Yup. Broadcast (ATSC) uses Mpeg-2. However both DirecTV and Dish use h.264 for many of their HD channels. That's why old DirecTiVo HDs miss out on a lot of programming.
Interestingly, prior art to all 3 of those patents exists as many TVs and VCRs from Phillips/Magnavox had this feature built in as far back as 1993: http://www.magnavox.com/index.cfm?event=about
Sony also included it in a few TV models back in 2001.
I suspect it has not seen more widespread adoption because of battery life and penny-pinching in a cutthroat market.
Unless you have a Docsis 3.0 modem, it is physically impossible for you to get 40-60Mbps. The calculation code is wrong somewhere.
Docsis 2.0 (North American annex-B) uses a single 6Mhz channel, with the highest available encoding being QAM-256 (8 bits per sample). The standardized sample rate is 5.36Msps, so *=8 gives the true peak throughput of 42.88 Mbps, of which ~38Mbps is usable data.
I see the same thing when starting a batch download with DownThemAll, which reports 6+ MBps regularly when starting, but it is simply inaccurate.
(That said, I am very happy with Comcast's "Blast" tier, which regularly gives me 20-30Mbps sustained rates as long as multiple connections are in use, like DownThemAll does. For whatever reason a single stream peaks out around 16Mbps sustained.)
-------
Gory details on 42.88 vs ~38Mbps usable, if anyone cares:
The standard specifies Reed-Solomon FEC (Forward Error Correction), which mean 6 samples per 128 are unavailable (-4.7%). The MPEG-2 transport stream structure further has a 4 byte header per 188 bytes (-2.1%), Ethernet eats another 18/1500 (-1.2%), IP headers and Docsis map traffic account for about -3% more overhead. It all adds up to ~11%, meaning 42.88*0.89, or ~38.16Mbps -- assuming you are the only person on that downstream, because there is other per-node chatter.
(EuroDOCSIS annex-A spec, btw, uses an 8-Mhz channel with 6.95Msps and correspondingly higher bandwidth.)
I was thinking the same thing, but this one may not be a flash in the pan.
Call/Recall has been around for 20+ years, and the co-founder / CTO is Dr. Peter Rentzepis, former head of Bell Labs and by all accounts a brilliant optical scientist.
Nichia is a huge chemical manufacturer that specializes in LED and laser technology and currently supplies over 2 million lasers for Blu-Ray per month.
Mod this AC up. He is dead on. Several other people mentioned the R&D and transformation costs (parts to product), but retailer margins are also a big deal. Apple plays this smart -- it uses products like this partially to give good margins to the retailers so that they will, in return, push products with very little margin like AppleTV.
It is all about what the consumer is willing to pay.
Though I certainly agree that HDMI is a good thing, the statement about it allowing better color depth is patently false. The analog signals in component can carry an effectively infinite amount of color variation, something the digital signal can not. Even considering the fact that the source is digital and display is digital, there are high enough resolution DACs/ADCs that an analog signal could convey a higher range of color than HDMI -- if sources and displays to exploit this actually existed.
That is a moot point, of course, since they don't exist.
The real quality advantage of HDMI, as I mentioned somewhere upthread, comes from the fact that the signal never enters the analog domain. Even though modern D/ACs are basically irrelevant to conversion loss at video bit depths, the rest of the signal after entering the analog domain sure isn't.
Potential for noise ingress due to sketchy DAC power, signal filtering, proper Y to Pb/Pr sync etc, etc. make doing quality analog circuitry a real PITA.
Achieving near-perfect quality on analog component circuitry takes a hell of a lot of work, extensive testing and tweaking with calibration equipment (e.g. Tek VM5000), and is damn near a black art. In contrast, a digital signal path clean enough to make sure all the 1's and 0's are right could be laid out on a board by a hung-over first year EE with with half a clue.
Input processing is certainly a possible culprit, but the source is equally likely. One of the best things about DVI/HDMI is that they are far easier to generate an accurate signal from.. you basically just have to get the timing right. With component you have to deal with noise filtering, varying levels, synchronization between Y and the half-rate PbPr channels, all in the analog domain where there are a lot more variables to make things go wrong.
A well designed source with extensive parametric analysis (Tek VM5000, etc) and tuning can output a stellar signal through component -- but a lot of equipment manufacturers just get it 'good enough' and ship it.
