But it is a very significant factor. Increasing the output resolution (either through pure display, or by supersample antialiasing*) gives you a linear decrease in performance. Double the number of pixels, halve your framerate. So if a $200, 3840x2160 monitor were to come out tomorrow, most gamers would be getting about 15fps using the same hardware and settings they are now.
* SSAA uses larger buffers for everything - color buffers, z-buffers, stencil buffers, etc. The more common MSAA (multisample antialiasing) algorithm only uses larger depth buffers (and maybe stencil?), but uses a normal-sized color buffer. This removes aliasing from polygon edges, but not from within textures - but by running much faster, it's an acceptable tradeoff.
Yes, but at this point AMD needed the publicity win. I'm on a few gaming-oriented sites, and opinion on AMD has pretty much pulled a 180 from a year ago. The standard responses have gone from "they're doomed" and "they suck" to "they're winning in some areas" and genuine interest in their future plans. Bulldozer CPUs aren't popular, except for some video editors, but their GPUs are on the upswing (they're fast going from "OK but cheap" to "good and cheap") and there's growing interest in Jaguar chips in low-power stuff (going against Intel's oft-neglected Atom instead of Core probably helps AMD gain footing). Even their APUs are considered good for ultra-low-budget gaming PCs. Add the hype about Mantle (which I don't believe), the hype about TrueAudio (which I might believe), and regular news of them hiring back some "famous" CPU designer, and they're winning in the publicity war.
I'm actually just as impressed with their business wins of late. They've gone from posting massive losses with no signs of anything on the horizon to getting all the processors for the next-gen consoles except the Wii U's CPU, plus a heavy feature in the new Mac Pro, plus a growing tablet side. And while Bulldozer still seems to be an overall failure, GCN is very competitive and Jaguar seems to be pretty powerful.
If they can fix the IPC problems with Bulldozer, or otherwise get a decently competitive desktop CPU, they would basically be neck-and-neck with Intel again.
There's a difference between "supported" and "actually usable".
The 780 has a theoretical single-precision compute rate of 4.0TFLOPS, comparable to the Titan's 4.5TFLOPS. Go up to double-precision, and the 780 plummets to 165GFLOPS (1/24th the power), while the Titan remains high at 1.5TFLOPS (1/3rd the power).
I'm not sure what the exact reason for that discrepancy in performance is, whether it's an actual hardware difference, some hardware being disabled during binning, or even just a driver change (although someone would have patched drivers to work around it if it were just a driver thing by now), but it is a massive performance difference.
He actually brings this up in TFA. He makes a note of two things that are "beyond the scope of this article": verifying that the compiler is not inserting any backdoors, and verifying that the source code has no security holes.
As a direct continuation of the original series, it hits the mark. Same sort of story, same sort of style.
Whether that is a good or a bad thing is up to you, but if you like the original series, you'll like this, and if you hated it, you'll hate this.
Personally, I think it did well by doing a character-oriented and fundamentally optimistic story. Far, far too much science fiction these days is depressing - grimdark futures, endless dystopias, and a constant cycle of getting darker and edgier. It avoided both that, and the second persistant flaw, overly raised stakes - at worst, the Enterprise was in danger, but never more, and rarely even that. While this continuation failed to correct the third major flaw I see in modern scifi - the use of pseudoscience and handwaved technology as magic, failing to try to actually predict or speculate where existing science and technology are taking us - it did so in devotion to the original series. It would not have been Star Trek without the often unscientific plots.
That is really the only failure of this attempt. They tried to exactly continue Trek, instead of trying to improve upon it. I'm pretty sure they spent quite a bit of time trying to get the CGI Enterprise to look like a plastic model, making the lighting look like an old TV show. Hell, they even cut to commercial breaks, even used 4:3 instead of 16:9. The only signs this was made in 2013 instead of 1973 are the occasional nods to The Next Generation.
Perhaps expecting them to try to do *better* than Star Trek isn't realistic. Maybe they decided nostalgia was better than some marginal technical improvements. I don't know.
I'm legitimately impressed with everything else. By fan-vid standards, this was amazing. It was even entertaining as a work on its own. I'll probably watch the next episode, if they make one. I just feel like there's still a void out there that needs filling - a piece of optimistic science fiction that actually cares for the science in its own right.
Someone's probably going to tell me that if I want something like that, I should write it myself. Maybe I will.
Valve plays an incredibly long game compared to most tech companies (hell, most companies, period). They started Steam because they could see where constantly-increasing bandwidth was leading. They missed on some of the particulars, but by getting the main point correct early on, they were able to gather the momentum to overcome minor obstacles before anyone else could seize initiative. So not only did they avoid being tied down to another company's proprietary platform, but they managed to become the de facto digital distribution system while still being a relatively minor player.
SteamOS is a defensive move. They're concerned that Microsoft may lose its Windows dominance, or might try to move it to an Apple-like locked store (they sort of have, with RT). So they ported Steam and their own games to both OS X and Linux.
That was enough to spur an initial kick of OS X games following after them. It's not nearly universal now, but it's respectable, and growing.
Linux didn't get the same kick, mainly because they don't have as much market share. So Valve is giving it more support, and perhaps more importantly, lending it a more prestigious (among gamers) brand name.
