It was RIM, not Verizon, who made the deal with the devil, er, Microsoft, to replace Google with Bing! And apparently (not a Blackberry user), you don't have an easy way to change default search in the Blackberry OS. Apple is rumored to be contemplating (eg, parlaying with Microsoft over just how much cash this will require) doing this to all iPhone and iPod Touch users. That rumor also suggests you'll be able to change back, but I'm sure that feature is also for sale.
Verizon, in fact, could NOT legally impose Bing! on any DROID users. The DROID is a "Google Experience" phone. Neither the hardware vendor nor the network company is allowed to modify the operation of a device in this class. I have a DROID myself, and I can assure you, there has never been a single Bing! search conducted on that device. Google works just dandy.
Given the large footprint of the iPhone, HTML/CSS/Javascript is the only "alternate app" hole they haven't closed. Given Palm's WebOS using this stuff as their "native" application development environment, it does rather suggest that some applications can be written this way without big problem. Yeah, it's a mess, but less work than four or five smartphone-specific versions? Perhaps not.
Apple refused Google Voice in the app store. So Google's not dealing with the app store.
So they just rewrite it for the web: Javascript, HTML, and CSS. This happens to work on the iPhone, and if they add some enhancements for iPhone users who want this program, how's that anything bad? This also runs on Palm's WebOS, and perhaps other smart phones with modern browers. This is a good thing... many people want this, and if Google had to write a phone-specific version for every phone, some people might be left behind. And in fact, this is the future... many apps will be written this way. WebOS, in fact, is largely based on using Javascript, HTML, and CSS to deliver applications. With Palm and Apple and various others fighting to get better Javascript benchmarks, this was only a matter of time.
They have a nice and very functional Google Voice app for Android, which will work just dandy, and better than an iPhone app would anyway, since it can run background servers. If you can run the program you want on your iPhone, aren't you better served? Why should you have to put up with Apple's plans.
It's kind of amazing... Microsoft, for years, did stupid little things to ensure their future dominance. They were usually keel-hauled for it, in forums like this. Didn't change anything.. they still did it. Well, up though Vista, which is where this "we're building an OS for us, but charging you for it" really caused them problems. So they backed off a bit.
Apple, on the other hand, is taking a hard-line approach, with draconic censoring of applications. So you can't run a Commodore 64 emulator on your iPhone, because its ability to run "programs Apple doesn't get paid for" is a major threat to Apple's future. And you can't run Java programs, for the same reason. And you'll never get Flash or Shockwave, for the same reason... it doesn't even matter that this makes iPhone a second-class web browsing engine.. Apple cares more about a few more pennies from users than it does about you getting what you think you paid for (eg, the often touted best pocket web browsing experience... which it's not anymore, not by a long shot).
Javascript was the only loophole... the only method of code execution that Apple didn't cut out of your typical web browser experience. And they made it fast... last year, they were faster than Android and twice as fast as WebOS, even though most WebOS needed the speed (this changed in WebOS 1.3 and, more still, in WebOS 1.4). Palm has pretty much shown the way... while there won't be a serious level of video games done this way, for many pocket-sized applications, web-based apps work fine. They're going to run on Palm, on Android, on Nokia, and, unless Apple further works to break their support of the Web's official and de-facto standards, on iPhone.
And the funny thing... Apple is pushing developers toward this kind of development, through their approval policies.
It wasn't simply "rendering engines optimized for x86"... it was the simple fact that Apple sold 6 million computers per year, which made them middle of the top 10 PC vendors. There was no way IBM or Motorola/Freescale or anyone else could build competitive PowerPC chips. Apple fell behind, then behind some more, and that was a continual thing. They were lucky to ever even match the PC (briefly, the PowerPC 970 was close to the AMD Opteron in performance, but only briefly).
Making the Mac into Just Another PC was Apple's only way to keep the platform relevant.
Comparing Premiere to Final Cut Pro is not even fair... Premiere is perhaps Adobe's weakest application. Lots of people were leaving Premiere in the late 1990s and early 2000s... some for FCP, some for Vegas, some for Avid, some elsewhere. Premiere 6 wasn't just flakey, it was about ten years behind the times. It still required video proxy files, it didn't have first class audio features, much less arbitrary audio buses, it didn't have unlimited video tracks, nested compositing, etc. It was primitive.
I hear some of that's been fixed in Premiere Pro, but it was never interesting enough to look at again.
More correctly, both Verizon CDMA, AT&T GSM/HSPA, and T-Mobile GSM/HSPA have chosen LTE as their "4G" cellular technology.
But all is not butterflies and puppies. Sprint didn't... they're part of the Clear consortium (also including the old Clearwire, Comcast, Intel, Google, and a few others) who have chosen WiMax as their 4G technology.
And frequencies... Verizon won the big 20-something-MHz chunk on 700MHz, which is going to be their LTE. AT&T is also on 700MHz, but they only got a 12MHz chunk. I have not heard what T-Mobile plans to do, but they're still finishing up the 3G network. Sprint/Clear are on 2500MHz, so they're going to need many more towers. But they do have lots of bandwidth.
As for the networks... the nice thing about CDMA is that their 3G technology, EvDO (Evolution Data-Only) used the same 2.5MHz bandwidth they use for 2G. This is why virtually every Verizon and Sprint cell is a 3G cell, and was years back. This was a Good Decision... this is why Verizon 3G coverage is so much better than AT&T's. Not faster at peak, but once you're off a cell, they pretty much settle down in the same practical ranges. No, you can't do voice over EvDO, other than VoIP on your own phone. There's no voice protocol for LTE yet, either, but some companies are proposing one... like HSPA, a separate layer, not VoIP. We'll see... the original plan was for LTE to be all IP based.
Thus, Sprint and Verizon already with the 4G. Sprint's got WiMax out in 30 cities; Verizon will go hot, all at once, sometime this summer in about as many places. AT&T's waiting for 2011 for 4G.... and still upgrading some 3G towers. AT&T's also behind because of the AT&T Mobility and Cingular merger of years back. AT&T Mobility originally used DAMPS (the "just like GSM but not GSM" system, they called it TDMA), and after the merger, they replaced DAMPS cells with GSM cells. They finished that in 2008.
HSPA needed new spectrum, usually. So that's why only about 20% of AT&Ts cells are 3G. Even fewer are the really cool HSPA+, but that's on two paired cells, for a total of 20MHz of spectrum. Where they have them (they'll have outfitted about... you guessed it... 30-40 cities as of this summer), you can get 7.2Mb/s down, 2Mb/s up... at least standing right by the cell tower. And not on your iPhone... they only do 384kb/s up, all models. Sprint is only claiming a 5Mb/s typical connection on their 4G service. Though Comcast claim 8-12Mb/s, and Clear claims 10-12Mb/s... BUT IT'S ALL THE SAME NETWORK. Go figure.
It has very little to do with storage or CPU. Yeah, Google could re-encode the whole YouTube library, probably in a short time, and pretty cheap. It would drop in quality, transcoding from the H.264. And don't discount H.264 on YouTube... phone-cam stuff isn't encoded at high bitrates. You only get HD options on YouTube from HD sources.... I have several of these up ("hazydave" channel), from full HD sources. They wouldn't look good on my 71" DLP, but for Web Video... doesn't suck. YouTube serves up "HD" at 2Mb/s, always cut to 720p if you don't downrez yourself (they actually accept 1080i/p uploads, but don't save them). Then again, the phone-cam on my phone produces better quality video that what you get from many consumer camcorders (Motorola DROID... it's just ok, but there are plenty of consumer SD camcorders with tiny 1/8" sensors and not even full DV quality resolution).
