There's nothing these kids are likely to do on a low-end Macbook that wouldn't work just as well on a Netbook... but even with Windows, you can buy a Netbook for as low as $200. My kids' High School (Woodstown, NJ) has been providing Netbooks to kids for years now. We have to pay insurance, that's it.. but even if we had to buy them, that's far more affordable, and just as functional.
Were my kids in this situation, I'd start a Hackintosh club in town.
Actually, no, at the typical bitrates used, MJPEG quality is much lower. The old-fashioned DV camcorders recorded at 25Mb/s, in 720x480 (NTSC) standard definition... the DV format is just a video optimized and standardized version of MJPEG. Compare this to the ~8Mb/s used for video on DVDs -- essentially the same quality from MPEG-2, less than 1/3 the bitrate.
The K-x records 720/24p in MJPEG... they would really need 65-75Mb/s for the same quality video you get with a camcorder in SD. They're using SDHC cards, Class 6 or better...they don't specify their video bitrate... the K-7 uses the same encoder, and offers modes up to 45Mb/s or more, but they also enable a higher resolution on the K-7, their odd 1,536 x 1,024/30p mode. Still, less than half of what it ought to be. What you get with this camera will be substantially lower quality than a DSLR (like the Canon EOS 5D Mk II or EOS T2i) that offer AVC at similar bitrates. You'll get more general artifacting than with an AVC camcorder.
The one exception might be very fast motion... interframe compression hits a weak spot there, of course. But then again, if you're shooting sports or other fast moving subjects, you want 720/60p or 1080/60p... 24p is weak for high motion video. And it's compounded by the fact the K-x locks the shutter speed at 1/24th in video mode, which is just plain nuts. And of course, you still have the typical rolling shutter distortion (the "jello effect") of any CMOS sensor camera, and in particular, the K-x has a pretty pronounced version of this effect, even for a DSLR (which are generally worse than dedicated camcorders, presumably due to the larger sensors).
And as well, they're doing 24p here. Real 24p. The video standard is actually 23.976fps. Not that adjusting it is hard, but WTF? There's absolutely no good reason not to record video as real video.
Memory cards are getting cheaper, but the current 32GB is the most you'll get for SDHC... and there's no support for SDXC in this model, for the 64GB cards (out now, but pricey). As well, Class 6 cards at 32GB are still pretty pricey... class 4 is around $80 nowadays. Not a shooting problem, because there's an upper limit of something like 24minutes to any shoot... like most DSLRs, it stops when it hits the 4GB boundary of a FAT32 file, rather than starting another file like a camcorder would.
Pentax uses MJPEG on the K-x because it's a beginner's ("entry level") DSLR, and they don't want to pay for either the licensing or additional video DSP to encode H.264. The computational requirements for AVC are on the order of 100x more than for MJPEG... and obviously, any digital still camera already had the JPEG engine.
There's actually nothing for you to sign. The disclaimer in the manual of an AVC camcorder is a description of the license that the camcorder manufacturer has signed with the MPEG-LA... this is NOT the same idea as a software shrinkwrap license, despite what some people have mis-understood.
In my case, Panasonic has paid royalties on a license with the MPEG-LA. This license allows them to record AVC, and it allows them to play it back, from the camera, but for personal use only. Beyond this, Panasonic is not involved anymore. If you do anything with the AVC files on your PC, that's yet another license. And the main point of the disclaimer is to let you know that, while you may have AVC files in hand, you're not licensed to use them for publication -- you need another license for that, just as if you had shot in any other format and published in AVC.
No, that's entirely not true. Your camcorder includes the license to encode video, and it stops there. What the licensing agreement in your camcorder manual is trying to say is that, simply because the camcorder records in AVC, you are not absolved of additional AVC licenses needed for public distribution. But if you distribute in another format, the licenses obviously do not apply. This license is not written very clearly, but that's what it means.
Well, there are two essential evils related to the MPEG-LA, and one is actually real.
First, the second one... the idea that the MPEG-LA claims license to your video. That's bunk, but it's easy to misunderstand given the license agreement you find along with any AVC-based camcorder. The license seems to claim that video shot with that camcorder is only legal for personal use, even for professional models. What they're really saying is that the AVC license included with that camcorder is only valid for personal use. In short, you can encode AVC, and you can play it back on the camcorder for personal use only. If you move the encoded file onto a PC, obviously the camcorder license no longer applies, and you'll presumably be under the license paid for that PC, your video editing application, etc. For public distribution in AVC, you need yet another license... you don't automatically get that by using the AVC camcorder, anymore than you'd get a free AVC distribution license if you shot in DV or MPEG-2 or some other format.
The other MPEG-LA factor is fairly pure evil -- the commercial nature of the MPEG-LA pretty much ensures that any new MPEG specification will tend toward the most patents possible to work into any new specification... even if existing, unpatented technology will do the job just as well. The nature of the licensing group and the current MPEG process has each member jockeying to get as much of their patented IP into any new spec as possible.... it's pretty close to the opposite of the way a proper industry standards body (W3C, for example) ought to work.
That's not really what the MPEG-LA license says. Having several H.264 camcorders, including a Panasonic Professional AVCCAM model, I looked at this in detail. What they're trying to say in the license is that, simply because you're capturing the video in AVC, you do not automatically have a license to use that AVC for any kind of commercial rebroadcast. They're really not trying to claim any ownership of my video (at least not yet).
So I can shoot in AVC, release in some other format (let's say, MPEG-1, to be certain we're clear of any lingering patent entanglements), and there's no need for licenses of any kind for any use of that end-product, commercial or otherwise. If, rather, I release the video on Blu-Ray, then I need the same licensing I would had I shot that video in any other format, even if I don't re-encode the AVC to make the Blu-Ray (as unlikely as that is). That's the point they're trying to make... Panasonic only licensed AVC for their use in the camcorder, which includes recording and only personal playback (eg, me looking at the camera screen, previewing on a monitor, or playback in-house for family videos), not for any possible commercial use in AVC format, without an additional license for that use.
I'll agree, they could have stated this more clearly.
The C128 was joined by the C64E, in the new casework, when the C128 debuted. At that point, it cost something like $35 to build and package a C64, and they were still selling well. In fact, the C128 introduction boosted C64 sales pretty dramatically. The C128 didn't sell C64 numbers, but it did sell about 5 million units. I don't think that's exactly "never sold well".
No, the C64 was never produced by licensees.... the chips were only available from Commodore, and once Commodore went into bankruptcy, all that stuff was locked up until ESCOM bought the Commodore assets.
