphire's press release. Therefore, you probably need an HDMI receiver with video pass-through/output. But, there's always the analog connections on an Audigy or M-Audio card. It works for the Toshiba.

Sapphire PR for HDMI card

Cjplay.

Thanks Cjplay, I understand you need a HDMI receiver, but what I'm wondering is, will you need a "content protection" sound card in order to get TrueHD/DTS-HD over HDMI? What confuses me is, CE devices cannot do TrueHD/DTS-HD over SPDIF, but HDMI video cards only have SPDIF passthrough like the Sapphire PIR doc mentions.

Can you confirm that analog output from a regular soundcard that supports the bitrate and sampling rate of TrueHD/DTS-HD would work like the Toshiba? From what I've read, both camps are very "scared" of open systems like a PC, and from what I read of MS's Vista documentation the audio path is encrypted as well, which sounds like a "HDCP-like requirement for Audio.

Maybe Amir can comment as he has an employee that runs a HD-DVD laptop setup?
post #1780 of 4623
Quote:
Originally Posted by DaViD Boulet
pip and "features" comingled with primary 1080p video streams should *always* be converted to the primary feature's frame-rate prior to merging. IMO all playback hardware... HD DVD and BD... should be designed this way regardless of disc-encoding strategy.

If I select 1080p24 output from my HD box, that's what I should get when watching a 1080p24 movie... regardless of the pip mode for Peter Jackson's commentary.
Note that the BD spec requires the secondary (PiP) video to match the framerate and interlace mode of the primary video.

- Talk
post #1781 of 4623
Quote:
Originally Posted by Talkstr8t
Note that the BD spec requires the secondary (PiP) video to match the framerate and interlace mode of the primary video.

- Talk
Right. So if you want/demand "1080P" output what would you like to happen in that case?
post #1782 of 4623
Quote:
Originally Posted by Tom McMahon
...If it were up to me I would omit all the PiP and interactive crap and just do great video and great audio on these discs and leave it at that!
Wow!
And I thought it's just me thinking this way.
Nice to have company!

Diogen.
post #1783 of 4623
Amir, can you comment on this report from MS Senior PM Steve Riley:

"Next-generation high definition content will not play in [32-bit Windows Vista] at all,†said Riley. "This is a decision that the Media Player folks made because there are just too many ways right now for unsigned kernel mode code [to compromise content protection]. The media companies asked us to do this and said they don’t want any of their high definition content to play in x32 at all, because of all of the unsigned malware that runs in kernel mode can get around content protection, so we had to do this."
post #1784 of 4623
I've been away from these Blu-Ray and HD-DVD forums for a long time, but came back PRECISELY because of this announcement. Congratulations everyone, unless you upgraded your system VERY recently, you're in for another round! For example, I don't think the Core Duo can handle Vista 64, but the Core 2 Duo can... but please correct me if I am wrong.

Then of course there is this...

I had honestly figured that HD-DVD's head start and PS3's delays were costing Blu-Ray the race. And at this point I'd just like to see a winner despite my technical preference for Blu-Ray. But I guess you ought never underestimate the ability of the market leader to shoot itself in the foot! Hope they can still stumble across the tape.
post #1785 of 4623
Ouch.

The upside is, pretty much any processor you want to buy right now is 64-bit capable. The downside is, upgrading to Vista for HTPC people with 32-bit processors just became a lot less useful.

Edit to add response to Michael:
Core Duo (Yonah) is indeed 32-bit. Core 2 Duo is 64-bit.
post #1786 of 4623
Thanks for the edit. As of today, no HTPC manufacturer with integrity ought to sell HTPC's without 64-bit support---at least without a BOLD disclaimer about their inability to run the new HD formats under Vista.
post #1787 of 4623
Quote:
Originally Posted by Brian Miller
Amir, can you comment on this report from MS Senior PM Steve Riley:

"Next-generation high definition content will not play in [32-bit Windows Vista] at all,†said Riley. "This is a decision that the Media Player folks made because there are just too many ways right now for unsigned kernel mode code [to compromise content protection]. The media companies asked us to do this and said they don’t want any of their high definition content to play in x32 at all, because of all of the unsigned malware that runs in kernel mode can get around content protection, so we had to do this."
Yes, I will shortly. I am trying to find out what is going on as we have made no policy changes at all.
post #1788 of 4623
Well, in many cases (not all), Merom (Core 2 Duo mobile chip) can be dropped into the Yonah socket with no other changes. So there's still an upgrade path available, but it means cash you didn't expect to have to spend... :(
post #1789 of 4623
Oh, I didn't realize that. That will simplify things for a lot of people whose processors are soldered in. Honestly, that's not so bad for those people for whom that truly is the case.
post #1790 of 4623
Quote:
Originally Posted by Michael Grant
Oh, I didn't realize that. That will simplify things for a lot of people whose processors are soldered in. Honestly, that's not so bad for those people for whom that truly is the case.
Yeah, people with surface-mounted Yonah's are kinda screwed unless they have access to SMT equipment or are really good with a soldering iron. Let's wait for a response from Amir before we get out the tar and feathers, though.
post #1791 of 4623
Quote:
Originally Posted by Michael Grant
Then of course there is this...

