Download the content add-in for the 360 (you can find it on the market-place, it's free). It allows you to stream DivX/XviD and MP4. You still have to live with the media player library though, unless you use TVersity.
The Touch HD (running WM, currently a 6.5 ROM) is my daily phone and it's a great piece of hardware. At the time, HTC hadn't put anywhere near that quality hardware in an Android phone and I found it rather disappointing that they hadn't because I think it would've given Android a good start. More competition will mean better phone software.
That said, I don't think WM deserves the kind of beat down it was getting from a lot of people here who haven't used a current generation WM phone.
But I'm touched you were willing to go through my old comments. That's a special kind of commitment.
But that doesn't change the fact that the system for allocating addresses wasn't designed with "fairness" in mind (misguided or otherwise) as the original poster indicated. The system was designed as an oligopoly benefiting the initial investors. Which is fair (as you indicated), because it was their network.
Certain members of that same oligopoly have a vested interest in people continuing to use IPv4 and that's one of the things that is standing in the way of a market driven solution (and IPv6 for that matter). Of course the original designers couldn't have foreseen this problem, but that doesn't mean it isn't one.
What makes you think it was an attempt at fairness? If they had tried to make things *fair* there wouldn't be class A in the first place. The whole allocation system was pretty much designed so the early players (defense/government, certain universities and certain large corporations) were guaranteed absurdly large chunks of continuous address range.
A better analogue for cloud computing would be "outsourcing". It's not a client server model, it's a way to outsource the server part of the client server model. But if they called it outsourcing, people would think that all your data would be in an intelligible accent that kept telling you to check if the modem was plugged in.
I heard from a source who worked for DEC at the time that Windows NT was ported to Alpha as part of an under the table settlement due to patent infringements (mainly on the virtual memory subsystem) and was never expected to make a profit.
The NT kernel (including the virtual memory subsystem) was designed to be abstracted enough for easy porting by re-writing the device specific parts, so I don't think handling virtual page tables will be that hard (although, code above that relies on 4k page assumptions etc is not going to be happy if you use different page sizes etc). You're not easily going to be able to port to ARM cores that don't have an MMU, but I don't think anyone is talking about targeting Windows to somewhere without an MMU.
Why bother doing all of that when each version of Windows has a way of getting the input from multiple mice / keyboards in user space. You can use Raw Input (XP and later)/DirectInput (prior to XP)/opening the ports yourself (prior to NT 4.0, back when you could get keyboards and mice that worked that way). If you're determined, you can write a device driver (which may live in kernel space, but doesn't modify the kernel) and pass the input via a side port.
In user space you can filter keyboard/mice messages and post your own behind the scenes and other similar hacks. Most of these you don't have to do (Windows supports multiple user contexts which have their own user state).
Of course, it's not as nice or seamless or even as easy as changing the kernel, but it can be done.
Windows NT 3.1 was multi-user and able to run multiple tasks as different users simultaneously and they haven't removed that feature since. That's all that all that's required to run with multiple simultaneous users at kernel level really.
Unlike D3D10, D3D11 is actually backward compatible all the way to D3D9 shader model 2.0 hardware. That is, you get a lot of D3D10/11 features that give you big speed boosts (like the D3D10 style resource management/buffers and D3D11 deferred device contexts) on D3D9 hardware. In fact, except for XP support there is no reason to write anything but D3D11 code any more once it's released (with different shader paths).
Deferred contexts allow you to build up a list of commands on a separate thread, shunt them to kernel space in a single call (very low thunking overhead) and then render them on the main thread in a single call. In terms of distributing the CPU cost of rendering over multiple cores and going further to avoid CPU thunking costs, this is a huge feature (and like I said, it's available on D3D9 hardware).
Finally D3D11 is pretty almost a strict super-set (as is the updated WDDM) of Direct3D 10 at driver level. If you have the March DirectX SDK, you can run D3D11 applications on D3D9/10 hardware on Vista now at almost full speed (obviously, D3D11 itself isn't fully optimized yet as this is pre-release). So your driver concerns are pretty much unfounded.
Not all HTC Phones are crappy hardware though... the HTC Touch HD is an amazing and well built piece of hardware (even if you need to be a power user to actually get software on it doing what you want) that even has a proper headphone jack.
Now, if they made a better version of the Touch HD with Android instead of WM, then that would be super damn awesome.