I certainly agree with you that, done right, it can be difficult to tell. Considering the signal source is digital and limited in resolution anyhow, modern DACs and ADCs are more than good enough that a component signal can be completely indistinguishable from a digital one. Unfortunately it is all those other wonderful PITA variables inherent in analog that can lead to wide variations.
Um, no. They are commonplace enough. HDTV sales will outpace SD sales for the first time this year. Morgan Stanley estimates that approximately 26% of households will have at least one set by the end of the year. That number rises to ~68% in 2010.
You can say that 26% this year and 33+% next year isn't wide spread enough, but I beg to differ. Those are also the households with the disposable income to afford not only the console, but the real expense of accessories and games for it.
Nintendo is making a mistake. Don't get me wrong, it doesn't mean the games won't be fun, but based on perception alone they are missing a major marketing 'checkmark'.
If the group responsible kept up to date by adding "certified" software to keep up with the "joe average" software needs, this could work.
As someone experienced who regularly uses Linux and Winderz (only with FFox and otherwise properly locked down), I wouldn't use it -- but I would likely switch the parents and other non-savvy friends and family to it quickly.
Are there not some developers who can say how (in)complete the development systems are? If people don't have dev kits, there won't be games, so who cares when the console launches.
The PS3, while potentially more powerful than the 360, seems almost certain to require substantially more effort to extract that power -- making a good dev environment even more critical.
Uh, Funny? He is dead-on, and there's nothing wrong with that. Dell does well because they make decent stuff cheaply.. They are the Honda of computer makers, and though they ain't perfect, they are pretty darn good and TONS of people like 'em.
Honda also spends tons of money to research reducing production costs, so they can put 'formerly-high-end-luxury-car-feature' into their cars at a reasonable cost. Ditto Dell.
I don't hate them, but don't buy 'em either. Start mixing AMD in, and we'll talk.
I know advanced anti-ship missles (and presumably torps) are programmed to pentrate the hull by a settable amount prior to detonation. It certainly seems like that would require some sophistication.
Neat idea, but doesn't it seem like it would be easy to program torpedoes to ignore? Ok, if it crushes them that's one thing, but false-detonation seems easily avoidable. Either way it seem like a fairly easy thing to engineer around, so I question how long a lifetime it would have.
p.s. to PETA, etc: Sure, take care in testing it, but if my sub is about to be hit by a torpedo, the whale can kiss my ass.
Think how easy it would be... unless it had force feedback, a cheap wheel with no electronics - just slide the controller into a dock in the middle and blammo.
I hope they have a communications capability (dock connector) on the control, so things like force feedback are also simple.
Amen! Doesn't even have to be the same car. Jumping on a crotchrocket is even worse.
That said, SHHH.. don't give the %#^@ media more ammo.. I have yet to see a story about a big wreck 'caused' by this, not the person, noooo. It seems inevitable, unfortunately.
Some Denon receivers can adjust audio delay (AVR3805+ at least), I suspect plenty of others do as well.
It is interesting how well you can tune the delay in to feel 'right', but a lot of mpeg programs have crappy sync anyway.. I finally settled on a bit more than one frame delay (~40ms) as the right average -- and forced myself to quit paying such close attention.
Yeah, RIGHT. Anyone who has been involved in or around a situation like this knows the real way it goes:
" Ok, you can 'resign' now, keep your preferred stock, take your 3-6 months of severance/vacation pay, etc.. or we can make up some 'cause' to get rid of you " -- and you get jack shit unless you spend way more on attorneys than it would get you.
Slashdot Mirror
5252 comes from:
radians per second is (2*pi) / 60 = 0.10472 radians per second
The standard for horsepower is 550 ft. lbs / second
550 / 0.10472 = 5252
Higher RPM with the same torque curve will give you higher acceleration because the engine will pull with said torque for *longer*.
Yup. Broadcast (ATSC) uses Mpeg-2. However both DirecTV and Dish use h.264 for many of their HD channels. That's why old DirecTiVo HDs miss out on a lot of programming.
Too late: (and these are just a few examples)
http://www.freepatentsonline.com/5598143.html
http://www.freepatentsonline.com/6879254.html
http://www.patentstorm.us/patents/6404349/description.html
Interestingly, prior art to all 3 of those patents exists as many TVs and VCRs from Phillips/Magnavox had this feature built in as far back as 1993: http://www.magnavox.com/index.cfm?event=about
Sony also included it in a few TV models back in 2001.
