They're making a map using gravitational lensing. I'm not familiar with the technique of mapping with this. It seems like they'd have to know the location of the background objects which are only visible "through the lens", and since we don't know the composition of the lens... It seem like 2 unknowns and one measurement. Is that the case?
Next, do astrophysicists consider the effects of thermodynamics? You know, expansion, cooling, heating, enthalpy, etc... They seem to be assuming radiation~=temperature, so cold matter would be "dark". Do these guys really understand where the hot regions would be in such a collision?
A better solution is to make a really really good digital copy. I thought the idea was that your TV could display the signal with macrovision noise added, but your VCR would lose sync and get all garbled because it was unable to make a good copy of the "noise". If you make as perfect a copy as you can, wouldn't you be able to play it back? It may still violate copyright, but you wouldn't be circumventing anything - the macrovision would be intact.
You'll love this: My daughter was tapping her hand to the beat at 2 and 1/2 WEEKS of age. I just about fell over when my wife said "look at this" and there she was tapping away in the air. She's shown that only a few times since. Musical prodigy? Don't know, don't care. She'll be good at something, some day. I took a few years of guitar lessons, but never could do great solos - had the feel, but not the speed. I did have an excellent sense of timing (my teacher thought that came naturally) and probably the fact that we played electric piano "demos" to get my daughter to sleep while gently bouncing around to the beat, had something to do with what she did so early. If she gets interested in music, I hope she does better at it than me. If she does something completely different, that will be fine too, so long as she like SOMETHING. I only hope that she enjoys life and gets the most out of it. Some folks worry too much about who is good at what, and why.
also seem to think that stars orbiting a galactic center are supposed to obey Keplers laws... The discrepancy between the observed galactic rotation curves and the "predicted" one are then attributed to "Dark Matter" rather than someones poor understanding of basic physics.
After the saw manufactures refused to pay his unreasonable licensing free (3-8% of the saw sale price)for his patented tchnology he moved on to lobbying for a law to make it mandetory (and still pay his licensing fee)
I don't know how much saws are marked up, but pushing that 3% of sales back down onto the manufacturers price shouldn't be that much an increase for a good safety feature. Sounds more like greedy sawmakers than a greedy inventor. Now if he wanted to require they put it on all their saws, that's a greedy inventor. Or if he wanted them to pay for all saw even if they don't have it, that's a greedy inventor. Having to pay 3 to 8 percent for this feature sounds reasonable to me as a potential end user - assuming it wouldn't false trigger of course...
You can use a zener diode reverse biased right at the knee. Amplify the output for a nice random signal full of noise from a quantum source - the diode junction. This was used for noise generation on a lot of old arcade games (Space War, Star Castle, etc...). It's low frequency, but it is truely random and costs well under a dollar.
For starters, I don't think MS would ever do that and if they did they wouldn't use GPL but some other kind of license they came up with.
I switched to the XBOX instead of keeping with TiVo so people wouldn't chime in with "but TiVo doesn't have the volume to get a custom chip from Intel". Suppose Sony wanted to use a variant of VLC for a media player in the PS3 rather than write their own. The point was that you can TiVoize the software by running on custom hardware where the user can't recompile for that hardware - regardless of the specific thing we call DRM.
I'm still waiting for a 64 bit processor that treats all registers the same. i.e. one load and one store instruction, but you can do fpadd or regular add on the same registers. This IMHO will reduce the number of opcodes needed, and you usually don't use a lot of FP registers and a lot of integer registers at the same time. Pure stacks suck BTW - you really need a swap (up to some depth) or a copy instruction to bring data to the top. A pure stack is too destructive. I do like the idea of a pure return stack with separate data stack.
Suppose MS wanted to run Free Software on the next XBOX and didn't want people to mess around with it. They could have Intel modify a processor any number of ways (change the opcodes for a SIMPLE example) and provide a proprietary tool chain to compile the code. No DRM, yet the users have no way to modify and run the code on that hardware. Does GPL need to require a complete tool chain be provided when binaries are provided? It seems overkill, but custom (closed) hardware running free software defeats the GPL in the same way as DRM. I need to read the new draft, but I think it suggests the broader concept of denying freedom more than DRM in particular. Thoughts?
