I can imagine that the cost and energy consumption of the machine is exorbitant. Even all the dishes you need can probably fit in a space as large as that dish maker. Most dishes stack very well and they clean with little energy. Heck, that video glosses over how they are cleaned, there's nothing on that machine that cleans the plastic before it's recycled, I can imagine that anything left on the plastic when it's reheated will embed itself into the plastic.
I think this transition with rapid prototyping is going to be a lot slower than printers for several reasons, assuming it happens.
The skills necessary to make anything interesting in 3D seem to be a lot less common than the skills necessary to slap together a home movie or a garage music recording.
Also, increasing the quality of reproductions means reducing the size of the steps. Printers are lucky enough to be stuck in 2D. If we have 1000 dpi resolution in 3D, that's one billion cells within a cubic inch. A printer would only have to deal with one million per square inch, and multiply the "print" time by a thousand. Rapid prototyping machines are still pretty crude and very slow. For the reason I just pointed out, I was told that a one cubic inch object might take four to five hours. This isn't any faster or much cheaper than when I asked five years ago. At least the machine outputs parts made of ABS plastic, so it makes a usable part. Lasers of the power necessary to do this well aren't getting any cheaper, and I don't see semiconductor lasers outputing the necessary power for a long time.
Besides, if Pamela Jones is just just an internet pseudonym for someone else, then what good would tracking down people named Pamela Jones in public records, if you already believe that.
It was definitely a nice looking car for the time, I watched the pilot and I know this might have been partly intentional, but all the other cars in the pilot were just ugly. Still, it's kind of dated now, in my opinion.
Aircraft flights are stressful enough as it is. The constant howl of the engines gets to me, and on one flight, my knees were basically jammed against the seat in front of me, despite it being in the fully vertical position.
Erring on the side of caution, as in, not wanting to give the slightest chance to crash a plane because a phone's signals confused the navigation equipment. I think the Mythbusters showed that most phones don't interfere, but that was only on one type of plane, and only a handful of phones. Their test rig did show a little bit of squirrely behavior with one band of GSM phones.
Avid is closer to 40/60 Mac/PC. Avid tried to axe the Mac port but that really hurt them.
I wouldn't write off FCP. It's too expensive to just use as a hobbyist editor. BBC and Al-Jezeera are two major networks switching to it. A lot of theatrical movies are being done with it, Zodiac is one. I know a guy that cuts TV ads in HD, and he prefers his own FCP setup to the TV station's Avid, and he's not that fond of the Mac platform. He could do Premiere too, but he's using After Effects in conjunction with FCP.
Premiere Pro CS3 is available for Mac Intel. I somehow doubt that they would be bothering if there wasn't enough of a market to justify adding Mac support to a program that didn't have it.
The very idea of AM and PM based off of the position of the sun in the sky, mid-day, and the middle of the day loses its original point of reference, so those are out the door too.
Computers and logging can and do run off GMT with or without the idea of time zones.
I think the 10x figure is without equivalent compatibility, as in, you junk your exising stuff. I think the number would be much lower if you could run all the software at native speeds. Keeping equivalent compatibility lowers that barrier by a significant margin.
The sacrifice isn't much. Last I heard is that less than 10% of a core (I assume excluding cache) is needed to do the decoding. Even assuming you eliminate all decoding costs by going RISC or VLIW, you get to be about 10% faster or 10% more efficient. Being 10% better is not worth dumping the entire hardware or software investment. The hardware is going to be 2x more expensive because there's no economy of scale, it would only be used for specialized tasks, ones where the software is not usually extended, upgraded or replaced by the owner.
The problem is that it horribly splits up development work. It isn't as if there are enough OSS developers as it is, and they seem to fork their way out of existence. These developers have to compete against multi billion dollar software companies that provide reasonably unified APIs, UIs and frankly, better backward compatibility. I'm still using a piece of expensive CAD software made in 1994, designed for Win32S on Win 3.1, and it still works fine on XP, for all I know, it might even work under Vista, I won't know because I don't plan to get it. Sure, statically built linux binaries from that time probably will work, but should it need a library, you are more likely than not stopped right there.
