1080 progressive on BBC HD channel

33 replies [Last post]
Paul Jordan
Offline
Joined: Apr 22 1999

I have just noticed some of the programmes on the BBC HD channel are being broadcast in 1080p rather than the usual 1080i. Not noticed any other HD channels doing this yet but good on the BBC for this.

Has anyone else noticed or maybe it has been happening for a while and I have just been blind!

Paul

steve
Offline
Joined: Apr 8 1999

How are you detecting this? They have been broadcasting 25psf for some time where the programme material is progressive.

Steve

Ron Jackson
Ron Jackson's picture
Offline
Joined: Aug 23 2007

And doesn't it depend on whether or not your TV is "1080p"?

Ron Jackson

Gavin Gration
Offline
Joined: Jul 29 1999

That's news to me. My understanding is that HD TV was broadcast 1440 x 1080i.

I know there was a call for it to be increased to 1920 x 1080i but not sure if they do that now.

AFAIK 1080P isn't currently broadcast.

steve
Offline
Joined: Apr 8 1999

My point was that if a set top box is being used, some can be setup to output 1080p/50, (upscaled from 1440x1080 to 1920x1080). If the incoming program is 1080psf/25, then the TV should be fed 1080p/50 via HDMI. The TV then thinks it is getting an external 1080p/50 input. A 1080psf/25 stream should require no more bandwidth than a 1080i/25 one.

Steve

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

All BBC HD is broadcast 1440x1080, i/25 or psf/25. There's no broadcast of true progressive as such. 'Film-look' programmes are made in 1080 psf/25 and broadcast as such. This really is progressive images, 25/second, but they're split into two fields for broadcast at 50/second. And that's exactly the way that film has been handled on TV si8nce 1936.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

Paul Jordan
Offline
Joined: Apr 22 1999

The TV is one with built in Freeview HD tuner so no external converting going on. When you bring up the programme info on what you are watching it tells you there and then whether the signal is 1080i or 1080p.

I had just never seen the 1080p before until now and as Alan says it appears on the "Film-look" programmes only. Only noticed this on BBC HD as ITV HD and 4HD show only 1080i for everything as far as I have seen.

steve
Offline
Joined: Apr 8 1999

Well Paul I suppose both terms are right, i.e. 25psf is originated as true progressive 25p and broadcast as a 1080i/25 stream. It's just a matter of how the TV designers choose to describe the programme.

Steve

dude
Offline
Joined: Nov 2 2010
Alan Roberts wrote:
All BBC HD is broadcast 1440x1080, i/25 or psf/25. There's no broadcast of true progressive as such.

Alan, there's someone on another forum who made a topic with that subject, stating that "new encoder software dynamically switches between 50i and 25p encoding". Moreover, he said that "shows shot 25p are encoded in 25p mode (even though the master tape is 50i - effectively 25psf)".
I would post the link if it's aloud to.

Gavin Gration
Offline
Joined: Jul 29 1999

I guess we're back to the old subject of labels, wrappers and contents. All different but all correct. ;)

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Yes, that's right, it's the most efficient way of coding the images. But, there are still no shows shot in p/25. All 'film-look' programmes are shot in psf/25 and edited in psf/25. There are currently no approved cameras which shoot in native p/25, simply because the distribution network is i/25 or psf/25.

the difference between coding i/ or psf/25 and p/25 is that in p/25 the whole frame is coded as a single image whereas in i and psf it's coded in pairs of fields. Provided the correct flags are used in the transport stream, it should all be transparent. But, you can be absolutely certain that you'll not see any material originated in p/25, yet.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

dude
Offline
Joined: Nov 2 2010
Alan Roberts wrote:
All 'film-look' programmes are shot in psf/25 and edited in psf/25.

So even shows like "Top Gear" have that look, they're shot in 25/psf. It looks so good, with what camcorders it's shot?

