The devices themselves haven't been that surprising. Keep in mind, the iPhone was such a poorly guarded secret/presumption that there were dozens of artists conception drawings of both the "all-screen" iPod (an extrapolation of the Palm T|X and T5 and some of the PocketPC hardware) and iPhone for over a year before the actual debut. Same with the "big fat iPod Touch", the iTablet, which of course because the iPad.
The price was hardly a shock to anyone paying attention. Ok, sure, the rumors has it as much as $1,000, but that would have been a MacOS driven tablet. The ARM tablets arrived as basically Netbooks with a bunch of stuff cut out (cheaper CPU, less RAM, no keyboard, fewer ports, half the battery, etc).. and of course, Apple would price theirs starting at about twice the proper entry-level. Which they did. Anyone who though the iPad was "priced aggressively" (and yeah, there were many, including some poor fools making the pricing decisions for a number of early Android tablet makers) was caught in the Reality Distortion Field and not paying attention to the man behind the curtain.
And yeah, I was ranting on those predictions before the iPad came out.
Microsoft also seems to have a talent for self-defeat. Yeah, Zune was a product released to compete with Apple's previous generation of PMP... that's the problem you get when you're reactionary, as Microsoft has been in practically everything they've ever done. And yeah, sometimes they were good at it, but still, they kept and keep getting surprised by others' successes they don't see coming (GUIs, office software, the internet, multimedia computing, livingroom computers, mobile computing, etc) and reacting to that. Those reactions are often not very organic... they have the menu of things their copying, and they check off each box. But that doesn't deliver an iPod or a PS3 or a Galaxy Note.
And it's even sometimes just stupid self-defeat. I had a 30GB Zune, once they were selling for $70 on Woot! In brown, to match my Martin D-15:-) The UI sucked (and sadly, grew up into Windows Phone and Metro), but the player was decent enough. Only, after years of bolstering "Plays4Sure" on other devices, they didn't support it. So the Zune was basically the only PMP on the market that was incompatible with just about everything else out there, just at the time that "runs on your portable device" became a selling point for DVDs, etc. Again, they seem to lack the fundamental conviction of a company that creates these devices out of a desire to make that cool thing you have in your mind... rather than to re-make Apple or Google's cool thing, but with a Microsoft stamp on it.
I think Microsoft is still in the denial phase of their transformation from essentially controlling the whole computer industry to being just another player. They got used to everyone pretty much just doing what they were told, and still haven't come to grips with the fact that, basically, that's the 20th Century Microsoft. That company, and those market conditions, no longer exist.
In a very real way, this is an extended legacy of Vista. Prior to Vista, Microsoft was the world's most valuable tech company, and both individuals and corporations pretty much did jump when Microsoft told them to do so. Sure, corporations were slow, usually waiting for SP1 of any new Microsoft OS. But there was never any serious attention given to the idea of "if" one would adopt the new MS-OS. It was only a matter of "when". The whole industry really just never entertained the "if" question -- of course you were going to upgrade to Windows, WIndows 95, 98, 2000, XP. And of course, all of the products that such new OSs enabled, all the upgrades those clever incompatibilities forced.
Then Vista. It didn't take long before many folks -- we Windows-using techies, the punditry, the press, even Mom and Pop, started to understand that Microsoft had in a new way overstepped the bounds of this odd de-facto agreement between Microsoft and 95% of the Computer Using Populace of Earth. Vista was evil, and evil in new ways. It was slow, it wanted fast new hardware, it was buggy, and it was just different, often in seemingly arbitrary ways. Vista sold, but in numbers that really did hurt Microsoft... by the height of the Vista era, Microsoft was no longer the world's most valuable tech company.
And then the realization set in... consumers and businesses eschewed Vista en masse... and didn't all suddenly turn into pumpkins. Their businesses didn't fail... in fact, in many cases, they worked just as well as they every had, and all of that money, time, and training for the new arbitrary and new obviously unneeded OS upgrade could be spent on things actually related to the success of one's business. Consumers, already pretty much not exceeding the powers of their present PCs, could save that cash for something else, and really not miss the fact that that shiny new Vista PC might go 75% unused, compared to their current only 50% unused XP machines.
Microsoft seemed to have learned their lesson with Windows 7... they fixed most of what was wrong. Windows 7 was for the end-user, not just a grab for more money and a means to advance other weird Microsoft agendas. But the damage was done. The cycle of mandatory upgrades was broken, and even today, plenty of individuals and corporations are happy with their older PC, or XP, or both. And this also started the first crack in a large number of folks (not just the slashdot regulars) even questioning whether Windows itself was important.
So it was actually Vista that, if not created, gave a big boost to the rise of mobile computing. The money I'm not spending on unnecessary PC upgrades can go to a smartphone or a tablet -- these things look like fun, and maybe they're even useful. And so, from Vista to Windows 8, we see a very significant percentage of computing go on vacation... it's now occurring on those pocket devices Microsoft, Palm, RIM, and a few others had pretty much decided were not mainstream but only for business folk, ages ago. Credit Apple with really cracking the consumer market, and Google/Linux for providing the go-to platform that could embrace the same kind of dynamic that made Microsoft big. And so, in 2012, Android outsells Windows, unit-wise. In 2013, it might actually come close to tripling Windows sales.
Microsoft understood the rise of mobile in much the same way they understood the rise of the Internet -- they were putting along at 10mph, and saw a big convoy flash past them doing 115. And much as in the Internet days, Microsoft pulled their usual "let's really over-react to this" card, and decided to make desktop Wi
Church marriages don't have any legal standing. It's quite common, of course, that one's chosen cleric is also empowered by the state to perform marriages, so s/he does the church marriage, then files the appropriate civil paperwork, and it's done. If that paperwork isn't done, the couple may consider themselves married, but the state doesn't.
This is precisely how plural marriage "gets away with it"... the couples are married within the church, but not legally. Utah had enough trouble with this, given that the Mormon founders were polygamists and made that part of their religion, that they have a special law, something to the order of that if you're living as if you're married, you're considered married, at least for the purposes of prosecuting polygamists.