There is at least one C64 clone. Jeri Ellsworth re-designed the C64 from scratch, yielding a single-chip C64 that was used for a video game stick in 2004, and perhaps elsewhere. And of course, C64 emulators abounded... there was even a fairly slow one for the first Amigas.
Kernel != OS. And as well, CMU only started Mach in 1985... Amiga was shipping in 1985. Did you maybe mean the Accent kernel? I thought so. But no, the Accent kernel was not a better OS than AmigaOS.
Yes, the TRIPOS stuff was unfortunate, but really just an annoyance for developers. It didn't really limit anything the AmigaOS could do. VM was a good idea, sure... but not for the kind of small computers Commodore was doing in 1985. And turning AmigaOS into UNIX would have ruined it... UNIX needed another 15 years before it could handle realtime multimedia work as well as AmigaOS in 1985. If you follow any of the UNIX machines doing something related to Multimedia back in the 80s and 90s, like SGI or even NeXT, they were offloading the realtime work to accessory hardware. Like DSPs for audio. Nothing wrong with that... but when you want to build a $500 or $1000 computer in the 80s, then yeah, something wrong in a big way with that.
And as far as 68020 + 68851, yeah, that became a solution. It wasn't a solution in 1985, the 68851 didn't work yet. Everyone doing UNIX workstations on 68K were building their own MMUs. Not to mention the fact that the 68020 was a $300 chip... that doesn't fit at the Amiga price point. Apple got $5,000 for their first '020 machines.
Having worked both at Amiga and at ESCOM's Amiga Technologies spinoff, I do think that was the last real chance the Amiga had. They actually took their time to study the problem, enlisted me and Andy Finkel to run hardware and software development groups, respectively, and had the right idea about how many people and how much time this was going to take, the right place in the market for new Amigas, etc. No guarantees... as I said, it was a chance.
I don't think any effort to resurrect the Amiga beyond that had a real chance. No one had the right expertise, the right target price point, or could even answer simple things like "how do we sell this new thing to people who've never heard of Amiga Computers".
Yes, there really were Commodore C65 prototypes. Numbers vary, but they did have units for developers... can't swear they actually had any developers. After the Commodore bankruptcy in 1994, these were bought by lot and sold openly. So yeah, some people really have these. No idea what they do with them... unlike the C128, the C65 has very low compatibility with the C64.
The C128 was the successor to the C64. The CBM-II series had nothing to do with the C64, they were just updated CBM systems. The C65 was primarily done just to keep Bill Gardei from bothering any of the people working on Amiga systems. The C128 made some sense in 1985 -- we sold somewhere around 5 million C128s. The C65 made absolutely no sense in 1993. But sure, it's an interesting collector's item.
The WDC 65816 was technically an 8/16-bit version of the 6502, but it was a pretty ugly hack. This wasn't a 6502 with 16-bit instructions and register modes added, as you might expect. No, this ran 6502 instructions in either 8-bit or 16-bit mode, depending on a bit in the status register. Evil. And the IIGS shipped over a year after the C128, after the Amiga shipped. It was a poor attempt to take on the Amiga, though you can given the designers some credit for building a multimedia computer from the old 8-bit stuff faster than the Mac people did.
The C128's MMU actually made things much easier to deal with, in the specific case of 128KB of memory, than either 65816 or 8088s and their segmentation registers (the 65816 had a similar mechanism, though yeah, it did have a very few "long" instructions). This wasn't something that would have been as useful going beyond 256K, but it was very easy to deal with memory management on the C128. It was nothing like the crazy stuff they did in the CBM line.
It was pretty obvious the C128 was the end of the line for 8-bit at Commodore, so there wasn't a great deal of looking forward in the C128... also not the budget we really wanted for it. But it was a very successful product.
As for the Amiga, yeah, the Amiga was clearly the way to go. Without the Amiga, Commodore would have gone forward with the Commodore 900, which was based on the Z8000 processor family. We had a megapixel monochrome display for it, a novel stacking expansion bus (similar in concept to PC/104), and it ran Coherent, a UNIX clone. This was cancelled after Commodore bought Amiga. Some of the ideas were used, er, stolen for the Atari ST.. Jack brought some of the early C900 people with him. They never got it working... that was George Robbins and Bob Welland, the third engineering team to work on the C900. They also went on to develop the Amiga 500. I was working on the C128 at the time, and went on to develop the Amiga 2000 and other Amiga systems.
Oh... but did you get the original C128D, with plastic case, clip-in keyboard, and carrying handle, or the revamped metal-case C128. I had the former... of course, I was on the C128 project when I worked at Commodore. The plastic version was just too expensive for production.
You're nuts! I ran full GNU Emacs on Amigas with AmigaOS and Amiga UNIX with only 4MB of memory.... ok, a bit of disc thrashing in the latter case. But at least this gets me into at least two cults in one shot.
You would HOPE that H.264 is better than VP8 today, particularly at things like motion estimation. There's been a crazy amount of work on H.264, getting it acceptable and all. Go shoot some video with an H.264 camcorder from 2-3 years ago, and the video will be pretty horrible. Buy a 2010 model, and it's very good. Same with the encoders. MPEG-2 went through the same kind of evolution.
While it remains to be seen if VP8 does anything better than H.264, it wins if it proves to be un-patent-encumbered (still a big question) and pretty much as good as H.264. "Better" is just gravy. It's also interesting that, in that now famous x264 guy analysis, very few of the supposed advantages of VP8 were even acknowledged. One has to wonder just how much homework he did, and how much was politics.
Actually, 1080p is still a hit or miss on most camcorders. Among major vendors, Sony and JVC don't support any progressive modes in their high-end consumer models. Canon does 24p and 30p on some models, Panasonic's been doing 24p for awhile and just added 60p.
Of course, if you're using a Flip-style webcam or a cell phone, you're shooting in progressive, but the HD value is questionable at best, even when you get an HD file out of the thing.
Obviously, someone's got to de-interlace for web delivery otherwise, but don't assume that progressive video is a common thing among consumers quite yet.
The original claim for VP8 was that, for low bitrate video (nothing fully defined, but obviously intended to mean "web video"), it would deliver 40%-60% better coding efficiency than H.264. That doesn't really suggest they mean low resolution video, only highly compressed video.
And really, none of the "this is only for low resolution video" claims have ever been correct. The MPEG started their MPEG-3 project with intent of delivering a new video CODEC for high-definition, only to be shown that MPEG-2 was just dandy for this. H.264 was created for telecommunications and video on cell-phone sized screens, now it's the runaway favorite for HD content.
