HDMI 1.4 Mystery Solved at CES!
-
- 3D Angel Eyes (Moderator)
- Posts: 4340
- Joined: Sat Dec 22, 2007 3:38 am
HDMI 1.4 Mystery Solved at CES!
However, in the case of the upcoming Nvidia 3DTV Play software, and AMD's HD3D solution, HDMI 1.4A displays are limited to 1280X720P for gaming purposes. With all that horsepower under the hood, why is this barrier so difficult to break?
Read full article...
-
- Certif-Eyed!
- Posts: 642
- Joined: Sat Sep 22, 2007 3:11 am
- Location: Valencia (Spain)
Re: HDMI 1.4 Mystery Solved at CES!
- Fredz
- Petrif-Eyed
- Posts: 2255
- Joined: Sat Jan 09, 2010 2:06 pm
- Location: Perpignan, France
- Contact:
Re: HDMI 1.4 Mystery Solved at CES!
- android78
- Certif-Eyable!
- Posts: 990
- Joined: Sat Dec 22, 2007 3:38 am
Re: HDMI 1.4 Mystery Solved at CES!
- tritosine5G
- Terrif-eying the Ladies!
- Posts: 894
- Joined: Wed Mar 17, 2010 9:35 am
- Location: As far from Hold Display guys as possible!!! ^2
Re: HDMI 1.4 Mystery Solved at CES!
- cybereality
- 3D Angel Eyes (Moderator)
- Posts: 11407
- Joined: Sat Apr 12, 2008 8:18 pm
Re: HDMI 1.4 Mystery Solved at CES!
- DmitryKo
- Diamond Eyed Freakazoid!
- Posts: 776
- Joined: Tue Jan 08, 2008 2:25 am
- Location: Moscow, Russia
Re: HDMI 1.4 Mystery Solved at CES!
If the community entirely missed each and every of my posts, well, that's a pity.
GDC Online's 3D Gaming Panel, Part II (Oct 2010)
MTBS-TV: Interview with Richard Laberge, Sensio Co-Founder (Jan 2010)
The Signal That Stops Your Old TV Supporting 3D, but PS3 ok (Feb 2010)
Here seems to be the way that the PS3 does 3D over HDMI 1.3 (Apr 2010)
Technical questions: 1920x1080@120Hz on a 3D tv?!? (May 2010)
ATi and stereo 3D, ATI speaks clearly (french interview) (Jul 2010)
etc.
- Likay
- Petrif-Eyed
- Posts: 2913
- Joined: Sat Apr 07, 2007 4:34 pm
- Location: Sweden
Re: HDMI 1.4 Mystery Solved at CES!
There are also plenty of other solutions standing in line to replace hdmi. HDBaseT and displayport are good candidates. Nowadays consoles are screwed for full-hd content but honestly in the early era of console-3d other things are more important.
- DmitryKo
- Diamond Eyed Freakazoid!
- Posts: 776
- Joined: Tue Jan 08, 2008 2:25 am
- Location: Moscow, Russia
Re: HDMI 1.4 Mystery Solved at CES!
No. HDMI 1.3/1.4 has enough bandwidth for 1080p60 per eye, it's the actual silicon that didn't catch up (unlike DisplayPort which supported full bandwidth right from the start and supported frame-sequential stereoscopic 3D since version 1.1).Likay wrote:the hdmi standard itself is limited. There's no point making chipsets with higher bandwidth until a new standard is made.
- Likay
- Petrif-Eyed
- Posts: 2913
- Joined: Sat Apr 07, 2007 4:34 pm
- Location: Sweden
Re: HDMI 1.4 Mystery Solved at CES!
- Zerofool
- Cross Eyed!
- Posts: 119
- Joined: Tue May 06, 2008 3:50 pm
- Location: Sofia, Bulgaria
Re: HDMI 1.4 Mystery Solved at CES!