Or maybe he thinks that decriminalization will reduce the street prices for the drugs. Decriminalization means that the lower tiers of the distribution network and the using individuals carry less risk, which means easier access, which should mean lower street prices and more competition based on quality. In the end, that's going to mean less money for the cartels.
Before you herald this progress, spare a thought for the poor memory-trick savants who will now have to spend half of their lives memorizing another trillion digits of Pi.
I searched for "why is google so expensive?" on google (well, google.com.au) and the first result that came back was: "Why Is Windows So Expensive on NewsTrendz - Snapshot Results..."
So, Microsoft is really just trying to compete with google search on the slams.
They should write it's power output in terms of sun, in which case it looks really puny next to nuclear. For example, the Tsar Bomba (largest human utilized explosive device, which was detonated at half the possible yield to prevent fallout) actually got into whole number percentages: "Since 50 Mt is 2.1*10^17 joules, the average power produced during the entire fission-fusion process, lasting around 39 nanoseconds, was about 5.4*10^24 watts or 5.4 yottawatts. This is equivalent to approximately 1.4% of the power output of the Sun.[9]" (Wikipedia).
If by considerably higher you mean lower. The 7800 GTX 256 (not the faster 512) had 38.4 gigs a second, vs the RSX's 22.4 gigs. Of course, the RSX can read from main memory at a good 15.5 gigs a second, in optimal circumstances when you're not relying on the 22.4 from the local memory. But even if the RSX could do both (full speed from main memory and local memory), 22.4 + 15.5 is only 37.9.
The Cell SPE's aren't crippled DSPs, they're regular DSPs and have a lot of trouble with things like branching or algorithms requiring scatter-gather across a large datastructure.
Do you have paranoid schizophrenia by any chance? I've been on the protest path and have many friends who are rather radical in their beliefs and no one would agree with what you're saying.
Slashdot Mirror
Actually, you might need to install the optional content pack from the marketplace (it's free). I had to do this to get divx to work.
Download the content add-in for the 360 (you can find it on the market-place, it's free). It allows you to stream DivX/XviD and MP4. You still have to live with the media player library though, unless you use TVersity.
It'll actually be called the John Holmes.
The Touch HD (running WM, currently a 6.5 ROM) is my daily phone and it's a great piece of hardware. At the time, HTC hadn't put anywhere near that quality hardware in an Android phone and I found it rather disappointing that they hadn't because I think it would've given Android a good start. More competition will mean better phone software.
That said, I don't think WM deserves the kind of beat down it was getting from a lot of people here who haven't used a current generation WM phone.
But I'm touched you were willing to go through my old comments. That's a special kind of commitment.
Don't show them a *good* Windows Mobile phone. That will make the slashdotters insecure about all their trash talk!
But that doesn't change the fact that the system for allocating addresses wasn't designed with "fairness" in mind (misguided or otherwise) as the original poster indicated. The system was designed as an oligopoly benefiting the initial investors. Which is fair (as you indicated), because it was their network.
Certain members of that same oligopoly have a vested interest in people continuing to use IPv4 and that's one of the things that is standing in the way of a market driven solution (and IPv6 for that matter). Of course the original designers couldn't have foreseen this problem, but that doesn't mean it isn't one.
What makes you think it was an attempt at fairness? If they had tried to make things *fair* there wouldn't be class A in the first place. The whole allocation system was pretty much designed so the early players (defense/government, certain universities and certain large corporations) were guaranteed absurdly large chunks of continuous address range.
A better analogue for cloud computing would be "outsourcing". It's not a client server model, it's a way to outsource the server part of the client server model. But if they called it outsourcing, people would think that all your data would be in an intelligible accent that kept telling you to check if the modem was plugged in.
I heard from a source who worked for DEC at the time that Windows NT was ported to Alpha as part of an under the table settlement due to patent infringements (mainly on the virtual memory subsystem) and was never expected to make a profit.
The NT kernel (including the virtual memory subsystem) was designed to be abstracted enough for easy porting by re-writing the device specific parts, so I don't think handling virtual page tables will be that hard (although, code above that relies on 4k page assumptions etc is not going to be happy if you use different page sizes etc). You're not easily going to be able to port to ARM cores that don't have an MMU, but I don't think anyone is talking about targeting Windows to somewhere without an MMU.
Travel to another country, listen to your music. Enjoy your $20000 roaming data charge.