So, given that religious marriages on their own have no legal standing in the USA, the only one really enabling marriages is the state. Thus, it's perfectly correct to keep calling this "marriage" -- the only real marriages are state sanctioned. The religious folk can make up a new name if they want to. Or move to the Middle East or some other place where the religious folks run the country.
Naa... let the stupid churches come up with their own name if they don't like the current use of the word "marriage"... but it's well established as in civil law. There's zero reason for that to change, just to make some knuckle-dragging religious nutbag more secure about his sexless "marriage" that got the magical seal of approval from the big grandpa in the sky.
Marriage has many important civil benefits between the two people involved. There are zero religious benefits, particularly to the non-religious (who have a much lower divorce rate than these crazed red staters). They can make up new words for the religious version of this if they like, and even define that new things to not include gays or biracial or any other couples they don't like. Not the State's business. In fact, I wish they'd get busy with that and either stop trying to redefine marriage in religious terms, or show me the clause in the Constitution that defines marriage their way. I already know that, in the Christian Bible, it's that arrangement between a man and any number of women, who will of course be stoned to death by the good townspeople if suspected of infidelity. None of that crap has a place in modern society.
Thing is, Microsoft isn't sharing applications -- not really. They're sharing a UI. The UI is the same between Windows 8 and Windows RT, similar on Windows Phone and X-Box. But Windows 8 applications don't run on Windows RT. Windows RT applications don't run on Windows Phone. Windows Phone applications don't run on X-Box.
And that's actually less useful than the other guys. Apple broke app compatibility between desktop and tablet, so they could make each of these optimal for those environments. Desktop apps don't run on the tablet. Tablet apps don't run on the phone, but the phone apps do run on the tablet, albeit with a letter/window box.
Android has actually done it best of all. There's just one Android, tablet or phone. Identical apps on each, but the UI adjusts to the device in question. Sure, this has let a few developers get sloppy, leaving tablet users with only an expended phone app. Then again, that's often all the app really needed to do. Android also did the integration right... data syncing automatically, just happens. The device is a first-class thing, not needed a tether to a PC. Apple's found that to be the same solution, years later.
And there's a good reason for this: it's an inherent compromise to make a useful computer that fits in your pocket or in the space of a book. Touch is a compromise, full screen apps are a compromise, etc. Dandy for when mobility is the number one priority, stupid when things get more involved and there's no need for mobility.
Metro itself isn't really a GUI anymore. The point of a GUI is that everything you can do ought to be either intuitive or suggested by some other element of the UI. Metro has severely broken that. Obviously, anyone used to the quirks of a text shell (much less the dozen plus I've used) won't have a problem learning these quirks... but there's also the "why bother" aspect. The main way Metro is broken is the app model -- all full screen, all the time. Ok, sure, you can panel 2 or 3 in Windows 8.1. But advanced users are using dozens of windows on multiple screens. Metro is a tablet/phone UI and it shows... full screen is generally the right answer for "consumption only" computing. Same reason you can't do native "For Metro on Metro" development... these are special powers reserved to Microsoft on a pure Metro machine.
And after all, Metro is just a design scheme, the Start Screen and all that is just a [semi-]graphical shell. There's no reason on earth Microsoft needed to torture desktop users with this mess. They didn't need to eliminate "Aero" themes, etc. They could easily have kept Metro apps, but had them come up in Windows on the desktop. They could have shipped Windows 8 with a desktop users shell, an improved Windows 7 shell.. or even just the Windows 7 shell, made this a user choice, and avoided nearly all the flack. But they seem convinced that forcing all the tablet stuff down the desktop user's throat will endear Windows RT or Phone devices to said user. I can't see that ever happening, but that's apparently the theory.
It'll be interesting to see what becomes of professional software, if Microsoft really does keep marginalizing the desktop user into oblivion. Most professional tools don't have any business as Metro apps; some can't even run as Metro apps -- no compilers.
And of course, there are two factors pushing this. Both are money. One is simply Microsoft once again wanting to follow Apple. Metro apps can only be sold through Microsoft's store, MS gets the same walled garden, the same 30% per app or whatever, that Apple's getting today. If Windows really goes this way, that's big bucks to Microsoft.
The other, of course, is the failing of PC sales against cheaper tablets. Perhaps some of that's even due to Windows 8, but even before that, it's significant. Android devices outshipped Windows devices in 2012. The last time any PC or OS outshipped Microsoft versions, the early 80s I think. This year, it's likely that Android outships Windows 3:1 or nearly so. iOS might even outship Windows in the next year or two. I think Microsoft brought most of this on themselves, but they're clearly being reactionary.
And that's the usual Microsoft thing. They missed the internet in the 80s, so Windows 98 didn't just ship with Internet Explorer, the whole Windows UI had to look like bad web pages. But at least you could switch that off. Microsoft's "all-in" approach is often in the right direction, but always a stupid overreaction. Like the bad fit of DRM throughout Vista, reacting to the idea of computing devices becoming media devices. And now everything looks like a tablet. That's bound to fail -- in many ways it already has, though the Vista failure was a worse one -- it taught Industry that MS upgrades are totally optional. And hurt Microsoft, cash-wise... in the XP heyday, they had been the most valuable tech company, worth even more, inflation-adjusted, than Apple at their peak.
We'll see. In short, Microsoft isn't wrong to want to extend Windows to work well on tablets... they just need to not force that option on every keyboard, mouse, and/or desktop user. Touch screens are a huge fail on the desktop, just as they were in the 80s (CAD industry, not mainstream). Windows 7 was Vista re-designed for the user, and it did well. If Microsoft does that again -- concentrates on what's important to users (CHOICE, for one) and not what's advancing MS's agenda by brute force, they might just fix this problem with Windows 9. And on the way, Microsoft: hire some actual GUI designers with real cognitive psych backgrounds. Metro is just so wrong... even on a tablet.
Also, it's reportedly just new chips going to TSMC. That makes some sense -- there's work and money to get an existing design moved to a new process. So Samsung's still going to be making A5s or A6s for the near future. And maybe other stuff -- there's certainly a contract between Apple and Samsung that has to be run out. And it's pretty unlikely Apple would bet on a new process. They haven't so far.. the chips Samsung did for them usually started out in a more conservative process. Whether that's Apple being cautious, or Samsung keeping the best for themselves, is anyone's guess.