Also, I've never heard of Compiz until this story.
One major reason that the newer notebooks might consume less power on average is because the desknote type system has been marginalized. Those systems were just monsters in weight and power consumption, I've seen some that used desktop Pentium 4 chips. One guy I knew used carry-on type luggage to haul it around rather than a notebook bag, because it was too big to fit in any existing notebook bag, being 2" thick, 17" screen and requiring two or three honking fans.
It is silly, but I think it would be nice if you could pick a time period. I know that the photos are updated in a patchwork manner, but if there was a way to "peel back" the history, I think it would be very interesting.
True. Actually, there were very, very few Firewire enclosures that could power a 3.5" drive. Actually, the only one I've seen was from Weibetech and they discontinued it. Those enclosures were too expensive. The 2.5" drives are too slow and low capacity to justify using just to save a power brick in a desktop situation, for mobile needs, a 2.5" drive is almost necessary, the 3.5" drives were usually too big.
Maybe if you are designing a budget car, but I don't think that money is a valid excuse for a BMW's user interface. They can afford to use industrial grade encoders and slides.
The ISS probably has significant heating and cooling systems, but I doubt a fan would be that hard to use. Astronauts in microgravity are supposed to exercise often, so I think they probably have that sort of problem taken care of.
Re:I hope they've fixed the memory hogging.
on
Firefox 3.0 Preview
·
· Score: 1
I think the core execution engine is supposed to be improved, so this problem might be fixed. One way to know is to try the alpha build and see what happens. It's probably fine, I've used the nightlies without too much issue. Memory isn't a problem for me anyway, my system has more memory than I've ever seen it use.
When it comes to GameStop vs. anyone else, I think my biases would lean to whoever that other party is. Let's just say that GameStop's retail practices aren't totally upstanding.
In the US, abolishing copyright would probably mean amending the constitution.
I make a distinction between personal use and redistribution. Copying what you bought, for personal use, is fine with me, but I don't think redistribution is acceptable. I think that used to be well within line of what was granted through copyright, before it was turned into a monster.
This ease of copying thing is one reason I really don't bother making software. I can't afford to fund myself, I don't think anyone else will, and how I make it worth my time is with the users paying for the software. If a person that would otherwise pay for the software can just download the software, then that doesn't support my work. The type of software I would make is specialized enough that tip-ware won't work, but then, I'm not convinced that tip-ware funding would make it worth my time developing software, there's quite a bit of risk there.
That's why I just make my money with physical objects. My products are hard to duplicate without expensive equipment or a lot of time, and I try to price it such that it's not worth trying to copy it. I suppose that might go away if FAB quality goes up and it's affordable to buy for home use, but that is a long ways away, all forms of rapid prototyping produce crude objects, especially the stuff made with the FAB project.
I think using breeder reactors is a very good idea, but I do understand the aversion to allowing this technology to proliferate because it's easy to change to make weapons grade plutonium too.
I'm not convinced symmetrical multithreading is a gimmick. The designers of the Alpha chip were puting in SMT onto the EV8 core and generally those designers weren't prone to including gimmicks into their chips. I think the difference was that Alphas had six functional units rather than three. Most of those same designers went on to work for AMD after they left Compaq.
However, it's worth noting, that these are clearly AMD ideas. * On die memory controller - AMD's idea - and it's been in use for quite a while now * Embedded GPU - a rip off of the AMD fusion idea, announced shortly after the aquisition of AMD.
Neither on-die memory controler, nor embedded GPU were AMD's idea, they came from elsewhere. For example, the DEC Alpha 21066 chip had an on-die memory controller, introduced in 1994. I'm sure someone else can submit where there was a CPU with an on-die graphics controller, maybe some ARM chips did that.