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Much of it is EX3 I think. Not sure though.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

rone01
Offline
Joined: Feb 9 2009

With the likes of Wallander being Red and Red having no interlace mode whatsoever - is this just encoded psf for broadcast?

To be honest this whole psf thing may be efficient and how it's done but I long for the day it's all done away with. I like the idea of native whole frames. For instance on blu-ray naitive 24p exists as a standard but for us in the UK we have to make do with producing psf/25 blu-ray discs which is very frustrating. Yes it may be seamless but some flat panel displays get in a right tizz trying to work out which mode to switch to.

Red EPIC / Scarlet - Panasonic AC 90 -  besq.co.uk

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

OK, the RED records progressive, as does ARRI D20/D21/Alexa, Dalsa, and all the other cameras (including most of the DSLRs and other sub-broadcast cameras) that are technically aiming to be film-replacement rather than video cameras. But, the productions they provide input for, when destined for television,, will always handle the material as psf. There's no image difference between progressive and psf, both capture progressive frames from the sensor, the only difference is in the delivery mode.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

rone01
Offline
Joined: Feb 9 2009

Alan, do you think we will ultimately move to full progressive delivery be it 24/25p or 50p and do away with interlace? Would you favour that?

Or is the bandwidth not there?

Red EPIC / Scarlet - Panasonic AC 90 -  besq.co.uk

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

The publicly stated aim of the EBU is to get to 1920x1080 p/50 broadcast for the whole of Europe. That aim has been there for at least 10 years to my certain knowledge. There's a fair amount of evidence that the broadcast data rate for 1080 p/50 will not need to be higher than for the current 1080 i/25 and psf/25 transmissions. This is because the half-second GoP will contain 25 frames instead of 12 frames (or 24 fields). Since the inter-frame differences will be smaller, the GoP will be easier to predict and so easier to code.

The real issue is capture of p/50 at the higher end of the scale. Drama and high-end production likes to use intra-frame coding, for which the data rate will double. Few cam-corders can do this as yet; although many of the system (studio) cameras will run at p/50, there are few recorders which will run at such speeds, and that's one reason why Sony are launching their new system this year.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

dude
Offline
Joined: Nov 2 2010
Alan Roberts wrote:
The real issue is capture of p/50 ...

And another one, also bigger I think, is receiving it.

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

But receiving it is a doddle, like I already said. the data rate probably won't change, and it's possible that some of the Freesat boxes will already work in p/50. But I expect we'd be in for another range of boxes anyway because coders are continually getting getter and new tricks being built in, so the decoders often don't keep up all that well.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

dude
Offline
Joined: Nov 2 2010
Alan Roberts wrote:
But receiving it is a doddle, like I already said. the data rate probably won't change, ...

Not quite, I think. I agree that data rate won't pose a threat (since EBU, trying to promote 720p way back, stated that compression of that is a 20% save over 1080i), but a new decoder is needed inside the box.
Yet another, actual HDMI transmitters 'know' only a pixel clock frequency of 74,25 Mhz, but for 1080p/50(60) one with 148,5 MHz is needed. It's worth mentioning that transmitting 3D frame packing format needs the same frequency, i.e. 148,5 MHz.

So, like you've said about capturing it - none, I think - it could take more than a while until mass deployment of new boxes.

steve
Offline
Joined: Apr 8 1999

1080p/60 has been in the HDMI standard since version 1.0 and chip manufacturers have offered transmitter devices in their ranges that provide the bandwidth. The difference in cost at 1000+ quantities can be less than $0.25 and sometimes they are available in the same style packaging. So upgrading HD STBs at the HDMI interface would probably be swept up in a six-monthly PCB tidying-up exercise.
It is probable that current boxes, after firmware updates, would be able to offer a default 1080i/50 or 1080p/25 mode when receiving the 1080p/50 broadcast transport stream. If the stream was suitably formatted, boxes with limited processing power could decode a subset of the data to provide video at current HDTV standards.
So the likelihood is that the constraint to introduction would be as Alan says, i.e. at the front end with the production kit.
T2 boxes have only been in production for just over a year, whereas HDMI hardware is now gt least 5 years in the market, so it is unlikely that the domestic kit would be a significant delaying influence.