Likay, yes, "the hdmi standard itself is limited", but it's artificially limited, that's the point. If they made the 1080p60 3D mode a mandatory mode for v1.4a, it would have been implemented already (the cable bandwidth is enough since v1.3 as DmitryKo said), and these transmitting chips would have been created in time to fit the 2010 line of TVs (yes, probably more expensive), but as I previously said - each generation of TVs has to have a selling point. Obviously, this 1080p60 3D mode will be one of the mandatory modes in v1.4b or whatever they decide to call it, and will probably still use exactly the same cables (then, in v1.5 they might decide to include higher than 1080p 3D modes... this would finally require more bandwidth and, therefore, new cable), but most likely it won't make it for the 2011 TVs (because these TVs are almost done, and the transmitter chips are still months away, as this person in the interview said), we'll have to wait for 2012 TVs supporting the latest HDMI version, to have that feature (or earlier, if any of the TV manufacturers decide to use DisplayPort).
I think that's what Neil should have asked - WHEN will this mode be available in consumer products (3DTVs). And being a part of the S3DGA in my oppinion there should be some talks (and I wonder why there haven't been any yet) with Panasonic and LG, where they need to be informed that this feature is really essential to PC S3D Gamers, and support for it (even outside of mandatory HDMI modes, or through DisplayPort) would be the most important thing such a gamer would look for when buying a 3DTV.
P.S.
I'm sorry, but I can't agree - it's not silicon's fault, it's the people that make chips with it. Just think about it - all the transmitter chips used today have more or less the same capabilities as the chips used in the first HDTVs that support 1080p60 (2d) input (right?). And that's .. I don't know - 2-3 years already? And now suddenly.. in few months these new chips will come out, that finally will have the capability to rout double the bandwidth... I don't buy that, for all that time, we should have had 4K p60 3D capable chips by now .DmitryKo wrote:it's the actual silicon that didn't catch up
-
- Binocular Vision CONFIRMED!
- Posts: 320
- Joined: Tue May 15, 2007 8:11 am
Re: HDMI 1.4 Mystery Solved at CES!
- tritosine5G
- Terrif-eying the Ladies!
- Posts: 894
- Joined: Wed Mar 17, 2010 9:35 am
- Location: As far from Hold Display guys as possible!!! ^2
Re: HDMI 1.4 Mystery Solved at CES!
this panel with dual dvi would be cool. They would be crazy not to sell it as monitor imho.
- DmitryKo
- Diamond Eyed Freakazoid!
- Posts: 776
- Joined: Tue Jan 08, 2008 2:25 am
- Location: Moscow, Russia
Re: HDMI 1.4 Mystery Solved at CES!
It's rather the other way round. Silicon Image is both the largest designer of HDMI parts and the founding member of HDMI LLC, so they knew what they were doing when they made it an optional feature. I bet it was done on purpose, they didn't feel the need to design this higher bandwidth into their silicon because they didn't anticipate the impact of PC gaming in the first place.Zerofool wrote:If they made the 1080p60 3D mode a mandatory mode for v1.4a, it would have been implemented already ... and these transmitting chips would have been created in time to fit the 2010 line of TVs (yes, probably more expensive)
I never said that silicon is able to design itself...DmitryKo wrote:I can't agree - it's not silicon's fault, it's the people that make chips with it.
Exactly. It will not happen until at least 2012, and by that time, we'd rather start seeing DisplayPort connectors in 3D TVs, since the TV industry is going replace internal LVDS links with Internal DisplayPort (iDP) starting with 2013 and will phase out VGA and DVI by 2015.I think that's what Neil should have asked - WHEN will this mode be available in consumer products (3DTVs).
most likely it won't make it for the 2011 TVs (because these TVs are almost done, and the transmitter chips are still months away
No, looks like HDTV makers missed that chance, they continue to treat the PC stereo gaming market as insignificant.ssiu wrote:did you see any 2011 3D HDTVs announced at CES supporting 1080p/60Hz 3D gaming via DisplayPort (to get around HDMI limitation)
However, as I said above, 2012 will be absolutely the last year PC industry uses LVDS in display panels, and involved companies like LG Display and Samsung should move their design and production to use Embedded DisplayPort (for monitors) and Internal DisplayPort (for TVs)... so the times are coming.
There's also pressure from WHDI 2.0 which supports 1080p60 stereo 3D and the Wireless Gigabit Alliance (WiGig) who teamed with VESA to create Wireless DisplayPort 1.2, which should also support 1080p60 stereo.
- Chiefwinston
- Diamond Eyed Freakazoid!
- Posts: 712
- Joined: Fri Jun 19, 2009 8:05 pm
Re: HDMI 1.4 Mystery Solved at CES!
cheers everyone
i7
DDD
PS3
Panasonic Plasma VT25 50" (Full HD 3D)
Polk Audio- Surround 7.1
Serving up my own 3D since 1996.