Re:Babylon 5 / Firefly / Star Blazers
on
Lost Ends
·
· Score: 1
Yeah, sure, the whole story, at least in a detailed outline, makes a better series. But Babylon 5 also demonstrates why that's a difficult thing, on many levels. First of all, you don't get a guy like JMS to write and direct that much of the series. Secondly, the series itself is inherently at the mercy of the network, which can cut it short or demand it run longer. Things are changing slowly, but the idea of an N-year run series that's guaranteed those N-years to run... that's about as likely as, I dunno... polar bears on tropical islands.
I don't think I really wanted any more Island Answers anyway. That's not to say I would have felt this way had the whole story been written in full before the filming started, Babylon 5 style.
But I guess I don't trust TV SciFi very far. Writers can write very compelling character stories, and I'm happy they kept Lost that way. I think too much additional explanation, and we would have had Star Trek: NTG style deus ex machina type explanations just masquerading as good science fiction. I'm happier with the lingering mystery than a crappy explanation.
The fear of lurking patents hasn't really been the problem with Ogg Theora. Ok, sure, that's one of many reasons why folks like Microsoft and Apple can claim for not supporting it, but they'd happy find more, if this one were eliminated. The big problem with Theora is that it's taken forever to be delivered in any usable form. There's no need for it off the web, and on the web, it's only useful in a browser that supports it. Which are fairly recent developments.
And in the time between VP3 going open source and Ogg Theora being quasi-usable online, H.264 took over most of the world.
We'll have to see about VP8. I'll admit it sounded fantastic when I read the material on the On2 site, and sounded even better when Google bought On2... it was very much in their SOP for Google to release the code in some way that made the open source folks happy. Now that's done, but the questions are just beginning.
For one, sounds like their "spec" is bad... not a real spec, but full of refrences to a single C-code implementation. That's a horrible way to write a specification. Clearly, this wasn't a problem for On2's original purposes, but it's not the way you build a good industry standard. And did they do any real review of this, if not with the world, at least with the others on the list, like Open, Mozilla, nVidia, AMD, etc? If you take any of the other specs, even Microsoft's VC-1 spec, you can build a compliant encoder or decoder directly from the spec. That's a good thing, particularly if there are aspects of this headed for silicon at some point.
And that's assuming there's enough difference between VP8 and H.264 to need any new hardware. Sounds like yes, in a few places anyway, and of course, only on devices... PCs will be fine as-is. And there's so much H.264 related stuff in there, there's still reason to fear patent lawsuits. Presumably, Google's lawyers have reviewed this against the H.264 patents that might be applicable, but there's nothing really reassuring about that on the WebM site. Given the long list of partners, though, you'd kind of hope this has been done.
I never bough the "OS as a browser" argument.. if there even is an argument. I mean, I can run a web browser on Windows or on Android or Linux, and yeah, that suggests that Web-centric applications will, within reason, be kind of cool. But what possible advantage to I gain in something like ChromeOS? Why not run that same level of web browser within Android, and get the best of Web-only (like Chrome), Web-agumented (like many Android apps), and stand-alone apps as well.
It's clear from many things they do that Google is thinking Web.. and they should, that's where they were born and where they thrive. The great thing about Android versus iPhoneOS or PalmOS is that I don't ever need to cable up to a PC and sync my phone or tablet... any important syncs can just happen, invisibly, via the net. This also means that "sync to a PC" itself is only interesting if I want to also use a PC... these devices, unlike most others, can live entirely without the need for PC support. That's critical in advancing the state of computing on these devices.
But going web-only doesn't seem to be a step forward. If web apps are really that great, folks will use them, on devices, on PCs, whereever. If they're not so great yet, you're not betting the whole device on web-only functionality and perfectly persistent internet connections.
Google didn't do Android to make a vast fortune selling smart phones. Their primary concern was to ensure their primary profit center, internet search, extended to small devices. Given that, prior to Android, every smart phone platform was proprietary and closed source (some, like Symbian, have gone open source since then), it was easy to believe that at least, default search could leave them out. In fact, that's just what happened... RIM made a search deal with Microsoft, and overnight, Blackberries started using Bing! as their default search engine.
Google is actually making money on Android, via search. In fact, depending on just how they package Android, some of the carriers are getting profit sharing from Google based on that income.
Search will absolutely become increasingly important on mobile devices. Things like Google Goggles, the various tools for finding things in unfamiliar areas, other augmented reality toys, they all show pretty clearly that this is going to replace a great deal of the stuff people once did on the desktop. Google was smart enough to not only see this coming, but release Android in such a way that ensured it would eventually dominate the small device market.
Also, I don't think it's a matter of Apple allowing other carries in the US. It was originally a matter of AT&T not allowing them... Apple had to give AT&T exclusivity in order to get them to carry the iPhone. By all accounts that's over, but the real question is, can anyone else deal with Apple. Verizon and Apple have apparently had serious talks at least three times now, and so far, no iPhone on Verizon. The story as I hear it is that Verizon has been interested, at least in the past, and were willing to deal with Apple much as they dealt with every other vendor. Apple, on the other hand, wanted the same kind of special treatment, like profit sharing, they got from AT&T.
Going forward, though, I think it's more of a longshot. The fact is, Verizon is doing very nicely with Android now. But it's not just that... they can only offer so many phones that do basically the same thing before it's just not worth their carrying the phones. As they have in the past, Verizon likes to brand the models they carrry. It's not that big of a deal to most people.. the Verizon Droid is essentially just a CDMA2000 Motorola Milestone. But when you say "Droid", it's automatically Verizon. All their other Android phones are marketed under the "Droid" name.
Look at the Google Nexus One deal... supposedly, Verizon was set to carry the Nexus One. Only problem... that was going in with Google primary branding, not the secondary branding you get with the Droid (being a "Google Experience" phone, the Droid is just as untouched by Verizon as the N1 would have been, but the Google logo's on the back, it says "Motorola" and "Verizon" on the front). Put an iPhone into Verizon, and it's still an iPhone, still Apple branded, with maybe a tiny "Verizon" on the back. I'm not sure they want that.
And if you can't have Verizon, that's at least currently 30% of the smart phones sold in the USA, and growing much faster than AT&T. So Apple would have to tag both T-Mobile and Sprint for the same coverage. And really, are they interested? Would Apple do a special CDMA2000 + WiMax phone for Sprint... that's probably what they'll be requiring for their high-end phones, like the EVO 4G, from now on. Given that AT&T and Verizon are both going LTE at 700MHz, this is probably a non-starter for Apple. So that leaves T-Mobile. Well, they're already GSM/HSPA, just on different 3G channels. But they're also the smallest carrier.
