Yup. The USB standard is that any powered port provided a single "unit load", 100mA at +5Vdc (150mA for USB 3.0). A device can request power for 5 unit loads, for a total of 500mA (6 unit loads/900mA in USB 3.0). There is also a fairly recent "Battery Charging Specification", which allows up to 1.8A from a simple charger (though curiously, standard USB 2.0 "A" connectors are rated for 1.5A power). Such a device isn't smart, but puts a low resistance between the D+ and D- pins, to allow attached devices to understand what they're attached to.
4-Band GSM manages voice and very low data (at high prices). That's the only effective standard supported by any US phones. 3G isn't handled in common even within the USA. Ok, technically, Verizon and Sprint use the same frequencies and technology, but they lock their networks based on the ID of the phone.. no SIM cards, so they know precisely which phones are sold into their networks and which come from elsewhere. The laws that govern phone unlocking didn't address this issue.
And in fact. some phones are built using just such a module. Sure, they're proprietary, and the phone itself needs specific software to support it.
This would work.. but it's not the whole story. You could add the module to your phone, sign up with a provider, and then hit the net via WiFi. You would need at least OS-specific, if not device-specific, new code to support the add-in module. And of course, for GSM phones, you need a place to put the SIM.
The harder actual problem is antennas. That's the frequency issue. Everyone doing Wifi has it standard -- 2.4GHz band, 5.6GHz if you want to support the "A" frequencies as well (which 802.11n can also use). But for cellphones... it's an issue.
For 2G, it's easy... all USA cells use 1900Mhz for 2G/voice. AT&T and Verizon have most of the original 850MHz AMPS band as well. In Europe, it's 900MHz and 1800MHz... thus the "quad band" for universal support of voice calls. Ok, sure, Nextel used a non-AMPS segment of the 850MHz band.. but the whole IDEN thing is phasing out.
Once you get to 3G, it's a mess. Well, a small mess, but a mess. CDMA2000 phones use the same channels they used for voice, so Verizon and Sprint work just as they did in 2G. AT&T needs wider channels for HSPA, but thanks to being in there early, they had both 850MHz and 1900MHz slots... 3G phones want to use both. T-Mo only has 1900MHz spectrum for 2G, to they had to wait to roll out 3G until a spectrum auction got them a chunk of 1700MHz and 2100MHz.. their 3G connections use both at the same time, like AT&T.
Going to 4G, it even more complicated. T-Mobile has decided to just call their 3G HSPA+ network "4G", and be done with it. This is why they can say "largest 4G network", despite the fact that AT&T has more HSPA+ coverage. Sprint got in with Clear and Comcast (and Google and Intel and a few others) to create WiMax networks at 2500MHz. Verizon and AT&T are both supporting LTE (the preferred 4G technology of the 3GPP group, the guys driving the future of the GSM-related technologies), and they both bought chunks of the 700MHz spectrum. Of course, neither WiMax nor LTE are 4G yet.. but future versions will meet the specs.
As for power plugs. just about everyone is using micro-USB. Enough user so that it's absolutely now the accepted standard. Except Apple, but you wouldn't expect Apple to worry about common standards or anything.
This is a new thing, and not aimed squarely at techies like Android was, so it may take some time. I guess the lack of a stellar turn out was a fail in light of the $100,000,000 or whatever they're spending on Windows Phone Ads.
In the Bad Way, a Windows Phone is like anything else in tech -- "civilian" buyers are apprehensive, just because it's a new thing they haven't seen before. Techies are apprehensive because it's really Windows Phone OS 1.0, and will need a few service packs before it's at all acceptable. Or at least, that's what you have to expect. And perhaps, we recall the crazy impact Microsoft made with the Zune, or how Windows Mobile has been hemorrhaging market share, or what a boondoggle the Kin was.
And then, outside of certainly reality distortion fields, phones are worse than other tech items. Most people who want to buy a new mobile phone have an old one... and won't buy a new one until they're off contract. Windows Phone has the additional liability of being sold only directly against the iPhone 4... only AT&T has them now. That'll keep people who don't want AT&T away as well. And of course, there's the two models... the unit with speakers. Really? Did they have to launch with a silly gimmick. People in the know, at least, understand that in a few months, there will be more models (unless this completely tanks, ala Kin) and maybe even other carriers signed up.
Apple is a special case, too. The bar they set for "New iPhone Day" will never be met by anyone else. Apple's cultivated this for years. They only do one new iPhone per year, they always release it in June. Savvy people who want the iPhone (assuming that's not an oxymoron) stop buying in late winter or early spring, and wait for the new model. Users who want to upgrade do too, or they're already on the update-in-June schedule, with their subsidized cell contract. There's no hope or concern of a better model coming out next month (next week, tommorrow) as there is in the Android world. So Apple creates an event that can never be matched in one day or one weekend by anyone else.