And no, I don't imagine a machine turning raw silicone into CPUs. Maybe breast implants and the better quality diving masks. But even when you start with silicon, it's still an involved process.
As for competing with Samsung's high-end.. Apple's got some multitasking issues. Android definitely taps four cores... seen that all the time on my tablet's performance monitor. iOS... supposedly, not so much right now. As long as Apple customers are happy that way, Apple's going to be spending less overall money per chip and selling that gear for more than Samsung. Not a bad business if you can get it. Then again, maybe not -- gaming has become so critical to Apple, they tend to ship their new devices with the fastest GPU rigs around (PowerVR, not their own, but still, more/faster cores).
Claiming this is market share, that implies this is for new sales only, not installed base. Sounds like they screwed up the terminology... otherwise, that doesn't say much at all good about Windows 8.... and also suggests some crazy people are still buying new Windows Vista systems.
If this really does mean installed base, then you have to ask how that's actually computed. If it's just based on sales figures, it's likely very skewed in Windows 8's favor. On the day that Windows 8 shipped, all of the enterprise licensees started buying Windows 8 licenses. These are the licenses that let the IT department clone their standard disc for all new PCs and just pay MS for each one. These licenses, of course, include full downgrade rights, and most of them are still being used for Windows 7 or Windows XP... but they come up as Windows 8 for the purpose of sales figures. The last study on this I saw showed that less than 60% of the actual sold Windows 8 licenses were actually being used for Windows 8. Some detail on which set of assumptions (lies, etc) this is used for would be interesting.
And the real news... earlier this year, late last year, etc. many different similar installed base reports put MacOS growing from 4.8% last year to just over 5% earlier this year -- this is internationally, Apple of course does much better domestically. It's probably just a difference in their calculations versus the various other industry numbers people.. but if MacOS really did jump 2% in one quarter, in installed base rather than just quarterly sales, that would be big news. Of course, that growth might come as much from a failing PC market as some rally of Apple products.
Incorrect. A single x1 PCI Express 3.0 lane does 8Gb/s. Thunderbolt 1.0 has two 10Gb/s lanes, while 2.0 is exactly the same, except it can aggregate two lanes to deliver a single 20Gb/s connection, just like PCI Express has done all along.
Assuming each of those ports actually connects full speed to the main system (I'm fairly certain it doesn't, because no one will use it that way, but it could), that's essentially 12 PCI x1 lanes to play around with. That's not horrible, given you already have the two GPUs accounted for. Of course, those are your GPUs for life, but they'll be decent for awhile yet.
And the GPUs are the reason there's not likely full 6-channel Thunderbolt to/from the main board. You've seen an HDMI port, but no others... that's because you're expected to hook your main monitors (and anyone using a system like this with only one monitor is a fool) to Thunderbolt ports. So, in a basic configuration, you'll probably eat two of those right away. And there's a pretty good chance the system has a cross point switch that allows the DisplayPort connection directly from each GPU to be routed directly to one of the Thunderbolt ports, and put it into DisplayPort mode. DisplayPort already supports full 60p 4K displays, no need to wait for a Thunderbolt monitor or live on HDMI alone (1.4 supports 4K at 24p).
I concur.. the E5 V2 has already been announced going to 12 cores. Not out yet.. but Apple's certainly waiting on something. As well, you do not see a place in that box for a second processor. Nope, it ain't there.
At 50-100GB per hour for video, things can get really busy, really fast, in a video editing house without large local storage. Particularly if everything's on GigE, which starts getting really slow when you're not the only one schlepping video across it. Users of this for serious media work will need a local Thunderbolt drive in addition to everything else.
Thunderbolt doesn't store a single bit. Maybe you meant "eSATA is dead"? You're still going to need a place for that Thunderbolt cable to find some storage, and you're probably going to want it to be redundant. Sure smells like a RAID to me. And there's such little storage in the system, external working drives will not be optional.
And for lots of media creation, it's not just the single stream throughput of the drive, but the aggregate performance across dozens, maybe hundreds of files. RAID isn' always a better answer here than multiple, independent drives. Of course, you can configure most RAID controllers to just to JBOD if that's what you need, but either way, it's still one more box to deal with. Companies may have a SAN, but then you're bottlenecked at GigE.
While it depends on what you're doing, if Apple's still selling to the professional media content market, they'll need some big RAM in this thing. I actually used an 8GB system for some years, with all the RAM I ever needed, doing HD video, electronics CAD, all kinds of things. It was seemingly mundane photo editing that got me to upgrade to 16GB, and has me drooling a little at new systems that'll go to 32GB or more. That's not for a single photo, but when you start shooting with 18-20+Mpixel cameras, in RAW, in HDR, and doing 20-60+ shot panoramas, that memory goes fast. I have individual composited photos well over a gigabyte in size.
And that's another issue here... PCI Express SSD may be wicked fast, but the sizes available will make it expensive and still totally unsuitable to media work. An external working HDD will be necessary, as well as the LAN, or RAID you're using for archival purposes.... or, oh, never mind, Apple discontinued their SAN years ago.
Depends on what you're looking for. Each Thunderbolt cable delivers two 10Mb/s links, slightly faster than an PCI Express 3.0 x1 link, or the new 20Mb/s single link, which is, not surprisingly, slightly faster than a PCI Express 3.0 x2 aggregate. That's enough for many things, but not everything.
Yeah, Apple's been all about the "laptop for the desktop". True, they're not managing to sell you monitors along with this, but everything but. The display cards look replaceable, but they're a proprietary design, and might not easily be duplicated by a third party, should Apple ever sell enough of these to make that interesting. Not to mention how tightly controlled the thermal footprint on this must be, dealing with that chimney and single fan.
For Apple fans in general, though, this is good news. This is certainly a more capable computer than anything Apple sells today. There was a very legit concern that Apple under Jobs was throwing all of their pro users under the bus... I mean, the Mac Pro of today (which, presumably, no one will every buy again) is 3.5 years old. and it's only ever been shipped with an AMD gamers's GPU card from 2009... though, given the drivers, one could always swap in a good one. And they did offer a dual-headed Mac Pro... though the upgrade to a current Xeon is certainly the better move, if you had to choose just one.