Did you notice what happened when someone was tainting heroin with fentanyl, and the customers were dying? They identified how it got into the supply, and tracked down the source of the fentanyl in no time. Did they do anything about the heroin? No. One problem is that if the "war on drugs" is won, the whole group of people fighting it would be out of a job. That's not really a position you should put people in if you want them to be effective - the incentives aren't right.
It's like here in MI where the DNR is supposed to "manage the herd of deer" by regulating hunting. Unfortunately they make money from hunting permits. Now we have car-deer accidents all the time because of the huge number of deer. How about when the PTO becomes a revenue source for the government? Or when... You get the idea. It doesn't take a conspiracy, just a failure to consider what the incentives are for the people involved in these organisations. It's like asking me to train a group of guys from India to write software;-)
Fedora Core 4 shipped with a broken pwc driver. OK fine, there was a lot of controversy about the open source driver at the time. Fedora 5 again shipped with a broken pwc driver even though currently shipping kernels come with the new one. When I say "broken" I don't mean that it's lacking the part that used to be proprietary, I mean it just doens't work. FC5 also shipped with a broken kernel because someone applied a late patch that IMHO really didn't belong in there. The kernel didn't work with nVidia drivers (and others) and some blamed the drivers when in fact it was a Fedora specific patch that caused the breakage. It took several releases of kernel before everyone seemed happy (other stuff broke with each release). The policy on other packages is that all work should be done upstream, yet Fedora seems to be letting a few guys fiddle around with the kernel and drivers at will. Having to update a fresh install to make it work right is unacceptable.
So the questions: What is Fedora doing to improve the quality of the kernels and drivers? What is the purpose of all the tweaking? Some folks use stock kernels with Fedora, why all the messing around?
If there were a way to cause light to bend/reflect/refract around an object from any direction so as to appear ivisible, the person inside such bubble would by definition be blind - at least for the affected wavelengths. Another interesting property of such a bubble is that no light could get out - the path of light is reversible so if nothing gets in then nothing gets out. If all light is directed around, then none gets in or out, simple as that.
In my humble opnion, if the FSF starts adding restrictions to its licenses, it will be loosing the 'Free' part of its name.
The FSF wants to protect USERS freedom. That means they try to prevent developers from denying users freedom. BSD gives developers freedom - including the ability to deny their users freedom.
IMHO, Richard is still misguided in his anti-DRM efforts. You don't need DRM to prevent people from running their own version of software on proprietary hardware. Intel could sell TiVo a processor with modified op-codes and a modified assembler to match (or insert some other mod that renders GCC unusable on their hardware). TiVo could run all the free software they want and give you source code, but without the "special" tool chain you'd be screwed. No keys or DRM required. DRM is just a special case of the problem FSF would really like to address. Is FSF prepared to require that people distribute hardware specs with the software? How about just a tool chain to build from source and run on the hardware supplied? How easy does the build process have to be? It's not an easy goal, but ranting about DRM in particular isn't the solution.
The obvious use for a large number of cores is graphics. After buying ATI, AMD will likely work at a new set of instructions: SSE4 or SSE5 which will bring pixel shader functionality back to the CPU where it belongs (one core can already handle vertex ops). You'll also have the extra cores in case someone actually has a nicely parallel application. System cost will go down, but a larger share of the PC price will go to AMD. With this purchase going on it should be absolutely obvious to everyone, but the stupid tech writers don't seem to get it yet. If this is AMDs plan, it's brilliant - if they just want to put a GPU on the same die it was a waste of money and will be the end of the company.
OK, while we're at it... When I was a teenager, I worked in a lighting store warehouse. We had a pegboard witha bunch of tools on the wall. One day I came in and all the tools were painted orange. Turns out one of the guys (chris) painted all the tools orange for who knows why. So after work, I drop by my buddy Don's house. I said something like "you'd never believe what happened at work today..." He stopped me abruptly and said some thing along the lines of "WAIT. I just got this strange feeling and I'm thinking of the word 'orange'.?!?!" Don didn't know Chris. Don didn't even have a reason to have tried calling me at work, and had never done so before - so he didn't have the number. It is plausible that he made the effort to get the number and call. It is exceedingly unlikely that he ended up talking to Chris instead of me - other people answered the phone and paged us. It is also unlikely that Don would continue this joke for 20 years - I have asked him about it. In all likelyhood that is exactly what happened, but I honestly don't believe it. I don't have an explanation that satisfies me.