That's not a terrible thing, either. It means that, for other phones, there's no boom in June, but no dry season either. You might get spots of that... I'll probably be waiting to see about what Motorola does or doesn't do on a Droid 3 before I trade up from my Droid next year, but there are plenty of other models that are completely acceptable already.
Guess I buy a different class of printer... I have several, some years old, and never had a problem with jets clogging. Incidently, only Epson sells printers with fixed jet nozzles. When you buy an HP or Lexmark, you get new jets with every cartridge. Canon splits the difference... the ink tanks are sold individually, but you can replace the nozzle assembly after it starts to fail.
Ink matters, too, if you're printing things that need to last. Cheap ink can fade in less than a year or two; high quality inks can last 100+ years. Obviously, there's no sense in paying extra for 100 year ink for a printed memo (do people still sent those) that's in use for a day, but for many uses, ink really does matter.
iPhone names don't have much to do with performance. Apparently, "3G" and "3GS" only meant "3G download" anyway, since iPhones were stuck at 384kb/s uploads until the iPhone 4... which isn't a "4G" at all, just a "4". At least Apple's not false advertising that part of it. They have decided so, in fact, get entirely out of the "G" race. They're all about more pixels these days...
No one's sharing spectrum. Sprint has gobs of it, like 90MHz in many cities (well, it's actually Sprint, Clear, Comcast, and even a little Intel and Google in there as well), but it's unfortunately at 2500MHz. AT&T and Verizon are both doing LTE at 700MHz, but on different chunks (Verizon's chunk is about twice as fat as AT&T's).
The big advantage of LTE is that they changed the upstream modulation from OFDM to SC-FDMA (Single Carrier Frequency Division Multiple Access). This offers a variety of advantages, but the big one is a reduction in PAPR (Peak to Average Power Ratio). Basically, traditional OFDM, COFDM, and similar modulation schemes have a very high crest factor. This means that the peaks are sometimes 10dB or more higher than the average power. That's bad, particularly for battery powered gear, because I have to design an amplifier that can handle the peaks. Lower that PAPR, and you get a device with a longer battery life.
Sprint users have probably noticed that WiMax tends to suck down battery power. LTE devices will not have this problem.. in fact, in heavy use, they'll eventually use less battery on LTE than on their 3G mode (it has yet to be seen if the first generation LTE chips themselves aren't power hogs, but this will work itself out in a few years).
It's not fourth generation... T-Mobile's HSPA+ is a tweak to the 3G HSPA standard. At least LTE and WiMax are actual next generation technologies, even if they don't yet meet the ITU's spec for "4G". You can also tell they're next generation spec because they're continuing to evolve. HSPA+ is pretty much the end of the line for the 3GPP's set of 3G technologies. Sprint could boost things again, if they have the bandwidth, by bonding two cells (this is also supported... but you need 10MHz per channel, rather than 5MHz), but that's as far as it'll ever go. LTE and WiMax will be real 4G within a few years, by virtue of some of those "in-generation tweaks".
100Mb/s peak for highly mobile links, 1Gb/s peak for stationary links. All packet-switched IP data-only, some other things. The qualifying specifications, so far, are LTE Advanced (new version of LTE) and 802.16m (new version of WiMax).
Well, some of it's that. Sprint's flavor of WiMax technically allows 40Mb/s per channel, which is right near the HSPA+ level of 42Mb/s per channel. But no one's allowing those levels, even if there were devices that ran that fast on the client end. And actually, yeah, being different is at least part of the qualification the ITU puts on their "generation" specs. If you're on the same frequencies, and using the same kind of radio protocols, and backwards compatible, you're considered an evolution of the existing generation. Thus, HSPA+ is very soundly part of 3G... even the 3GPP (the folks behind the spec) don't dispute this even slightly.
Sprint will get 4G speeds in their next update, 802.16m. But even then, you may not actually get that. For one, every network owner can limit the actual speeds available, peak, to a device. For example, T-Mobile is allowing 21Mb/s (half of what's possible) on their HSPA+ devices, AT&T limits it to 7.2Mb/s.
The other problem for Sprint, and also T-Mobile, is their high frequencies. Sprint has a ton of bandwidth, but it's at 2500MHz... it doesn't go through foliage or buildings very well. In fact, we're testing WiMax at my office, and had to put the modem in a window to keep a good link.. in Philly, right in town, in one of Sprint's flagship cities (because Comcast is a partner in the Sprint/Clear/Comcast rollout of WiMax, and they're based in Philly). T-Mo's not much betetr, at 1700MHz and 2100MHz. AT&T and Verizon will be able to deliver much more reliable high-speed at 700MHz. Of course, they'll probably cap it low, particularly Verizon... even 6Mb/s is about double what you can get on EvDO.
