The executives running those companies are probably MBAs. If they know anything about computer hardware, programming, or tying their own shoelaces, it'd be a goddamn miracle. MBAs are, for the most part, some of the stupidest people you'll ever meet.
1) Workstation class hardware is exactly the same as desktop hardware. Same CPUs except with more cores, same FSB architectures, same memory busses. The days when there was a big schism between the capabilities of consumer-class CPUs and server-class CPUs ended years ago: they're almost one and the same now. There are differences in I/O architecture, which is exactly the sort of thing that gaming doesn't really stress.
2/3) I did in fact watch the video, though I'll admit that I didn't have the patience to watch more than the first question in the Q/A session. OK, so there's custom hardware involved... score one for the technical viability of the idea I suppose, while taking one from the economic viability of it. Now you're expecting me to believe that not only will they have high-end GPUs in every machine (which, as others have mentioned, aren't virtualizable in any useful sort of way) but they'll also have custom-fabbed video accelerator boards? Technically possible, perhaps: economically sensible, hardly.
Wow. Just wow.
Kids, don't drink and post.
I still maintain that this simply can't work, and that it's an absolutely braindead money pit of an idea if it's not a total scam.
Idea: let's take the most latency sensitive, computationally demanding, and visually intensive thing you can do with a modern computer and try to apply the thin client model to it.
A single instance of the application in question will demand the full resources of the most powerful PC you can throw at it, but we'll just wave our hands and mutter something about virtualization to convince stupid investors that we have magic at our disposal. Because they are morons and because we put on a good show, they'll believe that you can somehow run many instances of a game on the equivalent of a single PC. We'll also be encoding 720p video in realtime at a quality / bandwidth ratio that no codec today can deliver; this will presumably happen on the same computing hardware that's already running multiple instances of cutting edge 3D games.
Finally, we'll throw in some shit about the iphone, because people can't stop fellating apple lately.
Anyone who believes this is technically feasible, much less economically viable, is fucking *retarded*.
Your
It went something like "Outlook not so good. Neat, next I'll ask it about Exchange Server."
I've never seen a baseless anti-Microsoft sentiment so badly expressed as the steaming little turd you call a
gb2hugbox, aspie.
Thermal IR imaging isn't magic.
First off, "FLIR" is kind of a misnomer. It stands for "Forward Looking InfraRed," but that just means that it, well, looks forward... like a camera. When thermal IR detectors were first invented, they were expensive, bulky, cryogenically cooled devices. As a result, making a high resolution grid of them was impractical. Instead they made a single row of them, mounted them to the bottom of an airplane, pointed them downward, and used the forward motion of plane itself to form a 2D image. It's sort of like how a flatbed scanner works: a 2D scanning element moved over the area or interest.
When thermal infrared detectors advanced enough to form a proper 2D grid at a semi-reasonable cost, you could aim them straight ahead as you would any other camera; thus, they became known as forward-looking infrared, a designation that persists to this day.
Further confusing the matter is that there's a company called FLIR that makes and sells thermal imaging equipment. But it's important to keep in mind that they're just far-infrared cameras. They detect thermal IR, emitted by every surface at a frequency and intensity proportional to its temperature. Thermal IR is emitted by everything and absorbed by pretty much everything: there are very few materials transparent to it. Let's put it this way: nothing your house is made of is transparent to IR at thermal wavelengths.
Bottom line, you can't use thermal IR to look inside anything. It's a tool for measuring surface temperatures, and that's it. Put another way, if you properly insulate your grow-op, thermal IR won't see it.
Actually, you don't. Know me, I mean.
There's seriously a program aimed at developing and deploying a fleet of nuclear bomb detectors at every port in the United States?
What kind of ridiculous bullshit is this? Did someone at the DHS watch a few episodes of 24 to come up with this? It's movie-plot anti-terrorism at its absolute worst: imaging ridiculously specific scenarios and spending enormous amounts of money to guard against them.