I suspect it has not seen more widespread adoption because of battery life and penny-pinching in a cutthroat market.
Unless you have a Docsis 3.0 modem, it is physically impossible for you to get 40-60Mbps. The calculation code is wrong somewhere.
Docsis 2.0 (North American annex-B) uses a single 6Mhz channel, with the highest available encoding being QAM-256 (8 bits per sample). The standardized sample rate is 5.36Msps, so *=8 gives the true peak throughput of 42.88 Mbps, of which ~38Mbps is usable data.
I see the same thing when starting a batch download with DownThemAll, which reports 6+ MBps regularly when starting, but it is simply inaccurate.
(That said, I am very happy with Comcast's "Blast" tier, which regularly gives me 20-30Mbps sustained rates as long as multiple connections are in use, like DownThemAll does. For whatever reason a single stream peaks out around 16Mbps sustained.)
-------
Gory details on 42.88 vs ~38Mbps usable, if anyone cares:
The standard specifies Reed-Solomon FEC (Forward Error Correction), which mean 6 samples per 128 are unavailable (-4.7%). The MPEG-2 transport stream structure further has a 4 byte header per 188 bytes (-2.1%), Ethernet eats another 18/1500 (-1.2%), IP headers and Docsis map traffic account for about -3% more overhead. It all adds up to ~11%, meaning 42.88*0.89, or ~38.16Mbps -- assuming you are the only person on that downstream, because there is other per-node chatter. (EuroDOCSIS annex-A spec, btw, uses an 8-Mhz channel with 6.95Msps and correspondingly higher bandwidth.)
All good points. Just wanted to point out that it is 'only' $240 million a month, not 24.
I was thinking the same thing, but this one may not be a flash in the pan.
Call/Recall has been around for 20+ years, and the co-founder / CTO is Dr. Peter Rentzepis, former head of Bell Labs and by all accounts a brilliant optical scientist.
Nichia is a huge chemical manufacturer that specializes in LED and laser technology and currently supplies over 2 million lasers for Blu-Ray per month.
Mod this AC up. He is dead on. Several other people mentioned the R&D and transformation costs (parts to product), but retailer margins are also a big deal. Apple plays this smart -- it uses products like this partially to give good margins to the retailers so that they will, in return, push products with very little margin like AppleTV.
It is all about what the consumer is willing to pay.
Moof!
Though I certainly agree that HDMI is a good thing, the statement about it allowing better color depth is patently false.
The analog signals in component can carry an effectively infinite amount of color variation, something the digital signal can not.
Even considering the fact that the source is digital and display is digital, there are high enough resolution DACs/ADCs that an analog signal could convey a higher range of color than HDMI -- if sources and displays to exploit this actually existed.
That is a moot point, of course, since they don't exist.
The real quality advantage of HDMI, as I mentioned somewhere upthread, comes from the fact that the signal never enters the analog domain. Even though modern D/ACs are basically irrelevant to conversion loss at video bit depths, the rest of the signal after entering the analog domain sure isn't.
Potential for noise ingress due to sketchy DAC power, signal filtering, proper Y to Pb/Pr sync etc, etc. make doing quality analog circuitry a real PITA.
Achieving near-perfect quality on analog component circuitry takes a hell of a lot of work, extensive testing and tweaking with calibration equipment (e.g. Tek VM5000), and is damn near a black art. In contrast, a digital signal path clean enough to make sure all the 1's and 0's are right could be laid out on a board by a hung-over first year EE with with half a clue.
Input processing is certainly a possible culprit, but the source is equally likely. One of the best things about DVI/HDMI is that they are far easier to generate an accurate signal from.. you basically just have to get the timing right. With component you have to deal with noise filtering, varying levels, synchronization between Y and the half-rate PbPr channels, all in the analog domain where there are a lot more variables to make things go wrong.
A well designed source with extensive parametric analysis (Tek VM5000, etc) and tuning can output a stellar signal through component -- but a lot of equipment manufacturers just get it 'good enough' and ship it.
I certainly agree with you that, done right, it can be difficult to tell.