The ITU didn't "come and say no". They put out their next generation requirements in 2005-2006, long before any of the chips used in these networks existed (the spec is actually called IMT-Advanced, just as the "3G" spec was actually called "IMT-2000"). Everyone knows this was "4G", since the ITU had set every prior standard. No, they didn't (and probably couldn't) trademark "4G"... if anyone could have, Apple would have, year ago. Both LTE and WiMax were known to not meet the ITU 4G specs long before any of these networks were rolled out. And in fact, last year everyone knew that future versions of these, LTE Advanced and WiMax 802.16m, are actual 4G standards, though not yet delivered.
And particularly, the different generations are, well, "generational"... that means new tech, leaving the old behind. Even in this simple case, WiMax and LTE qualify (new protocols, all packet-switched IP, new frequencies, etc), but HSPA+ certainly doesn't.
Given that AT&T has more HSPA+ coverage than Sprint, they're the natural company to actually call them on this... and I guess they have just started doing this. But they're also kind of stuck -- they have way too many customers to enable T-Mobile like speeds on their network (they could, but that would really clog things... they throttle HSPA+ connections to 7.2Mb/s peak). And if they decided to call HSPA+ 4G, what would they do next summer when they roll out LTE?
T-Mobile was similarly stuck. They had just finished rolling out HSPA... they were really behind on 3G because they had to wait for 1700MHz/2100MHz to be auctioned, which they got in 2006... Verizon had already built their 3G network before T-Mo even started. So now, like AT&T, they've decided to roll out HSPA+ strategically... AT&T finished their HSPA+ upgrades last summer, and now they're working on LTE. T-Mobile doesn't own any spectrum for LTE (AT&T got the smaller 700MHz block, Verizon the larger, in that auction back in 2008). I had wondered what, if any, strategy they had for 4G... I never would have guessed "just lie about it".
1G-4G aren't standards per se, they're performance metrics. Your specific standard either meets the appropriate ITU standard, or it doesn't. Everyone knows what the "Gs" mean (or at least used to), but the ITU doesn't formally use "G" in their specification names. So there's not much likelihood anyone can call Sprint or T-Mobile on this... other than informally, like here, in articles, etc.
It'll be interesting to see when someone actually does roll out real 4G. Will they take on the others and promote that they have the only "real" 4G, or will they give it a different name?
EvDO = Evolution, Data-Only. There's no actual 3G voice mode, in the sense of UMTS. They should have done what, hopefully, the LTE networks will do, and support mixed "voice" and "data" via VoIP. The only problem is that put an additional burden on the handsets, so they didn't, for 3G. But any old smartphone can run VoIP these days, as a "simple matter of software". So there's no real excuse any longer.
They didn't. The ITU started defining "beyond 3G" in 2002, and more formally, as IMT-Advanced, back in 2005. The only technologies so far proposed to meet this standard have been WiMax and LTE, neither actually making the 4G standards in their current form. Regardless of that, the "4G" name has been applied to LTE systems going back some years now (they're only being rolled out in the USA this year, but they've been available in other locations).
I think the problem is that the ITU's use of 2G, 3G, and 4G has been fairly informal. After all, the real ITU name for "3G" is IMT-2000... I don't think anyone was trying to sell 2G technologies as meeting 3G specs back then, but they probably could have. Same deal now... only the marketroids are intentionally distorting the truth here. And T-Mobile most of all.. HSPA+ is very much a 3G technology, regardless of the performance they're allowing on it. At least the other guys are backing technologies that, in another development cycle or two, will be actual 4G standards. HSPA is finished; it's a 3G tech and will remain so.
This isn't Sprint's fourth network, anyway. They have offered HSPA access as a "3G" service, same as AT&T, for quite some time. This is part of the 3GPP (3rd Generation Partnership Project) specifications, which define the set of technologies generally used by GSM-aligned companies. This is actually part of the 3GPP's WCDMA specification... officially "3G". Version 7 of the WCDMA specification allows for a higher speed option, HSPA+, still very much a 3G technology. AT&T has had HSPA+ available for quite some time as well.
The only real difference in Sprint's plan are the bitrate caps. Each technology has a maximum per client download rate, but they never allow any client access to the full rate, simply because that stresses both the number of clients on a cell, the backhaul, and overall client expectations (eg, if I get 10Mb/s on rare occasions, I'm much less happy when my speed drop to 1Mb/s than if I normally got 3Mb/s). AT&T limits their client devices to 3.6Mb/s on an HSPA node and 7.2Mb/s on an HSPA+ node. T-Mo recently upped their caps, so it's technically possible to get something like 20Mb/s down on an HSPA+ node. Actual limits are 14Mb/s for HSPA and 42Mb/s for HSPA+. Higher bitrates are already fairly common on GSM-based networks in other parts of the world. This is just an upgrade of existing technology, not a new network... except maybe to T-Mobile, who didn't have any 3G service before the 1700MHz/2100MHz bands were opened a few years ago.
