Well, I see artefacts that look like deinterlace problems on the running track lines, especially on the thick one which is the innermost boundary, and on all of them as the camera sweeps around the final bend to the winning post (if that's what it's called). The close-ups of athletes do look fine.
I'm going to have to fiddle with the upscaling settings (reon HQV) to see if I can do anything about it with a recording of the problem pictures. In the meantime, what TV do you have? Did you see any of these sort of shots? I'm quite willing to believe it's my set-up (Panny PZ80), especially as no-one else seems to be commenting on these artefacts.
I am seeing them too, although not obvious from all cameras.
Using VP50pro and PZ80, so we do have a common denominator.
However I am not convinced it's the plasma as we are both feeding a progressive signal to it from different deinterlacers.
I will take the vp50 out of circuit and see what happens.
Maybe a moot point on moving images as the encoding will filter those high frequencies to a degree, but it gives an idea of what the plasma can resolve.
I'm using a Fortec Passion +, so effectively freesat without the copy restrictions. It has its own (functionality) problems, but the pq seems good. My first impression was that it was better than the non-HDR Foxsat I had previously, but at any rate it's not a weak link in the chain. Overscan off on the Panny.
The Kuro is not full HD, not that that is a bad thing in itself, but it does mean the input will be downscaled.
Not quite as no pictures are transmitted on broadcast as full hd anyway as it takes too much bandwidth. They're all transmitted as 1080i. Full HD is 1080P. The Kuro is 1080i so they appear on the panel at full resolution.
Maybe a moot point on moving images as the encoding will filter those high frequencies to a degree, but it gives an idea of what the plasma can resolve.
I haven't seen the BBC test card so can't comment, sorry.
BTW if you check out the 428XD's reviews then right across the board you'll find that for PQ it beat all the other comparable panels including those with Full HD. Sometimes sheer number of pixels isn't everything! I don't think you can blame the Kuro for not seeing aretfacts. If anything in the Tv is to blame its superior processing.
Not quite as no pictures are transmitted on broadcast as full hd anyway as it takes too much bandwidth. They're all transmitted as 1080i. Full HD is 1080P. The Kuro is 1080i so they appear on the panel at full resolution.
I haven't seen the BBC test card so can't comment, sorry.
BTW if you check out the 428XD's reviews then right across the board you'll find that for PQ it beat all the other comparable panels including those with Full HD. Sometimes sheer number of pixels isn't everything! I don't think you can blame the Kuro for not seeing aretfacts. If anything in the Tv is to blame its superior processing.
I was not in any way criticising the Kuro. I know it's a good plasma. There are indeed many aspects which go to make up overall picture quality. Resolution is just one.
Your Kuro according to Pioneers website is 1024h x 768v. So it does not display the full (static) 1440 * 1080i resolution broadcast by BBCHD.
Although broadcast interlaced, plasma and lcds de-interlace pictures before displaying them, so you are always watching a progressive image that is scaled to fit the native resolution of the panel. In your case 768p.
This de-interlacing along with downscaling will have an effect on the picture, and yes different processing in different TVs will produce different results.
Thats all I was saying.
You will almost certainly see what I mean if you do take a look at the BBC test card. I will be surprises if you can clearly see the gratings in the bottom two boxes.
Is it superior if it's hiding things that are in the original image?
For reference my deinterlacing is done by an external VP50pro processor. Mwardys is also done by a different Reon HQV processsor.
I also see from your plasmas spec sheet it has various noise reduction features including block, mpeg (mosquito???) and digital noise reduction.
Unless you can specifically disable all of these in the menu, they could be helping to mask some of the issues currently exhibited by the new BBCHD encoders.
BTW the issue myself and mwardy are referring to I don't think has anything to do with the new encoders. Just looks like a classic interlace/deinterlacing issue.
I'm not denying there are some problems with the current picture given the number of report from people that say that there are, I'm just saying that I'm not seeing all of them which means there might be an equipment factor in here. Which in turn might help point to the exact cause.
I'm not denying there are some problems with the current picture given the number of report from people that say that there are, I'm just saying that I'm not seeing all of them which means there might be an equipment factor in here. Which in turn might help point to the exact cause.
Absolutely.
Humax have confirmed that my HDR has block and other noise reduction features.
The problems, although I can see them, appear less obvious to me, than they do to other posters.
This is the issue, are our set-ups masking the issues? Are the other posters setups more transparant and just showing the picture for what it is?
Just out of curiosity look carefully whenever there is a mix of pictures (not an instant cut from one camera to another) or a fade to/from black.
You will often see blocking. It's obvious when you know what to look for.
It's a lot better than it was when the new encoders first went on air, but was still apparant last time I looked.
I'm not denying there are some problems with the current picture given the number of report from people that say that there are, I'm just saying that I'm not seeing all of them which means there might be an equipment factor in here. Which in turn might help point to the exact cause.
