Still, patents should have variable term lengths depending on the industry. At this rate, each generation of wireless is patent-encumbered the whole time it's current-gen and beyond.
If you've ever given access to WiFi settings or Bluetooth to an app, those can correlate locations without any additional location data. The MAC address of a nearby WiFi access point is possibly in a database. Any app besides Google could share that info with an advertising network. That's not to say that Google isn't using it for their own network by default.
Location services don't usually even use GPS due to battery concerns. And location services APIs are not used for placing 911 calls. This could happen just as easily without a GPS chip.
I've tried both. On the low end - Home Mini vs Dot - the Echo device wins by a mile on hearing its wake word in a noisy room while music is playing and still picking up what you want to say. Plus they have audio out to better speakers. Even still, it's useless compared to a Google Home product on what it can actually do.
To near black very quickly, but the first fields persists until after the second field of the frame is drawn.
Strobing the backlight is to hide the pixel transition because the pixels do not fade and there is a short period where the liquid crystal is responding to voltage changes and the old frame doesn't go away instantly. Similar idea, but it's not because of the black. It's because your brain doesn't process the black inter-frame period and you only see whole frames. It's not that the blackness gives your brain the ability to perceive motion so much that it hides incomplete frames with mangled motion from view.
Nobody's talking about that here. Comcast deinterlaces one interlaced frame into two progressive frames. So instead of 2 fields, 29.97 times per second there are individual progressive frames played at 59.94 times per second. If you're going down to just 29.97 progressive frames per second there's a resolution and temporal quality loss, but not here.
It's obvious when you look at the maps. If you're at the fringes of reception on a 4G tower, you're marked as 4G. Even if your signal strength and SNR are barely good enough for dial up speeds, it's happening with 4G tech. And just past that fringe, there's a hard cut off.
This. You want perspectives to change in these areas? Give them exposure to the outside world. In a lot of cases, it's a cure for xenophobia if treatment it's started early.
It is not physically possible to maintain temporal quality unless the viewer is happy seeing only half an image per field. Since they would never accept that you MUST eat into time domain in order to display a coherent image. The algorithms to do this are vast and varied. The one commonality they ALL share every last one of them are tradeoffs. You never ever get 60 fps out of interlaced to an actual persistent display.
You're blaming a lot on Comcast that has nothing to do with them. You're both complaining that they aren't sending an interlaced signal for an interlaced source and simultaneously blaming them for the trade-offs involved with the conversion to progressive for ALL modern displays. Look, either Comcast is doing it or your TV is doing it - so you can't blame them for LCDs being progressive displays. And MPEG-4 doesn't compress interlaced content as well, so the choice there is obvious. If you've ever seen a UI being transmitted at hi-def with an interlaced signal, you know that the receiver is going to need to output progressive to prevent flickering on edges when in menus or when an OSD is active. So it still can't be run to the TV as interlaced without a lot of annoyance. And the TV can't do anything but convert an interlaced signal anyway.
And you don't see half an image per field. When converting to progressive, the image has already expanded this to a full frame with de-combing and starting with 540 lines per field and 1920 pixels across on the 1080i channels, which is plenty of information to work with. You don't have to eat into the time domain if you convert each field into a full picture and run that at 60fps. The trade-off is all about not having to encode interlaced video and losing quality to bitrate.
The standard most certainly exists. Failure to use it is all Comcast.
The standard IS 3:2 pulldown. Because not all TVs accept a 24p source, but anything that can will also be able to do reverse pulldown. The TV can automatically detect the frame doubling.
And this of course in addition to the ridiculously low bandwidth allocations to remaining subscribers of their TV service.
No argument there. This is precisely why I don't subscribe to any live TV service, but still use my ATSC tuner - 20Mbps works just fine with MPEG-2 so long as you're not trying to run 3-4 subchannels.
By today's standards a non-interlaced frame is more or less the same. The whole sensor is sampled, but then reduced to alternating fields. Either way, there are 60 temporal frames in the video, and only the quality of their vertical resolution varies between interlaced and progressive.