And the 3GPP themselves (not just the ITU) consider their 4G technology to be LTE. The original LTE rollout by Verizon this year and AT&T next year isn't there yet. LTE Advanced, to be finalized next year, is the first one that meets the ITU's actual specifications (at least in part) for 4G.
And yeah, these "G" standards have always been official, globally recognized standards defined by the ITU, not just things that marketing people invent. It was only a matter of time before Sprint's "4G" marketing of not-really-4G WiMax got everyone telling tall tales. Hopefully they'll all have to eat a little crow on this, but more than likely, they'll get away with it.
HD bitrates jump between 2600kb/s and 3800kb/s, SD bitrates jump between 300kb/s, 500kb/s, 1000kb/s, and 1500kb/s... at least based on the last direct information I have from Netflix. They're higher than YouTube for the same class video, and rather than limiting things to 10 minutes or less, the typical use is 2 hour films. Plus, Netflix via mail was successful enough to pretty much single handedly put Blockbuster, Hollywood Video, and other rentals out of business (I guess Red Box helped, but only fairly recently). So, they clearly have the customer based and the bandwidth demands to eat up that much bandwidth.
And they're certainly want more, if the typical pipe-to-the-home could take it. They originally played with higher bitrates, but didn't have enough success in delivering a viewable video. If the average increases enough, you can bet they'll be there with higher bitrates to eat that up, too. They still have to compete with other forms of rental/PPV, and increasing quality will make them more competitive.
Something's wrong there... Netflix claim their "HD" is encoded at 2.600Mb/s and 3.800Mb/s (they switch streams on the fly, based on available bandwidth). Maybe they've added some lower bitrate streams since then (http://blog.netflix.com/2008/11/encoding-for-streaming.html). Did you verify this is actually HD at that point? You're awfully close to their peak SD rate, 1.5Mb/s (which other sources claim is 1.6Mb/s, anyway). Some Netflix players let you fix the bitrate, others only automatically adapt.
New material encoded for SD is using VC-1 (eg, WMV9, SMPTE 421M), just as the HD stuff. While the HD bitrates vary between 2.6Mb/s and 3.8Mb/s, the SD stuff is apparently encoded at a fixed rate of 320kb/s, 500kb/s, 1000kb/s, and 1500kb/s... I gather all bitrates are available, and your connection determines which you get. They deliver 640x480 video, like most online streaming, so it's slightly lower rez than DVD just to being. A well encoded AVC stream will have twice the encoding efficiency of MPEG-2; VC-1 is a bit less, but even assuming it was as good as AVC, you're getting video that's more than twice as compressed, visually speaking, as a DVD. This ought to be comparable to some cable/satellite SD feeds (I have only recently seen HD Netflix, and the older WMV3 based SD, so I can't personally comment on today's SD), but it's not DVD class.
That's why I have a 71" TV.. which always displays in 1080p, regardless of the input.
For your 55" TV, the average person will not see any difference between 720p and 1080p at 13ft. They won't see the full 1080p resolution until they move to 7.2ft from the screen, but closer than 13ft and you're seeing an improvement over 720p. The SMPTE recommended viewing distance for a 55" screen is 7.5ft, the THX recommended viewing distance is 6.1ft. So if you're concerned about a good home theater experience, you already have your seats positioned for very good 1080p viewing.
My main problem with Netflix isn't that it's half the resolution of proper Blu-Ray (720/60p is good for high motion -- I shoot sports in 720/60p or 1080/60p, but for most content, you want 1080i or 1080p), but that it's at best about 1/10th the bitrate of Blu-Ray. So it's not just the resolution, but also the fact that they apply a global low-pass filter prior to encoding to minimize DCT artifacts, and then compress a bitrate 1/10th that of your average Blu-Ray, 1/5th that of ATSC broadcast. Now sure, if you watch nothing but such overcompressed view, your brain will adapt and you'll slowly start ignoring the compression artifacts -- same reason people were relatively happy with analog or first-generation TiVo, until better things came along.
The US also ranks 23rd in "Most Livable Places", 33rd in education, 37th in healthcare, 20th for child well-being... and 78th in Homicide (but the sale is from lowest to highest). And we're in a 26-way tie for the 21rst slot in literacy. Guess you can't win 'em all...
Their HD content SUCKS compares to OTA or Cable. If you don't have a 1080i/p TV, you'll notice it less, but even based on compression artifacts, it's pretty annoying. Ok, sure, I'm a video guy, I know what to look for, and I would need months of looking at Netflix video, rather than full 1080p from camcorders, before my brain started ignoring the flaws (which it will... eventually).
Slashdot Mirror
The plug is not a charger, just a power supply. The charging method is a private agreement between the device and its battery.