Will it be a success? Perhaps. At the very least, it's enough a threat to Microsoft that they're not going to try to take over the digital distribution market, because if they do, Valve will just drop Steam on Windows and enough publishers will follow them to wherever they lead that Microsoft will ultimately have lost. So in one sense, it's a deterrent. But it could become a legitimate gaming platform in its own right, particularly if they get enough console-like games for Steam Machines to go up against the PS4/Xb1 in the coming generation.
Yes, I agree that Apple isn't back at '95 levels yet. But the point is that they're sliding in that direction. They're gaining more and more models that don't seem to serve a strong purpose.
While the iPad line does boil down to a few simple choices, the *naming* of that line does not in any way indicate it. The Google comparison was just a demonstration that Google is managing a more Jobs-like naming convention than Apple is.
Remember the mid-90s, when Apple had dozens upon dozens of Macintoshes, Power Macs, Quadras, and so on? And how one of the first things Jobs did when he returned was slash all of those, which put them back on the road to their current success?
Yeah, they're up to four different iPads now, all currently being sold. The iPad Mini, "iPad Mini with Retina", iPad 2, and iPad Air, and I'm sure they still have some "Fourth-Generation iPads" to sell off. Each of these has a few variants for WiFi/3G and storage. And they also have a trio of iPhones - the 4S, 5C and 5S - again with storage capacity variations.
On the desktop, a pair of laptops (the Macbook Air and Macbook Pro) with a few size options, and trio of desktops (Mini, iMac (two sizes) and Mac Pro) has worked pretty well for them. They really need to cut down on their other models - using the old iPhone as the "cheap" model worked, discontinuing the old one in favor of a low-cost second model would also have worked, but as it is I see little purpose to keeping both the 4S and 5C around. And for the iPad? A Mini and a Pro would have been fine. Google is actually being smarter than they are on this - they have a Nexus 7 and a Nexus 10, updated as needed. Clear product differentiation - you want a small, cheap tablet? Nexus 7. Larger and more powerful tablet? Nexus 10. Apple is less clear - their high-end "Mini" costs the same as their low-end "full-size". They could probably make the iPad models make sense (iPad Mini, iPad Pro Mini, iPad, iPad Pro), but the way they currently are is crap.
Not a bad analysis, but I think it misses something. Right now, the watch phones have too poor a battery life to have significant processing power. The watch might make a decent display, but that's about all it can do with any quality. So it's in the same realm as the glasses.
We're going to need person-area networks. Put a big battery and a powerful computer system in your pocket, have it connect to the watch and glasses for user I/O. Problem is, at that point you may as well slap a screen on the computer part, and then you've got a full smartphone, which reduces the necessity of the other two.
The BC2 campaign (PC gamer, don't have BC1) was actually pretty damn fun. It was mostly a CoD clone, but it did enough differently (cover destruction, writing was a lot of witty banter and less drama and shouting) that it was entertaining.
Also, I've put roughly 400 hours into BF2. Less than a quarter of that is in actual online multiplayer - rest is either LAN, or against bots (again, mods fix the bots pretty well). Is it the "right" way to play? Maybe not by your definition. But I'm having fun, which means it's "a" right way to play the game.
EA said with BF3 "hey, how do we get the CoD players?" rather than asking the question "hey, how do we improve Battlefield?" Personally, I'm done with the franchise. Nothing more frustrating than trying to win a team game when 90% of players are stat whoring.
Precisely. BF3 is just a CoD clone, and BF4 is aiming to be the same.
The sad part is, it isn't even a good CoD clone. I bought it off the EA Humble Bundle, just to check it out. The singleplayer campaign is weak - the levels are perfectly linear (even CoD has better levels), the guns all work the same, the spectacle is weak (in a genre sometimes called "spectacle shooter"), the special sections are boring (MW1 gunship level? Fun. BF3 jet level? It's a turret level with seeker missiles) and even the writing is worse. Bad Company 2 was a better single-player game - it had some genuinely funny characters, and the story seemed deliberately silly, like they were poking a bit of fun at CoD and all the Clancy or Clancy-esque "military action thrillers".
BF3 took the same hackneyed "[bad guys] have a [superweapon] and plan to blow up AMERICA" plot, and instead of playing it off as funny, they tried to do it straight-faced and seriously. Despite that, they still used the SAME GODDAMN MODEL for the superweapon - the "Carrington device" (nice bit of writing - in WW2 they wouldn't have called them EMPs, after all) became a suitcase nuke, despite being about the size of a large thermos with cooling fins. But hey, they made the enemy Iran this time! Slightly different shade of brown and sightly different accent!
I didn't even bother with the multiplayer - my connection was spotty the one weekend I played it, and they don't seem to have an option for playing against bots. Seems I did not miss much by doing so.
Well, Intel and VIA have such things integrated into their processors now. Unfortunately, they (at least Intel - not sure how VIA's implementation worked) decided to whiten the data in firmware - you run a certain instruction that gives you a "random" number, instead of just polling the diode. With all the current furor over the NSA stuff, many people are claiming that it *is* hacked.
