The switch to a full channel is to coincide with an upscaled ITV1 simulcast when there's no "native" HD programming, so there'll always be something to watch on there, and it will be better than normal PQ.
From what I can gather elsewhere, ITVHD will be joining BBCHD on that transponder to allow it to become a full channel on Freesat.
If that is really true then I don't why Andy Quested is saying the bitrate of BBC HD won't vary. Surely once you have at least two HD channels together, the most efficient use of the bandwidth is to stat mux it, this is what occurs with all the Sky HD channels apart from the ones where the HD channel is on their own on a transponder with SD channels i.e. Rush and Luxe HD.
If that is really true then I don't why Andy Quested is saying the bitrate of BBC HD won't vary. Surely once you have at least two HD channels together, the most efficient use of the bandwidth is to stat mux it, this is what occurs with all the Sky HD channels apart from the ones where the HD channel is on their own on a transponder with SD channels i.e. Rush and Luxe HD.
It's something to do with a switch to DVB-S2 or something like that I think. I don't understand the technology
It's something to do with a switch to DVB-S2 or something like that I think. I don't understand the technology
No, all S2 transponders used by Sky use stat muxing, there is really no sensible reason why they won't be able to do it, my only guess is Andy knew that by saying they would vary the bitrate, he knew some people (i.e. me:D) would know that would 100% mean that other HD channels were on the way and he isn't meant to be letting that cat out of the bag (yet):cool:?
No, all S2 transponders used by Sky use stat muxing, there is really no sensible reason why they won't be able to do it, my only guess is Andy knew that by saying they would vary the bitrate, he knew some people (i.e. me:D) would know that would 100% mean that other HD channels were on the way and he isn't meant to be letting that cat out of the bag (yet):cool:?
Maybe, but ITV have publicly announced a full-time channel so it shouldn't be "secret" any more.
I dont see why BBC should waste one of their transponders so that the other broadcasters can just piggyback on them and dont make a proper effort by themselves :yawn:.
I dont see why BBC should waste one of their transponders so that the other broadcasters can just piggyback on them and dont make a proper effort by themselves :yawn:.
For the greater good. Of course, I have no evidence to back this up, but I suspect the BBC would rather help ITV than Sky.
24 & 25 fps quite simply isn't fast enough to handle fast pans etc. without judder, except for people who are insensitive to judder. Not sure about the relationship between judder and smear but it seems to me that HD needs higher frame rates.
30fps may be fast enough - I recently downloaded a satellite feed TS from a live concert, transmitted at 35 mbps/1080i/29.97fps/MPEG2/NTSC (or whatever it's called in HD) and there is no sign of judder or smearing... so it could perhaps be done, without straining the technology by having to go to 50p.
Recording at 24 fps in HD is plain bonkers to my eyes and I wonder why we even stick with 25 fps in Europe for HD - would the 50 Hz mains frequency really cause a problem these days when playing 29.97/30 fps HDTV/video? Ideally, given a choice, I'd go for 720p/50 any day of the week rather than what we get now... but I don't know if it is technically feasible to transmit that... anyone? If not, 29.97/30?
I still don't really understand why we have people moaning about the WBA v Newcastle game on BBC HD, stating that there is a lack of definition when the camera pans. I hope you've read this thread and perhaps understand why this is so and it's nothing to do with the BBC.
I still don't really understand why we have people moaning about the WBA v Newcastle game on BBC HD, stating that there is a lack of definition when the camera pans. I hope you've read this thread and perhaps understand why this is so and it's nothing to do with the BBC.
Actually fast movement and action sends the bitrate up in stat muxing, so the fact BBC HD cannot currently vary its bitrate is part of the reason why the image quality is not as good as it should be.
Actually fast movement and action sends the bitrate up in stat muxing, so the fact BBC HD cannot currently vary its bitrate is part of the reason why the image quality is not as good as it should be.
Rather irrelevant in the overall scheme of thinks regarding the issue of fast panning cameras though.
HD is always a bit of a con. It's higher definition but of course it could have been a lot higher if someone wanted to pay for the bandwidth and higher standards had been agreed.
Increasing the horizontal and vertical resolution is all well and good but an increase in the temporal resolution would have been good. I've seen images recorded and dispalyed at 120 frames per second and I was most impressed when the camera did a whip pan. No loss of definition at all.
Just stop moaning at the BBC they didn't decide on all these constraints by themselves.
Wasn't this the same benevolence that gave us Five on Freesat, let alone Five on Freesat with excellent picture quality? Good old auntie Beeb!