It's quality and bandwidth that makes Ogg Theora a problem right now. You need more bits for the same quality from an original source, plain and simple -- Theora is not as good as H.264. Google may be mighty, but they do pay for bandwidth. Lots. That's why YouTube is still losing money. Going to Theora would cost more in bandwidth than they would save not paying the H.264 licenses. Also not a good thing for the internet in general, considering how much volume of global traffic is video these days.
Google may well be going in yet-another-direction. They did just buy On2, the guys who did the work behind VP3, which begat Theora. And VP6, which was the original video CODEC in Flash video. And VP7 and VP8. The On2 people claim VP8 is 40% more efficient (eg, bits at the same perceived quality) than H.264 for low bitrate video. If that's true, Google could save themselves hundreds of millions a year by re-encoding to VP8.
That would be interesting. Even more interesting would be if Google open sourced VP8, assuming it's clear of any MPEG-LA or Microsoft or other video compression patents. Eliminating H.264 advantages of quality and bitrate means real money to any of these buy guys, and you don't have to beat H.264 on HDTV to beat it streaming at a megabit or two.
I would have liked a "must carry" CODEC or two as part of the VIDEO standard, that's no big deal. What is a big deal is Mozilla saying, no, we're not going to support H.264, but we're also not going to support any of your system CODECs already. Which means you won't get any video CODEC they don't decide to compile in themselves (well, you could build your own). That's kind of evil... I have been playing around with Dirac, which would be a reasonable web format as well, and, well... there are plenty of other browsers. It's a shame that, having built back at least a reasonable chunk of the former Netscape glory, Mozilla would fall on their sword this way.
Of course, Microsoft is so asleep at the wheel, they haven't even committed to HTML5 or the tag. And they're still around 50%. I guess they think they can get SOMEONE to run Silverlight...
Yeah, actually, most of these "Intel-pushed" standards, like USB, PCI, AGP, PCI Express, etc. are patented, but you're granted free license to all patents if you follow the specs. This is because Intel just wants to sell the chips, and adding new interfaces sells more chips. It's a good thing.
In fact, USB was born largely because Apple was asking $1.00 per port for Firewire, and Intel though that was nuts, given the margins in the PC industry. At the same time, Compaq and some others were trying to figure out some kind of "desktop bus" standard, like DEC's Access.Bus or Apple's ADB. The two forces got together, and the result was very good for us. Even for Apple, eventually. And Firewire was mitigated to special purpose applications.
Well, sometimes there are folks who stick to religious convictions, regardless of the impact. Some religions reject the use of modern medicine, others, apparently, reject the use of some specific operating system interfaces (video), but curiously, not others (images, files, etc).
And then there's occasionally someone who just wants to get the frickin' job done, and perhaps in a way that people will use it. Which is why everyone says "Linux" today, not "GNUix" or whatever..... Linus just got it done. I'm sure he's got plenty of principles, but they didn't get in the way of building something people would actually use.
Mozilla looks like they are trying very hard to make themselves a moot point. I rather expect they'll just keep playing videos via a Flash plug-in, years after HTML5 takes off, with increasingly fewer users. Or we'll be getting the popular version of Firefox from somewhere else, the one that adds the external video CODEC option to Mozilla's base.
They seem to be missing the point entirely here... popularizing Theora would lead to people using it. Force-feeding will not. And until Theora doesn't suck, and does get general support in video applications that are actually creating video (eg, via those same OS-level video CODECs that Mozilla abhors) it's not going to get much use. And I say this as a guy who does both engineering and video.
It would be for two reasons. For one, the big problem with the Flash Player hasn't been that it's a plug-in, but that it was a proprietary, closed source, even closed spec thing. A big mystery, and something very difficult to replace with a similar plug-in from another company, much less open source. Defaulting to OS video CODECs is nowhere near as complex. Even if H.264 has to be licensed, it's an international standard -- nothing's hidden, you just have to pay the patent people. That IS a big difference already.
Also, with Flash, you define the whole player architecture based on this one plug-in. For HTML5, using a system video CODEC changes nothing about interactions between user and browser, just the kind of video that can be played from within the browser.
Video users are used to different video plug-ins. It's a standard OS piece, like a device driver, not something bound to just one guy's web browser, either. Far less objectionable, IMHO, than something like Flash. Supporting these frameworks, Video for Widows, DirectShow, Quicktime, gstreamer, etc. was a big step forward for video integration is operating systems. "Built-in" on the browser is a big step backward... it makes the browser big and stupid again, unable to move with video technology advances.
Second one is that Theora could very definitely be included, free (at least until someone turns up a patent it offends that hasn't expired yet), ideally also as an OS-level CODEC. That's actually somewhat superior to "hidden within the web browser", as it allows that piece to also be something replaced, improved, accelerated... and used for other purposes. If it's installed by default, I could immediately crank out a Theora video, upload it to my web site, and be serving up video. This very much would help popularize its use.
This also opens the way for the industry and web community to ultimately decide which forms should be popular, rather than saying, use for these formats (Theora, H.264), but hey, any other format (MPEG, Dirac, etc) and you're still forced, forevermore, to use a browser plug-in.
For one, you WANT to use the OS-available CODEC if it exists. This is the best way to ensure you're tapping into the available hardware resources.
Mozilla CAN support these things. They can go ahead and build-in a Theora decoder, defaulting to an OS-level replacement if one exists (which might use your GPU, either via OpenCL or an OS-specific video acceleration API, or an add-in like the SPURS Engine). Or only support the external CODECs, but supply the latest Theora for the system at hand. Also support H.264, but not as a built-in, only via an external CODEC. Then offer a closed-source, pay-for CODEC of their own, if they're really that concerned that users won't have them.
But I think it's important to note that a built-in decoder is only just better than none at all. It should be the last resort... OS level decoders should be used if available. Yeah, it's a little more work for the developers.
It's stupid that this became a format war by the browser people. The real way to support innovation is to recommend that hooks into an OS multimedia system, if available. And that fully compliant browsers will supports some set of CODECs... such as H.264 and Theora, others optional. This allows content providers more freedom, and lets the whole industry get their act together on the formats everyone wants to support.
This has happened before... with DVD and then Blu-Ray.
For one, there's this whole H.264 vs. Theora debacle in HTML5. Google and many other big companies are in support of H.264 largely because of the patent pool -- it's safe. While it's a fact that Ogg Vorbia and Theora live in the open source community, there's plenty of concern that they're not clear, patent-wise. Theora was adapted from On2's VP3 CODEC, which they donated to the FOSS community. But that doesn't make it immune to infringement of others' patents.
On2 VP6 was the basis for the old standard definition Flash Video. But modern flash is done in H.264, and Google's not paying for flash encoding anyway. But VP7 or VP8 is the standard video CODEC for JavaFX, so there's going to be life here yet.
One good bet: Google might want to offer an update to the Theora, too. On2's current video CODEC is VP8, which is far more competitive to H.264. If the H.264 fees were high, Google might have considered improving Theora with VP8 as a way to save costs on YouTube. As it standards, going to Theora today on YouTube would be bad news... they'd either lower quality or raise the one really expensive part of it... bandwidth demands. But H.264 licensing is $10,000 per web site per year... I think Google spends more in replacement foosballs for a year. On the other hand, On2 has been claiming that, at least for lower bitrates ("HD" video on YouTube is only 2Mb/s, which may quality), VP8 outperforms H.264, and is less compute intensive on playback. While many are skeptical, it would be very worthwhile to Google's bottom line from YouTube is they could actually maintain quality and lower bitrate for all that video. That would normally be a problem -- who wants another plug-in? But if they also influence the HMTL standard... no problem. Google has shown before they have no problem buying software and open sourcing it if it's going to save them money and/or protect a market (AndroidOS for example).