If the display connections transmute into actual DisplayPort connections when attached to a DisplayPort device, that's not going to be a problem. And I think they must, given there's only the one HDMI output, probably intended to drive a preview TV/monitor -- that's the usually need for that in a video workstation. DisplayPort links support quad-HD monitors (3840x 2160, "4K class") at 60p, HDMI 1.4 only supports quad HD at 30p and full 4K at 24p.
Anyone using professional computer monitors left DVI and HDMI behind awhile back. My system here (non-Mac... not a Mac fan for many reasons) runs two 2560x1440 monitors on DisplayPort links, one 1920x1200 on DVI, and an optional 32" HDTV on HDMI. That's not exactly a typical consumer configuration. Keep in mind, Apple left DVI behind long ago, except on the current Mac Pro, and that only because it's only ever been shipped with a consumer GPU from 2009. Thunderbolt was designed to mutate into DisplayPort (the Thunderbolt connector IS the mini-DisplayPort connector), so presumably, any monitors shipped from Apple or used with Apple PCs in recent times will plug right into this new PC.
I'm not really defending Apple here, just laying some facts on y'all.
Well, there are limits. Thunderbolt 2 isn't any faster than Thunderbolt 1, it just allows the two 10Gb/s links to be aggregated, as PCI Express has always allowed. So, assuming they're all independently connected Thunderbolt ports (eg, aggregate throughput of 20Gb/s x 6, each way), this equivalent to about 15 PCI Express x1 links, in total. Well, that's enough for one external GPU, if you're not working it hard, and some external drives. And given that it's a virtual certainly some of these Thunderbolt ports will be used in DisplayPort mode, I'll bet you don't get anywhere near 120Gb/s to and from (it's a simultaneously two-way channel, just like PCI Express) main memory.
The real question I would have is on the bus architecture. There's got to be some Thunderbolt/DisplayPort switch internally, to route the output of the GPUs to the Thunderbolt port(s). A typical AMD GPU will have at least two DisplayPort outputs, but of course that's not crossing a main bus, that's driven directly out. That lone HDMI 1.4 port is only for one monitor, and HDMI 1.4 can only drive quadHD at 30p or true 4K at 24p... not what you'd want for a workstation-class system. The rationale for Thunderbolt 2.0 was to allow faster graphics, since Thunderbolt 1.0 can't support 4K in Thunderbolt mode, only DisplayPort mode (DisplayPort 1.2 can hit 17.2Gb/s, ever-so-slightly less than Thunderbolt 2.0). So you're going to be using monitors on at least two of these ports. And again, that's going to want to switch directly from the GPU modules, not cross with anything else.
So I bet they have a big old cross point switch, which the output of the GPUs, the Thunderbolt ports, and a link to the main system (via PCI Express most likely), to keep this thing going.
Microsoft has a problem. Only 100M Windows 8 licenses sold, less than 60M actually being used for Windows 8 (all enterprise and some pro licenses include downgrade rights... big companies buy Windows 8, but they're still installing XP or Windows 7). They also have the problem of being seen big, slow, and stupid. Apple's doing one iOS and usually one MacOS upgrade per year, and not charging big bucks. Android has been even faster, with usually two major releases per year.
Microsoft can't take them on head to head on a 3-year upgrade schedule. So the old "Service Pack" sounds like a bug fix. Now it's going to be a new incremental OS update.. which they thought, briefly, about charging for. They did this before, with Windows 98SE... no idea why, but there's history. But they have clearly decided that, whatever's in this update, the value to Microsoft in getting this out (perhaps selling more Windows 8.x copies, necessary to get developer support) is worth far more than they'd make selling 60M update packages.
The danger, of course, is that this is seen as just another service pack. They need to deliver the kind of punch you see from Android 4.1 to 4.2 or the numerous MacOS upgrades. They might... but some of that's going to be selling it, and most of those to new users who avoided Windows 8. Though given how bad Windows 8 is, after the lesson of Vista, I wouldn't place any bets...
There are only about 76 million X-Box 360s, total. There were 480 million Android devices sold just last year. I don't know how many people hold MS Accounts... though it's certainly less than Facebook or Google account holders. But I don't think X-Box is at all significant in boosting those numbers.
So yeah, they decided they can't charge for a service pack. Not news.
The change in name... 8.1 vs. 8sp1 is no doubt aimed at, very least, suggesting Microsoft is doing real yearly OS updates. This is needed because of the competition. Google usually does two new Android major releases in a year. Apple does one iOS and about one MacOS release every year.
Microsoft, on their typical 3-year schedule, simply can't compete with the pace of technology. Even today, not so much. Windows Phone OEMs are already complaining that the HW requirements of Windows Phone 8 is, once again, making it impossible to compete head-to-head with Android and iOS. I don't disagree.. though part of that is the artificial split between Windows Phone and Windows on tablets; iOS and Android don't do desktop, but they do tablet and phone. On Android, apps routinely adjust themselves for the different screen resolutions, so adding new ones is no big deal. Windows has a small number of very specific requirements, and hardware is changing faster than those numbers.
Sure... really long term, we're doomed either way. If there's the critical mass for a Big Crunch (these days, it's a question of dark energy vs. dark mass, given that even today, galaxies are still accelerating from one another), everything crunches together, whatever that really means. If not, eventually, all stars die out, and it's the heat death scenario.
But in practical terms, that's a "high class" problem. Any given species on earth is good for a few million years at best. We expect our big brains will automagically solve this problem, but another big dino-killing-sized asteroid could easily offer a different opinion. If we establish multiple home planets, we're trading some many thousands of years for many billions of years, potentially, of humanity.