I could be wrong but five now has public service obligations after getting extra coverage with freeview etc so it's recomended to be on as many platform as possible, maybe even forced? requierment to get the full pbs go ahead from ofcom...
And when ITV1HD is on 2D no reason for it to not be on Sky!
So ofcom should then force BBC but not Sky? way to regulate the market, im pretty sure they could agree on a cost for the transponders sky has aswell. IF they'd actually do something about it and faced the problem instead of stuffing channels on and cutting quality.
I still don't really understand why we have people moaning about the WBA v Newcastle game on BBC HD, stating that there is a lack of definition when the camera pans. I hope you've read this thread and perhaps understand why this is so and it's nothing to do with the BBC.
I'm not moaning about the BBC and (I think) I never have done. My grumble is with professionals... video makers, broadcasters and standards setters (of whom the BBC are of course part) who have over-hyped spatial resolution (because it is easy to hype by using still or relatively static images) when they have done little or nothing about the equally important temporal resolution, judder etc. Professionals who record video at 24 fps are among the worst culprits.
It just so happens that the subject came up in a BBC thread, although they could set a good example by placing less emphasis on spatial resolution and at least broadcasting some material in 720p/50 (if technically feasible). We should ideally be moving towards 1080p/50 as soon as that is feasible but for now, some 720p/50 would be nice. How would the bandwidth requirements compare with 1080i/25 (frames)?
I still don't really understand why we have people moaning about the WBA v Newcastle game on BBC HD, stating that there is a lack of definition when the camera pans. I hope you've read this thread and perhaps understand why this is so and it's nothing to do with the BBC.
We've seen no proof its nothing to do with the BBC, in fact by your own admission further down, you've seen how 120fps improves motion.
However, what you didn't mention is that it would have to be reduced to 50fps anyway because thats PAL spec - 50fps max.
I too as someone who shoots HD, am well aware of the affect of higher frames. However there are many factors to a HD picture with motion:
1. Frame Rate
2. Shutter Speed
3. Compression
Plus it could be some tv's don't handle the new compression as well as others.
You can't just blame the cameras and not the BBC without knowing where the problem lies because it could be compression just as much as camera settings. I'm not saying it is the BBC but I'm not saying it isn't either.
Also, don't think all you've got to do is set a high shutter and frame rate.
As I pointed out above, any frame rate over 50fps needs to reduced to 50fps anyway, so you're just discarding frames when you edit it to PAL standards.
Set it the frame rate too high and you get a jumpy picture where the missing frames become apparent. Set it too low and you get blur - 24fps is well noted for blur.
Same with shutter speed, set it too high and the information that's not recorded between frames becomes apparent in just the same way. I've seen racing car footage shot at 1/500th where the car literally stutters around the track in a series of small jumps.
JefG1 seems to have sorted itself out as far as MCE & MS Decoder. Just finished watching the Manson Doc on 5 and switched over to BBChd with dan cruckshank and death architecture and no more BlockiVision Lets hope it holds.
Might I add the Manson documentary was rather good definately a good watch if they repeat it, hats off to 5 or the american co that comissioned it Excellent.
JefG1 seems to have sorted itself out as far as MCE & MS Decoder. Just finished watching the Manson Doc on 5 and switched over to BBChd with dan cruckshank and death architecture and no more BlockiVision Lets hope it holds.
Might I add the Manson documentary was rather good definately a good watch if they repeat it, hats off to 5 or the american co that comissioned it Excellent.
Regards
I saw blockivision on Dan Cruckshank I'm afraid.
I only watched it for 5 mins though
<SNIP>
It just so happens that the subject came up in a BBC thread, although they could set a good example by placing less emphasis on spatial resolution and at least broadcasting some material in 720p/50 (if technically feasible). We should ideally be moving towards 1080p/50 as soon as that is feasible but for now, some 720p/50 would be nice. How would the bandwidth requirements compare with 1080i/25 (frames)?
720p50 requires pretty much the same bandwidth as 1080i25 (in uncompressed form 1280*720*50*3 == 132Mbytes/sec, 1920*1080*25*3 == 148MBytes/sec - assuming 8bit pixel depth). 1080p50 would require considerably more bandwidth, though probably not quite double. The H.264 encoders in use are already running at pretty close to 100:1 compression ratio when HD is broadcast at 10MBits/s, which is very impressive for something working on-the-fly.
