Search CE Pro






Print  |  Email  |  Share  |  News  |  Follow on Twitter, Facebook, Google+ or RSS

HDMI 1.4 Requires Increased Bandwidth: True or False?

It's unbelievable how much confusion and paranoia HDMI Rev 1.4 has stirred up.


image
The amount of confusion and paranoia HDMI Rev 1.4 has stirred up is just unbelievable.

There are a heck of a lot of great features that Rev 1.4 is bringing to the table. The main features include:
  • Ethernet capabilities for networking
  • Bi-directional audio transmission
  • Content enhancement for 3D TV
  • Increased resolution supporting 4096x2160
  • Increased color space specifically for DSLRs
It wasn't long before ridiculous claims came. However, they often aren't accompanied by real facts.

One major claim is about the increased bandwidth needed to support Rev 1.4. After all, HDMI is now offering the ability to support increased resolutions up to 4Kx2K, right?

Well, not really.

HDMI Rev 1.4 supports bandwidths up to 340 MHz or 3.4 GB for each color, 10.2 GB in total just like Rev 1.3. However, if 1080p 16-bit deep color consumes the entire 10.2 GB, how can it possibly support 4Kx2K under the same rules?

If you think about it, you will realize that 1080p 16-bit uses a frame rate of 60 Hz. However, 4Kx2K comes in at 30 Hz, not 60 Hz, which requires only half the bandwidth it would normally need. Doubling the resolution would in fact double the bandwidth by dividing the frame rate in half and allows for operation under 10.2 GB.

Will these people ever grow up? Like my father use to say, some people grow up while others just grow.




Subscribe to the CE Pro Newsletter

Article Topics

News · Product News · Wire and Cable · HDMI · Wire And Cable · Hdmi · All topics

About the Author

Jeff Boccaccio, President, DPL Labs
Jeff Boccaccio, president of DPL Labs, can be reached at jeff@dpllabs.com.

13 Comments (displayed in order by date/time)

Posted by George Tucker  on  10/01  at  10:46 AM

Can you show the ‘math’ on this? I love the concept of the article but it leaves me wanting to know more about the WHY the bandwidth would go from 60Hz to 30. The statement does not help to clarify how or why the claim is not true for a beginner or HD neophyte.

Posted by AACTrent  on  10/01  at  11:41 AM

“The amount of confusion and paranoia HDMI Rev 1.4 has stirred up is just unbelievable.”

Jeff, with all due respect, HDMI has rightfully earned every bit of confusion and paranoia it garners in this industry.

“Will these people ever grow up?”

Our dealers have been asking the same thing about HDMI for years smile

Trent Davis

Posted by Simon Westenholz  on  10/01  at  01:05 PM

4096 x 2160 = 8,847,360
1920 x 1080 = 2,073,600
Last time I checked that’s not double, but quadruple.
Lets assume 16 bits per pixel, and 30 fps.

8,847,360 x 16 x 3 x 30

That equals 12.74 Gb/s (note lowercase b to indicate bits rather than uppercase B to indicate bytes i.e. 8 bits) for 4096 x 2160. Well above the max video bandwidth of 8.16 Gb/s.

Simply stated: Passing 16 bit per pixel at that resolution is impossible. Devide by two, however, and you end up with 6.37 Gb/s at 8 bits per pixel, which is within the spec. Another matter is however that 30 fps isn’t supported, but 24 fps is.

I hate to rain on your parade, but your information is wrong.

Posted by Simon Westenholz  on  10/01  at  01:14 PM

A small addition: HDMI 1.4 increases the maximum resolution to 4K × 2K (3840×2160p at 24Hz/25Hz/30Hz and 4096×2160p at 24Hz

Kind regards,
Simon Westenholz

Posted by Bryce Palmer  on  10/01  at  05:12 PM

This “article” could have used more research, as indicated by Simon. And does this sentence make any sense, grammatically or otherwise?

“Doubling the resolution would in fact double the bandwidth by dividing the frame rate in half and allows for operation under 10.2 GB.”

Posted by cm  on  10/03  at  07:13 AM

Can CEPro pls clear up these points?  The intent of the article was apparently to reduce confusion and promote understanding around HDMI 1.4, but so far it seems to have done the opposite.

Posted by Simon Westenholz  on  10/03  at  07:52 AM

For the record, I am not disputing the fact that 4K x 2K with deep color (in one form or another) is possible. Just that most of the info in this article is plain wrong.

The max color depth 4096 x 2160p can pass at is 12 bits per channel at 24 fps, which is still a significant improvement over 8 bit. It’s 4096 levels compared to 256, in raw bit terms.

The calculation goes:
8,847,360 x 12 x 3 x 24

That equals 7.64 Gb/s. And that IS within the spec.
And just for the record 1080p60 with 16 bits per channel comes in at 5.97 Gb/s.

