I'll agree. It wasn't the visual design that did it for me, it was the animation.
Looking at the CG face, it *looked* perfectly human, but it didn't move very naturally, and that moved it from perfect into uncanny valley for me.
From behind-the-scenes imagery, it looks like they used a multi-camera head rig with visible-light tracking points (not many cameras, though). There are some shortcomings to this sort of system, and it seems like on top of that they over-interpolated the motion data...
From what I've seen of Depth Analysis's MotionScan system (ironically, the first and so far only use of this system is for a videogame), it does a much better job at producing realistic motion. If Tron Legacy had come out a few months later, it's possible that they might have been able to use MotionScan, and avoided the uncanny valley entirely.
Don't think of them as anecdotes, think of them as case studies.
Look at Netflix's contest. It seems to have been a big success. Netflix set a difficult goal in terms of improving on their recommendation algorithm. 41 thousand teams competed for the prize. Of those, over 5000 submitted valid entries, over 44 thousand in total. The teams published their progress and improved on each others' work. The winning team even published a variety of papers during the process. Eventually (three years later), the goal (a 10% improvement) was reached, and the prize handed out.
The prize was $1m. With over 44 thousand entries, it's likely that millions of man-hours were spent on the contest. Clearly it would not have been possible to pay for the development that happened. On the other hand, a point might be raised as to if Netflix could not have paid to assemble a specific team to solve the problem with a $1m budget? $1m can pay for some pretty smart researchers for three years...
VirtualGL is strictly for 3D-based stuff, though. Useless in the typical desktop use case (unless you can remote an OpenGL compositor with it?). TurboVNC is still ultimately just sending JPEG frames over the internet, which is both CPU intensive, and bandwidth intensive.
The big advantage of NX is that you're sending the underlying XDMCP data, executing the draw instructions directly, rather than trying to detect changes on the screen and then sending chunks of the screen over the internet as images. NX's primary speedup comes by caching (not compression), as it eliminates most of the round-trips that XDMCP normally requires.
In my personal experience, I've found NX to be enormously faster than VNC, more and more dramatically so on slower and slower connections.
A more interesting solution would be VirtualGL/NX, although it doesn't sound like VirtualGL is terribly useful over the internet.
That data is stored on S3, which itself is typically defined as cloud storage. If the concept of your data being broken into tiny pieces and stored in chunks on different machines in different places is not "the cloud", then perhaps you should just call it "geographically distributed clustered storage" or some such thing. But "cloud" works well enough.
90% of all people are in jail because they talked. What percentage of those people are there because they talked about how they're guilty of the crime they were accused of?
Few satellites use an RTG. Too high a risk of the radioactive material being released into the atmosphere if something goes wrong. Really hard for private companies to get the clearance to use them. It looks to me like most satellites using RTGs are government-backed.
If you combine all the claims of new technologies that enable 10x the capacity for batteries, we should be seeing AA cells that can hold enough power to run the planet for a year.
I'll believe it when I can buy it from Canadian Tire.
Re:That's one heck of a "long goodbye"
on
Goodbye, VGA
·
· Score: 1
Wups, didn't notice that it was female DVI.
Re:That's one heck of a "long goodbye"
on
Goodbye, VGA
·
· Score: 2, Informative
More like the adapters are defined in the DisplayPort specs rather than just being after-market addons like a DVI to component adapter would be. You can't add DRM to VGA (although you can degrade it, as you pointed out).
Re:That's one heck of a "long goodbye"
on
Goodbye, VGA
·
· Score: 1
You obviously haven't bought a graphics card recently:)
Most graphics cards sold in the past 5-10 years ship with at least one of 'em. Nowadays, they all do, since videocards rarely have VGA ports anymore. They normally have some combination of DVI-I, DisplayPort, and HDMI, and provide an adapter to use the DVI-I port as VGA.