How "open" is AMD as far as providing specs, documentation, info and code goes?
I have a set of x86-64 books I ordered from AMD for free a few years ago. AMD also backed the effort to get GCC building code for the new architecture, as well as providing a simulator to get Linux booting on it. They also funded (was it SUSE?) some people to make sure it actually happened. Linux was the only 64bit OS that ran on those chips when they first became available. OTOH, if they had not been so helpful there wouldn't have been anything 64bit to run on them at first. Lets hope they continue to be open and helpful even though there may not be a "need" to do so with ATI.
In my understanding, the feeling of Deja Vu is its own feeling, not the regular, everyday familiarity feeling.
The strange "feeling" of Deja Vu is simply due to experiencing familiarity when you believe there should be none. There is nothing strange about being in a familiar environment. It's when something seems familiar and you know it should not, that there is an extra feeling that comes along with it.
Personally, I had an experience where some friends and I were riding our bikes through some trails and found some tennis courts in the woods. I know for a fact that I had not been in this place before, but it sure seemed familiar. I said to my friend that it seemed really familiar and then I said there should be a drinking fountain "over there" and I pointed off about 90 degrees to the left... There it was, right where I pointed. We were both amazed. Could be that the layout of the area was the same as someplace I'd been before (unlikely - I don't play tennis) or more likely, it could be that I had seen the water fountain on the way in and hadn't conciously made note of it (i.e. it got classified as a "memory" rather than a current event) though as I rememeber it, I had not looked over that way until after I pointed. BTW, no "substances" were involved. My mind was rather stunned at there being tennis courts in this particular location (in a wooded area in the middle of a large subdivision) so it may have just played some tricks with me.
you need to store the power during the day for your peak night time usage. The most cost effective way of doing this currently is deep cycle lead acid batteries
No. For now the cheapest thing to do isn't to store the energy but push it onto the grid. Now in many states they screw you either disallowing this practice, or allowing the power company to pay you less than you pay them for the energy. That's more a political issue than practical though. Long term there is a problem - if everyone did it, there would be no place for the peak power generated to go because all homes would be sources not loads. In reality we are a long way from having to solve this energy storage issue. At present, the most efficient place to put the energy is on the grid.
You miss the point. In 6 years we will be doing ray tracing on the CPU(s) in real time. That's better graphics than GPUs can do. At that point, the GPU is truely redundant. When is the last time anyone tried to do rendering on the CPU? Quake 1? It's come a long way since then, and now we've got extra cores that no one knows what to do with. Besides, for the majority of business computers a GPU is already unnecessary - windows can be painted in software, even with fancy compositing.
Processors are getting fast enough to do rendering in software again. GPUs are trying to become general purpose CPUs. People will soon have 2 cores as standard and 4 or more are on the way. What are people supposed to do with all those CPU cores? Replace the GPU with them of course. Why would a CPU maker need to buy a GPU maker? Not sure, but perhaps just to gain the graphics expertise to write the software, and possibly to make some suggestions for the instruction set and hardware. I certainly hope they don't just integrate the GPU onto the same die - that's so unnecessary.
IMHO a much better - and cheaper - buy would have been that company (don't remember name) that has a really fast DirectX implementation in software. You don't even need a frame buffer any more, just put some circuitry in the north bridge that pulls data from main memory and spits it out over DVI-D. Let software on multiple cores take care of rendering.
For reference, I think we're still on track for software realtime raytracing by 2012. If we can do that, certainly software can rasterize polygons fast enough before then. The GPU as we know it is dead - good for ATI finding an exit strategy, bad for AMD spending so much money. What's nVidia's exit strategy? Intel already knows a lot about graphics.
"The worst part is the conspiracy theorists claiming the landing never occurred are going to go nuts over this."
Hey man, I remember that time lapse shot of the "earth rise" from the moon. They better destroy that one because the earth doesn't move relative to a guy on the moon.
Actually it was probably from orbit, but we can't check if the footage is missing...
I always said teaching game programming would be a good way to get students interested and to maintain their interest beyond the classroom. Simple board games can be used to teach data structures and search algorithms. Simple 70's or 80's style arcade games teach real-time methods and basic cooperative multitasking. OOP anyone? The best part is that when the class is over, students are more likely to continue on their own. With a little thought, you can cover most of the CS spectrum using various games.