The universe doesn't need us to survive. But *we* need us to survive.
Sure, that's probably just evolution talking - species who don't consider their species important enough to protect probably don't last too long. But just because we live in a cold, uncaring universe that would just as soon kill us and forget we ever existed, doesn't mean we have to accept it. Just because it's humanity, alone, against a universe that is larger than we can even comprehend, let alone conquer, doesn't mean we need to just give up.
How much would such a project cost? In the trillions of dollars, easily.
We can afford that. The world has a collective GDP of 70 trillion. Both the US and the EU produce about 15 trillion individually. Hell, we spent nearly a trillion on Iraq alone, and that's not counting the long-term costs of that war (because you know nobody factored that in when they started that war). Even if it costs us a trillion dollars per year for a generation, we can totally afford that. And given the potential costs of failing to do so, I'm not sure we can afford not to.
No, we need human spaceflight now more than ever. We need a self-sustaining colony, off-planet, ASAP. I don't care if it's a lunar colony, Mars, an asteroid, or even a city-sized space station at a Lagrange point, as long as it can sustain itself indefinitely. There's relatively little scientific gain to be made from this, but that's not why we should do it.
Tell me, what is rule #1 of computing? "Always keep a backup". Well, right now we're running on a single, non-redundant biosphere, and we seem to be actively sabotaging it. But even outside human-caused damage, there are easily dozens of things that could wipe out our planet's ability to sustain human life. Asteroids. Supervolcanoes. Major climate shift of any sort - anthropogenic or natural, warming or cooling. Oh, and don't forget we have enough nukes to murder ourselves quite efficiently.
Are these slim chances? Yes, but not as slim as I'd like, and considering that a lack of redundancy means the complete annihilation of the human race, I think we can afford a few trillion dollars to get things running.
No, no, no need to waste fuel sending them into a long-term stable orbit. A low orbit that decays in, say, a decade, or even just a year or so, and disintegrates on reentry, ought to be more than enough.
An app that allows you to write and publish a tweet through it, or an app that allows you to tweet things at your choosing (eg. "share this with your Twitter followers"), requires the same permissions. And that is probably what most people expected - the app would, say, have a listing of all the booths and such, and allow you, among other things, to send "I'm at the _____" tweets. Nobody really expects an app to just send out advertising tweets. This is perhaps a flaw in the permissions system, having two very different things be bound to the same permission, but there's not really a technical way to tell if a tweet was written (or at least approved) by the user, or if it's automatic.
Fortunately, there is likely a legal solution. According to the Twitter TOS, section 9, subsection v, which prohibits spamming, this may be cause to ban NYCC from Twitter. It may also fall under their rules regarding impersonation or automation. And even if their rules do not currently cover this, I expect Twitter will add rules to prevent such in the future, because it's not in their interests to have unwarranted and unpaid-for advertising being fraudulently sent through users' accounts.
Yes, most of what the NIF does is actually weapons research. The fusion power stuff seems to be mainly a political ploy to get the Democrats to sign off on it - they're never going to get actual fusion power, meaning actually turning this power back into electricity, at NIF, and unless they know something big I don't, I doubt they ever will at any inertial confinement reactor. I only hope that they're able to do some solid fundamental research for fusion power using this.
ATI/AMD has actually been consistent for several years now - they're literally just-now changing their scheme
The old system was a four-digit number. First digit is generation - a 7950 is newer than a 6870, and way newer than a 4830 or a 2600. The next two digits are how powerful it is within the generation - roughly, the second digit is the market segment, and the third is which model within that segment, but that's rough. They did tend to inflate numbers over time - the top-end single-GPU cards of each generation were the 2900 XT, the 3870, the 4890, the 5870, the 6970, and the 7970GE. Put simply, if you sort by the middle two digits within a generation, you also order by both power and price.
The fourth digit is always a zero. Always. I don't know why they bother.
Sometimes there's a suffix. "X2" used to mean it's a dual-GPU card, cramming two processors onto one board, but now those get a separate model number (they also only do that for the top GPU now, because they've found it's not worth it to use two weaker processors). "GE" or "Gigahertz Edition" was used on some 7xxx models, because Nvidia beat them pretty heavily with their 6xx series release so AMD had to rush out some cards that were essentially overclocked high enough to beat them. "Eyefinity Edition" used to be a thing, mainly it just meant it had a shitload of mini-DP outputs so you could do 3x2 six-monitor surround setups, which AMD was (and is) trying to push. And there were some "Pro" or "XT" models early on, but those were not significant.
Now forget all that, because they're throwing a new one out.
It's now a two-part thing, rather like what Intel does with their CPUs. "R9" is their "Enthusiast" series, for people with too much money. Within that, you have six models: the 270, 270X, 280, 280X, 290 and 290X. They haven't fully clarified things, but it seems that the X models are the "full" chip, while the non-X model has some cores binned off and slightly lower clocks. Other than that, it's a fairly straightforward list - the 290 beats the 280X beats the 280 beats the 270X and so on. Under those are the "R7" "gamer" series, which so far has the 240 through 260X, and an R5 230 model is listed on Wikipedia even though I've not seen it mentioned elsewhere.