Heck, my Radeon 9700 Pro shipped with two such adapters, since it only had two DVI-I ports;)
Re:When PS2 is better - one example
on
Goodbye, VGA
·
· Score: 1
There's an option in most BIOSes about if they should handle USB keyboard support themselves until the OS takes over. You probably need to flip the setting.
Re:That's one heck of a "long goodbye"
on
Goodbye, VGA
·
· Score: 2
For existing computers, sure. Although PS2's main problem is that it's not hot-pluggable, so if you don't have a keyboard plugged in at boot, you'll need to reboot the machine to get it detected (at least with Windows).
More to the point, modern motherboards tend not to have PS2 ports, and I don't believe any laptops have for what, at least half a decade?
FYI, Google's using Foxit for the built-in PDF viewer. So, you know, this is kind of like you using Foxit, but with less bloat, since you don't need a completely separate application and UI to get the Foxit PDF rendering engine.
People defending Comcast in this case proves how little they know about how the internet truly works
Hmm...
Level 3 in fact sells connections to Verizon (who happens to pay them a monthly fee for their customers to access the internet) as does Time warner, AT&T, Sprint and nearly every single other US ISP except other Tier 1's
You don't seem to know as much as you think you do. Verizon, AT&T, and Sprint are three of the eleven tier 1 networks. They don't pay Level3 a dime.
Seven of the eleven tier 1 networks are based in the US. Four of those seven (the three you mentioned, plus Qwest) also operate major consumer ISPs.
Yes, and that's my underlying point. This is a simple peering issue, there's nothing out of the ordinary here. Level3 is doing to Comcast what Cogent tried to do to Level3.
When Cogent did it to Level3, Level3 cut off Cogent. When Level3 does it to Comcast, they expect a different result for some reason.
If Comcast cut all connections to Level3, comcast customers would still be able to reach Netflix just fine... Not only does Netflix not rely exclusively on Level3, but Comcast would be able to use other routes to Level3.
The problem is that there is no "core" to be closer to or farther from. If the first ISP in the chain needs to get a packet to the last ISP in the chain, it picks an ISP that it connects to who both advertises that it has a route to the last ISP, and is closer to the last ISP. Through this manner, the packet eventually gets where it's going. There is no obvious point along the way where it makes sense to switch which party is paying.
If you want to define "retail customer" in terms of who pays who, and define "pay" to be strictly cash, and want to define the class of ISPs that have no such customers, then that is the definition of a tier 1 ISP. There are currently 11 ISPs that are agreed to be tier 1. Level3 is one of them, as are most of ma bell. Comcast is not.
It's likely that one of the reasons why Level3 does not want to pay Comcast is because Level3 would lose their status as a tier 1 ISP, although that doesn't necessarily mean anything important.
Why is it already the case? Can you point out any retail customers for, for example, Savvis?
In your analogy, since all costs are paid by retail customers, there is no transit (since an ISP has no reason to move data between two different networks since they won't be paid for it) and now every ISP in the world is required to interconnect with every other ISP in the world. Very few companies could afford to do that, and we'd likely end up left with a small handful of ISPs per continent. No network operator could survive unless they had millions upon millions of customers to cover the costs of connecting to everybody else.
Your ISP is a Comcast customer. Why should I have to pay for your ISP's connection? If you want me to upload a file to your web server, should you have to pay for my cable line?
And two Bell customers both have to pay Bell for the same phone call. What's your point? Comcast is trying to get two parties who wish to exchange data to pay for this exchange. Comcast provides a pipe between the two entities, and wants both ends of the pipe to pay for it.
People keep talking about how they want ISPs to behave as "dumb pipes", but then object when they actually do.
Slashdot Mirror
I'm not seeing how a 200MHz overclock puts a phone with a 1000MHz aggregate clockspeed advantage to shame...
Sure, performance doesn't scale linearly, but I'll still take a dual-core 1GHz Cortex A9 over a 1.2GHz single-core Cortex A8-alike any day.