Another one: Google is now the parent of two operating systems: Android and ChromeOS. Most OS companies (Microsoft and Apple in particular) have their own video CODECs as well as supporting the industry standards. Owning your own may offer some advantages... not sure about $100 million of them.
Another bet is China. They have standardized in a very big way on the On2 VP6 CODECs for web video. There's a flurry of activity around VP7 for mobile video now, too. This might also mesh with On2's not-so-long-ago purchase (for $58 million) of Hantro, a mobile video company. Google is in the mobile device market now in a big way. VP7 is supposedly lower complexity, thus, longer battery life on playback... assuming the video hardware on your typical Android device offers as much boost for VP7 as it does for H264. That was a big issue (in theory, at least) over the whole Theora thing: today's video acceleration hardware is tuned for H.264, MPEG, and occasionally VC-1.
H.264 had already won. It wins because it does the job the best. Yeah, it's a commercial thing, and that IS bad for the web in ways that's not really much of a big deal for other things. Similarly, delivering the best compression per bit is critical for video delivery over most other media, but particularly... well, the places it's already used. Every major still or video camera maker uses H.264. Every Blu-Ray member, satellite companies like Echostar and DirecTV. Every major PMP maker.
The simple fact is that H.264 is to the early 21rst century was MPEG-2 was to the 1990s... THE video standard format.
The correct way to manage this the way the video players (DVD, Blu-Ray) have done, but that still sucks a little if you're Mozilla. Basically, every video format has both mandatory and optional formats it must decode. This makes this a bit harder on the player producers, but much easier on the content delivery folks.... the only have to chose one of the above. So if I make a Blu-Ray (I do, this isn't just hypothetical), I have the choice of H.264/AVC, VC-1 (WMV9), or MPEG-2 as the video format, and half a dozen audio formats, including a few flavors of AC-3 and plain old uncompressed.
So the right consumer solution here would have been to demand that support H.264 and Ogg Theora as mandatory formats. Toss in a few optional formats if you like... they might not be popular, but you never know (MPEG Layer 2 audio was an optional format for Region 1 DVD, yet nearly every players supports it).
To support open source, make the H.264 piece into a closed source, pay-for plug-in. This is how a number of companies (Nero, Archos, etc) have dealt with the extra cost of supporting AVC, at least at some time in the past. It might have a bad taste without those Fundamentalists among us who want to refuse even the possibility of closed source or proprietary formats getting into their faces, but for those who just want a practical engineering solution, this works. None of the Theora backers were EVER going to get the H.264 proponents to back down and embrace Theora as the only HTML5 video format. Look at the list... they're all the guys with the money, and most of them have vast investments already in H.264 content. Taking them on the way this was done, it was a guaranteed fail.
But of course, this is the computer industry, where everyone has to fight over their one preferred solution, rather than take the "what's best for the most people" approach, which is really what the CE type solution comes down to. And of course, the CE industry NEVER has such format wars [ducking....]
Apple users embrace the "free-thinking" mantra because that's the image Apple's served up. In short, they were told that using a Mac makes them free-thinking. And no, I'm certain the irony is not lost of those of us who abhor Apple's general policies, which are nothing of the kind.
Apple found themselves, entering the early 1990s, as the lone major computer platform other than Windows, and they had arguably better graphics and a few pretty good music applications, which were struggling to actually function on the PC/Windows until well into the Windows 95 era (UNIX-like OSs didn't do audio well at all... you needed a DSP subsystem, as on the SGIs and the NeXT machines, to do audio at all in the very non-realtimey, who-cares-about-interrupt-latency versions of UNIX/Linux at the time).
So they used this as a sales pitch. The PC equals Windows, it's from IBM, and it's used in business... thus, its only uses are business-related. They weren't selling Macs to computer experts who knew this to be false, and certainly not those of us who actually did the PC work as well, then better than the Mac on media content creation of all sorts. They're selling to users who are fairly clueless about PCs.
Apple always had very good marketing, and both that, and their message, continue today. They were selling a slightly more capable 8-bit machine, back in the early 80s, versus Commodore and Atari machines at 1/5th the price (they had slots... that's the "more capable" part). The Mac came in, with hardware so oversimplified it was actually kind of creepy (the "Ready" pin on the SCSI controller drover/DTACK on the 68000, for any bitheads in the crowd) and cheap, but got huge margins. Today, a Mac is exactly a PC in a fancy case without a battery door... there's nothing different about an Apple PC, and yet they still get 2x-3x the cash. That pays for a ton of brainwashing.
And it's also something like human nature. As some may know, I was a senior hardware designer at Commodore on several high-end models of the Amiga computer. There was a time when the Amiga was the best (only) personal computer for color graphics or video work. Like, the mid-to-late 1980s. Today I do my video stuff on a PC running Windows 7 and Sony Vegas, with 8GB of RAM, a Quad-core CPU, and Terabytes of storage. But I still hear from people talking about how the Amiga IS better (not was, but IS).
When you join an exclusive club, you immediately embrace all the positive memes associated with club membership, and you employ these to justify your decision. This isn't restricted to computers, it's found in Video software (Vegas vs. Avid vs. Premiere vs. FCP, etc), cars (Ford vs. Chevy vs. Dodge), still cameras (Canon vs. Nikon), videocameras (Sony vs. Canon vs. Panasonic vs. JVC), soft drinks (Coke vs. Pepsi... sorry, Rock Star rules here, folks), etc. And sure, the cultier that club's culture becomes, the more the users grab hold of it.
Apple is one of the few remaining exclusive clubs in computing, and they're perhaps the cultiest and most exclusive there is in just about any discipline. Ok, Amiga fans could have given them a run for their money back in the early 1990s, but not since... the Mac hardcores have expended to embrace the iPhone. The iPhone has delivered new converts to the alter of Mac. There's a persistent meme that "Windows is hard", bug ridden, full of viruses, and of course, MacOS is impervious to any and all problems, the only way to do media content in computing, and so simple your cat can use it without reading a manual. Apple works very hard to keep these memes alive, in the general population to an extent, too, not just among the Apple Faithful.
Another factor, among those in a successful cult, is that they reinforce one another and don't pay much attention to the outside world. You can stay blissfully within your world of Apple -- magazines, web sites, etc. and never hear more than frightening stories about the world outside. This is also something that Apple cultivates...they were among t
Ok, yeah, $400,000 is a pretty insane sum of money. But you have to feel at least a little schadenfreude at Universal having to pay this. After all, they're part of the music industry machine that has worked tirelessly to erode and obfuscate Fair Use laws. Most of these laws are random exclusions in otherwise restrictive copyright laws. Any time there's a push for a true proactive Fair Use bill... a Fair Use Bill of Rights that really needs to be written... these guys are spending millions in lobbying money to such that down, pronto. So a little of what goes around finally did come around. I can't be even a little bit disturbed at this outcome... if the law were crystal clear, the bill would have been much, much smaller I'm sure.
Plus, if Universal's equally-ultra-expensive lawyers are at all good, Universal will spend even more money in legal hours trying to delay the payout as long as they possibly can.