Sadly, it's still a bit more complicated. See, some of the "new" ones are just the old ones relabeled. They're all the same fundamental "Graphics Core Next" architecture, but some of them have the new audio DSP stuff people are excited about. And it's not even a simple "everything under this is an old one lacking new features" - the 290X and 260X have the new stuff, but the 280X and 270X do not. And it gets worse still, because the 260X actually is a rebadge, it's just that they're enabling some hardware functionality now (the 290X actually is a genuine new chip as far as anyone can tell). So far, everything is 2__, so I would assume the first digit in this case is still the generation.
Oh, and there actually are some 8xxx series cards. There were some mobile models released (forgot to mention - an M suffix means mobile, and you can't directly compare numbers between them. A 7870 and 7870M are not the same.), and it looks like some OEM-only models on the desktop.
But yeah, it is a bit daunting at first, especially since they're transitioning to a new schema very abruptly (people were expecting an 8xxx and 9xxx series before a new schema). But not much has really changed - you just need to figure out which number is the generation, and which is the market segment, and you're good.
Read TFA, turns out summary was accurate for once.
I also realized that there's a third breakeven point that needs to be reached - the economic one, where the amount of money spent on fuel and operations is exceeded by the profits generated. NIF-type "reactors" may have some issues there (with the expensive "lens" of sorts that gets destroyed during the process), but even for them, I think just getting to the practical breakeven is harder than getting to the economic one (and for more continuous designs, it's even easier to get from practical breakeven to economic breakeven).
Nonetheless, I think this is a decent milestone. While the reactor design itself is unlikely to ever break even, hopefully they're at least learning enough about efficiently triggering a fusion reaction that they can apply it to more productive designs, whether they be tokamaks or some other design.
Scientific "break even", or practical "break even"
on
Fusion Reactor Breaks Even
·
· Score: 4, Interesting
One of the big criticisms of the NIF is that the design is basically unsuited to capture more than a slim percentage of the energy released. It's good for weapons research because it works vaguely the same way a bomb does - rapidly compressing fuel in a burst. But it doesn't really have a mechanism for capturing that energy, unlike tokamak-based designs.
Based on the summary (still reading TFA itself), it sounds like they broke even in terms of the energy input into the fuel being less than the total amount released from the reaction. But to be a self-sustaining, practical fusion power source, it needs to extend that two directions - first, by breaking even in terms of power into the entire system being less than that released, and second by breaking even in terms of power captured, not just power generated. The former is straightforward - more efficient lasers, more efficient reactions - but, and this is from a non-engineer's perspective, I don't think the latter will be simple.
Midlevel? For non-gaming usage, perhaps. For gaming they're strictly low-end, or unusable.
There are three Intel GPUs on the desktop side - the HD 4600, the Iris 5100, and the Iris Pro 5200. In raw processing power, the first gets you 430GFLOPS, and the latter two get 830GFLOPS. For comparison, the *weakest* GPU in these Steam Machines pumps out 1880GFLOPS, and the top end maxes out around 4.5 TFLOPS.
And that's a spec that's biased towards Intel - they're more compute-heavy than bandwidth-heavy, and unfortunately most graphics tasks are bound by memory bandwidth. For Intel, the first two have a mere 25.6 GB/S of bandwidth, with Iris Pro adding an on-chip cache to bring it up to 75GB/S. But even the GeForce 660 beats that at 144GB/S, and the Titan doubles that. For those who may not be familiar, the 660 Ti (and the new-gen rebadge-with-enhancements, the 760) was considered a good medium-end card, with the vanilla 660 being for those a bit more budget-minded. The Titan, of course, is their "luxury" card, costing a full $1000, but it's currently the most powerful single-GPU card, period.
That's just their theoretical performance - the real test, of course, is actual game benchmarks. Nvidia is currently the best at getting the most performance from their hardware in actual games. AMD has more raw power, but their drivers aren't as efficient so Nvidia beats them more often than not. Intel's far worse than either - while Iris Pro should be able to go head-to-head with a GeForce 650, it actually tends to benchmark closer to the GeForce 640. Go look it up on Anandtech, if you're interested.
Now, is it impressive how much power Intel managed to get out of an IGPU? Yeah, it is. Honestly, I would be interested in seeing them scale up the design further - go from 40 EUs to 200 EUs, bolt on the memory controller from the Xeon Phi, and sell it as a dedicated card. Might be something they can do with the 22nm fabs once they move to 14nm? But in any case, calling their current offerings "medium-end" is misleading at best, and downright wrong at worst.
Quite simply, this product is not for you. This product is for people who want a console-like simplicity of installation and use, but want the power and library of a PC and are willing to pay for it.
I do not think the high-end ones are going to be at all successful. I think the low-end ones may be useful for streaming games to a TV, since most Steam gamers already have a powerful desktop, and this will bypass the Windows/Linux incompatibilities. And, since it's there, it will probably see more and more ports coming out. But it's not going to be a huge success, at least at first. They're trying to topple Windows - we've been trying to do that for how long now?
Now, the controller is more exciting. I'd buy one of those. But the Steam Machines, even SteamOS, have minimal initial utility.