"Not since the quantum crisis have scientists been that arrogant to assume that their theories are set in stone"
My own observation is that ever since relativity and quantum mechanics, scientists are happy to make up wild-ass theories to explain stuff that doesn't quite fit. Witness Dark matter, Dark energy, "unseen" dimensions, time varying fundamental constants... No sir, I think physics has more arrogance now than before - people are willing to assert strange or non-intuitive explanations for everything, usually without covering all the bases first.
Slashdot Mirror
Next, do astrophysicists consider the effects of thermodynamics? You know, expansion, cooling, heating, enthalpy, etc... They seem to be assuming radiation~=temperature, so cold matter would be "dark". Do these guys really understand where the hot regions would be in such a collision?
A better solution is to make a really really good digital copy. I thought the idea was that your TV could display the signal with macrovision noise added, but your VCR would lose sync and get all garbled because it was unable to make a good copy of the "noise". If you make as perfect a copy as you can, wouldn't you be able to play it back? It may still violate copyright, but you wouldn't be circumventing anything - the macrovision would be intact.
You'll love this: My daughter was tapping her hand to the beat at 2 and 1/2 WEEKS of age. I just about fell over when my wife said "look at this" and there she was tapping away in the air. She's shown that only a few times since. Musical prodigy? Don't know, don't care. She'll be good at something, some day. I took a few years of guitar lessons, but never could do great solos - had the feel, but not the speed. I did have an excellent sense of timing (my teacher thought that came naturally) and probably the fact that we played electric piano "demos" to get my daughter to sleep while gently bouncing around to the beat, had something to do with what she did so early. If she gets interested in music, I hope she does better at it than me. If she does something completely different, that will be fine too, so long as she like SOMETHING. I only hope that she enjoys life and gets the most out of it. Some folks worry too much about who is good at what, and why.
The only dark matter is in these guys heads.
Old school wins again...
Yeah, F*** those guys. It's like trying to make laws to protect people from other people. They must be clueless...
I'm still waiting for a 64 bit processor that treats all registers the same. i.e. one load and one store instruction, but you can do fpadd or regular add on the same registers. This IMHO will reduce the number of opcodes needed, and you usually don't use a lot of FP registers and a lot of integer registers at the same time. Pure stacks suck BTW - you really need a swap (up to some depth) or a copy instruction to bring data to the top. A pure stack is too destructive. I do like the idea of a pure return stack with separate data stack.
Suppose MS wanted to run Free Software on the next XBOX and didn't want people to mess around with it. They could have Intel modify a processor any number of ways (change the opcodes for a SIMPLE example) and provide a proprietary tool chain to compile the code. No DRM, yet the users have no way to modify and run the code on that hardware. Does GPL need to require a complete tool chain be provided when binaries are provided? It seems overkill, but custom (closed) hardware running free software defeats the GPL in the same way as DRM. I need to read the new draft, but I think it suggests the broader concept of denying freedom more than DRM in particular. Thoughts?
It's like here in MI where the DNR is supposed to "manage the herd of deer" by regulating hunting. Unfortunately they make money from hunting permits. Now we have car-deer accidents all the time because of the huge number of deer. How about when the PTO becomes a revenue source for the government? Or when... You get the idea. It doesn't take a conspiracy, just a failure to consider what the incentives are for the people involved in these organisations. It's like asking me to train a group of guys from India to write software ;-)
So the questions: What is Fedora doing to improve the quality of the kernels and drivers? What is the purpose of all the tweaking? Some folks use stock kernels with Fedora, why all the messing around?
If there were a way to cause light to bend/reflect/refract around an object from any direction so as to appear ivisible, the person inside such bubble would by definition be blind - at least for the affected wavelengths. Another interesting property of such a bubble is that no light could get out - the path of light is reversible so if nothing gets in then nothing gets out. If all light is directed around, then none gets in or out, simple as that.
IMHO, Richard is still misguided in his anti-DRM efforts. You don't need DRM to prevent people from running their own version of software on proprietary hardware. Intel could sell TiVo a processor with modified op-codes and a modified assembler to match (or insert some other mod that renders GCC unusable on their hardware). TiVo could run all the free software they want and give you source code, but without the "special" tool chain you'd be screwed. No keys or DRM required. DRM is just a special case of the problem FSF would really like to address. Is FSF prepared to require that people distribute hardware specs with the software? How about just a tool chain to build from source and run on the hardware supplied? How easy does the build process have to be? It's not an easy goal, but ranting about DRM in particular isn't the solution.