It's a cool toy. That's it. Some article last week claimed that demographically, the iPad was bought largely by older, rich men. In short... yeah, it's "oooh... shiny...." all over again. Just as when the iPhone first came out... it wasn't even a PDA at that point, just a cellphone with a PMP and Web browser built-in.
Having viewed the dozen-or-so keyboardless tablets shown off at CES this year, as well as the iPad two weeks later, I have an idea of why I'd like something like this, and particularly why it's not the iPad. But I'll need a valid reason.
Smartphones are a valid reason to buy such a device. Before my Droid, I had a Palm Treo. The big advantage of the smartphone is that it's really not a phone, it's an application processor, a PDA, something that replaces a dozen small electronic devices. My Droid is a phone, sure, and it's just dandy as a phone. It's actually more than that, since it supports pretty much every form of modern communication: Skype, Twitter, IM, SMS, Facebook, interactive blogs, etc. About all that's missing is a front-facing camera for videoconferencing... and those are coming in the newer phones. Next time an important new form of communications comes along, there's just a new app to download.
It also replaces, and sometimes extends, the slew of small devices I'm likely to carry. It does the MP3 player/PMP thing, but does it better than any MP3 player or PMP I've had. First of all, a better-than-SD screen just rocks for video, and network access to YouTube and other sources make it more useful for video than, say, a plain old iPod. Same with music... the MP3 player function is dandy, but there fact of apps means you have better alternatives. On the Droid, I use Museek for listening to locally stored music... it adapts to my current mood. Or Pandora and other "internet radio" options... I recently drove from South Jersey to Boston and back, with Pandora running the whole way, perfectly glitch-free.
Next is GPS.. this is a better GPS than many units I've owned, even if you never use the live satellite overlays. There may be issues if you lose the network, but that generally doesn't happen. It's also a functional pocket camera and camcorder. No, it's no replacement for my Canon EOS or Panasonic HMC40, but since I have it anyway, it's ok. Video on this device is actually better than on many consumery SD camcorders... sure, the audio is just as bad if not worse, but again, it's one more device function.
I get more, via apps. I used to carry around one of those Sony mini-tape recorders, for writing ideas. I used voice notes on the Treo, these days I have both voice and text notes, and automatic, invisible sync to my PC via Evernote.. again, not just the old function, but an improved version of that old function. It's also a fully functional flashlight, a guitar tuner, a games machines, a ssh client, etc.
Then, of course, email and web... the web browser is good enough that, for a simple "hey, I want to check that out" kind of use, I don't bother with a PC anymore... the phone's always here. I have email set up though my server to go to the phone and the PC, and of course, work email via the crufty Microsoft stuff works as well as anywhere, I guess.
So... on to the tablet. I need a number of larger format devices replaced by the tablet, or I'm not interested. What could these devices be? Well, some of that's maybe basic laptop stuff.... if I'd take this tablet where I wouldn't take the laptop, but still get some needed functionality, I might find that useful. Doing CAD and Video, I'm always going to have a desktop, so for me, a functional enough tablet might replace the laptop entirely.
First thing I might drag along -- portable video player. I have a portable DVD player, I might like a portable Blu-Ray player to replace it, but given enough function in a tablet, maybe that'll do the job. The big problem there is storage... most of these don't have a single Blu-Ray's worth of storage. But if you can use SDHC/SDXC cards or USB sticks/hard drives, op
Also, Apple will get a large boost in sales due to iPhone upgrades. Quite a substantial number of upgrades are on the same cycle as the phone releases. So yeah, right now, the numbers are down, artificially low, as everyone knows the new iPhone announcement is coming in June or July. Once it's out, there's an almost guaranteed spike in iPhone sales, as all of those with expired 2G contracts rush out to buy the new phone. At least, that's expected... if it really sucks, maybe there's less of that, but I'd still bet on Android carrying 2Q2010, with iPhone probably back in the #2 or even #1 slot for 2H2010.
And that's just dandy, and not an Android problem. I think the fact Android hit #2 this early, and in the USA rather than globally, was kind of a surprise, if not a downright anomaly. Apple's situation helped, but it's definitely a sign that Android is growing, and growing fast. My personal opinion is that, unless Google manages to do something stupid on an epic scale, the dominance of Android over iPhoneOS is only a matter of when, not it. But with folks like China Mobile embracing Android, while Apple's still weak in much of the world outside the USA, I really expected this to be a global win before it hit stateside. And I still think it's going to flip-flop from quarter to quarter for awhile. Android will be more consistent, simply because there's a "hot new thing" introduced in the Android ecosystem every few months. But this will also allow Apple's sales spikes to have it keep winning sales from time to time, even after Android dominance is assured.
Given Apple's various policies, it's impossible they're not fully aware that they can't hold their current #2 position in the US forever. They probably don't care... selling high priced PCs hasn't been at all bad to Apple's bottom line. They're not going to change that policy anytime soon in the devices market... iPhones will continue to compete with the high-end of the Android and other smart phone markets, while the Android world will encompass an increasingly wide range of devices, getting to low price points Apple will never even bother with.
And in fact, Apple can't bother with those lower-end markets. It's the same reason Mercedes and BMW only get so cheap... they have to protect their hard won position as high-end models. Apple's selling much the same thing in the PC, PMP, Smartphone, and Tablet market as everyone else, but at twice the price, give or take. They're not about to give that up for increased market share... why would they? From a techie point of view, lots of us are bothered by this... I've been a little bothered by Apple since the 80s, as I was designing computers with the same parts for Commodore, delivering more performance, and Apple was still considered "high-end", even with often inferior hardware designs out of Apple.
But from a sales and marketing viewpoint, you have to love how successful they've been at maintaining this market perception... and for decades. This is the reason Apple's flush with cash, while many PC companies and cellphone companies alike have been struggling.
Not only that, but while AT&T has good 3G coverage in cities, only about 1/5 of their network, by area, is 3G. So when you get rural coverage, it's much less likely to be fast. By it's nature, ever Verizon cell is 3G... you only drop back to EDGE speeds when you're too far from a cell to support 3G. As with all radio, this can vary by conditions... I get 3G on my Droid in my house about 80% of the time; it's occasionally dropping back to 2G speeds, but that's pretty rare. I'm in a rural area (eg, internet and TV only by satellite...).