So the next problem... leaving the Universe for the one next door. If that's a real thing, anyway, as modern physics suggests. And yes, I do like Stephen Baxter's books, thanks for asking:-)
Slashdot Mirror
The devices themselves haven't been that surprising. Keep in mind, the iPhone was such a poorly guarded secret/presumption that there were dozens of artists conception drawings of both the "all-screen" iPod (an extrapolation of the Palm T|X and T5 and some of the PocketPC hardware) and iPhone for over a year before the actual debut. Same with the "big fat iPod Touch", the iTablet, which of course because the iPad.
The price was hardly a shock to anyone paying attention. Ok, sure, the rumors has it as much as $1,000, but that would have been a MacOS driven tablet. The ARM tablets arrived as basically Netbooks with a bunch of stuff cut out (cheaper CPU, less RAM, no keyboard, fewer ports, half the battery, etc).. and of course, Apple would price theirs starting at about twice the proper entry-level. Which they did. Anyone who though the iPad was "priced aggressively" (and yeah, there were many, including some poor fools making the pricing decisions for a number of early Android tablet makers) was caught in the Reality Distortion Field and not paying attention to the man behind the curtain.
And yeah, I was ranting on those predictions before the iPad came out.
Microsoft also seems to have a talent for self-defeat. Yeah, Zune was a product released to compete with Apple's previous generation of PMP... that's the problem you get when you're reactionary, as Microsoft has been in practically everything they've ever done. And yeah, sometimes they were good at it, but still, they kept and keep getting surprised by others' successes they don't see coming (GUIs, office software, the internet, multimedia computing, livingroom computers, mobile computing, etc) and reacting to that. Those reactions are often not very organic... they have the menu of things their copying, and they check off each box. But that doesn't deliver an iPod or a PS3 or a Galaxy Note.
And it's even sometimes just stupid self-defeat. I had a 30GB Zune, once they were selling for $70 on Woot! In brown, to match my Martin D-15 :-) The UI sucked (and sadly, grew up into Windows Phone and Metro), but the player was decent enough. Only, after years of bolstering "Plays4Sure" on other devices, they didn't support it. So the Zune was basically the only PMP on the market that was incompatible with just about everything else out there, just at the time that "runs on your portable device" became a selling point for DVDs, etc. Again, they seem to lack the fundamental conviction of a company that creates these devices out of a desire to make that cool thing you have in your mind... rather than to re-make Apple or Google's cool thing, but with a Microsoft stamp on it.
I think Microsoft is still in the denial phase of their transformation from essentially controlling the whole computer industry to being just another player. They got used to everyone pretty much just doing what they were told, and still haven't come to grips with the fact that, basically, that's the 20th Century Microsoft. That company, and those market conditions, no longer exist.
In a very real way, this is an extended legacy of Vista. Prior to Vista, Microsoft was the world's most valuable tech company, and both individuals and corporations pretty much did jump when Microsoft told them to do so. Sure, corporations were slow, usually waiting for SP1 of any new Microsoft OS. But there was never any serious attention given to the idea of "if" one would adopt the new MS-OS. It was only a matter of "when". The whole industry really just never entertained the "if" question -- of course you were going to upgrade to Windows, WIndows 95, 98, 2000, XP. And of course, all of the products that such new OSs enabled, all the upgrades those clever incompatibilities forced.
Then Vista. It didn't take long before many folks -- we Windows-using techies, the punditry, the press, even Mom and Pop, started to understand that Microsoft had in a new way overstepped the bounds of this odd de-facto agreement between Microsoft and 95% of the Computer Using Populace of Earth. Vista was evil, and evil in new ways. It was slow, it wanted fast new hardware, it was buggy, and it was just different, often in seemingly arbitrary ways. Vista sold, but in numbers that really did hurt Microsoft... by the height of the Vista era, Microsoft was no longer the world's most valuable tech company .
And then the realization set in... consumers and businesses eschewed Vista en masse... and didn't all suddenly turn into pumpkins. Their businesses didn't fail... in fact, in many cases, they worked just as well as they every had, and all of that money, time, and training for the new arbitrary and new obviously unneeded OS upgrade could be spent on things actually related to the success of one's business. Consumers, already pretty much not exceeding the powers of their present PCs, could save that cash for something else, and really not miss the fact that that shiny new Vista PC might go 75% unused, compared to their current only 50% unused XP machines.
Microsoft seemed to have learned their lesson with Windows 7 ... they fixed most of what was wrong. Windows 7 was for the end-user, not just a grab for more money and a means to advance other weird Microsoft agendas. But the damage was done. The cycle of mandatory upgrades was broken, and even today, plenty of individuals and corporations are happy with their older PC, or XP, or both. And this also started the first crack in a large number of folks (not just the slashdot regulars) even questioning whether Windows itself was important.
So it was actually Vista that, if not created, gave a big boost to the rise of mobile computing. The money I'm not spending on unnecessary PC upgrades can go to a smartphone or a tablet -- these things look like fun, and maybe they're even useful. And so, from Vista to Windows 8, we see a very significant percentage of computing go on vacation ... it's now occurring on those pocket devices Microsoft, Palm, RIM, and a few others had pretty much decided were not mainstream but only for business folk, ages ago. Credit Apple with really cracking the consumer market, and Google/Linux for providing the go-to platform that could embrace the same kind of dynamic that made Microsoft big. And so, in 2012, Android outsells Windows, unit-wise. In 2013, it might actually come close to tripling Windows sales.
Microsoft understood the rise of mobile in much the same way they understood the rise of the Internet -- they were putting along at 10mph, and saw a big convoy flash past them doing 115. And much as in the Internet days, Microsoft pulled their usual "let's really over-react to this" card, and decided to make desktop Wi
Church marriages don't have any legal standing. It's quite common, of course, that one's chosen cleric is also empowered by the state to perform marriages, so s/he does the church marriage, then files the appropriate civil paperwork, and it's done. If that paperwork isn't done, the couple may consider themselves married, but the state doesn't.
This is precisely how plural marriage "gets away with it"... the couples are married within the church, but not legally. Utah had enough trouble with this, given that the Mormon founders were polygamists and made that part of their religion, that they have a special law, something to the order of that if you're living as if you're married, you're considered married, at least for the purposes of prosecuting polygamists.