All mpeg4 streams from Comcast are 720p not 720i.
What's your point? I was talking about the source. And 720p60 is just 720i where the fields are split into separate frames and then interpolated to fit the full frame. It doesn't reduce the temporal quality unless you're going down to 720p30. And if you're going from 1080i to 720p60, then you're going from 540 vertical lines to 720 - you lose a bit going from 1920 to 1280 width, but vertically you come out fine. Progressive content compresses better, so de-interlacing at a higher frame rate just makes sense.
3:2 is not an integer multiple. This disparity introduces noticeable artifacts.
Which has nothing to do with Comcast - there are no 24p broadcast standards in use. If your TV actually can display 24p, then your TV can reverse the pulldown and get the original framerate back. Same as pulldown-tagged DVDs did all the way back in the late 90's.
60fps is just a convenient rounding that ignores the complexity and legacy of the original analog signal.
This is why fruits of what Comcast is doing is shit.
No, this means they are using 59.94 and calling it "60." Everyone calls it 60, because that's the nominal AC voltage frequency. Your computer might be able to do 60Hz, but I assure you that Comcast is using 59.94.
Even the electrolytic capacitors have longer lifespans now. But that's not your only option.
Replacing the power supply is plenty cheap. But for the same reason above, you may not have to for a long time. It's the CPU fan that you'll probably have to change first.
59.94i is 29.97 full resolution frames per second. But it's really two half-frames from two different moments in time. And digital sensors are going to sensor the whole CCD for each field pass rather than every other row for those half frames. You're still getting 59.94 frames per second. They're just packed two to the frame before display for archaic reasons.
Maybe you can convince my local Fox affiliate to stop broadcasting Line 21 closed captioning. I have no idea why they would even have that in there instead of just EIA-708, but if you view full frame, the top of the picture has a flickering black/white bar with CC data.
This is how video compression works. Store key frames and interpolate between them. The issue is just basically whether you interpolate 24 times a second, or 200, or whatever.
Sort of. Between key frames, there are P-frames and B-frames that actually tell you what pixel macroblocks changed and how much. That interpolation is based on hard recorded data, but lossy. Adding in more data than what's there will still give bad results.
Most TVs with motion interpolation remove motion blur and algorithmically try to fill in the gaps. Otherwise there wouldn't be much to fill in between frames, except more blur.
Human eyesight has motion blur, but reality doesn't. We just need 1000Hz light field cameras and playback systems so the eye can see it as real.
And that's without mentioning what it does to visual effects. They just look like video games - and you can easily tell apart the real and fake parts of the frame.
In theory, you could do decent interpolation at the player since it has access to the MPEG motion vectors, and it can keep the audio and video in sync
In theory because even though the fancy math could represent a higher source resolution - it doesn't. The encoder was fed the reduced resolution from the beginning. I'm surprised higher-quality encoders aren't using 4K to calculate the vector transforms and then rounding down from there - but I'm not aware of any so far.
I watched The Hobbit in 48fps and it was barely watchable. One of the few movies you could see that was shot in a high frame rate, not interlaced. That framerate is in an uncanny valley. You can "get used to it" the same way you could get used to the characters in the Polar Express. In that movie, nothing looks right - even though it has far more realistic human faces than movies before it. You need far more than a couple hundred Hz refresh rate to look like natural motion, so you need to be below the threshold of looking weird.
Going back 50+ years, most TV was shot on film, developed, and then telecined with 3:2 pulldown to get interlaced 60fps video. Soaps were always shot on video because of the fast turnaround time required for daily broadcast.
Even today, most primetime content is shot at 24fps progressive, while soaps are shot 60fps interlaced.
Yeah - I agree it's useful to sports - where immersion is not the goal. You need just a "sports" button on the TV remote to go into and out of it. Or just don't watch sports (my choice).
Slashdot Mirror
That's about as useful in the real world as write-only disks.
Still, patents should have variable term lengths depending on the industry. At this rate, each generation of wireless is patent-encumbered the whole time it's current-gen and beyond.