The obvious use for a large number of cores is graphics. After buying ATI, AMD will likely work at a new set of instructions: SSE4 or SSE5 which will bring pixel shader functionality back to the CPU where it belongs (one core can already handle vertex ops). You'll also have the extra cores in case someone actually has a nicely parallel application. System cost will go down, but a larger share of the PC price will go to AMD. With this purchase going on it should be absolutely obvious to everyone, but the stupid tech writers don't seem to get it yet. If this is AMDs plan, it's brilliant - if they just want to put a GPU on the same die it was a waste of money and will be the end of the company.
OK, while we're at it... When I was a teenager, I worked in a lighting store warehouse. We had a pegboard witha bunch of tools on the wall. One day I came in and all the tools were painted orange. Turns out one of the guys (chris) painted all the tools orange for who knows why. So after work, I drop by my buddy Don's house. I said something like "you'd never believe what happened at work today..." He stopped me abruptly and said some thing along the lines of "WAIT. I just got this strange feeling and I'm thinking of the word 'orange'.?!?!" Don didn't know Chris. Don didn't even have a reason to have tried calling me at work, and had never done so before - so he didn't have the number. It is plausible that he made the effort to get the number and call. It is exceedingly unlikely that he ended up talking to Chris instead of me - other people answered the phone and paged us. It is also unlikely that Don would continue this joke for 20 years - I have asked him about it. In all likelyhood that is exactly what happened, but I honestly don't believe it. I don't have an explanation that satisfies me.
Personally, I had an experience where some friends and I were riding our bikes through some trails and found some tennis courts in the woods. I know for a fact that I had not been in this place before, but it sure seemed familiar. I said to my friend that it seemed really familiar and then I said there should be a drinking fountain "over there" and I pointed off about 90 degrees to the left... There it was, right where I pointed. We were both amazed. Could be that the layout of the area was the same as someplace I'd been before (unlikely - I don't play tennis) or more likely, it could be that I had seen the water fountain on the way in and hadn't conciously made note of it (i.e. it got classified as a "memory" rather than a current event) though as I rememeber it, I had not looked over that way until after I pointed. BTW, no "substances" were involved. My mind was rather stunned at there being tennis courts in this particular location (in a wooded area in the middle of a large subdivision) so it may have just played some tricks with me.
You miss the point. In 6 years we will be doing ray tracing on the CPU(s) in real time. That's better graphics than GPUs can do. At that point, the GPU is truely redundant. When is the last time anyone tried to do rendering on the CPU? Quake 1? It's come a long way since then, and now we've got extra cores that no one knows what to do with. Besides, for the majority of business computers a GPU is already unnecessary - windows can be painted in software, even with fancy compositing.
IMHO a much better - and cheaper - buy would have been that company (don't remember name) that has a really fast DirectX implementation in software. You don't even need a frame buffer any more, just put some circuitry in the north bridge that pulls data from main memory and spits it out over DVI-D. Let software on multiple cores take care of rendering.
For reference, I think we're still on track for software realtime raytracing by 2012. If we can do that, certainly software can rasterize polygons fast enough before then. The GPU as we know it is dead - good for ATI finding an exit strategy, bad for AMD spending so much money. What's nVidia's exit strategy? Intel already knows a lot about graphics.
Hey man, I remember that time lapse shot of the "earth rise" from the moon. They better destroy that one because the earth doesn't move relative to a guy on the moon.
Actually it was probably from orbit, but we can't check if the footage is missing...
I always said teaching game programming would be a good way to get students interested and to maintain their interest beyond the classroom. Simple board games can be used to teach data structures and search algorithms. Simple 70's or 80's style arcade games teach real-time methods and basic cooperative multitasking. OOP anyone? The best part is that when the class is over, students are more likely to continue on their own. With a little thought, you can cover most of the CS spectrum using various games.
My own observation is that ever since relativity and quantum mechanics, scientists are happy to make up wild-ass theories to explain stuff that doesn't quite fit. Witness Dark matter, Dark energy, "unseen" dimensions, time varying fundamental constants... No sir, I think physics has more arrogance now than before - people are willing to assert strange or non-intuitive explanations for everything, usually without covering all the bases first.