Actually, if you'd bother reading the article, you'd find that the micro-controller is being used to do drive the electric motors, image process and maintain the turing machine's "state". that's it.
The thing is, since it's most likely a software problem.. they need to change their model to accommodate for hot-fixes. You shouldn't need to recall the car just to upgrade the firmware.
Maybe this sort of publicity will push towards a more modern servicing model.
Why? because it means that the amount of information you need to simulate will grow to infinity, which is impossible to contain.
example: Simulate the universe from within... when you reach the point where you simulate your own simulation, it must contain the simulated universe, which in turn contains another simulation of the universe and so forth.
You can't simulate that because you will quickly run out of memory. Regardless of how much memory you have for the simulation.
Assuming perfect mirrors....
When you bring to mirrors together, your information is in fact bounded by the reflected wave-length of light. i.e. you can't mirror any feature who's physical dimensions are smaller than that of the wave length.
Therefore you're not actually displaying an infinite amount of information. Furthermore, for imperfect mirrors, you are actually losing data with each 'recursion'. more so, if the mirror placement is not perfectly parallel, you will have the image crawl to an edge.
regarding VMWare... sure, you can create an OS that is horrible when ran outside VMWare... but then the simulation is imperfect, and it becomes completely irrelevant for the discussion.
You made so much sense in your previous post.. too bad you had to make this one as well.
Simulating the universe from within the universe is impossible - regardless of the rate, as your simulated universe should contain the simulation itself.... which is a positive feedback loop.
If they just used text instead of icons, the text would seem mashed together with adjacent text to the point where it would be difficult to distinguish where to stop and where to start reading.
Icons, while sometimes fail to convey the functionality properly, at least manage to display a clear separation between one function and the other.
Sounds like the submitter thinks there's a problem.
I'm not entirely convinced.
Somehow that big company survived as a business before he joined and made his observations. Whatever it is they're doing, seems to be working out for them. Maybe he specifically joined a department that's in loss or will be shut down if they don't start producing results, but i'm certain that it isn't indicative of the company as a whole.
I would assume he only needs to verify the last 120 billion digits.
Assuming his algorithm can support serialization of its state into a check-point, he can simply recalculate the last 120 billion digits a couple of times and compare.
Assuming linear time to compute each digit: 120e9/2.7e12 * 116 =~ 5 days. not too bad.
except for the distortion you get around the edges of each sensor...
Also note, that the denser pixels are going to get the less surface area you'll require for 26GP... though i suspect that there is a fundamental limit - the photo-diode will need to be larger than a some function of the wave length.
The size is dominated by the transistors, the photo-diode shares the same feature size are the transistors since it's manufactured under the same process.
These are two different things: 1. Partial decode due to missing data (i.e a picture is comprised of blocks, you can have some of the block but not all of them - hence partial decode) 2. A primitive decoder's inability to decipher/keep-real-time-rate-of an advanced encoding (i.e. trying to decompress something in realtime on a weak machine will at best result in frame drops -- or completely incompatible schemes of decoding: my decoder supports only mpeg-2 but i've got a h.264 movie)
You are talking about option 1, I'm talking about option 2.
Slashdot Mirror
Actually, if you'd bother reading the article, you'd find that the micro-controller is being used to do drive the electric motors, image process and maintain the turing machine's "state". that's it.
Ideally, It wouldn't be your fault at all.
Realistically, you should know better.
So it really depends on your POV... are you an insurance company trying to avoid paying a claim? or are you an Idealist trying to get justice?
Israel basically doesn't care about what they've done to these people because for them their war against Hamas justifies anything...
Ahh... without a single shred of evidence, you would still judge and convict Israel.
Lefties and their selective morality...
Spot on.
Searching for the rationale in an apple fanboi's statement is an exercise in futileness.
Of course, she's a woman so she doesn't have the 'the balls'...
Does that mean she insists on absolute safety and was trying to say that in a humoristic way?
The thing is, since it's most likely a software problem.. they need to change their model to accommodate for hot-fixes. You shouldn't need to recall the car just to upgrade the firmware.
Maybe this sort of publicity will push towards a more modern servicing model.