Slashdot Mirror
But it is a very significant factor. Increasing the output resolution (either through pure display, or by supersample antialiasing*) gives you a linear decrease in performance. Double the number of pixels, halve your framerate. So if a $200, 3840x2160 monitor were to come out tomorrow, most gamers would be getting about 15fps using the same hardware and settings they are now.
* SSAA uses larger buffers for everything - color buffers, z-buffers, stencil buffers, etc. The more common MSAA (multisample antialiasing) algorithm only uses larger depth buffers (and maybe stencil?), but uses a normal-sized color buffer. This removes aliasing from polygon edges, but not from within textures - but by running much faster, it's an acceptable tradeoff.
Revised headline: "Slashdot editors still drunk at work, approving spam".
Yes, but at this point AMD needed the publicity win. I'm on a few gaming-oriented sites, and opinion on AMD has pretty much pulled a 180 from a year ago. The standard responses have gone from "they're doomed" and "they suck" to "they're winning in some areas" and genuine interest in their future plans. Bulldozer CPUs aren't popular, except for some video editors, but their GPUs are on the upswing (they're fast going from "OK but cheap" to "good and cheap") and there's growing interest in Jaguar chips in low-power stuff (going against Intel's oft-neglected Atom instead of Core probably helps AMD gain footing). Even their APUs are considered good for ultra-low-budget gaming PCs. Add the hype about Mantle (which I don't believe), the hype about TrueAudio (which I might believe), and regular news of them hiring back some "famous" CPU designer, and they're winning in the publicity war.
I'm actually just as impressed with their business wins of late. They've gone from posting massive losses with no signs of anything on the horizon to getting all the processors for the next-gen consoles except the Wii U's CPU, plus a heavy feature in the new Mac Pro, plus a growing tablet side. And while Bulldozer still seems to be an overall failure, GCN is very competitive and Jaguar seems to be pretty powerful.
If they can fix the IPC problems with Bulldozer, or otherwise get a decently competitive desktop CPU, they would basically be neck-and-neck with Intel again.
There's a difference between "supported" and "actually usable".
The 780 has a theoretical single-precision compute rate of 4.0TFLOPS, comparable to the Titan's 4.5TFLOPS. Go up to double-precision, and the 780 plummets to 165GFLOPS (1/24th the power), while the Titan remains high at 1.5TFLOPS (1/3rd the power).
I'm not sure what the exact reason for that discrepancy in performance is, whether it's an actual hardware difference, some hardware being disabled during binning, or even just a driver change (although someone would have patched drivers to work around it if it were just a driver thing by now), but it is a massive performance difference.
He actually brings this up in TFA. He makes a note of two things that are "beyond the scope of this article": verifying that the compiler is not inserting any backdoors, and verifying that the source code has no security holes.
As a direct continuation of the original series, it hits the mark. Same sort of story, same sort of style.
Whether that is a good or a bad thing is up to you, but if you like the original series, you'll like this, and if you hated it, you'll hate this.
Personally, I think it did well by doing a character-oriented and fundamentally optimistic story. Far, far too much science fiction these days is depressing - grimdark futures, endless dystopias, and a constant cycle of getting darker and edgier. It avoided both that, and the second persistant flaw, overly raised stakes - at worst, the Enterprise was in danger, but never more, and rarely even that. While this continuation failed to correct the third major flaw I see in modern scifi - the use of pseudoscience and handwaved technology as magic, failing to try to actually predict or speculate where existing science and technology are taking us - it did so in devotion to the original series. It would not have been Star Trek without the often unscientific plots.
That is really the only failure of this attempt. They tried to exactly continue Trek, instead of trying to improve upon it. I'm pretty sure they spent quite a bit of time trying to get the CGI Enterprise to look like a plastic model, making the lighting look like an old TV show. Hell, they even cut to commercial breaks, even used 4:3 instead of 16:9. The only signs this was made in 2013 instead of 1973 are the occasional nods to The Next Generation.
Perhaps expecting them to try to do *better* than Star Trek isn't realistic. Maybe they decided nostalgia was better than some marginal technical improvements. I don't know.
I'm legitimately impressed with everything else. By fan-vid standards, this was amazing. It was even entertaining as a work on its own. I'll probably watch the next episode, if they make one. I just feel like there's still a void out there that needs filling - a piece of optimistic science fiction that actually cares for the science in its own right.
Someone's probably going to tell me that if I want something like that, I should write it myself. Maybe I will.
Slowly.
Valve plays an incredibly long game compared to most tech companies (hell, most companies, period). They started Steam because they could see where constantly-increasing bandwidth was leading. They missed on some of the particulars, but by getting the main point correct early on, they were able to gather the momentum to overcome minor obstacles before anyone else could seize initiative. So not only did they avoid being tied down to another company's proprietary platform, but they managed to become the de facto digital distribution system while still being a relatively minor player.
SteamOS is a defensive move. They're concerned that Microsoft may lose its Windows dominance, or might try to move it to an Apple-like locked store (they sort of have, with RT). So they ported Steam and their own games to both OS X and Linux.
That was enough to spur an initial kick of OS X games following after them. It's not nearly universal now, but it's respectable, and growing.