Yup. The USB standard is that any powered port provided a single "unit load", 100mA at +5Vdc (150mA for USB 3.0). A device can request power for 5 unit loads, for a total of 500mA (6 unit loads/900mA in USB 3.0). There is also a fairly recent "Battery Charging Specification", which allows up to 1.8A from a simple charger (though curiously, standard USB 2.0 "A" connectors are rated for 1.5A power). Such a device isn't smart, but puts a low resistance between the D+ and D- pins, to allow attached devices to understand what they're attached to.
4-Band GSM manages voice and very low data (at high prices). That's the only effective standard supported by any US phones. 3G isn't handled in common even within the USA. Ok, technically, Verizon and Sprint use the same frequencies and technology, but they lock their networks based on the ID of the phone.. no SIM cards, so they know precisely which phones are sold into their networks and which come from elsewhere. The laws that govern phone unlocking didn't address this issue.
And in fact. some phones are built using just such a module. Sure, they're proprietary, and the phone itself needs specific software to support it.
This would work.. but it's not the whole story. You could add the module to your phone, sign up with a provider, and then hit the net via WiFi. You would need at least OS-specific, if not device-specific, new code to support the add-in module. And of course, for GSM phones, you need a place to put the SIM.
The harder actual problem is antennas. That's the frequency issue. Everyone doing Wifi has it standard -- 2.4GHz band, 5.6GHz if you want to support the "A" frequencies as well (which 802.11n can also use). But for cellphones... it's an issue.
For 2G, it's easy... all USA cells use 1900Mhz for 2G/voice. AT&T and Verizon have most of the original 850MHz AMPS band as well. In Europe, it's 900MHz and 1800MHz... thus the "quad band" for universal support of voice calls. Ok, sure, Nextel used a non-AMPS segment of the 850MHz band.. but the whole IDEN thing is phasing out.
Once you get to 3G, it's a mess. Well, a small mess, but a mess. CDMA2000 phones use the same channels they used for voice, so Verizon and Sprint work just as they did in 2G. AT&T needs wider channels for HSPA, but thanks to being in there early, they had both 850MHz and 1900MHz slots... 3G phones want to use both. T-Mo only has 1900MHz spectrum for 2G, to they had to wait to roll out 3G until a spectrum auction got them a chunk of 1700MHz and 2100MHz.. their 3G connections use both at the same time, like AT&T.
Going to 4G, it even more complicated. T-Mobile has decided to just call their 3G HSPA+ network "4G", and be done with it. This is why they can say "largest 4G network", despite the fact that AT&T has more HSPA+ coverage. Sprint got in with Clear and Comcast (and Google and Intel and a few others) to create WiMax networks at 2500MHz. Verizon and AT&T are both supporting LTE (the preferred 4G technology of the 3GPP group, the guys driving the future of the GSM-related technologies), and they both bought chunks of the 700MHz spectrum. Of course, neither WiMax nor LTE are 4G yet.. but future versions will meet the specs.
As for power plugs. just about everyone is using micro-USB. Enough user so that it's absolutely now the accepted standard. Except Apple, but you wouldn't expect Apple to worry about common standards or anything.
This is a new thing, and not aimed squarely at techies like Android was, so it may take some time. I guess the lack of a stellar turn out was a fail in light of the $100,000,000 or whatever they're spending on Windows Phone Ads.
In the Bad Way, a Windows Phone is like anything else in tech -- "civilian" buyers are apprehensive, just because it's a new thing they haven't seen before. Techies are apprehensive because it's really Windows Phone OS 1.0, and will need a few service packs before it's at all acceptable. Or at least, that's what you have to expect. And perhaps, we recall the crazy impact Microsoft made with the Zune, or how Windows Mobile has been hemorrhaging market share, or what a boondoggle the Kin was.
And then, outside of certainly reality distortion fields, phones are worse than other tech items. Most people who want to buy a new mobile phone have an old one... and won't buy a new one until they're off contract. Windows Phone has the additional liability of being sold only directly against the iPhone 4... only AT&T has them now. That'll keep people who don't want AT&T away as well. And of course, there's the two models... the unit with speakers. Really? Did they have to launch with a silly gimmick. People in the know, at least, understand that in a few months, there will be more models (unless this completely tanks, ala Kin) and maybe even other carriers signed up.
Apple is a special case, too. The bar they set for "New iPhone Day" will never be met by anyone else. Apple's cultivated this for years. They only do one new iPhone per year, they always release it in June. Savvy people who want the iPhone (assuming that's not an oxymoron) stop buying in late winter or early spring, and wait for the new model. Users who want to upgrade do too, or they're already on the update-in-June schedule, with their subsidized cell contract. There's no hope or concern of a better model coming out next month (next week, tommorrow) as there is in the Android world. So Apple creates an event that can never be matched in one day or one weekend by anyone else.