Slashdot Mirror
There's nothing these kids are likely to do on a low-end Macbook that wouldn't work just as well on a Netbook... but even with Windows, you can buy a Netbook for as low as $200. My kids' High School (Woodstown, NJ) has been providing Netbooks to kids for years now. We have to pay insurance, that's it.. but even if we had to buy them, that's far more affordable, and just as functional.
Were my kids in this situation, I'd start a Hackintosh club in town.
Actually, no, at the typical bitrates used, MJPEG quality is much lower. The old-fashioned DV camcorders recorded at 25Mb/s, in 720x480 (NTSC) standard definition... the DV format is just a video optimized and standardized version of MJPEG. Compare this to the ~8Mb/s used for video on DVDs -- essentially the same quality from MPEG-2, less than 1/3 the bitrate.
The K-x records 720/24p in MJPEG... they would really need 65-75Mb/s for the same quality video you get with a camcorder in SD. They're using SDHC cards, Class 6 or better...they don't specify their video bitrate... the K-7 uses the same encoder, and offers modes up to 45Mb/s or more, but they also enable a higher resolution on the K-7, their odd 1,536 x 1,024/30p mode. Still, less than half of what it ought to be. What you get with this camera will be substantially lower quality than a DSLR (like the Canon EOS 5D Mk II or EOS T2i) that offer AVC at similar bitrates. You'll get more general artifacting than with an AVC camcorder.
The one exception might be very fast motion... interframe compression hits a weak spot there, of course. But then again, if you're shooting sports or other fast moving subjects, you want 720/60p or 1080/60p... 24p is weak for high motion video. And it's compounded by the fact the K-x locks the shutter speed at 1/24th in video mode, which is just plain nuts. And of course, you still have the typical rolling shutter distortion (the "jello effect") of any CMOS sensor camera, and in particular, the K-x has a pretty pronounced version of this effect, even for a DSLR (which are generally worse than dedicated camcorders, presumably due to the larger sensors).
And as well, they're doing 24p here. Real 24p. The video standard is actually 23.976fps. Not that adjusting it is hard, but WTF? There's absolutely no good reason not to record video as real video.
Memory cards are getting cheaper, but the current 32GB is the most you'll get for SDHC... and there's no support for SDXC in this model, for the 64GB cards (out now, but pricey). As well, Class 6 cards at 32GB are still pretty pricey... class 4 is around $80 nowadays. Not a shooting problem, because there's an upper limit of something like 24minutes to any shoot... like most DSLRs, it stops when it hits the 4GB boundary of a FAT32 file, rather than starting another file like a camcorder would.
Pentax uses MJPEG on the K-x because it's a beginner's ("entry level") DSLR, and they don't want to pay for either the licensing or additional video DSP to encode H.264. The computational requirements for AVC are on the order of 100x more than for MJPEG... and obviously, any digital still camera already had the JPEG engine.
Anyway... I'd do serious homework before selecting this camera (or any DSLR) if video is a significant target for its use. There's a long video-oriented thread about the D-x here: http://dvxuser.com/V6/showthread.php?s=a2fcac7736ec18bb3b1b69883720a0d8&t=183691
There's actually nothing for you to sign. The disclaimer in the manual of an AVC camcorder is a description of the license that the camcorder manufacturer has signed with the MPEG-LA... this is NOT the same idea as a software shrinkwrap license, despite what some people have mis-understood.
In my case, Panasonic has paid royalties on a license with the MPEG-LA. This license allows them to record AVC, and it allows them to play it back, from the camera, but for personal use only. Beyond this, Panasonic is not involved anymore. If you do anything with the AVC files on your PC, that's yet another license. And the main point of the disclaimer is to let you know that, while you may have AVC files in hand, you're not licensed to use them for publication -- you need another license for that, just as if you had shot in any other format and published in AVC.
No, that's entirely not true. Your camcorder includes the license to encode video, and it stops there. What the licensing agreement in your camcorder manual is trying to say is that, simply because the camcorder records in AVC, you are not absolved of additional AVC licenses needed for public distribution. But if you distribute in another format, the licenses obviously do not apply. This license is not written very clearly, but that's what it means.
Well, there are two essential evils related to the MPEG-LA, and one is actually real.
First, the second one... the idea that the MPEG-LA claims license to your video. That's bunk, but it's easy to misunderstand given the license agreement you find along with any AVC-based camcorder. The license seems to claim that video shot with that camcorder is only legal for personal use, even for professional models. What they're really saying is that the AVC license included with that camcorder is only valid for personal use. In short, you can encode AVC, and you can play it back on the camcorder for personal use only. If you move the encoded file onto a PC, obviously the camcorder license no longer applies, and you'll presumably be under the license paid for that PC, your video editing application, etc. For public distribution in AVC, you need yet another license... you don't automatically get that by using the AVC camcorder, anymore than you'd get a free AVC distribution license if you shot in DV or MPEG-2 or some other format.
The other MPEG-LA factor is fairly pure evil -- the commercial nature of the MPEG-LA pretty much ensures that any new MPEG specification will tend toward the most patents possible to work into any new specification... even if existing, unpatented technology will do the job just as well. The nature of the licensing group and the current MPEG process has each member jockeying to get as much of their patented IP into any new spec as possible.... it's pretty close to the opposite of the way a proper industry standards body (W3C, for example) ought to work.
That's not really what the MPEG-LA license says. Having several H.264 camcorders, including a Panasonic Professional AVCCAM model, I looked at this in detail. What they're trying to say in the license is that, simply because you're capturing the video in AVC, you do not automatically have a license to use that AVC for any kind of commercial rebroadcast. They're really not trying to claim any ownership of my video (at least not yet).
So I can shoot in AVC, release in some other format (let's say, MPEG-1, to be certain we're clear of any lingering patent entanglements), and there's no need for licenses of any kind for any use of that end-product, commercial or otherwise. If, rather, I release the video on Blu-Ray, then I need the same licensing I would had I shot that video in any other format, even if I don't re-encode the AVC to make the Blu-Ray (as unlikely as that is). That's the point they're trying to make... Panasonic only licensed AVC for their use in the camcorder, which includes recording and only personal playback (eg, me looking at the camera screen, previewing on a monitor, or playback in-house for family videos), not for any possible commercial use in AVC format, without an additional license for that use.
I'll agree, they could have stated this more clearly.
The C128 was joined by the C64E, in the new casework, when the C128 debuted. At that point, it cost something like $35 to build and package a C64, and they were still selling well. In fact, the C128 introduction boosted C64 sales pretty dramatically. The C128 didn't sell C64 numbers, but it did sell about 5 million units. I don't think that's exactly "never sold well".