Linux didn't get the same kick, mainly because they don't have as much market share. So Valve is giving it more support, and perhaps more importantly, lending it a more prestigious (among gamers) brand name.
Will it be a success? Perhaps. At the very least, it's enough a threat to Microsoft that they're not going to try to take over the digital distribution market, because if they do, Valve will just drop Steam on Windows and enough publishers will follow them to wherever they lead that Microsoft will ultimately have lost. So in one sense, it's a deterrent. But it could become a legitimate gaming platform in its own right, particularly if they get enough console-like games for Steam Machines to go up against the PS4/Xb1 in the coming generation.
Yes, I agree that Apple isn't back at '95 levels yet. But the point is that they're sliding in that direction. They're gaining more and more models that don't seem to serve a strong purpose.
While the iPad line does boil down to a few simple choices, the *naming* of that line does not in any way indicate it. The Google comparison was just a demonstration that Google is managing a more Jobs-like naming convention than Apple is.
Remember the mid-90s, when Apple had dozens upon dozens of Macintoshes, Power Macs, Quadras, and so on? And how one of the first things Jobs did when he returned was slash all of those, which put them back on the road to their current success?
Yeah, they're up to four different iPads now, all currently being sold. The iPad Mini, "iPad Mini with Retina", iPad 2, and iPad Air, and I'm sure they still have some "Fourth-Generation iPads" to sell off. Each of these has a few variants for WiFi/3G and storage. And they also have a trio of iPhones - the 4S, 5C and 5S - again with storage capacity variations.
On the desktop, a pair of laptops (the Macbook Air and Macbook Pro) with a few size options, and trio of desktops (Mini, iMac (two sizes) and Mac Pro) has worked pretty well for them. They really need to cut down on their other models - using the old iPhone as the "cheap" model worked, discontinuing the old one in favor of a low-cost second model would also have worked, but as it is I see little purpose to keeping both the 4S and 5C around. And for the iPad? A Mini and a Pro would have been fine. Google is actually being smarter than they are on this - they have a Nexus 7 and a Nexus 10, updated as needed. Clear product differentiation - you want a small, cheap tablet? Nexus 7. Larger and more powerful tablet? Nexus 10. Apple is less clear - their high-end "Mini" costs the same as their low-end "full-size". They could probably make the iPad models make sense (iPad Mini, iPad Pro Mini, iPad, iPad Pro), but the way they currently are is crap.
Not a bad analysis, but I think it misses something. Right now, the watch phones have too poor a battery life to have significant processing power. The watch might make a decent display, but that's about all it can do with any quality. So it's in the same realm as the glasses.
We're going to need person-area networks. Put a big battery and a powerful computer system in your pocket, have it connect to the watch and glasses for user I/O. Problem is, at that point you may as well slap a screen on the computer part, and then you've got a full smartphone, which reduces the necessity of the other two.
The BC2 campaign (PC gamer, don't have BC1) was actually pretty damn fun. It was mostly a CoD clone, but it did enough differently (cover destruction, writing was a lot of witty banter and less drama and shouting) that it was entertaining.
Also, I've put roughly 400 hours into BF2. Less than a quarter of that is in actual online multiplayer - rest is either LAN, or against bots (again, mods fix the bots pretty well). Is it the "right" way to play? Maybe not by your definition. But I'm having fun, which means it's "a" right way to play the game.
EA said with BF3 "hey, how do we get the CoD players?" rather than asking the question "hey, how do we improve Battlefield?" Personally, I'm done with the franchise. Nothing more frustrating than trying to win a team game when 90% of players are stat whoring.
Precisely. BF3 is just a CoD clone, and BF4 is aiming to be the same.
The sad part is, it isn't even a good CoD clone. I bought it off the EA Humble Bundle, just to check it out. The singleplayer campaign is weak - the levels are perfectly linear (even CoD has better levels), the guns all work the same, the spectacle is weak (in a genre sometimes called "spectacle shooter"), the special sections are boring (MW1 gunship level? Fun. BF3 jet level? It's a turret level with seeker missiles) and even the writing is worse. Bad Company 2 was a better single-player game - it had some genuinely funny characters, and the story seemed deliberately silly, like they were poking a bit of fun at CoD and all the Clancy or Clancy-esque "military action thrillers".
BF3 took the same hackneyed "[bad guys] have a [superweapon] and plan to blow up AMERICA" plot, and instead of playing it off as funny, they tried to do it straight-faced and seriously. Despite that, they still used the SAME GODDAMN MODEL for the superweapon - the "Carrington device" (nice bit of writing - in WW2 they wouldn't have called them EMPs, after all) became a suitcase nuke, despite being about the size of a large thermos with cooling fins. But hey, they made the enemy Iran this time! Slightly different shade of brown and sightly different accent!
I didn't even bother with the multiplayer - my connection was spotty the one weekend I played it, and they don't seem to have an option for playing against bots. Seems I did not miss much by doing so.
Well, Intel and VIA have such things integrated into their processors now. Unfortunately, they (at least Intel - not sure how VIA's implementation worked) decided to whiten the data in firmware - you run a certain instruction that gives you a "random" number, instead of just polling the diode. With all the current furor over the NSA stuff, many people are claiming that it *is* hacked.