As if a terrorist organization resourceful enough to obtain a *nuclear fucking weapon* would somehow have difficulty bringing it into the country. This is a nation into which several metric tonnes of cocaine and thousands of illegal immigrants are successfully smuggled every year, and someone imagines that they'll be able to erect a perfect wall to keep a few kilograms of metal out of the country?
What congressman's nephew is being paid to make these detectors?
Mercury tiltswitch from a thermostat + relay. Cuts the power if anyone tries to move your box. It wouldn't be hard to wire it such that it sends mains voltage to your hard drives instead, but I stopped short of that because I was just doing it for fun and didn't want an accidental kick to the tower to destroy all of my data.
Everything you wrote is flat out wrong. Please educate yourself on Ohm's law before commenting further.
I = V / R
A common battery doesn't shock you because your skin resistance is high enough that almost no current flows through you. Your average car battery can put out hundreds of amperes, the average capacitor, perhaps thousands. Given how little current it takes to kill you, they might as well be infinite sources... but at 12V, neither will do you any harm.
You have two computers right next to each other. You want to get a file from one to other... good luck with that. For some totally inexplicable reason, this common situation presents us with a problem that's never been adequately solved. I've seen people sitting next to each other with laptops log on to their webmail accounts to send a file. Only to find that they can't, because the file is too large. Etc.
Let's review your options:
USB's architecture means two hosts can't talk to each other.
Firewire isn't common enough a port, and there are two connector types to worry about.
Ethernet is universal, the cables are cheap, and people might actually carry them around. You no longer need to worry about crossover cables, it's the fastest external interface on the modern PC... do we have a winner?
802.11g/n is also universal. Making a peer-to-peer network in windows isn't exactly easy though, and then you have to convince the other guy to disconnect from whatever network he's on, search the area, connect to yours... several minutes of work and a huge pain in the ass.
And all of the above suffer from the problem that they set up TCP/IP connections. Even with the autoconfig addresses that you'll get after Windows gives up on DHCP, two machines connected over TCP/IP have no practical way to talk to each other. What are you going to do, set up an FTP server? Connect to the C$ share of the other machine? Even if you were to do anything like that, you'd still need to ask for the guy's IP address first. Have fun teaching Ted from accounting about ipconfig.
What we need is something that's more than just a TCP/IP connection... something that automatically discovers the devices around you and gives you the option to easily send them a file. The standard has to specify everything right up to the application layer.
So... we need bluetooth. This is exactly the kind of problem it was made to solve.
The potential of it was ruined by two factors. First is that bluetooth continues to be a $30 option (for a $0.30 chip) on a lot of laptops. Second, and more importantly, there's the matter of the windows bluetooth stack; god help us all. Make the machine discoverable, get the other guy to search for devices in the area, pair them, exchange passkeys... all through an interface that, at least on XP, confuses the shit out of everyone.
In order for Wi-Fi direct to be useful, it will have to be more than just another way to establish a TCP/IP connection, and it will have to let go of this ridiculous obsession with security: pairing and discoverability and pass keys and all that nonsense. Christ, just let two machines talk to each other.
Remember IRDA? It wasn't exactly popular, but it worked. Two computers get in range, windows makes a neat little sound, and you get a systray icon you can click to immediately send files. That's the way it should be. The one time I ever managed to use it, it was glorious.
The solution we've managed to come up with in the absence of this capability is sneakernet for the 21st century: the USB flash drive. At least they're cheap and common now... there was a time when two computers sitting next to each other really had *no* options at all. Now we have these... they're not particularly fast, you're likely but by no means guaranteed to have one lying around, and whatever disposable cereal-box prize you're likely to be using will always have just a little less capacity than you need.
Damn it, it's the future. I want to beam files from one computer to another. Why can't I?
Real Programmers don't eat quiche. They eat Twinkies and Szechwan food.