Steve

Steve

dude
Offline
Joined: Nov 2 2010
steve wrote:
1080p/60 has been in the HDMI standard since version 1.0

You're absolutely right, but only at par.6.3.2: "Optional Video Format Timings" which is obviously different from the par. 6.3.1: "Primary Video Format Timings".
That's why only recently came out BD players that output that format.

steve wrote:
It is probable that current boxes, after firmware updates, would be able to offer a default 1080i/50 or 1080p/25 mode when receiving the 1080p/50 broadcast transport stream.

That could be possible but, in the end, we're talking about millions of new boxes and that would not happen over night.

steve wrote:
So the likelihood is that the constraint to introduction would be as Alan says, i.e. at the front end with the production kit.

I totally agree, the lack of content is the main problem, but I also think that the 'push' will come from film industry.
James Cameron is preaching towards the 48 fps format for years saying that "A 2K image at 48 frames per second looks as sharp as a 4K image at 24 frames per second ... with one fundamental difference: the 4K/24 image will judder miserably during a panning shot, and the 2K/48 won't. Higher pixel counts only preserve motion artifacts like strobing with greater fidelity. They don't solve them at all."

Moreover, trying to put apart benefits of higher frame rate vs resolution, he also said that "4K is a concept born in fear. When the studios were looking at converting to digital cinemas, they were afraid of change, and searched for reasons not to do it. One reason they hit upon was that if people were buying HD monitors for the home, with 1080x1920 resolution, and that was virtually the same as the 2K standard being proposed, then why would people go to the cinema? Which ignores the fact that the social situation is entirely different, and that the cinema screen is 100 times larger in area. So they somehow hit on 4K, which people should remember is not twice the amount of picture data, it is four times the data. Meaning servers need to be four times the capacity, as does the delivery pipe to the theater, etc. "

And let's not forget about Douglas Trumbull and his 'showscan' process with 60 fps and, most important, the conclusion he drew that as the frame rate increased, so did the viewer's emotional reaction.

IMHO, James is no fool. Jammin' along with him on Avatar sets were Spielberg and Lucas (just like Clapton did with BB King on "Riding with the King" - btw, great album!) so I do believe that, along with 3D shoting, they might have try to put away all constrains related to judder, that Alan described: no rapid camera/objects movement, panning the camera to follow the action and use of short depth of field to soften the moving in the image. And might have work it, cause James also said that Avatar 2 shall be shot in 48 fps (and in 3D, obviously!) so the 'brave' concept of 'film look' is in absolutely no danger ... I think ;)

And now I'll close my eyes and try to imagine how would it look at 48 "The good, the bad and the ugly" :) ...

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Undoubtedly it'll need new boxes, that's the way of the world. But it really isn't that hard to do. And HDMI can carry data at much faster than 74.25Mhz, it's only an agreed protocol layer that fixes at 74.25.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

rone01
Offline
Joined: Feb 9 2009

Steve ... have you seen RED's 4k Magic Motion / HDRx system?

Red EPIC / Scarlet - Panasonic AC 90 -  besq.co.uk

steve
Offline
Joined: Apr 8 1999

I've looked it up since you mentioned it here. At a glance, it seems to be a 3 or 4K feature as far as Red is concerned. What JC is saying, i.e. frame rate is more important than resolution goes on a different route.

Steve

MattDavis
MattDavis's picture
Offline
Joined: Jul 1 2005

IIRC, I've asked Alan about this before, and I am still a little woolly about this because I'm not yet in the BBC HD owners club yet, but...

We have a broadcast image of 1440x1080 non-square pixels, of which most material is 25fps (stay with me, not to get embroilled in psf YET), because most HD cameras cannot actually record 1080p50 (my FS100 is on order), which is where the EBU, Mr Bilbo Jackson, that Avatar chap and a lot of other people want to go.