I don't understand the reasoning for not broadcasting 1080i25 or 720p50 as required - if the spec for the decoder incorporates the requirement to rescale to whatever to feed the telly there won't be any horrible visible glitches. But HEY what do I know?
The H.264 encoders in use are already running at pretty close to 100:1 compression ratio when HD is broadcast at 10MBits/s, which is very impressive for something working on-the-fly.
Perhaps so, but what doesn't quite add up for me is that an AVCHD camcorder can create superb 1440x1080/25i H.264 video (and AC3 audio) on the fly already, for example mine does that to an SD card for nearly 2 hours before the 1.3Ah battery runs out. The battery is not much bigger than an OXO cube and even at the lowest quality setting of 5mbps the results look excellent even in severe "hosing" situations. I find it very hard to find fault with the results, especially at the higher settings of 9mbps and 12mbps (the latter is full 1920x1080 HD). It is only when you freeze-frame and go closer to the screen that you can really pick out any artefacts.
So I'm wondering what the difficulty is when a camcorder which is not much larger than the cardboard core of a toilet roll is already able to do this job very well, portably, using little power, in more difficult "wobbly amateur" circumstances, and at a lower bit rate. I.e. is real-time H.264 encoding really that hard these days?
Reading this column, I'm glad that I'm not the only one that dislikes low motion temporal rates.
The broadcast equipment company Snell & Wilcox did some early HD experiments with 75Hz at 768 lines and found it gave better results than 1150 lines at 50Hz.
Doug Trumball in the States experimented with higher picture rates and found that 72fps was the maximum rate that produced the most EMOTIONAL response in viewers. He settled on 60fps for his 'Showscan' film format to harmonise with US TV.
Do the math; 720@72 or 1080@50. As someone mentioned, line structure becomes much less significant at high frame rates (and faster shutter speeds!) All for a less punishing bit rate! It seems that temporal resolution needs to increase in proportion with spacial definition.
With many Red Camera users shooting high frame rate stuff, and recoding to 24/5fps, its easy to download stuff captured at 72 or 75 fps clips and play them in VLC or quicktime at 3x speed to see what 'natural motion' looks like. Be warned though, because once seen, it'll spoil how you look at 24/5 or 50fps forever!
After all, 24fps came about purely to save on the cost of film. Regrettably a 'film look' is seen to be aspriational, attributed to big budget productions so therefore it must be 'better'.
A parallel can be drawn here. CB was orginally illegal. There was a boom in imported US AM CB tranceivers. When CB was legalised, FM was specified but most CBers hated it because it sounded 'boxy' compared to their preference for the warmer sounding AM, even though FM is technically superior.
Comments
Hallelujah, about flaming time!
Can't wait
I still don't really understand why we have people moaning about the WBA v Newcastle game on BBC HD, stating that there is a lack of definition when the camera pans. I hope you've read this thread and perhaps understand why this is so and it's nothing to do with the BBC.
Rather irrelevant in the overall scheme of thinks regarding the issue of fast panning cameras though.
HD is always a bit of a con. It's higher definition but of course it could have been a lot higher if someone wanted to pay for the bandwidth and higher standards had been agreed.
Increasing the horizontal and vertical resolution is all well and good but an increase in the temporal resolution would have been good. I've seen images recorded and dispalyed at 120 frames per second and I was most impressed when the camera did a whip pan. No loss of definition at all.
Just stop moaning at the BBC they didn't decide on all these constraints by themselves.
I could be wrong but five now has public service obligations after getting extra coverage with freeview etc so it's recomended to be on as many platform as possible, maybe even forced? requierment to get the full pbs go ahead from ofcom...
lower quality picture on the free one aswell...
So ofcom should then force BBC but not Sky? way to regulate the market, im pretty sure they could agree on a cost for the transponders sky has aswell. IF they'd actually do something about it and faced the problem instead of stuffing channels on and cutting quality.
I'm not moaning about the BBC and (I think) I never have done. My grumble is with professionals... video makers, broadcasters and standards setters (of whom the BBC are of course part) who have over-hyped spatial resolution (because it is easy to hype by using still or relatively static images) when they have done little or nothing about the equally important temporal resolution, judder etc. Professionals who record video at 24 fps are among the worst culprits.
It just so happens that the subject came up in a BBC thread, although they could set a good example by placing less emphasis on spatial resolution and at least broadcasting some material in 720p/50 (if technically feasible). We should ideally be moving towards 1080p/50 as soon as that is feasible but for now, some 720p/50 would be nice. How would the bandwidth requirements compare with 1080i/25 (frames)?