That's not a terrible thing, either. It means that, for other phones, there's no boom in June, but no dry season either. You might get spots of that... I'll probably be waiting to see about what Motorola does or doesn't do on a Droid 3 before I trade up from my Droid next year, but there are plenty of other models that are completely acceptable already.
Guess I buy a different class of printer... I have several, some years old, and never had a problem with jets clogging. Incidently, only Epson sells printers with fixed jet nozzles. When you buy an HP or Lexmark, you get new jets with every cartridge. Canon splits the difference... the ink tanks are sold individually, but you can replace the nozzle assembly after it starts to fail.
Ink matters, too, if you're printing things that need to last. Cheap ink can fade in less than a year or two; high quality inks can last 100+ years. Obviously, there's no sense in paying extra for 100 year ink for a printed memo (do people still sent those) that's in use for a day, but for many uses, ink really does matter.
iPhone names don't have much to do with performance. Apparently, "3G" and "3GS" only meant "3G download" anyway, since iPhones were stuck at 384kb/s uploads until the iPhone 4... which isn't a "4G" at all, just a "4". At least Apple's not false advertising that part of it. They have decided so, in fact, get entirely out of the "G" race. They're all about more pixels these days...
No one's sharing spectrum. Sprint has gobs of it, like 90MHz in many cities (well, it's actually Sprint, Clear, Comcast, and even a little Intel and Google in there as well), but it's unfortunately at 2500MHz. AT&T and Verizon are both doing LTE at 700MHz, but on different chunks (Verizon's chunk is about twice as fat as AT&T's).
The big advantage of LTE is that they changed the upstream modulation from OFDM to SC-FDMA (Single Carrier Frequency Division Multiple Access). This offers a variety of advantages, but the big one is a reduction in PAPR (Peak to Average Power Ratio). Basically, traditional OFDM, COFDM, and similar modulation schemes have a very high crest factor. This means that the peaks are sometimes 10dB or more higher than the average power. That's bad, particularly for battery powered gear, because I have to design an amplifier that can handle the peaks. Lower that PAPR, and you get a device with a longer battery life.
Sprint users have probably noticed that WiMax tends to suck down battery power. LTE devices will not have this problem.. in fact, in heavy use, they'll eventually use less battery on LTE than on their 3G mode (it has yet to be seen if the first generation LTE chips themselves aren't power hogs, but this will work itself out in a few years).
It's not fourth generation... T-Mobile's HSPA+ is a tweak to the 3G HSPA standard. At least LTE and WiMax are actual next generation technologies, even if they don't yet meet the ITU's spec for "4G". You can also tell they're next generation spec because they're continuing to evolve. HSPA+ is pretty much the end of the line for the 3GPP's set of 3G technologies. Sprint could boost things again, if they have the bandwidth, by bonding two cells (this is also supported... but you need 10MHz per channel, rather than 5MHz), but that's as far as it'll ever go. LTE and WiMax will be real 4G within a few years, by virtue of some of those "in-generation tweaks".
Depends on the frequencies. Trees are really effective at eating higher frequency radio signals, such as Sprint's 2500MHz WiMax.
100Mb/s peak for highly mobile links, 1Gb/s peak for stationary links. All packet-switched IP data-only, some other things. The qualifying specifications, so far, are LTE Advanced (new version of LTE) and 802.16m (new version of WiMax).
Get an iPhone. Problem solved.
Well, some of it's that. Sprint's flavor of WiMax technically allows 40Mb/s per channel, which is right near the HSPA+ level of 42Mb/s per channel. But no one's allowing those levels, even if there were devices that ran that fast on the client end. And actually, yeah, being different is at least part of the qualification the ITU puts on their "generation" specs. If you're on the same frequencies, and using the same kind of radio protocols, and backwards compatible, you're considered an evolution of the existing generation. Thus, HSPA+ is very soundly part of 3G... even the 3GPP (the folks behind the spec) don't dispute this even slightly.
Sprint will get 4G speeds in their next update, 802.16m. But even then, you may not actually get that. For one, every network owner can limit the actual speeds available, peak, to a device. For example, T-Mobile is allowing 21Mb/s (half of what's possible) on their HSPA+ devices, AT&T limits it to 7.2Mb/s.
The other problem for Sprint, and also T-Mobile, is their high frequencies. Sprint has a ton of bandwidth, but it's at 2500MHz... it doesn't go through foliage or buildings very well. In fact, we're testing WiMax at my office, and had to put the modem in a window to keep a good link.. in Philly, right in town, in one of Sprint's flagship cities (because Comcast is a partner in the Sprint/Clear/Comcast rollout of WiMax, and they're based in Philly). T-Mo's not much betetr, at 1700MHz and 2100MHz. AT&T and Verizon will be able to deliver much more reliable high-speed at 700MHz. Of course, they'll probably cap it low, particularly Verizon... even 6Mb/s is about double what you can get on EvDO.