I was at a set-top box company, based in Germany, in the late 1990s. We had every intention of building out multimedia set-top box on Linux, even though that wasn't as doable as it is with today's modern Linux kernel (and, well, faster CPUs doesn't hurt, either). The big problem: finding anyone who knew Linux and wanted to actually get paid. We advertised, we attended Linux shows, etc... no go. Might have been practical if we were based in California, but at the time, there was a real issue among many Linux hackers of working for pay. So we wound up using OS/2 on the same hardware. I was largely the hardware boss, but I also wrote some drivers -- tragically, one place Linux was way ahead (physical drivers had to be written in 16-bit code under OS/2, at least at the time).
Nice to see things change. Not that spare time, hobby, educational, etc. contributions are a bad thing at all. But commercial concerns doing all that work is a strong indicator that lots more work is being done. They also sometimes have a better sense of finishing a project. Not always, but sometimes.
Never used an Android phone, have you? Thought not... it does all that. And no need to dock to a PC of any flavor, ever. You can if you want (shows up as a USB storage device).
Nokia certainly had some plans for the GPS market. It was in 2007 that they scrambled to buy one of the two big turn-by-turn mapping companies. First they made an offer to Tele-Atlas, which scared Tom-Tom into buying them instead. Then they bough up Navteq. Oddly enough, this was the shift that got Google scared about their future plans in the maps and navigation business (they did, at one point, buy the mapping data from both companies).
You do have to wonder, though... this could help kill their external Navteq business, but it would take the actual destruction of the stand-alone GPS market. Presumably, they need this in the phones to complete with Google/Android, or at least think they do.
Anyway, there's little to fear about the updates... this is the same data Garmin and many of the others are using, and charging you $50-$100 every year or so in "new maps" charges. There are some advantages, still, to stand-alone units... the 5" screen on my Pioneer AVIC GPS is a bit more readable than the 3.7" on my DROID. The Pioneer also is integrated into my car stereo system, so for GPS advice and phone calls, no need to mess with volume knob. And Google Navigator (beta) has occasionally led me on "adventures", while the Pioneer, not so much. But there's nothing cooler than GPSing with satellite images, and certainly, if I already had the DROID, I probably would have thought twice about buying a stand-alone.
Sounds like monitor problems, or cheap monitors. I have dual 24" 1200p monitors here, usually on HDMI, though I use VGA and YPrPb ports occasionally. It's very obvious, particularly for computer graphics, when I'm on VGA. For video-level stuff, it's much harder to tell.. you usually only see what's left of analog effects, viewed on a digital monitor, at sharp pixel transitions.
VGA was perpetually being replaced... only problem was, it didn't really need to be. So, while there were plenty of alternatives, none of them took hold.
Once monitors went digital, though, it was clearly time to nix the VGA. DVI was designed to be transitional... it's got all the VGA signals and digital signals. It was intended to let you put just the one connector on your digital video card and connector to either analog or digital monitors. Hey, it worked.. most video cards carry one or two DVI connectors these days.
HDMI was something else... this was not a computer industry standard, but one for the consumer electronics industry. They added audio, which is certainly important in CE applications, perhaps something they forgot to put into DVI. Also, the HDMI connector is more consumer friendly than the D-Sub inspired DVI connector. Though they do fall out, which is somewhat annoying. HDMIs are showing up in computer gear simply because the connector size is better for many applications (notebooks, multi-head display cards). It's also a bit of an upgrade from DVI, in terms of data speeds.
DisplayPort is the replacement for DVI.. all digital, with all of the goodies that computer industry folks want to find in their next generation video-specific port.
Next might be Light Peak, which will do video speeds, but will be a general purpose interconnect, not video specific. This will be interesting, since that might bring about smarter displays... they could negotiate not just on resolutions but encodings and other fun stuff (well, DVI and HDMI do a bit of this, negotiating YUV or YCrCb vs RGB.
Actually, the plastic TOSlink cables are generally crap (the standard calls for plastic, you can't use glass). Much lossier than comparable copper, particularly at the low speeds used for digital audio. Mid-range users, maybe. High-end people don't use these. At least those who know what they're doing (or, occasionally, can't avoid them... if you run out of AES/EBU or S/PDIF ports, TOSlink is fine over short cables).
Expenses. The Sony/Philips optical is about the only consumer optical in common use, but that's over plastic, driven by LEDs. Pretty cheap. For optical at these rates, you'd need real lasers (LEDs peak around 500Mb/s) something like 10GBASE-R or 10GBASE-SR cable (LOMMF/OM3). None of that's crazy expensive... unless you compare it to electrical. And in particular, the electrical that the equipment makers are actually paying for.
Keep in mind, these are the industry guys who got together to create DisplayPort, at least in part because they got bent out of shape having to pay US$0.04 per device to use HDMI. They're not likely to replace a $0.50 electrical connector with a $2.00 optical connector and $5.00+ laser. And of course, lasers go one way... you actually need a laser at each end, if you want 2-way traffic. Or a custom cable, with electrical backchannel.
Well, why not.. I have some video cables around here with integrated optical audio channel.
So this is the next one up, after DisplayPort, but designed as a general purpose standard: Light Peak. I think this started out as an optical answer to Firewire at Apple, but rather than do it themselves, change too much for the spec, and have Intel (and the rest of the PC industry) go and create an alternative, this time Apple brought it to Intel. Maybe.. at least that's one story.
The nice thing about Light Peak... it's fast enough to do the HDMI/DisplayPort thing. And replace SATA, USB, Firewire, anything else you want. Of course, like all optical interconnects, the connectors are an issue (dirt kills), and unless they go to some kind of FDM, they'll need one cable in each direction, just like 10Ge uses in its various optical forms. Then there's the issue of power... we're kind of used to USB and Firewire cables providing power for small devices. But it's still a work in progress, 10Gb/s on launch, up to 100Gb/s on the roadmap.
DisplayPort itself uses a completely different kind of signalling than HDMI. An HDMI signal is basically just a digital version of the video stream... it runs three differential TMDS data links and one clock link at 10x the rate of the display's pixel clock, very tightly coupled to the video. Native DisplayPort sends packetized data, and the signal is over 1, 2, or 4 differential serial lines, with clocking information embedded in that signal.
HDMI 1.3 supports 10.2Gb/s per link (there's a dual link version, but it's not common), and resolutions up to 2560×1600p75 for 24-bits per pixel or 1920x1200p60 for 48-bits per pixel. HDMI 1.4 adds a 100Mb/s ethernet channel and support for displays up to 4096x2160p24 at 24 bits per pixel (same bandwidth, more formats).
DisplayPort 1.1 supports a throughput of 8.64 Gb/s across the four links, which goes to 17.2 Gb/s in version 1.2 (the 21.6Gb/s is the symbol rate over 8b/10b encoded links).
In short, they're completely different. However, the spec does allow a DisplayPort connector to switch and spit out HDMI/DVI compatible signals instead. There is no requirement that DisplayPort connectors do this, however.
So any in-cable passive DisplayPort to HDMI cable is counting on this HDMI bridge feature being in the device you're using. An "active" cable will convert the video signal (well, a simple one), but lose the 2-way data features.
Slashdot Mirror
It was RIM, not Verizon, who made the deal with the devil, er, Microsoft, to replace Google with Bing! And apparently (not a Blackberry user), you don't have an easy way to change default search in the Blackberry OS. Apple is rumored to be contemplating (eg, parlaying with Microsoft over just how much cash this will require) doing this to all iPhone and iPod Touch users. That rumor also suggests you'll be able to change back, but I'm sure that feature is also for sale.