If you've ever given access to WiFi settings or Bluetooth to an app, those can correlate locations without any additional location data. The MAC address of a nearby WiFi access point is possibly in a database. Any app besides Google could share that info with an advertising network. That's not to say that Google isn't using it for their own network by default.
Location services don't usually even use GPS due to battery concerns. And location services APIs are not used for placing 911 calls. This could happen just as easily without a GPS chip.
I've tried both. On the low end - Home Mini vs Dot - the Echo device wins by a mile on hearing its wake word in a noisy room while music is playing and still picking up what you want to say. Plus they have audio out to better speakers. Even still, it's useless compared to a Google Home product on what it can actually do.
What, Google doesn't support Unicode either?
Sounds more like GOTO hell.
The NSA??!
To near black very quickly, but the first fields persists until after the second field of the frame is drawn.
Strobing the backlight is to hide the pixel transition because the pixels do not fade and there is a short period where the liquid crystal is responding to voltage changes and the old frame doesn't go away instantly. Similar idea, but it's not because of the black. It's because your brain doesn't process the black inter-frame period and you only see whole frames. It's not that the blackness gives your brain the ability to perceive motion so much that it hides incomplete frames with mangled motion from view.
Nobody's talking about that here. Comcast deinterlaces one interlaced frame into two progressive frames. So instead of 2 fields, 29.97 times per second there are individual progressive frames played at 59.94 times per second. If you're going down to just 29.97 progressive frames per second there's a resolution and temporal quality loss, but not here.
It's obvious when you look at the maps. If you're at the fringes of reception on a 4G tower, you're marked as 4G. Even if your signal strength and SNR are barely good enough for dial up speeds, it's happening with 4G tech. And just past that fringe, there's a hard cut off.
This. You want perspectives to change in these areas? Give them exposure to the outside world. In a lot of cases, it's a cure for xenophobia if treatment it's started early.
50Hz is certainly in the range of good speakers. It may be feeling more than heard when loud enough, but I'm pretty sure it's in the range of hearing.
It is not physically possible to maintain temporal quality unless the viewer is happy seeing only half an image per field. Since they would never accept that you MUST eat into time domain in order to display a coherent image. The algorithms to do this are vast and varied. The one commonality they ALL share every last one of them are tradeoffs. You never ever get 60 fps out of interlaced to an actual persistent display.
You're blaming a lot on Comcast that has nothing to do with them. You're both complaining that they aren't sending an interlaced signal for an interlaced source and simultaneously blaming them for the trade-offs involved with the conversion to progressive for ALL modern displays. Look, either Comcast is doing it or your TV is doing it - so you can't blame them for LCDs being progressive displays. And MPEG-4 doesn't compress interlaced content as well, so the choice there is obvious. If you've ever seen a UI being transmitted at hi-def with an interlaced signal, you know that the receiver is going to need to output progressive to prevent flickering on edges when in menus or when an OSD is active. So it still can't be run to the TV as interlaced without a lot of annoyance. And the TV can't do anything but convert an interlaced signal anyway.
And you don't see half an image per field. When converting to progressive, the image has already expanded this to a full frame with de-combing and starting with 540 lines per field and 1920 pixels across on the 1080i channels, which is plenty of information to work with. You don't have to eat into the time domain if you convert each field into a full picture and run that at 60fps. The trade-off is all about not having to encode interlaced video and losing quality to bitrate.
The standard most certainly exists. Failure to use it is all Comcast.
The standard IS 3:2 pulldown. Because not all TVs accept a 24p source, but anything that can will also be able to do reverse pulldown. The TV can automatically detect the frame doubling.
And this of course in addition to the ridiculously low bandwidth allocations to remaining subscribers of their TV service.
No argument there. This is precisely why I don't subscribe to any live TV service, but still use my ATSC tuner - 20Mbps works just fine with MPEG-2 so long as you're not trying to run 3-4 subchannels.
Wrong...Most content is captured at either 59.94fps interlaced or 24fps.
A non-interlaced frame is not the same as an interlaced frame.
60 fps progressive = 60 fields, 60 fps interlaced = 30 fields.