The ITU didn't "come and say no". They put out their next generation requirements in 2005-2006, long before any of the chips used in these networks existed (the spec is actually called IMT-Advanced, just as the "3G" spec was actually called "IMT-2000"). Everyone knows this was "4G", since the ITU had set every prior standard. No, they didn't (and probably couldn't) trademark "4G"... if anyone could have, Apple would have, year ago. Both LTE and WiMax were known to not meet the ITU 4G specs long before any of these networks were rolled out. And in fact, last year everyone knew that future versions of these, LTE Advanced and WiMax 802.16m, are actual 4G standards, though not yet delivered.
And particularly, the different generations are, well, "generational"... that means new tech, leaving the old behind. Even in this simple case, WiMax and LTE qualify (new protocols, all packet-switched IP, new frequencies, etc), but HSPA+ certainly doesn't.
Given that AT&T has more HSPA+ coverage than Sprint, they're the natural company to actually call them on this... and I guess they have just started doing this. But they're also kind of stuck -- they have way too many customers to enable T-Mobile like speeds on their network (they could, but that would really clog things... they throttle HSPA+ connections to 7.2Mb/s peak). And if they decided to call HSPA+ 4G, what would they do next summer when they roll out LTE?
T-Mobile was similarly stuck. They had just finished rolling out HSPA... they were really behind on 3G because they had to wait for 1700MHz/2100MHz to be auctioned, which they got in 2006... Verizon had already built their 3G network before T-Mo even started. So now, like AT&T, they've decided to roll out HSPA+ strategically... AT&T finished their HSPA+ upgrades last summer, and now they're working on LTE. T-Mobile doesn't own any spectrum for LTE (AT&T got the smaller 700MHz block, Verizon the larger, in that auction back in 2008). I had wondered what, if any, strategy they had for 4G... I never would have guessed "just lie about it".
1G-4G aren't standards per se, they're performance metrics. Your specific standard either meets the appropriate ITU standard, or it doesn't. Everyone knows what the "Gs" mean (or at least used to), but the ITU doesn't formally use "G" in their specification names. So there's not much likelihood anyone can call Sprint or T-Mobile on this... other than informally, like here, in articles, etc.
It'll be interesting to see when someone actually does roll out real 4G. Will they take on the others and promote that they have the only "real" 4G, or will they give it a different name?
EvDO = Evolution, Data-Only. There's no actual 3G voice mode, in the sense of UMTS. They should have done what, hopefully, the LTE networks will do, and support mixed "voice" and "data" via VoIP. The only problem is that put an additional burden on the handsets, so they didn't, for 3G. But any old smartphone can run VoIP these days, as a "simple matter of software". So there's no real excuse any longer.
They didn't. The ITU started defining "beyond 3G" in 2002, and more formally, as IMT-Advanced, back in 2005. The only technologies so far proposed to meet this standard have been WiMax and LTE, neither actually making the 4G standards in their current form. Regardless of that, the "4G" name has been applied to LTE systems going back some years now (they're only being rolled out in the USA this year, but they've been available in other locations).
I think the problem is that the ITU's use of 2G, 3G, and 4G has been fairly informal. After all, the real ITU name for "3G" is IMT-2000... I don't think anyone was trying to sell 2G technologies as meeting 3G specs back then, but they probably could have. Same deal now... only the marketroids are intentionally distorting the truth here. And T-Mobile most of all.. HSPA+ is very much a 3G technology, regardless of the performance they're allowing on it. At least the other guys are backing technologies that, in another development cycle or two, will be actual 4G standards. HSPA is finished; it's a 3G tech and will remain so.
This isn't Sprint's fourth network, anyway. They have offered HSPA access as a "3G" service, same as AT&T, for quite some time. This is part of the 3GPP (3rd Generation Partnership Project) specifications, which define the set of technologies generally used by GSM-aligned companies. This is actually part of the 3GPP's WCDMA specification... officially "3G". Version 7 of the WCDMA specification allows for a higher speed option, HSPA+, still very much a 3G technology. AT&T has had HSPA+ available for quite some time as well.
The only real difference in Sprint's plan are the bitrate caps. Each technology has a maximum per client download rate, but they never allow any client access to the full rate, simply because that stresses both the number of clients on a cell, the backhaul, and overall client expectations (eg, if I get 10Mb/s on rare occasions, I'm much less happy when my speed drop to 1Mb/s than if I normally got 3Mb/s). AT&T limits their client devices to 3.6Mb/s on an HSPA node and 7.2Mb/s on an HSPA+ node. T-Mo recently upped their caps, so it's technically possible to get something like 20Mb/s down on an HSPA+ node. Actual limits are 14Mb/s for HSPA and 42Mb/s for HSPA+. Higher bitrates are already fairly common on GSM-based networks in other parts of the world. This is just an upgrade of existing technology, not a new network... except maybe to T-Mobile, who didn't have any 3G service before the 1700MHz/2100MHz bands were opened a few years ago.