I have been brought up to believe that, in our post CRT, web delivery era, we should be recording progressive to match displays. I have also been brought up to believe that, for broadcast (over the air, non-IP) systems, it utilises interlacing but we don't worry about that because the technical reasons for that are taken care of at the RX end and basically a progressive signal is what we actually watch on the screen. All progressive material in camera (progressive in that it could be psf, but not interlaced) becomes/is psf at TX, then reassembled as progressive after RX, so if we were to pause a frame with motion, there would be no inter-field flicker, and that the screen is not de-interlacing and therefore losing either 25% or 50% of the resolution.

Sorry if I oversimplify somewhat, and if I misunderstand, please can you wonderful people clarify. Like quadratic equations, however little good it may do me, I'd still like to understand the principle even if I have no strong desire to actually put it into practice.

Matt Davis - Director/Editor - Write, shoot, edit, publish - website & Blog
2x EX1R, FS100 & FS700 into FCPX & CS6

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

No Matt, that's just about spot on.

The only reason for continuing to broadcast 1080i or 1080psf (at 25 fps) now is to keep faith with legacy displays, the crt. There's lots of them still around, so it makes sense, even though there's very few of them actually showing HD as such. The current system is a transitory one, it's the best way to progress from interlaced SD at 50 fields/second to HD at 50 frames/second.

Transmission of 1080p/50 shouldn't be too hard, the current bit rates for 1080i/25 should be enough for 1080p/50 with a bit of luck (the bit rate might even go down slightly), but there will need to be a whole new generation of consumer decoders to do it, so don't hold your breath. It's the capture and edit of 50p that's going to be the problem, the increased data rate and storage problems are all a bit much for the broadcasters because they don't like to compress it too much before the actual broadcast compressor.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

dude
Offline
Joined: Nov 2 2010
Alan Roberts wrote:
It's the capture and edit of 50p that's going to be the problem ...

But thanks God we have Moore's law on our side. I think Peter Jackson counts on that since it's already shooting "The Hobbit" at 48 fps, obviously in 3D.

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Agreed. It's a logistics problem it a technological one. Just image what the response would have been, only 10 years ago, to a request to set up a broadcast server with dozens of Terrabyte running at HD speeds. But now you can just go out and buy them at almost sensible prices.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

dude
Offline
Joined: Nov 2 2010

PJ is shooting with Red Epic cameras, although I don't know what resolution (cameras are capable of 5K) he's at 2 years away to finish. JC is at (at least) 18 months away to start shooting Avatar 2.
I think time is working for them.

Alan Roberts
Alan Roberts's picture
Offline
Joined: May 3 1999

Absolutely right, time sorts it all out in the end.

Get my test cards document, and cards for 625, 525, 720 and 1080. Thanks to Gavin Gration for hosting them.
Camera settings documents are held by Daniel Browning and at the EBU
My book, 'Circles of Confusion' is available here.
Also EBU Tech.3335 tells how to test cameras, and R.118 tells how to use the results.

LittlePicture
Offline
Joined: Feb 20 2009

I remember Andy Quested saying recently that the encoders have been upgraded so that material shot true progressive would be displayed as progressive, not psf. This is done by way of introducing a 1/2 second delay so that fields A+B are displayed simultaneously. So receivers that can output progressive do output progressive where appropriate. This was effective of 24th March this year.

dude
Offline
Joined: Nov 2 2010
LittlePicture wrote:
I remember Andy Quested saying ...

Is that something written somewhere or is just saying? btw, read the post #9 on this topic.

LittlePicture
Offline
Joined: Feb 20 2009
dude wrote:
Is that something written somewhere or is just saying? btw, read the post #9 on this topic.

It was at a BBC presentation / seminar of Large Sensor Cameras for DOPs at Pinewood Studios earlier this year. I don't have an online source for it I'm afraid.

You could ask him..http://www.bbc.co.uk/blogs/bbcinternet/andy_quested/