By today's standards a non-interlaced frame is more or less the same. The whole sensor is sampled, but then reduced to alternating fields. Either way, there are 60 temporal frames in the video, and only the quality of their vertical resolution varies between interlaced and progressive.
All mpeg4 streams from Comcast are 720p not 720i.
What's your point? I was talking about the source. And 720p60 is just 720i where the fields are split into separate frames and then interpolated to fit the full frame. It doesn't reduce the temporal quality unless you're going down to 720p30. And if you're going from 1080i to 720p60, then you're going from 540 vertical lines to 720 - you lose a bit going from 1920 to 1280 width, but vertically you come out fine. Progressive content compresses better, so de-interlacing at a higher frame rate just makes sense.
3:2 is not an integer multiple. This disparity introduces noticeable artifacts.
Which has nothing to do with Comcast - there are no 24p broadcast standards in use. If your TV actually can display 24p, then your TV can reverse the pulldown and get the original framerate back. Same as pulldown-tagged DVDs did all the way back in the late 90's.
60fps is just a convenient rounding that ignores the complexity and legacy of the original analog signal.
This is why fruits of what Comcast is doing is shit.
No, this means they are using 59.94 and calling it "60." Everyone calls it 60, because that's the nominal AC voltage frequency. Your computer might be able to do 60Hz, but I assure you that Comcast is using 59.94.
Even the electrolytic capacitors have longer lifespans now. But that's not your only option.
Replacing the power supply is plenty cheap. But for the same reason above, you may not have to for a long time. It's the CPU fan that you'll probably have to change first.
59.94i is 29.97 full resolution frames per second. But it's really two half-frames from two different moments in time. And digital sensors are going to sensor the whole CCD for each field pass rather than every other row for those half frames. You're still getting 59.94 frames per second. They're just packed two to the frame before display for archaic reasons.
Maybe you can convince my local Fox affiliate to stop broadcasting Line 21 closed captioning. I have no idea why they would even have that in there instead of just EIA-708, but if you view full frame, the top of the picture has a flickering black/white bar with CC data.
This is how video compression works. Store key frames and interpolate between them. The issue is just basically whether you interpolate 24 times a second, or 200, or whatever.
Sort of. Between key frames, there are P-frames and B-frames that actually tell you what pixel macroblocks changed and how much. That interpolation is based on hard recorded data, but lossy. Adding in more data than what's there will still give bad results.
Most TVs with motion interpolation remove motion blur and algorithmically try to fill in the gaps. Otherwise there wouldn't be much to fill in between frames, except more blur.
Human eyesight has motion blur, but reality doesn't. We just need 1000Hz light field cameras and playback systems so the eye can see it as real.
And that's without mentioning what it does to visual effects. They just look like video games - and you can easily tell apart the real and fake parts of the frame.
In theory, you could do decent interpolation at the player since it has access to the MPEG motion vectors, and it can keep the audio and video in sync
In theory because even though the fancy math could represent a higher source resolution - it doesn't. The encoder was fed the reduced resolution from the beginning. I'm surprised higher-quality encoders aren't using 4K to calculate the vector transforms and then rounding down from there - but I'm not aware of any so far.
I watched The Hobbit in 48fps and it was barely watchable. One of the few movies you could see that was shot in a high frame rate, not interlaced. That framerate is in an uncanny valley. You can "get used to it" the same way you could get used to the characters in the Polar Express. In that movie, nothing looks right - even though it has far more realistic human faces than movies before it. You need far more than a couple hundred Hz refresh rate to look like natural motion, so you need to be below the threshold of looking weird.
Going back 50+ years, most TV was shot on film, developed, and then telecined with 3:2 pulldown to get interlaced 60fps video. Soaps were always shot on video because of the fast turnaround time required for daily broadcast.
Even today, most primetime content is shot at 24fps progressive, while soaps are shot 60fps interlaced.
Yeah - I agree it's useful to sports - where immersion is not the goal. You need just a "sports" button on the TV remote to go into and out of it. Or just don't watch sports (my choice).