(34) Patents
- Zerofool
- Cross Eyed!
- Posts: 119
- Joined: Tue May 06, 2008 3:50 pm
- Location: Sofia, Bulgaria
Re: HDMI 1.4 Mystery Solved at CES!
I took your statement:DmitryKo wrote:I never said that silicon is able to design itself...
as "it's the manufacturing process of silicon that didn't catch up", and that's why I replied as I did, meaning that if they wanted to, they had the ability to do it for quite some time (unlike 2-3 years back, when it was technologically impossible/very hard (I guess)).DmitryKo wrote:it's the actual silicon that didn't catch up
Thanks for your clarifications on the relationship between Silicon Image and HDMI (they are both to blame, it appears ).
- cybereality
- 3D Angel Eyes (Moderator)
- Posts: 11407
- Joined: Sat Apr 12, 2008 8:18 pm
Re: HDMI 1.4 Mystery Solved at CES!
- DmitryKo
- Diamond Eyed Freakazoid!
- Posts: 776
- Joined: Tue Jan 08, 2008 2:25 am
- Location: Moscow, Russia
Re: HDMI 1.4 Mystery Solved at CES!
Better manufacturing processes have no relevance for HDMI, since it uses 5V TMDS signalling which employs quite old manufacturing process (130 nm) and is not compabitle with today's low-voltage logic used in GPUs and video processors; that's why you always need an external transmitter chip (unlilke DisplayPort, I must add).Zerofool wrote:"it's the manufacturing process of silicon that didn't catch up", and that's why I replied as I did, meaning that if they wanted to, they had the ability to do it for quite some time (unlike 2-3 years back, when it was technologically impossible/very hard (I guess))
You will not be surprised to learn that TMDS was developed by none other but Silicon Image, who receive licensing fees for its use in both DVI and HDMI.
Exactly, it's in fact quite old. The physical HDMI connection is essentially a single-link DVI-D with a few additional wires, and DVI 1.0 was designed in the year 1998 (and have never gone through any further revision since).cybereality wrote:Why is it that HDMI is so far behind the times. I mean, you can easily do 1080P@120Hz over DVI with an Nvidia card as old as 8800. So a video card from over 4 years ago has more power than what they have today? Some real "next-generation" technology these HDMI guys came up with.
Of course LVDS would be much better (and royalty-free), but HDMI Founders just wanted backward compatibility with DVI to reduce implementation costs, not to mention the licensing fees associated with TMDS... Considering how we ended up with having up to 5 HDMI ports in modern TV sets, the HDMI Founders should have just added a separate DVI port right from the start, and used a more efficient protocol for the main link, like VESA did with DisplayPort.
DisplayPort has a fully compatible HDMI 1.3 emulation mode which only requires a passive signal-level conversion adapter, so I'd imagine future TVs will be fully based on DisplayPort protocol internally, even though they will still offer HDMI ports.Chiefwinston wrote: hdmi is and will remain the default display standard for hdtv's. Everything from manufacturing to software licencing is already in place. All other technologies even better higher bandwidth solutions will have an uphill battle to unseat hdmi.
- Chiefwinston
- Diamond Eyed Freakazoid!
- Posts: 712
- Joined: Fri Jun 19, 2009 8:05 pm
Re: HDMI 1.4 Mystery Solved at CES!
cheers everyone
i7
DDD
PS3
Panasonic Plasma VT25 50" (Full HD 3D)
Polk Audio- Surround 7.1
Serving up my own 3D since 1996.
(34) Patents
- Chiefwinston
- Diamond Eyed Freakazoid!
- Posts: 712
- Joined: Fri Jun 19, 2009 8:05 pm
Re: HDMI 1.4 Mystery Solved at CES!
hmm looks like $699 even more impressive
cheerws everyone
- cybereality
- 3D Angel Eyes (Moderator)
- Posts: 11407
- Joined: Sat Apr 12, 2008 8:18 pm
Re: HDMI 1.4 Mystery Solved at CES!
- Zerofool
- Cross Eyed!
- Posts: 119
- Joined: Tue May 06, 2008 3:50 pm
- Location: Sofia, Bulgaria
Re: HDMI 1.4 Mystery Solved at CES!