Considering the signal source is digital and limited in resolution anyhow, modern DACs and ADCs are more than good enough that a component signal can be completely indistinguishable from a digital one. Unfortunately it is all those other wonderful PITA variables inherent in analog that can lead to wide variations.
Um, no. They are commonplace enough. HDTV sales will outpace SD sales for the first time this year. Morgan Stanley estimates that approximately 26% of households will have at least one set by the end of the year. That number rises to ~68% in 2010.
You can say that 26% this year and 33+% next year isn't wide spread enough, but I beg to differ. Those are also the households with the disposable income to afford not only the console, but the real expense of accessories and games for it.
Nintendo is making a mistake. Don't get me wrong, it doesn't mean the games won't be fun, but based on perception alone they are missing a major marketing 'checkmark'.
If the group responsible kept up to date by adding "certified" software to keep up with the "joe average" software needs, this could work.
As someone experienced who regularly uses Linux and Winderz (only with FFox and otherwise properly locked down), I wouldn't use it -- but I would likely switch the parents and other non-savvy friends and family to it quickly.
Interesting idea.
Are there not some developers who can say how (in)complete the development systems are? If people don't have dev kits, there won't be games, so who cares when the console launches.
The PS3, while potentially more powerful than the 360, seems almost certain to require substantially more effort to extract that power -- making a good dev environment even more critical.
But what is the DRM, and how will it rootkit me?
Uh, Funny? He is dead-on, and there's nothing wrong with that. Dell does well because they make decent stuff cheaply.. They are the Honda of computer makers, and though they ain't perfect, they are pretty darn good and TONS of people like 'em.
Honda also spends tons of money to research reducing production costs, so they can put 'formerly-high-end-luxury-car-feature' into their cars at a reasonable cost. Ditto Dell.
I don't hate them, but don't buy 'em either. Start mixing AMD in, and we'll talk.
Surprised this has not been talked about more.
$300 for a viPod
$200 for a 9" screen it docks with in the car
$ 6 to put a few new kiddies shows on it just as you are heading out the door
Hours of bliss while driving to the parents for the holidays: Priceless
Most of the time you are going to use it just like a non video ipod, but having the feature added on does not suck.
Fair enough as far as the brick wall goes.
I know advanced anti-ship missles (and presumably torps) are programmed to pentrate the hull by a settable amount prior to detonation. It certainly seems like that would require some sophistication.
Personally, I don't block them until they a) blink b) slow down the page.
Animated crap and poorly designed pages that make the ad-links (ohh, and that damned javascript highlight words BS) get insta-adblock.
Sure, that policy has led to my adblock filter catching damn near all graphical ads -- that ain't my fault.
I still see Google's.
Yeah, and by God, it better be connected in the right direction too!
Neat idea, but doesn't it seem like it would be easy to program torpedoes to ignore? Ok, if it crushes them that's one thing, but false-detonation seems easily avoidable. Either way it seem like a fairly easy thing to engineer around, so I question how long a lifetime it would have.
p.s. to PETA, etc: Sure, take care in testing it, but if my sub is about to be hit by a torpedo, the whale can kiss my ass.
Think how easy it would be... unless it had force feedback, a cheap wheel with no electronics - just slide the controller into a dock in the middle and blammo.
I hope they have a communications capability (dock connector) on the control, so things like force feedback are also simple.
Planning for the near-future? Surfing, email, etc. are pretty cool on a nice big HDTV.
Amen! Doesn't even have to be the same car. Jumping on a crotchrocket is even worse.
That said, SHHH.. don't give the %#^@ media more ammo.. I have yet to see a story about a big wreck 'caused' by this, not the person, noooo. It seems inevitable, unfortunately.
Some Denon receivers can adjust audio delay (AVR3805+ at least), I suspect plenty of others do as well.
It is interesting how well you can tune the delay in to feel 'right', but a lot of mpeg programs have crappy sync anyway.. I finally settled on a bit more than one frame delay (~40ms) as the right average -- and forced myself to quit paying such close attention.
Yeah, RIGHT.
Anyone who has been involved in or around a situation like this knows the real way it goes:
" Ok, you can 'resign' now, keep your preferred stock, take your 3-6 months of severance/vacation pay, etc.. or we can make up some 'cause' to get rid of you " -- and you get jack shit unless you spend way more on attorneys than it would get you.