Verizon, in fact, could NOT legally impose Bing! on any DROID users. The DROID is a "Google Experience" phone. Neither the hardware vendor nor the network company is allowed to modify the operation of a device in this class. I have a DROID myself, and I can assure you, there has never been a single Bing! search conducted on that device. Google works just dandy.
Given the large footprint of the iPhone, HTML/CSS/Javascript is the only "alternate app" hole they haven't closed. Given Palm's WebOS using this stuff as their "native" application development environment, it does rather suggest that some applications can be written this way without big problem. Yeah, it's a mess, but less work than four or five smartphone-specific versions? Perhaps not.
That's not what Google did.
Apple refused Google Voice in the app store. So Google's not dealing with the app store.
So they just rewrite it for the web: Javascript, HTML, and CSS. This happens to work on the iPhone, and if they add some enhancements for iPhone users who want this program, how's that anything bad? This also runs on Palm's WebOS, and perhaps other smart phones with modern browers. This is a good thing... many people want this, and if Google had to write a phone-specific version for every phone, some people might be left behind. And in fact, this is the future... many apps will be written this way. WebOS, in fact, is largely based on using Javascript, HTML, and CSS to deliver applications. With Palm and Apple and various others fighting to get better Javascript benchmarks, this was only a matter of time.
They have a nice and very functional Google Voice app for Android, which will work just dandy, and better than an iPhone app would anyway, since it can run background servers. If you can run the program you want on your iPhone, aren't you better served? Why should you have to put up with Apple's plans.
It's kind of amazing... Microsoft, for years, did stupid little things to ensure their future dominance. They were usually keel-hauled for it, in forums like this. Didn't change anything .. they still did it. Well, up though Vista, which is where this "we're building an OS for us, but charging you for it" really caused them problems. So they backed off a bit.
Apple, on the other hand, is taking a hard-line approach, with draconic censoring of applications. So you can't run a Commodore 64 emulator on your iPhone, because its ability to run "programs Apple doesn't get paid for" is a major threat to Apple's future. And you can't run Java programs, for the same reason. And you'll never get Flash or Shockwave, for the same reason... it doesn't even matter that this makes iPhone a second-class web browsing engine.. Apple cares more about a few more pennies from users than it does about you getting what you think you paid for (eg, the often touted best pocket web browsing experience... which it's not anymore, not by a long shot).
Javascript was the only loophole... the only method of code execution that Apple didn't cut out of your typical web browser experience. And they made it fast... last year, they were faster than Android and twice as fast as WebOS, even though most WebOS needed the speed (this changed in WebOS 1.3 and, more still, in WebOS 1.4). Palm has pretty much shown the way... while there won't be a serious level of video games done this way, for many pocket-sized applications, web-based apps work fine. They're going to run on Palm, on Android, on Nokia, and, unless Apple further works to break their support of the Web's official and de-facto standards, on iPhone.
And the funny thing... Apple is pushing developers toward this kind of development, through their approval policies.
It wasn't simply "rendering engines optimized for x86"... it was the simple fact that Apple sold 6 million computers per year, which made them middle of the top 10 PC vendors. There was no way IBM or Motorola/Freescale or anyone else could build competitive PowerPC chips. Apple fell behind, then behind some more, and that was a continual thing. They were lucky to ever even match the PC (briefly, the PowerPC 970 was close to the AMD Opteron in performance, but only briefly).
Making the Mac into Just Another PC was Apple's only way to keep the platform relevant.
Comparing Premiere to Final Cut Pro is not even fair... Premiere is perhaps Adobe's weakest application. Lots of people were leaving Premiere in the late 1990s and early 2000s... some for FCP, some for Vegas, some for Avid, some elsewhere. Premiere 6 wasn't just flakey, it was about ten years behind the times. It still required video proxy files, it didn't have first class audio features, much less arbitrary audio buses, it didn't have unlimited video tracks, nested compositing, etc. It was primitive.
I hear some of that's been fixed in Premiere Pro, but it was never interesting enough to look at again.
More correctly, both Verizon CDMA, AT&T GSM/HSPA, and T-Mobile GSM/HSPA have chosen LTE as their "4G" cellular technology.
But all is not butterflies and puppies. Sprint didn't... they're part of the Clear consortium (also including the old Clearwire, Comcast, Intel, Google, and a few others) who have chosen WiMax as their 4G technology.
And frequencies... Verizon won the big 20-something-MHz chunk on 700MHz, which is going to be their LTE. AT&T is also on 700MHz, but they only got a 12MHz chunk. I have not heard what T-Mobile plans to do, but they're still finishing up the 3G network. Sprint/Clear are on 2500MHz, so they're going to need many more towers. But they do have lots of bandwidth.
As for the networks... the nice thing about CDMA is that their 3G technology, EvDO (Evolution Data-Only) used the same 2.5MHz bandwidth they use for 2G. This is why virtually every Verizon and Sprint cell is a 3G cell, and was years back. This was a Good Decision... this is why Verizon 3G coverage is so much better than AT&T's. Not faster at peak, but once you're off a cell, they pretty much settle down in the same practical ranges. No, you can't do voice over EvDO, other than VoIP on your own phone. There's no voice protocol for LTE yet, either, but some companies are proposing one... like HSPA, a separate layer, not VoIP. We'll see... the original plan was for LTE to be all IP based.
Thus, Sprint and Verizon already with the 4G. Sprint's got WiMax out in 30 cities; Verizon will go hot, all at once, sometime this summer in about as many places. AT&T's waiting for 2011 for 4G.... and still upgrading some 3G towers. AT&T's also behind because of the AT&T Mobility and Cingular merger of years back. AT&T Mobility originally used DAMPS (the "just like GSM but not GSM" system, they called it TDMA), and after the merger, they replaced DAMPS cells with GSM cells. They finished that in 2008.
HSPA needed new spectrum, usually. So that's why only about 20% of AT&Ts cells are 3G. Even fewer are the really cool HSPA+, but that's on two paired cells, for a total of 20MHz of spectrum. Where they have them (they'll have outfitted about... you guessed it ... 30-40 cities as of this summer), you can get 7.2Mb/s down, 2Mb/s up... at least standing right by the cell tower. And not on your iPhone... they only do 384kb/s up, all models. Sprint is only claiming a 5Mb/s typical connection on their 4G service. Though Comcast claim 8-12Mb/s, and Clear claims 10-12Mb/s... BUT IT'S ALL THE SAME NETWORK. Go figure.
It has very little to do with storage or CPU. Yeah, Google could re-encode the whole YouTube library, probably in a short time, and pretty cheap. It would drop in quality, transcoding from the H.264. And don't discount H.264 on YouTube... phone-cam stuff isn't encoded at high bitrates. You only get HD options on YouTube from HD sources.... I have several of these up ("hazydave" channel), from full HD sources. They wouldn't look good on my 71" DLP, but for Web Video... doesn't suck. YouTube serves up "HD" at 2Mb/s, always cut to 720p if you don't downrez yourself (they actually accept 1080i/p uploads, but don't save them). Then again, the phone-cam on my phone produces better quality video that what you get from many consumer camcorders (Motorola DROID... it's just ok, but there are plenty of consumer SD camcorders with tiny 1/8" sensors and not even full DV quality resolution).