No, the C64 was never produced by licensees.... the chips were only available from Commodore, and once Commodore went into bankruptcy, all that stuff was locked up until ESCOM bought the Commodore assets.
There is at least one C64 clone. Jeri Ellsworth re-designed the C64 from scratch, yielding a single-chip C64 that was used for a video game stick in 2004, and perhaps elsewhere. And of course, C64 emulators abounded... there was even a fairly slow one for the first Amigas.
Kernel != OS. And as well, CMU only started Mach in 1985... Amiga was shipping in 1985. Did you maybe mean the Accent kernel? I thought so. But no, the Accent kernel was not a better OS than AmigaOS.
Yes, the TRIPOS stuff was unfortunate, but really just an annoyance for developers. It didn't really limit anything the AmigaOS could do. VM was a good idea, sure... but not for the kind of small computers Commodore was doing in 1985. And turning AmigaOS into UNIX would have ruined it... UNIX needed another 15 years before it could handle realtime multimedia work as well as AmigaOS in 1985. If you follow any of the UNIX machines doing something related to Multimedia back in the 80s and 90s, like SGI or even NeXT, they were offloading the realtime work to accessory hardware. Like DSPs for audio. Nothing wrong with that... but when you want to build a $500 or $1000 computer in the 80s, then yeah, something wrong in a big way with that.
And as far as 68020 + 68851, yeah, that became a solution. It wasn't a solution in 1985, the 68851 didn't work yet. Everyone doing UNIX workstations on 68K were building their own MMUs. Not to mention the fact that the 68020 was a $300 chip... that doesn't fit at the Amiga price point. Apple got $5,000 for their first '020 machines.
Having worked both at Amiga and at ESCOM's Amiga Technologies spinoff, I do think that was the last real chance the Amiga had. They actually took their time to study the problem, enlisted me and Andy Finkel to run hardware and software development groups, respectively, and had the right idea about how many people and how much time this was going to take, the right place in the market for new Amigas, etc. No guarantees... as I said, it was a chance.
I don't think any effort to resurrect the Amiga beyond that had a real chance. No one had the right expertise, the right target price point, or could even answer simple things like "how do we sell this new thing to people who've never heard of Amiga Computers".
Yes, there really were Commodore C65 prototypes. Numbers vary, but they did have units for developers... can't swear they actually had any developers. After the Commodore bankruptcy in 1994, these were bought by lot and sold openly. So yeah, some people really have these. No idea what they do with them... unlike the C128, the C65 has very low compatibility with the C64.
The C128 was the successor to the C64. The CBM-II series had nothing to do with the C64, they were just updated CBM systems. The C65 was primarily done just to keep Bill Gardei from bothering any of the people working on Amiga systems. The C128 made some sense in 1985 -- we sold somewhere around 5 million C128s. The C65 made absolutely no sense in 1993. But sure, it's an interesting collector's item.
The WDC 65816 was technically an 8/16-bit version of the 6502, but it was a pretty ugly hack. This wasn't a 6502 with 16-bit instructions and register modes added, as you might expect. No, this ran 6502 instructions in either 8-bit or 16-bit mode, depending on a bit in the status register. Evil. And the IIGS shipped over a year after the C128, after the Amiga shipped. It was a poor attempt to take on the Amiga, though you can given the designers some credit for building a multimedia computer from the old 8-bit stuff faster than the Mac people did.
The C128's MMU actually made things much easier to deal with, in the specific case of 128KB of memory, than either 65816 or 8088s and their segmentation registers (the 65816 had a similar mechanism, though yeah, it did have a very few "long" instructions). This wasn't something that would have been as useful going beyond 256K, but it was very easy to deal with memory management on the C128. It was nothing like the crazy stuff they did in the CBM line.
It was pretty obvious the C128 was the end of the line for 8-bit at Commodore, so there wasn't a great deal of looking forward in the C128... also not the budget we really wanted for it. But it was a very successful product.
As for the Amiga, yeah, the Amiga was clearly the way to go. Without the Amiga, Commodore would have gone forward with the Commodore 900, which was based on the Z8000 processor family. We had a megapixel monochrome display for it, a novel stacking expansion bus (similar in concept to PC/104), and it ran Coherent, a UNIX clone. This was cancelled after Commodore bought Amiga. Some of the ideas were used, er, stolen for the Atari ST.. Jack brought some of the early C900 people with him. They never got it working... that was George Robbins and Bob Welland, the third engineering team to work on the C900. They also went on to develop the Amiga 500. I was working on the C128 at the time, and went on to develop the Amiga 2000 and other Amiga systems.
Oh... but did you get the original C128D, with plastic case, clip-in keyboard, and carrying handle, or the revamped metal-case C128. I had the former... of course, I was on the C128 project when I worked at Commodore. The plastic version was just too expensive for production.
You're nuts! I ran full GNU Emacs on Amigas with AmigaOS and Amiga UNIX with only 4MB of memory.... ok, a bit of disc thrashing in the latter case. But at least this gets me into at least two cults in one shot.
That's not really it.
You would HOPE that H.264 is better than VP8 today, particularly at things like motion estimation. There's been a crazy amount of work on H.264, getting it acceptable and all. Go shoot some video with an H.264 camcorder from 2-3 years ago, and the video will be pretty horrible. Buy a 2010 model, and it's very good. Same with the encoders. MPEG-2 went through the same kind of evolution.
While it remains to be seen if VP8 does anything better than H.264, it wins if it proves to be un-patent-encumbered (still a big question) and pretty much as good as H.264. "Better" is just gravy. It's also interesting that, in that now famous x264 guy analysis, very few of the supposed advantages of VP8 were even acknowledged. One has to wonder just how much homework he did, and how much was politics.
Actually, 1080p is still a hit or miss on most camcorders. Among major vendors, Sony and JVC don't support any progressive modes in their high-end consumer models. Canon does 24p and 30p on some models, Panasonic's been doing 24p for awhile and just added 60p.
Of course, if you're using a Flip-style webcam or a cell phone, you're shooting in progressive, but the HD value is questionable at best, even when you get an HD file out of the thing.
Obviously, someone's got to de-interlace for web delivery otherwise, but don't assume that progressive video is a common thing among consumers quite yet.
The original claim for VP8 was that, for low bitrate video (nothing fully defined, but obviously intended to mean "web video"), it would deliver 40%-60% better coding efficiency than H.264. That doesn't really suggest they mean low resolution video, only highly compressed video.