Would this be the MS concession for Hollywood to get the MMC nailed in AACS 1.0? Is the delay of final agreement result from the fighting over this?
post #1792 of 4623
Quote:
Originally Posted by lymzy
Would this be the MS concession for Hollywood to get the MMC nailed in AACS 1.0? Is the delay of final agreement result from the fighting over this?
Most definitely not. There is nothing of this kind going on here guys. AACS organization creates cross platform specifications. It can not depend on anything we do or don't do in Vista.....
post #1793 of 4623
Quote:
Originally Posted by amirm
Yes, I will shortly. I am trying to find out what is going on as we have made no policy changes at all.
Any comments on HD Audio formats on the PC?
post #1794 of 4623
Vista/32 was never going to REQUIRE signed drivers, was it? You'd lose AACS playback if you had one that you didn't un-install or you installed one, but that would be a decision for the consumer.

Gary
post #1795 of 4623
Gary---if that's the way it shakes down (i.e., no unsigned drivers for AACS compliance), then that's OK with me. That's just not how it is being reported. I'm encouraged that this catches Amir by surprise, becase it likely means there is more to the story than is being reported.
post #1796 of 4623
Quote:
Originally Posted by Michael Grant
Gary---if that's the way it shakes down (i.e., no unsigned drivers for AACS compliance), then that's OK with me. That's just not how it is being reported. I'm encouraged that this catches Amir by surprise, becase it likely means there is more to the story than is being reported.
Yes. What I said above was how Amir was previously describing things.

They trust everyone they know at first, until it's shown they can't be trusted. But, that REQUIRES a signature (certificate) to really know the driver.

It all made sense until today.

Gary
post #1797 of 4623
Tom,

I'm still confused. When you talk about live-cam content going to progressive it seems you cited companies filming in 720P. Most of us don't consider 1280 x 720p to be the holy-grail of HD image quality... we're assuming if we're talking about the future of HD that we're talking about 1920 x 1080 (or higher) so if you state that live-cam content is going to progressive at 60Hz please clarify if you only mean 720 or if you mean full 1080p resolution. Most of us wouldn't consider it an objective "win" to downres to 1280 x 720 just to attain progressive-encoding... despite the interlaced artifacts in a 1080i broadcast shot from cam.

Also, I'm trying to understand your comments responding to BD's requirement that PIP 2ndary image streams match the primary stream in frame rate and interlace/progressive encoding. Are you suggesting that's a problem in some way? You comment is unclear. Thanks!
post #1798 of 4623
Quote:
Originally Posted by dialog_gvf
Vista/32 was never going to REQUIRE signed drivers, was it? You'd lose AACS playback if you had one that you didn't un-install or you installed one, but that would be a decision for the consumer.

Gary
Agreed. MS should only disable the functionality in Vista32 if the system has unsigned drivers (not that I know how difficult that is; maybe the only option was to disable it altogether, but that would indeed suck; then again, the reports indicate that the studios are the one asking for protected HD discs playback to be disabled, so maybe MS is just acceeding to that demand, even though the better option would be to only disable it if unsigned drivers are present; also, the reports aren't clear as to what is disabled - they seem to imply that HD playback is impossible on Vists32, but they also seem to imply that only 1080p is disabled :confused: :confused: :confused: ).
post #1799 of 4623
Quote:
Originally Posted by DaViD Boulet
Tom,


Also, I'm trying to understand your comments responding to BD's requirement that PIP 2ndary image streams match the primary stream in frame rate and interlace/progressive encoding. Are you suggesting that's a problem in some way? You comment is unclear. Thanks!
I have that same question, (but I thought I was the only one who didn't "get it," so I was gonna keep my mouth shut! Thanks David.)
post #1800 of 4623
Can current SoC solutions process iHD sufficently to do all the things it does as well as the current Toshiba player?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: HDTV Software Media Discussion
This thread is locked  
AVS › AVS Forum › Blu-ray & HD DVD › HDTV Software Media Discussion › Industry Insiders Q&A Thread: only Questions to insiders please