It's quality and bandwidth that makes Ogg Theora a problem right now. You need more bits for the same quality from an original source, plain and simple -- Theora is not as good as H.264. Google may be mighty, but they do pay for bandwidth. Lots. That's why YouTube is still losing money. Going to Theora would cost more in bandwidth than they would save not paying the H.264 licenses. Also not a good thing for the internet in general, considering how much volume of global traffic is video these days.
Google may well be going in yet-another-direction. They did just buy On2, the guys who did the work behind VP3, which begat Theora. And VP6, which was the original video CODEC in Flash video. And VP7 and VP8. The On2 people claim VP8 is 40% more efficient (eg, bits at the same perceived quality) than H.264 for low bitrate video. If that's true, Google could save themselves hundreds of millions a year by re-encoding to VP8.
That would be interesting. Even more interesting would be if Google open sourced VP8, assuming it's clear of any MPEG-LA or Microsoft or other video compression patents. Eliminating H.264 advantages of quality and bitrate means real money to any of these buy guys, and you don't have to beat H.264 on HDTV to beat it streaming at a megabit or two.
I would have liked a "must carry" CODEC or two as part of the VIDEO standard, that's no big deal. What is a big deal is Mozilla saying, no, we're not going to support H.264, but we're also not going to support any of your system CODECs already. Which means you won't get any video CODEC they don't decide to compile in themselves (well, you could build your own). That's kind of evil... I have been playing around with Dirac, which would be a reasonable web format as well, and, well... there are plenty of other browsers. It's a shame that, having built back at least a reasonable chunk of the former Netscape glory, Mozilla would fall on their sword this way.
Of course, Microsoft is so asleep at the wheel, they haven't even committed to HTML5 or the tag. And they're still around 50%. I guess they think they can get SOMEONE to run Silverlight...
Yeah, actually, most of these "Intel-pushed" standards, like USB, PCI, AGP, PCI Express, etc. are patented, but you're granted free license to all patents if you follow the specs. This is because Intel just wants to sell the chips, and adding new interfaces sells more chips. It's a good thing.
In fact, USB was born largely because Apple was asking $1.00 per port for Firewire, and Intel though that was nuts, given the margins in the PC industry. At the same time, Compaq and some others were trying to figure out some kind of "desktop bus" standard, like DEC's Access.Bus or Apple's ADB. The two forces got together, and the result was very good for us. Even for Apple, eventually. And Firewire was mitigated to special purpose applications.
Flash actually does On2 VP6 or H.264, these days. The older SD video online is VP6 in a Flash container.
Well, sometimes there are folks who stick to religious convictions, regardless of the impact. Some religions reject the use of modern medicine, others, apparently, reject the use of some specific operating system interfaces (video), but curiously, not others (images, files, etc).
And then there's occasionally someone who just wants to get the frickin' job done, and perhaps in a way that people will use it. Which is why everyone says "Linux" today, not "GNUix" or whatever..... Linus just got it done. I'm sure he's got plenty of principles, but they didn't get in the way of building something people would actually use.
Mozilla looks like they are trying very hard to make themselves a moot point. I rather expect they'll just keep playing videos via a Flash plug-in, years after HTML5 takes off, with increasingly fewer users. Or we'll be getting the popular version of Firefox from somewhere else, the one that adds the external video CODEC option to Mozilla's base.
They seem to be missing the point entirely here... popularizing Theora would lead to people using it. Force-feeding will not. And until Theora doesn't suck, and does get general support in video applications that are actually creating video (eg, via those same OS-level video CODECs that Mozilla abhors) it's not going to get much use. And I say this as a guy who does both engineering and video.
If you're running Safari, you're already far more locked than any Android user.
It would be for two reasons. For one, the big problem with the Flash Player hasn't been that it's a plug-in, but that it was a proprietary, closed source, even closed spec thing. A big mystery, and something very difficult to replace with a similar plug-in from another company, much less open source. Defaulting to OS video CODECs is nowhere near as complex. Even if H.264 has to be licensed, it's an international standard -- nothing's hidden, you just have to pay the patent people. That IS a big difference already.
Also, with Flash, you define the whole player architecture based on this one plug-in. For HTML5, using a system video CODEC changes nothing about interactions between user and browser, just the kind of video that can be played from within the browser.
Video users are used to different video plug-ins. It's a standard OS piece, like a device driver, not something bound to just one guy's web browser, either. Far less objectionable, IMHO, than something like Flash. Supporting these frameworks, Video for Widows, DirectShow, Quicktime, gstreamer, etc. was a big step forward for video integration is operating systems. "Built-in" on the browser is a big step backward... it makes the browser big and stupid again, unable to move with video technology advances.
Second one is that Theora could very definitely be included, free (at least until someone turns up a patent it offends that hasn't expired yet), ideally also as an OS-level CODEC. That's actually somewhat superior to "hidden within the web browser", as it allows that piece to also be something replaced, improved, accelerated... and used for other purposes. If it's installed by default, I could immediately crank out a Theora video, upload it to my web site, and be serving up video. This very much would help popularize its use.
This also opens the way for the industry and web community to ultimately decide which forms should be popular, rather than saying, use for these formats (Theora, H.264), but hey, any other format (MPEG, Dirac, etc) and you're still forced, forevermore, to use a browser plug-in.
Perhaps both.
For one, you WANT to use the OS-available CODEC if it exists. This is the best way to ensure you're tapping into the available hardware resources.
Mozilla CAN support these things. They can go ahead and build-in a Theora decoder, defaulting to an OS-level replacement if one exists (which might use your GPU, either via OpenCL or an OS-specific video acceleration API, or an add-in like the SPURS Engine).
Or only support the external CODECs, but supply the latest Theora for the system at hand. Also support H.264, but not as a built-in, only via an external CODEC. Then offer a closed-source, pay-for CODEC of their own, if they're really that concerned that users won't have them.
But I think it's important to note that a built-in decoder is only just better than none at all. It should be the last resort... OS level decoders should be used if available. Yeah, it's a little more work for the developers.
It's stupid that this became a format war by the browser people. The real way to support innovation is to recommend that hooks into an OS multimedia system, if available. And that fully compliant browsers will supports some set of CODECs... such as H.264 and Theora, others optional. This allows content providers more freedom, and lets the whole industry get their act together on the formats everyone wants to support.
This has happened before... with DVD and then Blu-Ray.
Google buying On2 could mean a few things.
For one, there's this whole H.264 vs. Theora debacle in HTML5. Google and many other big companies are in support of H.264 largely because of the patent pool -- it's safe. While it's a fact that Ogg Vorbia and Theora live in the open source community, there's plenty of concern that they're not clear, patent-wise. Theora was adapted from On2's VP3 CODEC, which they donated to the FOSS community. But that doesn't make it immune to infringement of others' patents.
On2 VP6 was the basis for the old standard definition Flash Video. But modern flash is done in H.264, and Google's not paying for flash encoding anyway. But VP7 or VP8 is the standard video CODEC for JavaFX, so there's going to be life here yet.
One good bet: Google might want to offer an update to the Theora, too. On2's current video CODEC is VP8, which is far more competitive to H.264. If the H.264 fees were high, Google might have considered improving Theora with VP8 as a way to save costs on YouTube. As it standards, going to Theora today on YouTube would be bad news... they'd either lower quality or raise the one really expensive part of it... bandwidth demands. But H.264 licensing is $10,000 per web site per year... I think Google spends more in replacement foosballs for a year. On the other hand, On2 has been claiming that, at least for lower bitrates ("HD" video on YouTube is only 2Mb/s, which may quality), VP8 outperforms H.264, and is less compute intensive on playback. While many are skeptical, it would be very worthwhile to Google's bottom line from YouTube is they could actually maintain quality and lower bitrate for all that video. That would normally be a problem -- who wants another plug-in? But if they also influence the HMTL standard... no problem. Google has shown before they have no problem buying software and open sourcing it if it's going to save them money and/or protect a market (AndroidOS for example).