And really, none of the "this is only for low resolution video" claims have ever been correct. The MPEG started their MPEG-3 project with intent of delivering a new video CODEC for high-definition, only to be shown that MPEG-2 was just dandy for this. H.264 was created for telecommunications and video on cell-phone sized screens, now it's the runaway favorite for HD content.
Yeah, sure, the whole story, at least in a detailed outline, makes a better series. But Babylon 5 also demonstrates why that's a difficult thing, on many levels. First of all, you don't get a guy like JMS to write and direct that much of the series. Secondly, the series itself is inherently at the mercy of the network, which can cut it short or demand it run longer. Things are changing slowly, but the idea of an N-year run series that's guaranteed those N-years to run... that's about as likely as, I dunno... polar bears on tropical islands.
I don't think I really wanted any more Island Answers anyway. That's not to say I would have felt this way had the whole story been written in full before the filming started, Babylon 5 style.
But I guess I don't trust TV SciFi very far. Writers can write very compelling character stories, and I'm happy they kept Lost that way. I think too much additional explanation, and we would have had Star Trek: NTG style deus ex machina type explanations just masquerading as good science fiction. I'm happier with the lingering mystery than a crappy explanation.
The fear of lurking patents hasn't really been the problem with Ogg Theora. Ok, sure, that's one of many reasons why folks like Microsoft and Apple can claim for not supporting it, but they'd happy find more, if this one were eliminated. The big problem with Theora is that it's taken forever to be delivered in any usable form. There's no need for it off the web, and on the web, it's only useful in a browser that supports it. Which are fairly recent developments.
And in the time between VP3 going open source and Ogg Theora being quasi-usable online, H.264 took over most of the world.
We'll have to see about VP8. I'll admit it sounded fantastic when I read the material on the On2 site, and sounded even better when Google bought On2... it was very much in their SOP for Google to release the code in some way that made the open source folks happy. Now that's done, but the questions are just beginning.
For one, sounds like their "spec" is bad... not a real spec, but full of refrences to a single C-code implementation. That's a horrible way to write a specification. Clearly, this wasn't a problem for On2's original purposes, but it's not the way you build a good industry standard. And did they do any real review of this, if not with the world, at least with the others on the list, like Open, Mozilla, nVidia, AMD, etc? If you take any of the other specs, even Microsoft's VC-1 spec, you can build a compliant encoder or decoder directly from the spec. That's a good thing, particularly if there are aspects of this headed for silicon at some point.
And that's assuming there's enough difference between VP8 and H.264 to need any new hardware. Sounds like yes, in a few places anyway, and of course, only on devices... PCs will be fine as-is. And there's so much H.264 related stuff in there, there's still reason to fear patent lawsuits. Presumably, Google's lawyers have reviewed this against the H.264 patents that might be applicable, but there's nothing really reassuring about that on the WebM site. Given the long list of partners, though, you'd kind of hope this has been done.
I never bough the "OS as a browser" argument.. if there even is an argument. I mean, I can run a web browser on Windows or on Android or Linux, and yeah, that suggests that Web-centric applications will, within reason, be kind of cool. But what possible advantage to I gain in something like ChromeOS? Why not run that same level of web browser within Android, and get the best of Web-only (like Chrome), Web-agumented (like many Android apps), and stand-alone apps as well.
It's clear from many things they do that Google is thinking Web.. and they should, that's where they were born and where they thrive. The great thing about Android versus iPhoneOS or PalmOS is that I don't ever need to cable up to a PC and sync my phone or tablet... any important syncs can just happen, invisibly, via the net. This also means that "sync to a PC" itself is only interesting if I want to also use a PC... these devices, unlike most others, can live entirely without the need for PC support. That's critical in advancing the state of computing on these devices.
But going web-only doesn't seem to be a step forward. If web apps are really that great, folks will use them, on devices, on PCs, whereever. If they're not so great yet, you're not betting the whole device on web-only functionality and perfectly persistent internet connections.
Google didn't do Android to make a vast fortune selling smart phones. Their primary concern was to ensure their primary profit center, internet search, extended to small devices. Given that, prior to Android, every smart phone platform was proprietary and closed source (some, like Symbian, have gone open source since then), it was easy to believe that at least, default search could leave them out. In fact, that's just what happened ... RIM made a search deal with Microsoft, and overnight, Blackberries started using Bing! as their default search engine.
Google is actually making money on Android, via search. In fact, depending on just how they package Android, some of the carriers are getting profit sharing from Google based on that income.
Search will absolutely become increasingly important on mobile devices. Things like Google Goggles, the various tools for finding things in unfamiliar areas, other augmented reality toys, they all show pretty clearly that this is going to replace a great deal of the stuff people once did on the desktop. Google was smart enough to not only see this coming, but release Android in such a way that ensured it would eventually dominate the small device market.
Also, I don't think it's a matter of Apple allowing other carries in the US. It was originally a matter of AT&T not allowing them... Apple had to give AT&T exclusivity in order to get them to carry the iPhone. By all accounts that's over, but the real question is, can anyone else deal with Apple. Verizon and Apple have apparently had serious talks at least three times now, and so far, no iPhone on Verizon. The story as I hear it is that Verizon has been interested, at least in the past, and were willing to deal with Apple much as they dealt with every other vendor. Apple, on the other hand, wanted the same kind of special treatment, like profit sharing, they got from AT&T.
Going forward, though, I think it's more of a longshot. The fact is, Verizon is doing very nicely with Android now. But it's not just that... they can only offer so many phones that do basically the same thing before it's just not worth their carrying the phones. As they have in the past, Verizon likes to brand the models they carrry. It's not that big of a deal to most people.. the Verizon Droid is essentially just a CDMA2000 Motorola Milestone. But when you say "Droid", it's automatically Verizon. All their other Android phones are marketed under the "Droid" name.
Look at the Google Nexus One deal... supposedly, Verizon was set to carry the Nexus One. Only problem... that was going in with Google primary branding, not the secondary branding you get with the Droid (being a "Google Experience" phone, the Droid is just as untouched by Verizon as the N1 would have been, but the Google logo's on the back, it says "Motorola" and "Verizon" on the front). Put an iPhone into Verizon, and it's still an iPhone, still Apple branded, with maybe a tiny "Verizon" on the back. I'm not sure they want that.
And if you can't have Verizon, that's at least currently 30% of the smart phones sold in the USA, and growing much faster than AT&T. So Apple would have to tag both T-Mobile and Sprint for the same coverage. And really, are they interested? Would Apple do a special CDMA2000 + WiMax phone for Sprint... that's probably what they'll be requiring for their high-end phones, like the EVO 4G, from now on. Given that AT&T and Verizon are both going LTE at 700MHz, this is probably a non-starter for Apple. So that leaves T-Mobile. Well, they're already GSM/HSPA, just on different 3G channels. But they're also the smallest carrier.