So, given that religious marriages on their own have no legal standing in the USA, the only one really enabling marriages is the state. Thus, it's perfectly correct to keep calling this "marriage" -- the only real marriages are state sanctioned. The religious folk can make up a new name if they want to. Or move to the Middle East or some other place where the religious folks run the country.
Naa... let the stupid churches come up with their own name if they don't like the current use of the word "marriage"... but it's well established as in civil law. There's zero reason for that to change, just to make some knuckle-dragging religious nutbag more secure about his sexless "marriage" that got the magical seal of approval from the big grandpa in the sky.
Marriage has many important civil benefits between the two people involved. There are zero religious benefits, particularly to the non-religious (who have a much lower divorce rate than these crazed red staters). They can make up new words for the religious version of this if they like, and even define that new things to not include gays or biracial or any other couples they don't like. Not the State's business. In fact, I wish they'd get busy with that and either stop trying to redefine marriage in religious terms, or show me the clause in the Constitution that defines marriage their way. I already know that, in the Christian Bible, it's that arrangement between a man and any number of women, who will of course be stoned to death by the good townspeople if suspected of infidelity. None of that crap has a place in modern society.
Thing is, Microsoft isn't sharing applications -- not really. They're sharing a UI. The UI is the same between Windows 8 and Windows RT, similar on Windows Phone and X-Box. But Windows 8 applications don't run on Windows RT. Windows RT applications don't run on Windows Phone. Windows Phone applications don't run on X-Box.
And that's actually less useful than the other guys. Apple broke app compatibility between desktop and tablet, so they could make each of these optimal for those environments. Desktop apps don't run on the tablet. Tablet apps don't run on the phone, but the phone apps do run on the tablet, albeit with a letter/window box.
Android has actually done it best of all. There's just one Android, tablet or phone. Identical apps on each, but the UI adjusts to the device in question. Sure, this has let a few developers get sloppy, leaving tablet users with only an expended phone app. Then again, that's often all the app really needed to do. Android also did the integration right... data syncing automatically, just happens. The device is a first-class thing, not needed a tether to a PC. Apple's found that to be the same solution, years later.
And there's a good reason for this: it's an inherent compromise to make a useful computer that fits in your pocket or in the space of a book. Touch is a compromise, full screen apps are a compromise, etc. Dandy for when mobility is the number one priority, stupid when things get more involved and there's no need for mobility.
Metro itself isn't really a GUI anymore. The point of a GUI is that everything you can do ought to be either intuitive or suggested by some other element of the UI. Metro has severely broken that. Obviously, anyone used to the quirks of a text shell (much less the dozen plus I've used) won't have a problem learning these quirks... but there's also the "why bother" aspect. The main way Metro is broken is the app model -- all full screen, all the time. Ok, sure, you can panel 2 or 3 in Windows 8.1. But advanced users are using dozens of windows on multiple screens. Metro is a tablet/phone UI and it shows... full screen is generally the right answer for "consumption only" computing. Same reason you can't do native "For Metro on Metro" development... these are special powers reserved to Microsoft on a pure Metro machine.
And after all, Metro is just a design scheme, the Start Screen and all that is just a [semi-]graphical shell. There's no reason on earth Microsoft needed to torture desktop users with this mess. They didn't need to eliminate "Aero" themes, etc. They could easily have kept Metro apps, but had them come up in Windows on the desktop. They could have shipped Windows 8 with a desktop users shell, an improved Windows 7 shell.. or even just the Windows 7 shell, made this a user choice, and avoided nearly all the flack. But they seem convinced that forcing all the tablet stuff down the desktop user's throat will endear Windows RT or Phone devices to said user. I can't see that ever happening, but that's apparently the theory.
It'll be interesting to see what becomes of professional software, if Microsoft really does keep marginalizing the desktop user into oblivion. Most professional tools don't have any business as Metro apps; some can't even run as Metro apps -- no compilers.
And of course, there are two factors pushing this. Both are money. One is simply Microsoft once again wanting to follow Apple. Metro apps can only be sold through Microsoft's store, MS gets the same walled garden, the same 30% per app or whatever, that Apple's getting today. If Windows really goes this way, that's big bucks to Microsoft.
The other, of course, is the failing of PC sales against cheaper tablets. Perhaps some of that's even due to Windows 8, but even before that, it's significant. Android devices outshipped Windows devices in 2012. The last time any PC or OS outshipped Microsoft versions, the early 80s I think. This year, it's likely that Android outships Windows 3:1 or nearly so. iOS might even outship Windows in the next year or two. I think Microsoft brought most of this on themselves, but they're clearly being reactionary.
And that's the usual Microsoft thing. They missed the internet in the 80s, so Windows 98 didn't just ship with Internet Explorer, the whole Windows UI had to look like bad web pages. But at least you could switch that off. Microsoft's "all-in" approach is often in the right direction, but always a stupid overreaction. Like the bad fit of DRM throughout Vista, reacting to the idea of computing devices becoming media devices. And now everything looks like a tablet. That's bound to fail -- in many ways it already has, though the Vista failure was a worse one -- it taught Industry that MS upgrades are totally optional. And hurt Microsoft, cash-wise... in the XP heyday, they had been the most valuable tech company, worth even more, inflation-adjusted, than Apple at their peak.
We'll see. In short, Microsoft isn't wrong to want to extend Windows to work well on tablets... they just need to not force that option on every keyboard, mouse, and/or desktop user. Touch screens are a huge fail on the desktop, just as they were in the 80s (CAD industry, not mainstream). Windows 7 was Vista re-designed for the user, and it did well. If Microsoft does that again -- concentrates on what's important to users (CHOICE, for one) and not what's advancing MS's agenda by brute force, they might just fix this problem with Windows 9. And on the way, Microsoft: hire some actual GUI designers with real cognitive psych backgrounds. Metro is just so wrong... even on a tablet.
Also, it's reportedly just new chips going to TSMC. That makes some sense -- there's work and money to get an existing design moved to a new process. So Samsung's still going to be making A5s or A6s for the near future. And maybe other stuff -- there's certainly a contract between Apple and Samsung that has to be run out. And it's pretty unlikely Apple would bet on a new process. They haven't so far.. the chips Samsung did for them usually started out in a more conservative process. Whether that's Apple being cautious, or Samsung keeping the best for themselves, is anyone's guess.