OK, that's good to know. And it sounds like this can't be easily bypassed, so let me ask you this:DmitryKo wrote:Better manufacturing processes have no relevance for HDMI, since it uses 5V TMDS signalling which employs quite old manufacturing process (130 nm) and is not compabitle with today's low-voltage logic used in GPUs and video processors; that's why you always need an external transmitter chip (unlilke DisplayPort, I must add).
You will not be surprised to learn that TMDS was developed by none other but Silicon Image, who receive licensing fees for its use in both DVI and HDMI.
What changed "now" to allow them to create the new chips, the guy in the video is talking about (that will come in few months)? And what stopped them to do it earlier?
The price is great, but the native resolution of 1366 x 768 is too low for 50" display imo, I've seen such models from Samsung.
- DmitryKo
- Diamond Eyed Freakazoid!
- Posts: 776
- Joined: Tue Jan 08, 2008 2:25 am
- Location: Moscow, Russia
Re: HDMI 1.4 Mystery Solved at CES!
Nothing but lack of any foresight.Zerofool wrote:What changed "now" to allow them to create the new chips, the guy in the video is talking about (that will come in few months)? And what stopped them to do it earlier?
AFAIK there are no technical barriers to practically achieve 340 MHz clock defined in the HDMI 1.3/1.4 standards; other differential serial buses like DisplayPort and LVDS (which is used in HyperTransport, PCI Express, FireWire, and Serial ATA, besides internal display panel links) reliably run at up to 540 MHz.
- cybereality
- 3D Angel Eyes (Moderator)
- Posts: 11407
- Joined: Sat Apr 12, 2008 8:18 pm
Re: HDMI 1.4 Mystery Solved at CES!
Yeah, but if you want to use it for gaming, HDMI 1.4 maxes out at 1280x720. So the TVs resolution is still higher than the source material.Zerofool wrote: The price is great, but the native resolution of 1366 x 768 is too low for 50" display imo, I've seen such models from Samsung.
- Chiefwinston
- Diamond Eyed Freakazoid!
- Posts: 712
- Joined: Fri Jun 19, 2009 8:05 pm
Re: HDMI 1.4 Mystery Solved at CES!
cheers everyone
i7
DDD
PS3
Panasonic Plasma VT25 50" (Full HD 3D)
Polk Audio- Surround 7.1
Serving up my own 3D since 1996.
(34) Patents
- tritosine5G
- Terrif-eying the Ladies!
- Posts: 894
- Joined: Wed Mar 17, 2010 9:35 am
- Location: As far from Hold Display guys as possible!!! ^2
Re: HDMI 1.4 Mystery Solved at CES!
1. eye is more sensitive to luma (brightness) than contrast (blacks), plasma can't compete in 3d brightness department, white will be always way off, and going to ghost because of phosphor decay*
2. human visual system does biological anti aliasing after a distance (individual pixels blend together), thats why 720p 85" looks awesome at 3.5 meter
3. 2d perception is whole different deal than 3d perception, another reason 720p 85" looks awesome at 3.5 meter
4. depthQ is a rebadged , bog standard DLP light engine. Cheap DLP projector is worse because its less bright, on the other hand it has less color breakup for the same reason its less bright.
*they want to cure plasma ghosting with a new shorter burn phosphor type. (This also means current plasma screens are rebadged 2d) .
Because of shorter hold time , its going to be less bright.
New, dedicated 3d LCD panels will be good, panasonic said it is on par with plasma when it comes to ghosting, but its LCD. Because its LCD , it emits polarized light, in theory it will be much, much brighter than plasma, plus , they can use longer hold time. I wouldn't buy plasma right now.
- Zerofool
- Cross Eyed!
- Posts: 119
- Joined: Tue May 06, 2008 3:50 pm
- Location: Sofia, Bulgaria
Re: HDMI 1.4 Mystery Solved at CES!
So you agree with my initial statement that if they really wanted, they could have created such a chip by the time v1.4a was standardized and they could have made the 1080p60 3D mode a mandatory mode, right?DmitryKo wrote:Nothing but lack of any foresight.Zerofool wrote:What changed "now" to allow them to create the new chips, the guy in the video is talking about (that will come in few months)? And what stopped them to do it earlier?