What the STB and display can do often makes whatever the artefact is appear worse ... and often very different from the orignal Transmitted error -if any.
This makes the job of finding out what is happening rather difficult - and that is assuming that the display is set up correctly... e.g. not excessive sharpness or Saturated colours.
In my experience there is usually something which (eventually) you can track down as the trigger but how visable this is on another display (or CRT) is often marginal.
... and this is in a world where I have the material on a stream recorder ... so can loop the "error".
It was bad enough on MPEG 2 - MPEG4AVC has a different set of issues which seem to confuse some upscalers/ deinterlacers.
I've been distracted by a technical glitch so haven't done a proper comparison on the athletics yet, but to back up what technologist has said, one thing is obvious, which is that the Fortec upscaler adds a load of sharpness compared to the Reon. This immediately makes it look 'better'. (I inadvertently had the stb at 1080i so the Reon was just doing the deinterlacing.) When the Fortec is not upscaling I can suddenly see why people had been complaining that the onscreen graphics looked fuzzy whereas I'd found them pin sharp. But of course this may be the source of some of the problems with the lines on the running track, which at their worse pulsate like they are alive.
Getting the best picture will take some sorting out. Fun. :rolleyes:
I've been distracted by a technical glitch so haven't done a proper comparison on the athletics yet, but to back up what technologist has said, one thing is obvious, which is that the Fortec upscaler adds a load of sharpness compared to the Reon. This immediately makes it look 'better'. (I inadvertently had the stb at 1080i so the Reon was just doing the deinterlacing.) When the Fortec is not upscaling I can suddenly see why people had been complaining that the onscreen graphics looked fuzzy whereas I'd found them pin sharp. But of course this may be the source of some of the problems with the lines on the running track, which at their worse pulsate like they are alive.
Getting the best picture will take some sorting out. Fun. :rolleyes:
I am a little confused, wouldn't you want the STB to output 1080i on HD material, the native picture resolution? Then let the reon deinterlace?
I am a little confused, wouldn't you want the STB to output 1080i, the native picture resolution? Then let the reon deinterlace?
I assume the reon is in an AV amp?
Sorry, yes, of course you are right. Please ignore that total nonsense. I must have been thinking about the glitch I mentioned.
The only issue is whether the reon or the panel should deinterlace. As I'm sure you know, the PZ is always reported to have an underwhelming deinterlacer, but just for this artefact it seems to be a bit better, on a brief test. What happens without the VP50? It's such a particular issue though that I'm going to leave the reon doing the duties. And yes, the reon is in an Onkyo 875.
But between us it seems to just confirm that this is a classic deinterlacing issue.
BTW,it was only the BBC blue/black/grey graphics that looked like upscaled SD, not the main orange/brown ones.
Also the sharpness on the panny is turned all the way down to the left. I only add a very small amount of "fine detail" and edge enhancement on the vp50 scaler. 14 and 4 respectively on a scale of 100!
Sorry, yes, of course you are right. Please ignore that total nonsense. I must have been thinking about the glitch I mentioned.
The only issue is whether the reon or the panel should deinterlace. As I'm sure you know, the PZ is always reported to have an underwhelming deinterlacer, but just for this artefact it seems to be a bit better, on a brief test. What happens without the VP50? It's such a particular issue though that I'm going to leave the reon doing the duties. And yes, the reon is in an Onkyo 875.
But between us it seems to just confirm that this is a classic deinterlacing issue.
I would expect the reon to outperform the panny deinterlacer, although I admit to not having first hand experience of the chip.
I don't get jaggies as you described earlier but a sort of twittering on some lines. Not obvious on all shots.
One where I do see it is at the race finish where the camera is 90deg to the track and pulls back. As you say the track lines sort of "pulsate" slightly.
I put this down to a deinterlacing issue and is the same with the VP50 out of circuit.
One thing the VP50 does do nicely is bring the pannys gamma into line. Does the reon have a gamma adjustment?
I wish panasonic would put these controls in the UK menus.
Apparantly they are there on the same european models.
I don't get jaggies as you described earlier but a sort of twittering on some lines. Not obvious on all shots.
One where I do see it is at the race finish where the camera is 90deg to the track and pulls back. As you say the track lines sort of "pulsate" slightly.
I put this down to a deinterlacing issue and is the same with the VP50 out of circuit.
Exactly the same here on the affected shots and effects. Interesting to hear that the vp50 can't overcome it, but as I say, on a quick look the reon seems to exaggerate it a fair bit. It's worse on the slow motion replay. I only called it 'jaggies' because it's a better known term than twittering (I think). For real jaggies, I watch my old Panny CRT.
One thing the VP50 does do nicely is bring the pannys gamma into line. Does the reon have a gamma adjustment?