The universe doesn't need us to survive. But *we* need us to survive.
Sure, that's probably just evolution talking - species who don't consider their species important enough to protect probably don't last too long. But just because we live in a cold, uncaring universe that would just as soon kill us and forget we ever existed, doesn't mean we have to accept it. Just because it's humanity, alone, against a universe that is larger than we can even comprehend, let alone conquer, doesn't mean we need to just give up.
How much would such a project cost? In the trillions of dollars, easily.
We can afford that. The world has a collective GDP of 70 trillion. Both the US and the EU produce about 15 trillion individually. Hell, we spent nearly a trillion on Iraq alone, and that's not counting the long-term costs of that war (because you know nobody factored that in when they started that war). Even if it costs us a trillion dollars per year for a generation, we can totally afford that. And given the potential costs of failing to do so, I'm not sure we can afford not to.
No, we need human spaceflight now more than ever. We need a self-sustaining colony, off-planet, ASAP. I don't care if it's a lunar colony, Mars, an asteroid, or even a city-sized space station at a Lagrange point, as long as it can sustain itself indefinitely. There's relatively little scientific gain to be made from this, but that's not why we should do it.
Tell me, what is rule #1 of computing? "Always keep a backup". Well, right now we're running on a single, non-redundant biosphere, and we seem to be actively sabotaging it. But even outside human-caused damage, there are easily dozens of things that could wipe out our planet's ability to sustain human life. Asteroids. Supervolcanoes. Major climate shift of any sort - anthropogenic or natural, warming or cooling. Oh, and don't forget we have enough nukes to murder ourselves quite efficiently.
Are these slim chances? Yes, but not as slim as I'd like, and considering that a lack of redundancy means the complete annihilation of the human race, I think we can afford a few trillion dollars to get things running.
No, no, no need to waste fuel sending them into a long-term stable orbit. A low orbit that decays in, say, a decade, or even just a year or so, and disintegrates on reentry, ought to be more than enough.
An app that allows you to write and publish a tweet through it, or an app that allows you to tweet things at your choosing (eg. "share this with your Twitter followers"), requires the same permissions. And that is probably what most people expected - the app would, say, have a listing of all the booths and such, and allow you, among other things, to send "I'm at the _____" tweets. Nobody really expects an app to just send out advertising tweets. This is perhaps a flaw in the permissions system, having two very different things be bound to the same permission, but there's not really a technical way to tell if a tweet was written (or at least approved) by the user, or if it's automatic.
Fortunately, there is likely a legal solution. According to the Twitter TOS, section 9, subsection v, which prohibits spamming, this may be cause to ban NYCC from Twitter. It may also fall under their rules regarding impersonation or automation. And even if their rules do not currently cover this, I expect Twitter will add rules to prevent such in the future, because it's not in their interests to have unwarranted and unpaid-for advertising being fraudulently sent through users' accounts.
Yes, most of what the NIF does is actually weapons research. The fusion power stuff seems to be mainly a political ploy to get the Democrats to sign off on it - they're never going to get actual fusion power, meaning actually turning this power back into electricity, at NIF, and unless they know something big I don't, I doubt they ever will at any inertial confinement reactor. I only hope that they're able to do some solid fundamental research for fusion power using this.
ATI/AMD has actually been consistent for several years now - they're literally just-now changing their scheme
The old system was a four-digit number. First digit is generation - a 7950 is newer than a 6870, and way newer than a 4830 or a 2600. The next two digits are how powerful it is within the generation - roughly, the second digit is the market segment, and the third is which model within that segment, but that's rough. They did tend to inflate numbers over time - the top-end single-GPU cards of each generation were the 2900 XT, the 3870, the 4890, the 5870, the 6970, and the 7970GE. Put simply, if you sort by the middle two digits within a generation, you also order by both power and price.
The fourth digit is always a zero. Always. I don't know why they bother.
Sometimes there's a suffix. "X2" used to mean it's a dual-GPU card, cramming two processors onto one board, but now those get a separate model number (they also only do that for the top GPU now, because they've found it's not worth it to use two weaker processors). "GE" or "Gigahertz Edition" was used on some 7xxx models, because Nvidia beat them pretty heavily with their 6xx series release so AMD had to rush out some cards that were essentially overclocked high enough to beat them. "Eyefinity Edition" used to be a thing, mainly it just meant it had a shitload of mini-DP outputs so you could do 3x2 six-monitor surround setups, which AMD was (and is) trying to push. And there were some "Pro" or "XT" models early on, but those were not significant.
Now forget all that, because they're throwing a new one out.
It's now a two-part thing, rather like what Intel does with their CPUs. "R9" is their "Enthusiast" series, for people with too much money. Within that, you have six models: the 270, 270X, 280, 280X, 290 and 290X. They haven't fully clarified things, but it seems that the X models are the "full" chip, while the non-X model has some cores binned off and slightly lower clocks. Other than that, it's a fairly straightforward list - the 290 beats the 280X beats the 280 beats the 270X and so on. Under those are the "R7" "gamer" series, which so far has the 240 through 260X, and an R5 230 model is listed on Wikipedia even though I've not seen it mentioned elsewhere.