And no, I don't imagine a machine turning raw silicone into CPUs. Maybe breast implants and the better quality diving masks. But even when you start with silicon, it's still an involved process.
As for competing with Samsung's high-end.. Apple's got some multitasking issues. Android definitely taps four cores... seen that all the time on my tablet's performance monitor. iOS... supposedly, not so much right now. As long as Apple customers are happy that way, Apple's going to be spending less overall money per chip and selling that gear for more than Samsung. Not a bad business if you can get it. Then again, maybe not -- gaming has become so critical to Apple, they tend to ship their new devices with the fastest GPU rigs around (PowerVR, not their own, but still, more/faster cores).
Claiming this is market share, that implies this is for new sales only, not installed base. Sounds like they screwed up the terminology... otherwise, that doesn't say much at all good about Windows 8.... and also suggests some crazy people are still buying new Windows Vista systems.
If this really does mean installed base, then you have to ask how that's actually computed. If it's just based on sales figures, it's likely very skewed in Windows 8's favor. On the day that Windows 8 shipped, all of the enterprise licensees started buying Windows 8 licenses. These are the licenses that let the IT department clone their standard disc for all new PCs and just pay MS for each one. These licenses, of course, include full downgrade rights, and most of them are still being used for Windows 7 or Windows XP... but they come up as Windows 8 for the purpose of sales figures. The last study on this I saw showed that less than 60% of the actual sold Windows 8 licenses were actually being used for Windows 8. Some detail on which set of assumptions (lies, etc) this is used for would be interesting.
And the real news... earlier this year, late last year, etc. many different similar installed base reports put MacOS growing from 4.8% last year to just over 5% earlier this year -- this is internationally, Apple of course does much better domestically. It's probably just a difference in their calculations versus the various other industry numbers people.. but if MacOS really did jump 2% in one quarter, in installed base rather than just quarterly sales, that would be big news. Of course, that growth might come as much from a failing PC market as some rally of Apple products.
No, you are not. It's totally a rubbish bin.
Incorrect. A single x1 PCI Express 3.0 lane does 8Gb/s. Thunderbolt 1.0 has two 10Gb/s lanes, while 2.0 is exactly the same, except it can aggregate two lanes to deliver a single 20Gb/s connection, just like PCI Express has done all along.
Assuming each of those ports actually connects full speed to the main system (I'm fairly certain it doesn't, because no one will use it that way, but it could), that's essentially 12 PCI x1 lanes to play around with. That's not horrible, given you already have the two GPUs accounted for. Of course, those are your GPUs for life, but they'll be decent for awhile yet.
And the GPUs are the reason there's not likely full 6-channel Thunderbolt to/from the main board. You've seen an HDMI port, but no others... that's because you're expected to hook your main monitors (and anyone using a system like this with only one monitor is a fool) to Thunderbolt ports. So, in a basic configuration, you'll probably eat two of those right away. And there's a pretty good chance the system has a cross point switch that allows the DisplayPort connection directly from each GPU to be routed directly to one of the Thunderbolt ports, and put it into DisplayPort mode. DisplayPort already supports full 60p 4K displays, no need to wait for a Thunderbolt monitor or live on HDMI alone (1.4 supports 4K at 24p).
I concur.. the E5 V2 has already been announced going to 12 cores. Not out yet.. but Apple's certainly waiting on something. As well, you do not see a place in that box for a second processor. Nope, it ain't there.
At 50-100GB per hour for video, things can get really busy, really fast, in a video editing house without large local storage. Particularly if everything's on GigE, which starts getting really slow when you're not the only one schlepping video across it. Users of this for serious media work will need a local Thunderbolt drive in addition to everything else.
Thunderbolt doesn't store a single bit. Maybe you meant "eSATA is dead"? You're still going to need a place for that Thunderbolt cable to find some storage, and you're probably going to want it to be redundant. Sure smells like a RAID to me. And there's such little storage in the system, external working drives will not be optional.
And for lots of media creation, it's not just the single stream throughput of the drive, but the aggregate performance across dozens, maybe hundreds of files. RAID isn' always a better answer here than multiple, independent drives. Of course, you can configure most RAID controllers to just to JBOD if that's what you need, but either way, it's still one more box to deal with. Companies may have a SAN, but then you're bottlenecked at GigE.
While it depends on what you're doing, if Apple's still selling to the professional media content market, they'll need some big RAM in this thing. I actually used an 8GB system for some years, with all the RAM I ever needed, doing HD video, electronics CAD, all kinds of things. It was seemingly mundane photo editing that got me to upgrade to 16GB, and has me drooling a little at new systems that'll go to 32GB or more. That's not for a single photo, but when you start shooting with 18-20+Mpixel cameras, in RAW, in HDR, and doing 20-60+ shot panoramas, that memory goes fast. I have individual composited photos well over a gigabyte in size.
And that's another issue here... PCI Express SSD may be wicked fast, but the sizes available will make it expensive and still totally unsuitable to media work. An external working HDD will be necessary, as well as the LAN, or RAID you're using for archival purposes.... or, oh, never mind, Apple discontinued their SAN years ago.
I was thinking iWastebasket, but either way...
Only the chips that let you occupy two 80MHz channels. That's optional in 802.11ac.
Depends on what you're looking for. Each Thunderbolt cable delivers two 10Mb/s links, slightly faster than an PCI Express 3.0 x1 link, or the new 20Mb/s single link, which is, not surprisingly, slightly faster than a PCI Express 3.0 x2 aggregate. That's enough for many things, but not everything.
Yeah, Apple's been all about the "laptop for the desktop". True, they're not managing to sell you monitors along with this, but everything but. The display cards look replaceable, but they're a proprietary design, and might not easily be duplicated by a third party, should Apple ever sell enough of these to make that interesting. Not to mention how tightly controlled the thermal footprint on this must be, dealing with that chimney and single fan.