I wish panasonic would put these controls in the UK menus.
Apparantly they are there on the same european models.
No gamma adjustment on the 875. I tried to justify to myself getting an 876, which does have it, and per input white balance adjustments, but I couldn't. Nowadays I'm looking at a professional panel to get rid of these bugbears, though I wouldn't be surprised if the Euro controls appeared on the next gen TVs given the bad publicity that has been generated about the cut down UK versions.
Exactly the same here on the affected shots and effects. Interesting to hear that the vp50 can't overcome it, but as I say, on a quick look the reon seems to exaggerate it a fair bit. It's worse on the slow motion replay. I only called it 'jaggies' because it's a better known term than twittering (I think). For real jaggies, I watch my old Panny CRT.
.
Your right about the slo mos. Definitely nothing to do with our end. Thats in the transmission.
Personally I want to see 1080p300 as soon as possible, at the encoder it actually has some significant advantages because the difference between each frame is substantially smaller and thus the temporal motion estimation is made much easier. Plus because it is not interlaced you get a simpler image for modern compression.
Bob
Interestingly enough Elan got his wish.
I watched the Freerunning Championships on BBC HD today (recorded).
Whilst the general cameras on Trafalgar Square showed some movement blur (specifically they looked at a guy at the traffic lights in a Rolls Royce and as they zoomed out the blur was there), the Freerunners themselves were razor sharp.
Halfway through the presenter said "we are of course using special high tech cameras that shoot 300 frames per second!!!!"
Now whilst it perhaps couldn't be used live, it does raise the question as to why these cameras couldn't be used to record pre-recorded athletic meets or football highlights (assuming the latter could be processed by the evening slot)
I thought the freerunning thing looked really good when I was channel hopping the other night. Makes Bob_Cat's post seem less like sci fi or an academic argument (which, genuinely, I'd taken it to be) and more like insider knowledge. Cool!
Obviously there is no infrastructure to show these things as is, but it clearly helps for reencoding as Bob says. More please!
I thought the freerunning thing looked really good when I was channel hopping the other night. Makes Bob_Cat's post seem less like sci fi or an academic argument (which, genuinely, I'd taken it to be) and more like insider knowledge. Cool!
Obviously there is no infrastructure to show these things as is, but it clearly helps for reencoding as Bob says. More please!
Or Bob_cat has read teh BBC's R&D white paper on the subject?
Comments
I am seeing them too, although not obvious from all cameras.
Using VP50pro and PZ80, so we do have a common denominator.
However I am not convinced it's the plasma as we are both feeding a progressive signal to it from different deinterlacers.
I will take the vp50 out of circuit and see what happens.
Thanks for the info. But what's not plasma?
It would be good if you could see whether you do or don't have the 'jaggies'/twitter on the long shots of the running track.
mwardy, are you Sky?
Humax hdr here.
presume overscan turned off on the panny?
The Kuro is not full HD, not that that is a bad thing in itself, but it does mean the input will be downscaled.
That will have an effect on the appearance (or non appearance) of thin lines at high frequencies.
White Night, on the BBC test card can you see lines in the bottom frequency grating box on the right hand side?
Can you see this?
http://picasaweb.google.co.uk/lh/photo/ZDzSF7Z7n8oTmx62DTFP7Q?feat=directlink
Maybe a moot point on moving images as the encoding will filter those high frequencies to a degree, but it gives an idea of what the plasma can resolve.
I was wondering about the 720p on the Kuro.
What Hi Fi Magazine
http://whathifi.com/News/BBC-admits-to-problems-with-new-HD-encoders/
Not quite as no pictures are transmitted on broadcast as full hd anyway as it takes too much bandwidth. They're all transmitted as 1080i. Full HD is 1080P. The Kuro is 1080i so they appear on the panel at full resolution.
I haven't seen the BBC test card so can't comment, sorry.
BTW if you check out the 428XD's reviews then right across the board you'll find that for PQ it beat all the other comparable panels including those with Full HD. Sometimes sheer number of pixels isn't everything! I don't think you can blame the Kuro for not seeing aretfacts. If anything in the Tv is to blame its superior processing.
I was not in any way criticising the Kuro. I know it's a good plasma. There are indeed many aspects which go to make up overall picture quality. Resolution is just one.
Your Kuro according to Pioneers website is 1024h x 768v. So it does not display the full (static) 1440 * 1080i resolution broadcast by BBCHD.
http://www.pioneer.co.uk/uk/products/archive/PDP-428XD/index.html
Although broadcast interlaced, plasma and lcds de-interlace pictures before displaying them, so you are always watching a progressive image that is scaled to fit the native resolution of the panel. In your case 768p.
This de-interlacing along with downscaling will have an effect on the picture, and yes different processing in different TVs will produce different results.
Thats all I was saying.