ITV only announced theyre "looking into it" as far as freesat goes. Thats not quite the same as saying they're going to do it.
We've seen no proof its nothing to do with the BBC, in fact by your own admission further down, you've seen how 120fps improves motion.
However, what you didn't mention is that it would have to be reduced to 50fps anyway because thats PAL spec - 50fps max.
I too as someone who shoots HD, am well aware of the affect of higher frames. However there are many factors to a HD picture with motion:
1. Frame Rate
2. Shutter Speed
3. Compression
Plus it could be some tv's don't handle the new compression as well as others.
You can't just blame the cameras and not the BBC without knowing where the problem lies because it could be compression just as much as camera settings. I'm not saying it is the BBC but I'm not saying it isn't either.
Also, don't think all you've got to do is set a high shutter and frame rate.
As I pointed out above, any frame rate over 50fps needs to reduced to 50fps anyway, so you're just discarding frames when you edit it to PAL standards.
Set it the frame rate too high and you get a jumpy picture where the missing frames become apparent. Set it too low and you get blur - 24fps is well noted for blur.
Same with shutter speed, set it too high and the information that's not recorded between frames becomes apparent in just the same way. I've seen racing car footage shot at 1/500th where the car literally stutters around the track in a series of small jumps.
Might I add the Manson documentary was rather good definately a good watch if they repeat it, hats off to 5 or the american co that comissioned it Excellent.
Regards
I saw blockivision on Dan Cruckshank I'm afraid.
I only watched it for 5 mins though
720p50 requires pretty much the same bandwidth as 1080i25 (in uncompressed form 1280*720*50*3 == 132Mbytes/sec, 1920*1080*25*3 == 148MBytes/sec - assuming 8bit pixel depth). 1080p50 would require considerably more bandwidth, though probably not quite double. The H.264 encoders in use are already running at pretty close to 100:1 compression ratio when HD is broadcast at 10MBits/s, which is very impressive for something working on-the-fly.
I don't understand the reasoning for not broadcasting 1080i25 or 720p50 as required - if the spec for the decoder incorporates the requirement to rescale to whatever to feed the telly there won't be any horrible visible glitches. But HEY what do I know?
Perhaps so, but what doesn't quite add up for me is that an AVCHD camcorder can create superb 1440x1080/25i H.264 video (and AC3 audio) on the fly already, for example mine does that to an SD card for nearly 2 hours before the 1.3Ah battery runs out. The battery is not much bigger than an OXO cube and even at the lowest quality setting of 5mbps the results look excellent even in severe "hosing" situations. I find it very hard to find fault with the results, especially at the higher settings of 9mbps and 12mbps (the latter is full 1920x1080 HD). It is only when you freeze-frame and go closer to the screen that you can really pick out any artefacts.
So I'm wondering what the difficulty is when a camcorder which is not much larger than the cardboard core of a toilet roll is already able to do this job very well, portably, using little power, in more difficult "wobbly amateur" circumstances, and at a lower bit rate. I.e. is real-time H.264 encoding really that hard these days?
The camcorder is an old Panasonic 3CCD HDC-SD5.
Reading this column, I'm glad that I'm not the only one that dislikes low motion temporal rates.
The broadcast equipment company Snell & Wilcox did some early HD experiments with 75Hz at 768 lines and found it gave better results than 1150 lines at 50Hz.
Doug Trumball in the States experimented with higher picture rates and found that 72fps was the maximum rate that produced the most EMOTIONAL response in viewers. He settled on 60fps for his 'Showscan' film format to harmonise with US TV.
Do the math; 720@72 or 1080@50. As someone mentioned, line structure becomes much less significant at high frame rates (and faster shutter speeds!) All for a less punishing bit rate! It seems that temporal resolution needs to increase in proportion with spacial definition.
With many Red Camera users shooting high frame rate stuff, and recoding to 24/5fps, its easy to download stuff captured at 72 or 75 fps clips and play them in VLC or quicktime at 3x speed to see what 'natural motion' looks like. Be warned though, because once seen, it'll spoil how you look at 24/5 or 50fps forever!
After all, 24fps came about purely to save on the cost of film. Regrettably a 'film look' is seen to be aspriational, attributed to big budget productions so therefore it must be 'better'.
A parallel can be drawn here. CB was orginally illegal. There was a boom in imported US AM CB tranceivers. When CB was legalised, FM was specified but most CBers hated it because it sounded 'boxy' compared to their preference for the warmer sounding AM, even though FM is technically superior.