I'll agree. It wasn't the visual design that did it for me, it was the animation.
Looking at the CG face, it *looked* perfectly human, but it didn't move very naturally, and that moved it from perfect into uncanny valley for me.
From behind-the-scenes imagery, it looks like they used a multi-camera head rig with visible-light tracking points (not many cameras, though). There are some shortcomings to this sort of system, and it seems like on top of that they over-interpolated the motion data...
From what I've seen of Depth Analysis's MotionScan system (ironically, the first and so far only use of this system is for a videogame), it does a much better job at producing realistic motion. If Tron Legacy had come out a few months later, it's possible that they might have been able to use MotionScan, and avoided the uncanny valley entirely.
Don't think of them as anecdotes, think of them as case studies.
Look at Netflix's contest. It seems to have been a big success. Netflix set a difficult goal in terms of improving on their recommendation algorithm. 41 thousand teams competed for the prize. Of those, over 5000 submitted valid entries, over 44 thousand in total. The teams published their progress and improved on each others' work. The winning team even published a variety of papers during the process. Eventually (three years later), the goal (a 10% improvement) was reached, and the prize handed out.
The prize was $1m. With over 44 thousand entries, it's likely that millions of man-hours were spent on the contest. Clearly it would not have been possible to pay for the development that happened. On the other hand, a point might be raised as to if Netflix could not have paid to assemble a specific team to solve the problem with a $1m budget? $1m can pay for some pretty smart researchers for three years...
VirtualGL is strictly for 3D-based stuff, though. Useless in the typical desktop use case (unless you can remote an OpenGL compositor with it?). TurboVNC is still ultimately just sending JPEG frames over the internet, which is both CPU intensive, and bandwidth intensive.
The big advantage of NX is that you're sending the underlying XDMCP data, executing the draw instructions directly, rather than trying to detect changes on the screen and then sending chunks of the screen over the internet as images. NX's primary speedup comes by caching (not compression), as it eliminates most of the round-trips that XDMCP normally requires.
In my personal experience, I've found NX to be enormously faster than VNC, more and more dramatically so on slower and slower connections.
A more interesting solution would be VirtualGL/NX, although it doesn't sound like VirtualGL is terribly useful over the internet.
That data is stored on S3, which itself is typically defined as cloud storage. If the concept of your data being broken into tiny pieces and stored in chunks on different machines in different places is not "the cloud", then perhaps you should just call it "geographically distributed clustered storage" or some such thing. But "cloud" works well enough.
I'm surprised Microsoft is developing things outside of it's own ecosystem. It's my understanding they don't generally do this.
I think we passed that moment when Microsoft became Linux kernel developers by submitting some virtualization-related patches.
90% of all people are in jail because they talked. What percentage of those people are there because they talked about how they're guilty of the crime they were accused of?
Few satellites use an RTG. Too high a risk of the radioactive material being released into the atmosphere if something goes wrong. Really hard for private companies to get the clearance to use them. It looks to me like most satellites using RTGs are government-backed.
If you combine all the claims of new technologies that enable 10x the capacity for batteries, we should be seeing AA cells that can hold enough power to run the planet for a year.
I'll believe it when I can buy it from Canadian Tire.
Wups, didn't notice that it was female DVI.
More like the adapters are defined in the DisplayPort specs rather than just being after-market addons like a DVI to component adapter would be. You can't add DRM to VGA (although you can degrade it, as you pointed out).
You obviously haven't bought a graphics card recently :)
Most graphics cards sold in the past 5-10 years ship with at least one of 'em. Nowadays, they all do, since videocards rarely have VGA ports anymore. They normally have some combination of DVI-I, DisplayPort, and HDMI, and provide an adapter to use the DVI-I port as VGA.
Heck, my Radeon 9700 Pro shipped with two such adapters, since it only had two DVI-I ports ;)
There's an option in most BIOSes about if they should handle USB keyboard support themselves until the OS takes over. You probably need to flip the setting.