AFAIK there are no technical barriers to practically achieve 340 MHz clock defined in the HDMI 1.3/1.4 standards;
That doesn't matter, the thing that personally bugs me is the pixellated picture. I guess I'm too sensitive, and the pixels of the plasma with such size are too big for me (in the case of 1366x768 native resolution for 50" screen that we discussed).cybereality wrote:Yeah, but if you want to use it for gaming, HDMI 1.4 maxes out at 1280x720. So the TVs resolution is still higher than the source material.
The resolution of the content is different story - with lower rez signal you lose sharpness, but the picture doesn't get pixellated (if you have high display native resolution), I hope you understand what I'm trying to explain.
Projectors are different story, you can see the individual projected pixels much harder (even if it has lower native resolution).Chiefwinston wrote:For quite a few years the the best consumer level projector technology was the depthq. It ran in 720P. Price=$5,000. It was the most practical solution for high end 3D in the home up to 2010.
Because the separation is relatively low, and the director smartly avoids high contrast scenes.Chiefwinston wrote:full hd 3D blu-rays are incredible and ghost free.
It's not the software - just crank up the separation and it's a ghosting fest once again . Still better than other 3D techs, but it's nowhere near perfect. Once you lower it to decent level - the crosstalk almost disappears (a friend of mine has a VT20E, I know that form him).Chiefwinston wrote:However, gaming in stereo has minor ghosting issues. Some games are completely ghost free. Others well- lets just say its the games stereo software thats the problem. I'm sure of it.
I think you are talking about signal, not native resolution, see my answer to cybereality. If you're talking about the native rez - then in both cases it will be intolerable for me .Chiefwinston wrote:720P with great black levels will give a better 3D experience than lets say 1080P with mediocre black level reproduction.
I also am a plasma fan, but I think black levels are much more important in 2D viewing. In 3D, the glasses block enough light to compensate. The advantage the PDP tech has over LCD is the fast response time (low phosphorus lag), leaving less trails and, therefore, leads to less crosstalk. But I still hope for good enough HMDs (which have absolutely zero ghosting/crosstalk) in the (near) future (like that Sony 720p OLED prototype Silversurfer first wrote about for example).Chiefwinston wrote:Now I'm a plasma guy. My opinion is that many LCD do not match a plasma's image quality. And its all in the black level reproduction.
I think 1366x768 native resolution is acceptable for up to 32-37" TVs, not more. As I said, the perception is extremely individual. If you're talking about the signal now - then yes, I agree, I guess 720p is bearable for 1080p TVs, up to 50".Chiefwinston wrote:And finally, 720P is more suited for 50" and less. Above this and you'll most definately want 1080p60.
I think the human eye is more sensitive to luma (brightness levels - the whole scale (including the black level)) that to chroma (color) - that's why all modern compression algorithms use 1/4 resolution for the chroma channel (look for 4:2:0 or YV12). And FYI, when you adjust the "contrast" setting of your display you actually adjust the maximum brightness level, and when you adjust the "brightness" setting, you adjust the minimum brightness level (the black level), as crazy as it sounds (I learned that just few months ago myself ).tritosine wrote:facts:
1. eye is more sensitive to luma (brightness) than contrast (blacks), plasma can't compete in 3d brightness department, white will be always way off, and going to ghost because of phosphor decay*
I guess you're talking about projectors as well. Once again, I think it's pretty individual. I, for example, think that 2K DLP projectors (or whatever they are using) in IMAX theaters (or at least in the one in my city) are not enough for a screen that big (and even for smaller ones) - you can't see the individual pixels, but the picture is blurry ass hell.tritosine wrote:2. human visual system does biological anti aliasing after a distance (individual pixels blend together), thats why 720p 85" looks awesome at 3.5 meter
- cybereality
- 3D Angel Eyes (Moderator)
- Posts: 11407
- Joined: Sat Apr 12, 2008 8:18 pm
Re: HDMI 1.4 Mystery Solved at CES!
- 3D Blu-Ray runs at the full 1080P, while games are capped at 720P.
- 3D Movies have extremely conservative separation (only a few pixels in some cases) when games can be cracked up with 200 pixels or more of separation.
- 3D Movies have an "anti-ghosting" post-process done on them. This is probably not economical to do in real-time (although maybe an optimized version could work).