It's a cool toy. That's it. Some article last week claimed that demographically, the iPad was bought largely by older, rich men. In short... yeah, it's "oooh... shiny...." all over again. Just as when the iPhone first came out... it wasn't even a PDA at that point, just a cellphone with a PMP and Web browser built-in.
Having viewed the dozen-or-so keyboardless tablets shown off at CES this year, as well as the iPad two weeks later, I have an idea of why I'd like something like this, and particularly why it's not the iPad. But I'll need a valid reason.
Smartphones are a valid reason to buy such a device. Before my Droid, I had a Palm Treo. The big advantage of the smartphone is that it's really not a phone, it's an application processor, a PDA, something that replaces a dozen small electronic devices. My Droid is a phone, sure, and it's just dandy as a phone. It's actually more than that, since it supports pretty much every form of modern communication: Skype, Twitter, IM, SMS, Facebook, interactive blogs, etc. About all that's missing is a front-facing camera for videoconferencing... and those are coming in the newer phones. Next time an important new form of communications comes along, there's just a new app to download.
It also replaces, and sometimes extends, the slew of small devices I'm likely to carry. It does the MP3 player/PMP thing, but does it better than any MP3 player or PMP I've had. First of all, a better-than-SD screen just rocks for video, and network access to YouTube and other sources make it more useful for video than, say, a plain old iPod. Same with music... the MP3 player function is dandy, but there fact of apps means you have better alternatives. On the Droid, I use Museek for listening to locally stored music... it adapts to my current mood. Or Pandora and other "internet radio" options... I recently drove from South Jersey to Boston and back, with Pandora running the whole way, perfectly glitch-free.
Next is GPS.. this is a better GPS than many units I've owned, even if you never use the live satellite overlays. There may be issues if you lose the network, but that generally doesn't happen. It's also a functional pocket camera and camcorder. No, it's no replacement for my Canon EOS or Panasonic HMC40, but since I have it anyway, it's ok. Video on this device is actually better than on many consumery SD camcorders... sure, the audio is just as bad if not worse, but again, it's one more device function.
I get more, via apps. I used to carry around one of those Sony mini-tape recorders, for writing ideas. I used voice notes on the Treo, these days I have both voice and text notes, and automatic, invisible sync to my PC via Evernote.. again, not just the old function, but an improved version of that old function. It's also a fully functional flashlight, a guitar tuner, a games machines, a ssh client, etc.
Then, of course, email and web... the web browser is good enough that, for a simple "hey, I want to check that out" kind of use, I don't bother with a PC anymore... the phone's always here. I have email set up though my server to go to the phone and the PC, and of course, work email via the crufty Microsoft stuff works as well as anywhere, I guess.
So... on to the tablet. I need a number of larger format devices replaced by the tablet, or I'm not interested. What could these devices be? Well, some of that's maybe basic laptop stuff.... if I'd take this tablet where I wouldn't take the laptop, but still get some needed functionality, I might find that useful. Doing CAD and Video, I'm always going to have a desktop, so for me, a functional enough tablet might replace the laptop entirely.
First thing I might drag along -- portable video player. I have a portable DVD player, I might like a portable Blu-Ray player to replace it, but given enough function in a tablet, maybe that'll do the job. The big problem there is storage... most of these don't have a single Blu-Ray's worth of storage. But if you can use SDHC/SDXC cards or USB sticks/hard drives, op
Also, Apple will get a large boost in sales due to iPhone upgrades. Quite a substantial number of upgrades are on the same cycle as the phone releases. So yeah, right now, the numbers are down, artificially low, as everyone knows the new iPhone announcement is coming in June or July. Once it's out, there's an almost guaranteed spike in iPhone sales, as all of those with expired 2G contracts rush out to buy the new phone. At least, that's expected... if it really sucks, maybe there's less of that, but I'd still bet on Android carrying 2Q2010, with iPhone probably back in the #2 or even #1 slot for 2H2010.
And that's just dandy, and not an Android problem. I think the fact Android hit #2 this early, and in the USA rather than globally, was kind of a surprise, if not a downright anomaly. Apple's situation helped, but it's definitely a sign that Android is growing, and growing fast. My personal opinion is that, unless Google manages to do something stupid on an epic scale, the dominance of Android over iPhoneOS is only a matter of when, not it. But with folks like China Mobile embracing Android, while Apple's still weak in much of the world outside the USA, I really expected this to be a global win before it hit stateside. And I still think it's going to flip-flop from quarter to quarter for awhile. Android will be more consistent, simply because there's a "hot new thing" introduced in the Android ecosystem every few months. But this will also allow Apple's sales spikes to have it keep winning sales from time to time, even after Android dominance is assured.
Given Apple's various policies, it's impossible they're not fully aware that they can't hold their current #2 position in the US forever. They probably don't care... selling high priced PCs hasn't been at all bad to Apple's bottom line. They're not going to change that policy anytime soon in the devices market... iPhones will continue to compete with the high-end of the Android and other smart phone markets, while the Android world will encompass an increasingly wide range of devices, getting to low price points Apple will never even bother with.
And in fact, Apple can't bother with those lower-end markets. It's the same reason Mercedes and BMW only get so cheap... they have to protect their hard won position as high-end models. Apple's selling much the same thing in the PC, PMP, Smartphone, and Tablet market as everyone else, but at twice the price, give or take. They're not about to give that up for increased market share... why would they? From a techie point of view, lots of us are bothered by this... I've been a little bothered by Apple since the 80s, as I was designing computers with the same parts for Commodore, delivering more performance, and Apple was still considered "high-end", even with often inferior hardware designs out of Apple.
But from a sales and marketing viewpoint, you have to love how successful they've been at maintaining this market perception... and for decades. This is the reason Apple's flush with cash, while many PC companies and cellphone companies alike have been struggling.
Not only that, but while AT&T has good 3G coverage in cities, only about 1/5 of their network, by area, is 3G. So when you get rural coverage, it's much less likely to be fast. By it's nature, ever Verizon cell is 3G... you only drop back to EDGE speeds when you're too far from a cell to support 3G. As with all radio, this can vary by conditions... I get 3G on my Droid in my house about 80% of the time; it's occasionally dropping back to 2G speeds, but that's pretty rare. I'm in a rural area (eg, internet and TV only by satellite...).