Why bother doing all of that when each version of Windows has a way of getting the input from multiple mice / keyboards in user space. You can use Raw Input (XP and later)/DirectInput (prior to XP)/opening the ports yourself (prior to NT 4.0, back when you could get keyboards and mice that worked that way). If you're determined, you can write a device driver (which may live in kernel space, but doesn't modify the kernel) and pass the input via a side port.
In user space you can filter keyboard/mice messages and post your own behind the scenes and other similar hacks. Most of these you don't have to do (Windows supports multiple user contexts which have their own user state).
Of course, it's not as nice or seamless or even as easy as changing the kernel, but it can be done.
Windows NT 3.1 was multi-user and able to run multiple tasks as different users simultaneously and they haven't removed that feature since. That's all that all that's required to run with multiple simultaneous users at kernel level really.
Unlike D3D10, D3D11 is actually backward compatible all the way to D3D9 shader model 2.0 hardware. That is, you get a lot of D3D10/11 features that give you big speed boosts (like the D3D10 style resource management/buffers and D3D11 deferred device contexts) on D3D9 hardware. In fact, except for XP support there is no reason to write anything but D3D11 code any more once it's released (with different shader paths).
Deferred contexts allow you to build up a list of commands on a separate thread, shunt them to kernel space in a single call (very low thunking overhead) and then render them on the main thread in a single call. In terms of distributing the CPU cost of rendering over multiple cores and going further to avoid CPU thunking costs, this is a huge feature (and like I said, it's available on D3D9 hardware).
Finally D3D11 is pretty almost a strict super-set (as is the updated WDDM) of Direct3D 10 at driver level. If you have the March DirectX SDK, you can run D3D11 applications on D3D9/10 hardware on Vista now at almost full speed (obviously, D3D11 itself isn't fully optimized yet as this is pre-release). So your driver concerns are pretty much unfounded.
What he actually means is that your 570 dollar graphics card will be worth 1 dollar. That's a 570x better performance/price ratio!
Not all HTC Phones are crappy hardware though... the HTC Touch HD is an amazing and well built piece of hardware (even if you need to be a power user to actually get software on it doing what you want) that even has a proper headphone jack.
Now, if they made a better version of the Touch HD with Android instead of WM, then that would be super damn awesome.
Your physics teacher's story is highly embellished and full of inaccuracies.
Or maybe he thinks that decriminalization will reduce the street prices for the drugs. Decriminalization means that the lower tiers of the distribution network and the using individuals carry less risk, which means easier access, which should mean lower street prices and more competition based on quality. In the end, that's going to mean less money for the cartels.
Before you herald this progress, spare a thought for the poor memory-trick savants who will now have to spend half of their lives memorizing another trillion digits of Pi.
I searched for "why is google so expensive?" on google (well, google.com.au) and the first result that came back was: "Why Is Windows So Expensive on NewsTrendz - Snapshot Results ..."
So, Microsoft is really just trying to compete with google search on the slams.
THC is psychoactive and as such can be a hallucinogen in high enough quantities. That's why if you eat too many brownies you'll see the devil.
They should write it's power output in terms of sun, in which case it looks really puny next to nuclear. For example, the Tsar Bomba (largest human utilized explosive device, which was detonated at half the possible yield to prevent fallout) actually got into whole number percentages:
"Since 50 Mt is 2.1*10^17 joules, the average power produced during the entire fission-fusion process, lasting around 39 nanoseconds, was about 5.4*10^24 watts or 5.4 yottawatts. This is equivalent to approximately 1.4% of the power output of the Sun.[9]" (Wikipedia).
If by considerably higher you mean lower. The 7800 GTX 256 (not the faster 512) had 38.4 gigs a second, vs the RSX's 22.4 gigs. Of course, the RSX can read from main memory at a good 15.5 gigs a second, in optimal circumstances when you're not relying on the 22.4 from the local memory. But even if the RSX could do both (full speed from main memory and local memory), 22.4 + 15.5 is only 37.9.
The Cell SPE's aren't crippled DSPs, they're regular DSPs and have a lot of trouble with things like branching or algorithms requiring scatter-gather across a large datastructure.
Microsoft makes genital hair pieces?
Do you have paranoid schizophrenia by any chance? I've been on the protest path and have many friends who are rather radical in their beliefs and no one would agree with what you're saying.