For existing computers, sure. Although PS2's main problem is that it's not hot-pluggable, so if you don't have a keyboard plugged in at boot, you'll need to reboot the machine to get it detected (at least with Windows).
More to the point, modern motherboards tend not to have PS2 ports, and I don't believe any laptops have for what, at least half a decade?
What territories did Netflix get these rights? Because I doubt that we'll ever see any of this on netflix.ca.
FYI, Google's using Foxit for the built-in PDF viewer. So, you know, this is kind of like you using Foxit, but with less bloat, since you don't need a completely separate application and UI to get the Foxit PDF rendering engine.
I would suggest that two rights don't make a wrong ;)
People defending Comcast in this case proves how little they know about how the internet truly works
Hmm...
Level 3 in fact sells connections to Verizon (who happens to pay them a monthly fee for their customers to access the internet) as does Time warner, AT&T, Sprint and nearly every single other US ISP except other Tier 1's
You don't seem to know as much as you think you do. Verizon, AT&T, and Sprint are three of the eleven tier 1 networks. They don't pay Level3 a dime.
Seven of the eleven tier 1 networks are based in the US. Four of those seven (the three you mentioned, plus Qwest) also operate major consumer ISPs.
Yes, and that's my underlying point. This is a simple peering issue, there's nothing out of the ordinary here. Level3 is doing to Comcast what Cogent tried to do to Level3.
When Cogent did it to Level3, Level3 cut off Cogent.
When Level3 does it to Comcast, they expect a different result for some reason.
It's pure hypocrisy.
If Comcast cut all connections to Level3, comcast customers would still be able to reach Netflix just fine... Not only does Netflix not rely exclusively on Level3, but Comcast would be able to use other routes to Level3.
The problem is that there is no "core" to be closer to or farther from. If the first ISP in the chain needs to get a packet to the last ISP in the chain, it picks an ISP that it connects to who both advertises that it has a route to the last ISP, and is closer to the last ISP. Through this manner, the packet eventually gets where it's going. There is no obvious point along the way where it makes sense to switch which party is paying.
If you want to define "retail customer" in terms of who pays who, and define "pay" to be strictly cash, and want to define the class of ISPs that have no such customers, then that is the definition of a tier 1 ISP. There are currently 11 ISPs that are agreed to be tier 1. Level3 is one of them, as are most of ma bell. Comcast is not.
It's likely that one of the reasons why Level3 does not want to pay Comcast is because Level3 would lose their status as a tier 1 ISP, although that doesn't necessarily mean anything important.
then Time Warner can charge Level 3
They already do. Level3 pays either in money or bandwidth.
Then AT&T can charge Level 3.
They already do. Level3 pays either in money or bandwidth.
Then Verizon and Sprint will charge Level 3.
They already do. Level3 pays either in money or bandwidth.
Then these companies will charge each and every web page and web service to be allowed access so their customers can access web content.
They already do. Those web hosts pay for their connectivity, and each network along the way gets compensated.
You see where I'm going here?
Why is it already the case? Can you point out any retail customers for, for example, Savvis?
In your analogy, since all costs are paid by retail customers, there is no transit (since an ISP has no reason to move data between two different networks since they won't be paid for it) and now every ISP in the world is required to interconnect with every other ISP in the world. Very few companies could afford to do that, and we'd likely end up left with a small handful of ISPs per continent. No network operator could survive unless they had millions upon millions of customers to cover the costs of connecting to everybody else.
Your ISP is a Comcast customer. Why should I have to pay for your ISP's connection? If you want me to upload a file to your web server, should you have to pay for my cable line?
And two Bell customers both have to pay Bell for the same phone call. What's your point? Comcast is trying to get two parties who wish to exchange data to pay for this exchange. Comcast provides a pipe between the two entities, and wants both ends of the pipe to pay for it.
People keep talking about how they want ISPs to behave as "dumb pipes", but then object when they actually do.