Another one: Google is now the parent of two operating systems: Android and ChromeOS. Most OS companies (Microsoft and Apple in particular) have their own video CODECs as well as supporting the industry standards. Owning your own may offer some advantages... not sure about $100 million of them.
Another bet is China. They have standardized in a very big way on the On2 VP6 CODECs for web video. There's a flurry of activity around VP7 for mobile video now, too. This might also mesh with On2's not-so-long-ago purchase (for $58 million) of Hantro, a mobile video company. Google is in the mobile device market now in a big way. VP7 is supposedly lower complexity, thus, longer battery life on playback... assuming the video hardware on your typical Android device offers as much boost for VP7 as it does for H264. That was a big issue (in theory, at least) over the whole Theora thing: today's video acceleration hardware is tuned for H.264, MPEG, and occasionally VC-1.
About VP8: http://www.dspdesignline.com/showArticle.jhtml?articleID=214303691
H.264 had already won. It wins because it does the job the best. Yeah, it's a commercial thing, and that IS bad for the web in ways that's not really much of a big deal for other things. Similarly, delivering the best compression per bit is critical for video delivery over most other media, but particularly ... well, the places it's already used. Every major still or video camera maker uses H.264. Every Blu-Ray member, satellite companies like Echostar and DirecTV. Every major PMP maker.
The simple fact is that H.264 is to the early 21rst century was MPEG-2 was to the 1990s... THE video standard format.
The correct way to manage this the way the video players (DVD, Blu-Ray) have done, but that still sucks a little if you're Mozilla. Basically, every video format has both mandatory and optional formats it must decode. This makes this a bit harder on the player producers, but much easier on the content delivery folks.... the only have to chose one of the above. So if I make a Blu-Ray (I do, this isn't just hypothetical), I have the choice of H.264/AVC, VC-1 (WMV9), or MPEG-2 as the video format, and half a dozen audio formats, including a few flavors of AC-3 and plain old uncompressed.
So the right consumer solution here would have been to demand that support H.264 and Ogg Theora as mandatory formats. Toss in a few optional formats if you like... they might not be popular, but you never know (MPEG Layer 2 audio was an optional format for Region 1 DVD, yet nearly every players supports it).
To support open source, make the H.264 piece into a closed source, pay-for plug-in. This is how a number of companies (Nero, Archos, etc) have dealt with the extra cost of supporting AVC, at least at some time in the past. It might have a bad taste without those Fundamentalists among us who want to refuse even the possibility of closed source or proprietary formats getting into their faces, but for those who just want a practical engineering solution, this works. None of the Theora backers were EVER going to get the H.264 proponents to back down and embrace Theora as the only HTML5 video format. Look at the list... they're all the guys with the money, and most of them have vast investments already in H.264 content. Taking them on the way this was done, it was a guaranteed fail.
But of course, this is the computer industry, where everyone has to fight over their one preferred solution, rather than take the "what's best for the most people" approach, which is really what the CE type solution comes down to. And of course, the CE industry NEVER has such format wars [ducking....]
Apple users embrace the "free-thinking" mantra because that's the image Apple's served up. In short, they were told that using a Mac makes them free-thinking. And no, I'm certain the irony is not lost of those of us who abhor Apple's general policies, which are nothing of the kind.
Apple found themselves, entering the early 1990s, as the lone major computer platform other than Windows, and they had arguably better graphics and a few pretty good music applications, which were struggling to actually function on the PC/Windows until well into the Windows 95 era (UNIX-like OSs didn't do audio well at all... you needed a DSP subsystem, as on the SGIs and the NeXT machines, to do audio at all in the very non-realtimey, who-cares-about-interrupt-latency versions of UNIX/Linux at the time).
So they used this as a sales pitch. The PC equals Windows, it's from IBM, and it's used in business... thus, its only uses are business-related. They weren't selling Macs to computer experts who knew this to be false, and certainly not those of us who actually did the PC work as well, then better than the Mac on media content creation of all sorts. They're selling to users who are fairly clueless about PCs.
Apple always had very good marketing, and both that, and their message, continue today. They were selling a slightly more capable 8-bit machine, back in the early 80s, versus Commodore and Atari machines at 1/5th the price (they had slots... that's the "more capable" part). The Mac came in, with hardware so oversimplified it was actually kind of creepy (the "Ready" pin on the SCSI controller drover /DTACK on the 68000, for any bitheads in the crowd) and cheap, but got huge margins. Today, a Mac is exactly a PC in a fancy case without a battery door... there's nothing different about an Apple PC, and yet they still get 2x-3x the cash. That pays for a ton of brainwashing.
And it's also something like human nature. As some may know, I was a senior hardware designer at Commodore on several high-end models of the Amiga computer. There was a time when the Amiga was the best (only) personal computer for color graphics or video work. Like, the mid-to-late 1980s. Today I do my video stuff on a PC running Windows 7 and Sony Vegas, with 8GB of RAM, a Quad-core CPU, and Terabytes of storage. But I still hear from people talking about how the Amiga IS better (not was, but IS).
When you join an exclusive club, you immediately embrace all the positive memes associated with club membership, and you employ these to justify your decision. This isn't restricted to computers, it's found in Video software (Vegas vs. Avid vs. Premiere vs. FCP, etc), cars (Ford vs. Chevy vs. Dodge), still cameras (Canon vs. Nikon), videocameras (Sony vs. Canon vs. Panasonic vs. JVC), soft drinks (Coke vs. Pepsi... sorry, Rock Star rules here, folks), etc. And sure, the cultier that club's culture becomes, the more the users grab hold of it.
Apple is one of the few remaining exclusive clubs in computing, and they're perhaps the cultiest and most exclusive there is in just about any discipline. Ok, Amiga fans could have given them a run for their money back in the early 1990s, but not since... the Mac hardcores have expended to embrace the iPhone. The iPhone has delivered new converts to the alter of Mac. There's a persistent meme that "Windows is hard", bug ridden, full of viruses, and of course, MacOS is impervious to any and all problems, the only way to do media content in computing, and so simple your cat can use it without reading a manual. Apple works very hard to keep these memes alive, in the general population to an extent, too, not just among the Apple Faithful.
Another factor, among those in a successful cult, is that they reinforce one another and don't pay much attention to the outside world. You can stay blissfully within your world of Apple -- magazines, web sites, etc. and never hear more than frightening stories about the world outside. This is also something that Apple cultivates...they were among t
Ok, yeah, $400,000 is a pretty insane sum of money. But you have to feel at least a little schadenfreude at Universal having to pay this. After all, they're part of the music industry machine that has worked tirelessly to erode and obfuscate Fair Use laws. Most of these laws are random exclusions in otherwise restrictive copyright laws. Any time there's a push for a true proactive Fair Use bill... a Fair Use Bill of Rights that really needs to be written ... these guys are spending millions in lobbying money to such that down, pronto. So a little of what goes around finally did come around. I can't be even a little bit disturbed at this outcome... if the law were crystal clear, the bill would have been much, much smaller I'm sure.