And the 3GPP themselves (not just the ITU) consider their 4G technology to be LTE. The original LTE rollout by Verizon this year and AT&T next year isn't there yet. LTE Advanced, to be finalized next year, is the first one that meets the ITU's actual specifications (at least in part) for 4G.
And yeah, these "G" standards have always been official, globally recognized standards defined by the ITU, not just things that marketing people invent. It was only a matter of time before Sprint's "4G" marketing of not-really-4G WiMax got everyone telling tall tales. Hopefully they'll all have to eat a little crow on this, but more than likely, they'll get away with it.
HD bitrates jump between 2600kb/s and 3800kb/s, SD bitrates jump between 300kb/s, 500kb/s, 1000kb/s, and 1500kb/s... at least based on the last direct information I have from Netflix. They're higher than YouTube for the same class video, and rather than limiting things to 10 minutes or less, the typical use is 2 hour films. Plus, Netflix via mail was successful enough to pretty much single handedly put Blockbuster, Hollywood Video, and other rentals out of business (I guess Red Box helped, but only fairly recently). So, they clearly have the customer based and the bandwidth demands to eat up that much bandwidth.
And they're certainly want more, if the typical pipe-to-the-home could take it. They originally played with higher bitrates, but didn't have enough success in delivering a viewable video. If the average increases enough, you can bet they'll be there with higher bitrates to eat that up, too. They still have to compete with other forms of rental/PPV, and increasing quality will make them more competitive.
How about stuff I DON'T already have on Blu-Ray/DVD?
Something's wrong there... Netflix claim their "HD" is encoded at 2.600Mb/s and 3.800Mb/s (they switch streams on the fly, based on available bandwidth). Maybe they've added some lower bitrate streams since then (http://blog.netflix.com/2008/11/encoding-for-streaming.html). Did you verify this is actually HD at that point? You're awfully close to their peak SD rate, 1.5Mb/s (which other sources claim is 1.6Mb/s, anyway). Some Netflix players let you fix the bitrate, others only automatically adapt.
New material encoded for SD is using VC-1 (eg, WMV9, SMPTE 421M), just as the HD stuff. While the HD bitrates vary between 2.6Mb/s and 3.8Mb/s, the SD stuff is apparently encoded at a fixed rate of 320kb/s, 500kb/s, 1000kb/s, and 1500kb/s... I gather all bitrates are available, and your connection determines which you get. They deliver 640x480 video, like most online streaming, so it's slightly lower rez than DVD just to being. A well encoded AVC stream will have twice the encoding efficiency of MPEG-2; VC-1 is a bit less, but even assuming it was as good as AVC, you're getting video that's more than twice as compressed, visually speaking, as a DVD. This ought to be comparable to some cable/satellite SD feeds (I have only recently seen HD Netflix, and the older WMV3 based SD, so I can't personally comment on today's SD), but it's not DVD class.
That's why I have a 71" TV.. which always displays in 1080p, regardless of the input.
For your 55" TV, the average person will not see any difference between 720p and 1080p at 13ft. They won't see the full 1080p resolution until they move to 7.2ft from the screen, but closer than 13ft and you're seeing an improvement over 720p. The SMPTE recommended viewing distance for a 55" screen is 7.5ft, the THX recommended viewing distance is 6.1ft. So if you're concerned about a good home theater experience, you already have your seats positioned for very good 1080p viewing.
My main problem with Netflix isn't that it's half the resolution of proper Blu-Ray (720/60p is good for high motion -- I shoot sports in 720/60p or 1080/60p, but for most content, you want 1080i or 1080p), but that it's at best about 1/10th the bitrate of Blu-Ray. So it's not just the resolution, but also the fact that they apply a global low-pass filter prior to encoding to minimize DCT artifacts, and then compress a bitrate 1/10th that of your average Blu-Ray, 1/5th that of ATSC broadcast. Now sure, if you watch nothing but such overcompressed view, your brain will adapt and you'll slowly start ignoring the compression artifacts -- same reason people were relatively happy with analog or first-generation TiVo, until better things came along.
The US also ranks 23rd in "Most Livable Places", 33rd in education, 37th in healthcare, 20th for child well-being... and 78th in Homicide (but the sale is from lowest to highest). And we're in a 26-way tie for the 21rst slot in literacy. Guess you can't win 'em all...
Their HD content SUCKS compares to OTA or Cable. If you don't have a 1080i/p TV, you'll notice it less, but even based on compression artifacts, it's pretty annoying. Ok, sure, I'm a video guy, I know what to look for, and I would need months of looking at Netflix video, rather than full 1080p from camcorders, before my brain started ignoring the flaws (which it will... eventually).