You will almost certainly see what I mean if you do take a look at the BBC test card. I will be surprises if you can clearly see the gratings in the bottom two boxes.
Is it superior if it's hiding things that are in the original image?
For reference my deinterlacing is done by an external VP50pro processor. Mwardys is also done by a different Reon HQV processsor.
I also see from your plasmas spec sheet it has various noise reduction features including block, mpeg (mosquito???) and digital noise reduction.
Unless you can specifically disable all of these in the menu, they could be helping to mask some of the issues currently exhibited by the new BBCHD encoders.
BTW the issue myself and mwardy are referring to I don't think has anything to do with the new encoders. Just looks like a classic interlace/deinterlacing issue.
I'm not denying there are some problems with the current picture given the number of report from people that say that there are, I'm just saying that I'm not seeing all of them which means there might be an equipment factor in here. Which in turn might help point to the exact cause.
Absolutely.
Humax have confirmed that my HDR has block and other noise reduction features.
The problems, although I can see them, appear less obvious to me, than they do to other posters.
This is the issue, are our set-ups masking the issues? Are the other posters setups more transparant and just showing the picture for what it is?
Just out of curiosity look carefully whenever there is a mix of pictures (not an instant cut from one camera to another) or a fade to/from black.
You will often see blocking. It's obvious when you know what to look for.
It's a lot better than it was when the new encoders first went on air, but was still apparant last time I looked.
duplicate post
This makes the job of finding out what is happening rather difficult - and that is assuming that the display is set up correctly... e.g. not excessive sharpness or Saturated colours.
In my experience there is usually something which (eventually) you can track down as the trigger but how visable this is on another display (or CRT) is often marginal.
... and this is in a world where I have the material on a stream recorder ... so can loop the "error".
It was bad enough on MPEG 2 - MPEG4AVC has a different set of issues which seem to confuse some upscalers/ deinterlacers.
Getting the best picture will take some sorting out. Fun. :rolleyes:
I am a little confused, wouldn't you want the STB to output 1080i on HD material, the native picture resolution? Then let the reon deinterlace?
I assume the reon is in an AV amp?
Sorry, yes, of course you are right. Please ignore that total nonsense.
The only issue is whether the reon or the panel should deinterlace. As I'm sure you know, the PZ is always reported to have an underwhelming deinterlacer, but just for this artefact it seems to be a bit better, on a brief test. What happens without the VP50? It's such a particular issue though that I'm going to leave the reon doing the duties. And yes, the reon is in an Onkyo 875.
But between us it seems to just confirm that this is a classic deinterlacing issue.
BTW,it was only the BBC blue/black/grey graphics that looked like upscaled SD, not the main orange/brown ones.
Also the sharpness on the panny is turned all the way down to the left. I only add a very small amount of "fine detail" and edge enhancement on the vp50 scaler. 14 and 4 respectively on a scale of 100!
I would expect the reon to outperform the panny deinterlacer, although I admit to not having first hand experience of the chip.
I don't get jaggies as you described earlier but a sort of twittering on some lines. Not obvious on all shots.
One where I do see it is at the race finish where the camera is 90deg to the track and pulls back. As you say the track lines sort of "pulsate" slightly.
I put this down to a deinterlacing issue and is the same with the VP50 out of circuit.
One thing the VP50 does do nicely is bring the pannys gamma into line. Does the reon have a gamma adjustment?
I wish panasonic would put these controls in the UK menus.
Apparantly they are there on the same european models.
That's right. This is a bit of an odd case I think, which is why I'm leaving the reon doing the deinterlacing.
Exactly the same here on the affected shots and effects. Interesting to hear that the vp50 can't overcome it, but as I say, on a quick look the reon seems to exaggerate it a fair bit. It's worse on the slow motion replay. I only called it 'jaggies' because it's a better known term than twittering (I think). For real jaggies, I watch my old Panny CRT.
No gamma adjustment on the 875.
Your right about the slo mos. Definitely nothing to do with our end. Thats in the transmission.
Interestingly enough Elan got his wish.
I watched the Freerunning Championships on BBC HD today (recorded).
Whilst the general cameras on Trafalgar Square showed some movement blur (specifically they looked at a guy at the traffic lights in a Rolls Royce and as they zoomed out the blur was there), the Freerunners themselves were razor sharp.
Halfway through the presenter said "we are of course using special high tech cameras that shoot 300 frames per second!!!!"
Now whilst it perhaps couldn't be used live, it does raise the question as to why these cameras couldn't be used to record pre-recorded athletic meets or football highlights (assuming the latter could be processed by the evening slot)
Obviously there is no infrastructure to show these things as is, but it clearly helps for reencoding as Bob says. More please!
Or Bob_cat has read teh BBC's R&D white paper on the subject?
http://downloads.bbc.co.uk/rd/pubs/whp/whp-pdf-files/WHP169.pdf