Plus, if Universal's equally-ultra-expensive lawyers are at all good, Universal will spend even more money in legal hours trying to delay the payout as long as they possibly can.
I was at a set-top box company, based in Germany, in the late 1990s. We had every intention of building out multimedia set-top box on Linux, even though that wasn't as doable as it is with today's modern Linux kernel (and, well, faster CPUs doesn't hurt, either). The big problem: finding anyone who knew Linux and wanted to actually get paid. We advertised, we attended Linux shows, etc... no go. Might have been practical if we were based in California, but at the time, there was a real issue among many Linux hackers of working for pay. So we wound up using OS/2 on the same hardware. I was largely the hardware boss, but I also wrote some drivers -- tragically, one place Linux was way ahead (physical drivers had to be written in 16-bit code under OS/2, at least at the time).
Nice to see things change. Not that spare time, hobby, educational, etc. contributions are a bad thing at all. But commercial concerns doing all that work is a strong indicator that lots more work is being done. They also sometimes have a better sense of finishing a project. Not always, but sometimes.
Never used an Android phone, have you? Thought not... it does all that. And no need to dock to a PC of any flavor, ever. You can if you want (shows up as a USB storage device).
Nokia certainly had some plans for the GPS market. It was in 2007 that they scrambled to buy one of the two big turn-by-turn mapping companies. First they made an offer to Tele-Atlas, which scared Tom-Tom into buying them instead. Then they bough up Navteq. Oddly enough, this was the shift that got Google scared about their future plans in the maps and navigation business (they did, at one point, buy the mapping data from both companies).
You do have to wonder, though... this could help kill their external Navteq business, but it would take the actual destruction of the stand-alone GPS market. Presumably, they need this in the phones to complete with Google/Android, or at least think they do.
Anyway, there's little to fear about the updates... this is the same data Garmin and many of the others are using, and charging you $50-$100 every year or so in "new maps" charges. There are some advantages, still, to stand-alone units... the 5" screen on my Pioneer AVIC GPS is a bit more readable than the 3.7" on my DROID. The Pioneer also is integrated into my car stereo system, so for GPS advice and phone calls, no need to mess with volume knob. And Google Navigator (beta) has occasionally led me on "adventures", while the Pioneer, not so much. But there's nothing cooler than GPSing with satellite images, and certainly, if I already had the DROID, I probably would have thought twice about buying a stand-alone.
Sounds like monitor problems, or cheap monitors. I have dual 24" 1200p monitors here, usually on HDMI, though I use VGA and YPrPb ports occasionally. It's very obvious, particularly for computer graphics, when I'm on VGA. For video-level stuff, it's much harder to tell.. you usually only see what's left of analog effects, viewed on a digital monitor, at sharp pixel transitions.
VGA was perpetually being replaced... only problem was, it didn't really need to be. So, while there were plenty of alternatives, none of them took hold.
Once monitors went digital, though, it was clearly time to nix the VGA. DVI was designed to be transitional... it's got all the VGA signals and digital signals. It was intended to let you put just the one connector on your digital video card and connector to either analog or digital monitors. Hey, it worked.. most video cards carry one or two DVI connectors these days.
HDMI was something else... this was not a computer industry standard, but one for the consumer electronics industry. They added audio, which is certainly important in CE applications, perhaps something they forgot to put into DVI. Also, the HDMI connector is more consumer friendly than the D-Sub inspired DVI connector. Though they do fall out, which is somewhat annoying. HDMIs are showing up in computer gear simply because the connector size is better for many applications (notebooks, multi-head display cards). It's also a bit of an upgrade from DVI, in terms of data speeds.
DisplayPort is the replacement for DVI.. all digital, with all of the goodies that computer industry folks want to find in their next generation video-specific port.
Next might be Light Peak, which will do video speeds, but will be a general purpose interconnect, not video specific. This will be interesting, since that might bring about smarter displays... they could negotiate not just on resolutions but encodings and other fun stuff (well, DVI and HDMI do a bit of this, negotiating YUV or YCrCb vs RGB.
Actually, the plastic TOSlink cables are generally crap (the standard calls for plastic, you can't use glass). Much lossier than comparable copper, particularly at the low speeds used for digital audio. Mid-range users, maybe. High-end people don't use these. At least those who know what they're doing (or, occasionally, can't avoid them... if you run out of AES/EBU or S/PDIF ports, TOSlink is fine over short cables).
Expenses. The Sony/Philips optical is about the only consumer optical in common use, but that's over plastic, driven by LEDs. Pretty cheap. For optical at these rates, you'd need real lasers (LEDs peak around 500Mb/s) something like 10GBASE-R or 10GBASE-SR cable (LOMMF/OM3). None of that's crazy expensive... unless you compare it to electrical. And in particular, the electrical that the equipment makers are actually paying for.
Keep in mind, these are the industry guys who got together to create DisplayPort, at least in part because they got bent out of shape having to pay US$0.04 per device to use HDMI. They're not likely to replace a $0.50 electrical connector with a $2.00 optical connector and $5.00+ laser. And of course, lasers go one way... you actually need a laser at each end, if you want 2-way traffic. Or a custom cable, with electrical backchannel.
Well, why not.. I have some video cables around here with integrated optical audio channel.
So this is the next one up, after DisplayPort, but designed as a general purpose standard: Light Peak. I think this started out as an optical answer to Firewire at Apple, but rather than do it themselves, change too much for the spec, and have Intel (and the rest of the PC industry) go and create an alternative, this time Apple brought it to Intel. Maybe.. at least that's one story.
Anyway, read more here: http://en.wikipedia.org/wiki/Light_Peak
The nice thing about Light Peak... it's fast enough to do the HDMI/DisplayPort thing. And replace SATA, USB, Firewire, anything else you want. Of course, like all optical interconnects, the connectors are an issue (dirt kills), and unless they go to some kind of FDM, they'll need one cable in each direction, just like 10Ge uses in its various optical forms. Then there's the issue of power... we're kind of used to USB and Firewire cables providing power for small devices. But it's still a work in progress, 10Gb/s on launch, up to 100Gb/s on the roadmap.
Maybe or maybe not.
DisplayPort itself uses a completely different kind of signalling than HDMI. An HDMI signal is basically just a digital version of the video stream... it runs three differential TMDS data links and one clock link at 10x the rate of the display's pixel clock, very tightly coupled to the video. Native DisplayPort sends packetized data, and the signal is over 1, 2, or 4 differential serial lines, with clocking information embedded in that signal.
HDMI 1.3 supports 10.2Gb/s per link (there's a dual link version, but it's not common), and resolutions up to 2560×1600p75 for 24-bits per pixel or 1920x1200p60 for 48-bits per pixel. HDMI 1.4 adds a 100Mb/s ethernet channel and support for displays up to 4096x2160p24 at 24 bits per pixel (same bandwidth, more formats).
DisplayPort 1.1 supports a throughput of 8.64 Gb/s across the four links, which goes to 17.2 Gb/s in version 1.2 (the 21.6Gb/s is the symbol rate over 8b/10b encoded links).
In short, they're completely different. However, the spec does allow a DisplayPort connector to switch and spit out HDMI/DVI compatible signals instead. There is no requirement that DisplayPort connectors do this, however.
So any in-cable passive DisplayPort to HDMI cable is counting on this HDMI bridge feature being in the device you're using. An "active" cable will convert the video signal (well, a simple one), but lose the 2-way data features.
Oh man, now you're in Dolby 3D... I was using RealD glasses!