- Games frequently have high contrast solid colors that would never happen in a movie (for example a HUD element that is pure white)
There is nothing happening with the 3D drivers that causes ghosting. You can verify this by taking a JPS screenshot and then opening it with Photoshop. No ghosting. There is also no ghosting with HMDs, even using the same exact driver (iz3D) with the same settings.
- Chiefwinston
- Diamond Eyed Freakazoid!
- Posts: 712
- Joined: Fri Jun 19, 2009 8:05 pm
Re: HDMI 1.4 Mystery Solved at CES!
cheers everyone
i7
DDD
PS3
Panasonic Plasma VT25 50" (Full HD 3D)
Polk Audio- Surround 7.1
Serving up my own 3D since 1996.
(34) Patents
- DmitryKo
- Diamond Eyed Freakazoid!
- Posts: 776
- Joined: Tue Jan 08, 2008 2:25 am
- Location: Moscow, Russia
Re: HDMI 1.4 Mystery Solved at CES!
They probably could, but they didn't really want or need to make it mandatory in the first place.Zerofool wrote:So you agree with my initial statement that if they really wanted, they could have created such a chip by the time v1.4a was standardized and they could have made the 1080p60 3D mode a mandatory mode, right?
- cybereality
- 3D Angel Eyes (Moderator)
- Posts: 11407
- Joined: Sat Apr 12, 2008 8:18 pm
Re: HDMI 1.4 Mystery Solved at CES!
Yeah, they can probably make more money by coming out with a new half-baked standard every 2 years and then getting additional licensing revenue when all companies involved have to refresh their products. Plus the CEMs can get additional money from the consumer, now having them replace the TV they just bought with a brand new HDMI 1.5b display which supports the "Really-Full-HD-3D-This-Time" resolutions.DmitryKo wrote:They probably could, but they didn't really want or need to make it mandatory in the first place.
- Chiefwinston
- Diamond Eyed Freakazoid!
- Posts: 712
- Joined: Fri Jun 19, 2009 8:05 pm
Re: HDMI 1.4 Mystery Solved at CES!
cheers everyone
i7
DDD
PS3
Panasonic Plasma VT25 50" (Full HD 3D)
Polk Audio- Surround 7.1
Serving up my own 3D since 1996.
(34) Patents
- DmitryKo
- Diamond Eyed Freakazoid!
- Posts: 776
- Joined: Tue Jan 08, 2008 2:25 am
- Location: Moscow, Russia
Re: HDMI 1.4 Mystery Solved at CES!
I don't think their aim is to produce underperforming specifications, that would be bad strategy for any business. However this is what they've practically ended with through the series of wrong decisions.cybereality wrote:Yeah, they can probably make more money by coming out with a new half-baked standard every 2 years and then getting additional licensing revenue when all companies involved have to refresh their products.
You mean Itanium?Chiefwinston wrote: I mean look at intel. They have chips that function way way beyond whats on the market today. They're gonna milk it.
- Chiefwinston
- Diamond Eyed Freakazoid!
- Posts: 712
- Joined: Fri Jun 19, 2009 8:05 pm
Re: HDMI 1.4 Mystery Solved at CES!
Also, some of you referenced pixelated Plasma's. Step back untill you can't see the pixel structure. Your to close if you see it. It's not a computer monitor.
cheers everyone
i7
DDD
PS3
Panasonic Plasma VT25 50" (Full HD 3D)
Polk Audio- Surround 7.1
Serving up my own 3D since 1996.
(34) Patents
- Neil
- 3D Angel Eyes (Moderator)
- Posts: 6882
- Joined: Wed Dec 31, 1969 6:00 pm
- Contact:
HDMI 1.4 Mystery Solved at CES!
We all understand why Sony PlayStation 3 limits 3D performance to 1280X720P resolution in stereoscopic 3D gaming. There is clearly a performance limitation that creates a barrier of what can be realistically achieved, and 720P is pretty standard with traditional 2D console games anyway.
[caption id="attachment_15900" align="alignnone" width="514"]Left to right: Gabriele Collier, Charlene Wan, Steve Venuti from HDMI Licensing[/caption]
However, in the case of the upcoming Nvidia 3DTV Play software, and AMD's HD3D solution, HDMI 1.4A displays are limited to 1280X720P for gaming purposes. With all that horsepower under the hood, why is this barrier so difficult to break?