For Apple fans in general, though, this is good news. This is certainly a more capable computer than anything Apple sells today. There was a very legit concern that Apple under Jobs was throwing all of their pro users under the bus... I mean, the Mac Pro of today (which, presumably, no one will every buy again) is 3.5 years old. and it's only ever been shipped with an AMD gamers's GPU card from 2009... though, given the drivers, one could always swap in a good one. And they did offer a dual-headed Mac Pro... though the upgrade to a current Xeon is certainly the better move, if you had to choose just one.
If the display connections transmute into actual DisplayPort connections when attached to a DisplayPort device, that's not going to be a problem. And I think they must, given there's only the one HDMI output, probably intended to drive a preview TV/monitor -- that's the usually need for that in a video workstation. DisplayPort links support quad-HD monitors (3840x 2160, "4K class") at 60p, HDMI 1.4 only supports quad HD at 30p and full 4K at 24p.
Anyone using professional computer monitors left DVI and HDMI behind awhile back. My system here (non-Mac... not a Mac fan for many reasons) runs two 2560x1440 monitors on DisplayPort links, one 1920x1200 on DVI, and an optional 32" HDTV on HDMI. That's not exactly a typical consumer configuration. Keep in mind, Apple left DVI behind long ago, except on the current Mac Pro, and that only because it's only ever been shipped with a consumer GPU from 2009. Thunderbolt was designed to mutate into DisplayPort (the Thunderbolt connector IS the mini-DisplayPort connector), so presumably, any monitors shipped from Apple or used with Apple PCs in recent times will plug right into this new PC.
I'm not really defending Apple here, just laying some facts on y'all.
Well, there are limits. Thunderbolt 2 isn't any faster than Thunderbolt 1, it just allows the two 10Gb/s links to be aggregated, as PCI Express has always allowed. So, assuming they're all independently connected Thunderbolt ports (eg, aggregate throughput of 20Gb/s x 6, each way), this equivalent to about 15 PCI Express x1 links, in total. Well, that's enough for one external GPU, if you're not working it hard, and some external drives. And given that it's a virtual certainly some of these Thunderbolt ports will be used in DisplayPort mode, I'll bet you don't get anywhere near 120Gb/s to and from (it's a simultaneously two-way channel, just like PCI Express) main memory.
The real question I would have is on the bus architecture. There's got to be some Thunderbolt/DisplayPort switch internally, to route the output of the GPUs to the Thunderbolt port(s). A typical AMD GPU will have at least two DisplayPort outputs, but of course that's not crossing a main bus, that's driven directly out. That lone HDMI 1.4 port is only for one monitor, and HDMI 1.4 can only drive quadHD at 30p or true 4K at 24p... not what you'd want for a workstation-class system. The rationale for Thunderbolt 2.0 was to allow faster graphics, since Thunderbolt 1.0 can't support 4K in Thunderbolt mode, only DisplayPort mode (DisplayPort 1.2 can hit 17.2Gb/s, ever-so-slightly less than Thunderbolt 2.0). So you're going to be using monitors on at least two of these ports. And again, that's going to want to switch directly from the GPU modules, not cross with anything else.
So I bet they have a big old cross point switch, which the output of the GPUs, the Thunderbolt ports, and a link to the main system (via PCI Express most likely), to keep this thing going.
Microsoft has a problem. Only 100M Windows 8 licenses sold, less than 60M actually being used for Windows 8 (all enterprise and some pro licenses include downgrade rights... big companies buy Windows 8, but they're still installing XP or Windows 7). They also have the problem of being seen big, slow, and stupid. Apple's doing one iOS and usually one MacOS upgrade per year, and not charging big bucks. Android has been even faster, with usually two major releases per year.
Microsoft can't take them on head to head on a 3-year upgrade schedule. So the old "Service Pack" sounds like a bug fix. Now it's going to be a new incremental OS update.. which they thought, briefly, about charging for. They did this before, with Windows 98SE... no idea why, but there's history. But they have clearly decided that, whatever's in this update, the value to Microsoft in getting this out (perhaps selling more Windows 8.x copies, necessary to get developer support) is worth far more than they'd make selling 60M update packages.
The danger, of course, is that this is seen as just another service pack. They need to deliver the kind of punch you see from Android 4.1 to 4.2 or the numerous MacOS upgrades. They might... but some of that's going to be selling it, and most of those to new users who avoided Windows 8. Though given how bad Windows 8 is, after the lesson of Vista, I wouldn't place any bets...
There are only about 76 million X-Box 360s, total. There were 480 million Android devices sold just last year. I don't know how many people hold MS Accounts ... though it's certainly less than Facebook or Google account holders. But I don't think X-Box is at all significant in boosting those numbers.
So yeah, they decided they can't charge for a service pack. Not news.
The change in name... 8.1 vs. 8sp1 is no doubt aimed at, very least, suggesting Microsoft is doing real yearly OS updates. This is needed because of the competition. Google usually does two new Android major releases in a year. Apple does one iOS and about one MacOS release every year.
Microsoft, on their typical 3-year schedule, simply can't compete with the pace of technology. Even today, not so much. Windows Phone OEMs are already complaining that the HW requirements of Windows Phone 8 is, once again, making it impossible to compete head-to-head with Android and iOS. I don't disagree.. though part of that is the artificial split between Windows Phone and Windows on tablets; iOS and Android don't do desktop, but they do tablet and phone. On Android, apps routinely adjust themselves for the different screen resolutions, so adding new ones is no big deal. Windows has a small number of very specific requirements, and hardware is changing faster than those numbers.
Sure... really long term, we're doomed either way. If there's the critical mass for a Big Crunch (these days, it's a question of dark energy vs. dark mass, given that even today, galaxies are still accelerating from one another), everything crunches together, whatever that really means. If not, eventually, all stars die out, and it's the heat death scenario.
But in practical terms, that's a "high class" problem. Any given species on earth is good for a few million years at best. We expect our big brains will automagically solve this problem, but another big dino-killing-sized asteroid could easily offer a different opinion. If we establish multiple home planets, we're trading some many thousands of years for many billions of years, potentially, of humanity.
So the next problem... leaving the Universe for the one next door. If that's a real thing, anyway, as modern physics suggests. And yes, I do like Stephen Baxter's books, thanks for asking :-)