Sadly, it's still a bit more complicated. See, some of the "new" ones are just the old ones relabeled. They're all the same fundamental "Graphics Core Next" architecture, but some of them have the new audio DSP stuff people are excited about. And it's not even a simple "everything under this is an old one lacking new features" - the 290X and 260X have the new stuff, but the 280X and 270X do not. And it gets worse still, because the 260X actually is a rebadge, it's just that they're enabling some hardware functionality now (the 290X actually is a genuine new chip as far as anyone can tell). So far, everything is 2__, so I would assume the first digit in this case is still the generation.
Oh, and there actually are some 8xxx series cards. There were some mobile models released (forgot to mention - an M suffix means mobile, and you can't directly compare numbers between them. A 7870 and 7870M are not the same.), and it looks like some OEM-only models on the desktop.
But yeah, it is a bit daunting at first, especially since they're transitioning to a new schema very abruptly (people were expecting an 8xxx and 9xxx series before a new schema). But not much has really changed - you just need to figure out which number is the generation, and which is the market segment, and you're good.
Read TFA, turns out summary was accurate for once.
I also realized that there's a third breakeven point that needs to be reached - the economic one, where the amount of money spent on fuel and operations is exceeded by the profits generated. NIF-type "reactors" may have some issues there (with the expensive "lens" of sorts that gets destroyed during the process), but even for them, I think just getting to the practical breakeven is harder than getting to the economic one (and for more continuous designs, it's even easier to get from practical breakeven to economic breakeven).
Nonetheless, I think this is a decent milestone. While the reactor design itself is unlikely to ever break even, hopefully they're at least learning enough about efficiently triggering a fusion reaction that they can apply it to more productive designs, whether they be tokamaks or some other design.
One of the big criticisms of the NIF is that the design is basically unsuited to capture more than a slim percentage of the energy released. It's good for weapons research because it works vaguely the same way a bomb does - rapidly compressing fuel in a burst. But it doesn't really have a mechanism for capturing that energy, unlike tokamak-based designs.
Based on the summary (still reading TFA itself), it sounds like they broke even in terms of the energy input into the fuel being less than the total amount released from the reaction. But to be a self-sustaining, practical fusion power source, it needs to extend that two directions - first, by breaking even in terms of power into the entire system being less than that released, and second by breaking even in terms of power captured, not just power generated. The former is straightforward - more efficient lasers, more efficient reactions - but, and this is from a non-engineer's perspective, I don't think the latter will be simple.
Midlevel? For non-gaming usage, perhaps. For gaming they're strictly low-end, or unusable.
There are three Intel GPUs on the desktop side - the HD 4600, the Iris 5100, and the Iris Pro 5200. In raw processing power, the first gets you 430GFLOPS, and the latter two get 830GFLOPS. For comparison, the *weakest* GPU in these Steam Machines pumps out 1880GFLOPS, and the top end maxes out around 4.5 TFLOPS.
And that's a spec that's biased towards Intel - they're more compute-heavy than bandwidth-heavy, and unfortunately most graphics tasks are bound by memory bandwidth. For Intel, the first two have a mere 25.6 GB/S of bandwidth, with Iris Pro adding an on-chip cache to bring it up to 75GB/S. But even the GeForce 660 beats that at 144GB/S, and the Titan doubles that. For those who may not be familiar, the 660 Ti (and the new-gen rebadge-with-enhancements, the 760) was considered a good medium-end card, with the vanilla 660 being for those a bit more budget-minded. The Titan, of course, is their "luxury" card, costing a full $1000, but it's currently the most powerful single-GPU card, period.
That's just their theoretical performance - the real test, of course, is actual game benchmarks. Nvidia is currently the best at getting the most performance from their hardware in actual games. AMD has more raw power, but their drivers aren't as efficient so Nvidia beats them more often than not. Intel's far worse than either - while Iris Pro should be able to go head-to-head with a GeForce 650, it actually tends to benchmark closer to the GeForce 640. Go look it up on Anandtech, if you're interested.
Now, is it impressive how much power Intel managed to get out of an IGPU? Yeah, it is. Honestly, I would be interested in seeing them scale up the design further - go from 40 EUs to 200 EUs, bolt on the memory controller from the Xeon Phi, and sell it as a dedicated card. Might be something they can do with the 22nm fabs once they move to 14nm? But in any case, calling their current offerings "medium-end" is misleading at best, and downright wrong at worst.
Quite simply, this product is not for you. This product is for people who want a console-like simplicity of installation and use, but want the power and library of a PC and are willing to pay for it.
I do not think the high-end ones are going to be at all successful. I think the low-end ones may be useful for streaming games to a TV, since most Steam gamers already have a powerful desktop, and this will bypass the Windows/Linux incompatibilities. And, since it's there, it will probably see more and more ports coming out. But it's not going to be a huge success, at least at first. They're trying to topple Windows - we've been trying to do that for how long now?
Now, the controller is more exciting. I'd buy one of those. But the Steam Machines, even SteamOS, have minimal initial utility.
If I have learned anything from my childhood, its that the solution to giant monsters is more giant monsters.
Not necessarily true - sometimes the solution to giant monsters is giant robots.