If it has a special case to deal with sim-inside-sim Then it would not be an exact simulation of the top level.
err, no...
your analogy is so bad.
i don't have the will to get into this.
just have a long think... and figure out if a program printing itself is the same as a program simulating itself.
Why? because it means that the amount of information you need to simulate will grow to infinity, which is impossible to contain.
example:
Simulate the universe from within... when you reach the point where you simulate your own simulation, it must contain the simulated universe, which in turn contains another simulation of the universe and so forth.
You can't simulate that because you will quickly run out of memory. Regardless of how much memory you have for the simulation.
Assuming perfect mirrors....
When you bring to mirrors together, your information is in fact bounded by the reflected wave-length of light. i.e. you can't mirror any feature who's physical dimensions are smaller than that of the wave length.
Therefore you're not actually displaying an infinite amount of information. Furthermore, for imperfect mirrors, you are actually losing data with each 'recursion'. more so, if the mirror placement is not perfectly parallel, you will have the image crawl to an edge.
regarding VMWare... sure, you can create an OS that is horrible when ran outside VMWare...
but then the simulation is imperfect, and it becomes completely irrelevant for the discussion.
You made so much sense in your previous post.. too bad you had to make this one as well.
Simulating the universe from within the universe is impossible - regardless of the rate, as your simulated universe should contain the simulation itself.... which is a positive feedback loop.
This could be due to the fact that normally you'd find large breasts on fat women... which is objectively disgusting :P
Who are you fooling? this is Slashdot...
If they just used text instead of icons, the text would seem mashed together with adjacent text to the point where it would be difficult to distinguish where to stop and where to start reading.
Icons, while sometimes fail to convey the functionality properly, at least manage to display a clear separation between one function and the other.
It's called astroturfing, and it's an apple fanboi's favorite pastime after ... well, that's NSFW.
Sounds like the submitter thinks there's a problem.
I'm not entirely convinced.
Somehow that big company survived as a business before he joined and made his observations. Whatever it is they're doing, seems to be working out for them.
Maybe he specifically joined a department that's in loss or will be shut down if they don't start producing results, but i'm certain that it isn't indicative of the company as a whole.
I would assume he only needs to verify the last 120 billion digits.
Assuming his algorithm can support serialization of its state into a check-point, he can simply recalculate the last 120 billion digits a couple of times and compare.
Assuming linear time to compute each digit: 120e9/2.7e12 * 116 =~ 5 days. not too bad.
Hey guess what? You could submit a new story about your snake. There's plenty of room for both.
except for the distortion you get around the edges of each sensor...
Also note, that the denser pixels are going to get the less surface area you'll require for 26GP... though i suspect that there is a fundamental limit - the photo-diode will need to be larger than a some function of the wave length.
Read about CMOS Active Pixel Sensors:
http://en.wikipedia.org/wiki/Active_pixel_sensor
The size is dominated by the transistors, the photo-diode shares the same feature size are the transistors since it's manufactured under the same process.
Moore's law applies.
Actually, according to moore's law, we just need 20 years (for a factor of 1000)...
If we're gonna stitch photos together, i think Google Earth is probably by far "higher-resolution" than this.
Show me a SINGLE image sensor that can do 26GP and i'll be impressed!
These are two different things:
1. Partial decode due to missing data (i.e a picture is comprised of blocks, you can have some of the block but not all of them - hence partial decode)
2. A primitive decoder's inability to decipher/keep-real-time-rate-of an advanced encoding (i.e. trying to decompress something in realtime on a weak machine will at best result in frame drops -- or completely incompatible schemes of decoding: my decoder supports only mpeg-2 but i've got a h.264 movie)
You are talking about option 1, I'm talking about option 2.
> can't keep up and render an inferior picture.
It's not like there's half-way here. This is the digital age - if it can't keep up you won't see anything!
Sensationalist... i would expect this from a tabloid.
Title should have been: Unencrypted data broadcasted everywhere ... can be received by anyone!
The leap from that to "$26 of Software Defeats American Military" is quite a big leap in my opinion.
Forget about the non-gurus...
even gurus don't have the will to tinker about their settings for days on end just to get something trivial working.
We want to it to work already so that we could get around to doing our OWN work.