The last thing I will comment on is the fact that the author claims the max bandwidth for the video to be 10.2 Gb/s, when it in actual fact is “only” 8.16 Gb/s. The 10.2 Gb/s is TMDS which both contains the actual video and audio, AND control bits.

I am done now… I think.

Posted by George Tucker  on  10/03  at  10:45 AM

Simon;

Thanks! This is some good work toward clarifying.
I would like to see the author incorporate and expand his explanation. I presume that he was asked to make it succinct and general in order to not lose the general interest reader. I do not think updating the article would make it neither too large nor too complicated.

Posted by Tom  on  10/07  at  11:44 AM

Simon,

If you haven’t noticed, HDMI Corner is extremely short in length.  George is right.  Jeff B. had to make it general in nature in order to fit in the space he is allotted.  You used more space in your 3 postings to dispute ONE POINT than Jeff B. makes in the entire article.  A note to CE Pro – Give Jeff B. more space so he can explain HDMI topics in more depth.

I don’t understand why in you first posting that you say his information is wrong.  He said that 4K x 2K fits in HDMI 1.3/1.4.  And it does.  You proved it in your third posting.  The problem is that you made incorrect assumptions on the number of bits and the frame rate in your first posting.  You also spelled DIVIDE wrong!

Posted by Simon Westenholz  on  10/07  at  12:07 PM

Tom,

First of all, what he stated was that 4K x 2K was only double the resolution of 1080P, which it isn’t, and that was my initial point. He also states that 4K x 2K comes in at 30 fps, which is also wrong. He also “implies” that this is possible at 16 bits per channel, which again is wrong. In the article there is also a reference to max bandwidth, and that is ALSO wrong.

And if you feel the need to criticize my spelling, go ahead. I’m Danish, and as such don’t feel it is my duty to spell perfectly, or my shame if I make a mistake occasionally.

And as for my posts taking up more room, how can you criticize me for elaborating on my points? I find that most people just proclaim something, without even backing it up, and I don’t want to be such a person.

The article needn’t be longer, but it could certainly be more precise.

/Simon

Posted by Tom  on  10/07  at  05:15 PM

Simon,

If you had been to one of Jeff Boccaccio’s seminars (I just attended one at the Atlanta CEDIA show), you would know that he is quite knowledgeable on both the history and technical aspects of analog and digital video.  He has helped installers like myself understand this flawed interface which I am greatly appreciative. 

What I got out of the article is that HDMI 1.4 brings some new features to the table and that 4K x 2K is possible without an increase in bandwidth over what is specified in HDMI 1.3.  That means I probably will not have to change the cable I just installed for my biggest client when he decides to upgrade to 4K x 2K.  THAT IS THE POINT HERE!  Further, he did not imply that 4K x 2K called for 16 bit, you inferred it. 

And I did not criticize you for elaborating your points.  I am just pointing out that you took quite a bit of space to explain your point to the fullest extent (a luxury that Jeff B. did not have) and you unfairly criticized him. He was not writing a White Paper on the subject, he was making a quick point by simplifying the math.

Lighten Up.

Posted by cm  on  10/07  at  05:44 PM

The trouble I see with the article is that it was somewhat sloppy at best, and arguably somewhat erroneous and/or misleading.  The reason this is so important in this particular case is because the article was trying to address HDMI confusion and paranoia, which is precisely what erroneous or misleading statements promote.  (In fact, the article went so far as to belittle people who were confused or paranoid.)

I suspect the article could have been written in the same number of words without these troubles.

Posted by George  on  11/23  at  03:16 PM

Hey guys,

I see that there was a flurry of comments back in early October and I know that it is now late November, but it occurs to me that If all of you would take a moment to read the article from which I believe Jeff condensed this short blurb, you would see that he went into a bit more detail there about why he said what he said.  I’ve trusted Jeff’s math for a number of years and whenever I thought that I might have caught him in an error… the truth was that it wasn’t Jeff who was mistaken!

Here’s the article from his DPLrating site:  http://www.dplrating.org/dpl/article/hdmi-rev-14-4k-x-2k-resolution-mission-impossible

Hopefully, reading this other article will clear up some of that confusion everyone is talking about.  If not, instead of going on and on in these comments, maybe you could try contacting Jeff and chatting with him directly.  His contact information is listed here and he DOES talk to folks who ask for information.  Communication, give it a try.

Page 1 of 1 comment pages
Post a comment
Commenting is not available in this weblog entry.

Sponsored Links

  About Us Customer Service Privacy Policy Contact Us Advertise With Us Dealer Services Subscribe Reprints ©2013 CE Pro
  EH Network: Electronic House CE Ideas Store Commercial Integrator ChannelPro ProSoundWeb Church Production Worship Facilities Electronic House Expo Worship Facilities Expo