Slashdot Mirror
I can imagine that the cost and energy consumption of the machine is exorbitant. Even all the dishes you need can probably fit in a space as large as that dish maker. Most dishes stack very well and they clean with little energy. Heck, that video glosses over how they are cleaned, there's nothing on that machine that cleans the plastic before it's recycled, I can imagine that anything left on the plastic when it's reheated will embed itself into the plastic.
I think this transition with rapid prototyping is going to be a lot slower than printers for several reasons, assuming it happens.
The skills necessary to make anything interesting in 3D seem to be a lot less common than the skills necessary to slap together a home movie or a garage music recording.
Also, increasing the quality of reproductions means reducing the size of the steps. Printers are lucky enough to be stuck in 2D. If we have 1000 dpi resolution in 3D, that's one billion cells within a cubic inch. A printer would only have to deal with one million per square inch, and multiply the "print" time by a thousand. Rapid prototyping machines are still pretty crude and very slow. For the reason I just pointed out, I was told that a one cubic inch object might take four to five hours. This isn't any faster or much cheaper than when I asked five years ago. At least the machine outputs parts made of ABS plastic, so it makes a usable part. Lasers of the power necessary to do this well aren't getting any cheaper, and I don't see semiconductor lasers outputing the necessary power for a long time.
Besides, if Pamela Jones is just just an internet pseudonym for someone else, then what good would tracking down people named Pamela Jones in public records, if you already believe that.
It was definitely a nice looking car for the time, I watched the pilot and I know this might have been partly intentional, but all the other cars in the pilot were just ugly. Still, it's kind of dated now, in my opinion.
Aircraft flights are stressful enough as it is. The constant howl of the engines gets to me, and on one flight, my knees were basically jammed against the seat in front of me, despite it being in the fully vertical position.
Erring on the side of caution, as in, not wanting to give the slightest chance to crash a plane because a phone's signals confused the navigation equipment. I think the Mythbusters showed that most phones don't interfere, but that was only on one type of plane, and only a handful of phones. Their test rig did show a little bit of squirrely behavior with one band of GSM phones.
Avid is closer to 40/60 Mac/PC. Avid tried to axe the Mac port but that really hurt them.
I wouldn't write off FCP. It's too expensive to just use as a hobbyist editor. BBC and Al-Jezeera are two major networks switching to it. A lot of theatrical movies are being done with it, Zodiac is one. I know a guy that cuts TV ads in HD, and he prefers his own FCP setup to the TV station's Avid, and he's not that fond of the Mac platform. He could do Premiere too, but he's using After Effects in conjunction with FCP.
Premiere Pro CS3 is available for Mac Intel. I somehow doubt that they would be bothering if there wasn't enough of a market to justify adding Mac support to a program that didn't have it.
The very idea of AM and PM based off of the position of the sun in the sky, mid-day, and the middle of the day loses its original point of reference, so those are out the door too.
Computers and logging can and do run off GMT with or without the idea of time zones.
I've seen the show. It's on the Science Channel on occasion.
I think the 10x figure is without equivalent compatibility, as in, you junk your exising stuff. I think the number would be much lower if you could run all the software at native speeds. Keeping equivalent compatibility lowers that barrier by a significant margin.
The sacrifice isn't much. Last I heard is that less than 10% of a core (I assume excluding cache) is needed to do the decoding. Even assuming you eliminate all decoding costs by going RISC or VLIW, you get to be about 10% faster or 10% more efficient. Being 10% better is not worth dumping the entire hardware or software investment. The hardware is going to be 2x more expensive because there's no economy of scale, it would only be used for specialized tasks, ones where the software is not usually extended, upgraded or replaced by the owner.
The problem is that it horribly splits up development work. It isn't as if there are enough OSS developers as it is, and they seem to fork their way out of existence. These developers have to compete against multi billion dollar software companies that provide reasonably unified APIs, UIs and frankly, better backward compatibility. I'm still using a piece of expensive CAD software made in 1994, designed for Win32S on Win 3.1, and it still works fine on XP, for all I know, it might even work under Vista, I won't know because I don't plan to get it. Sure, statically built linux binaries from that time probably will work, but should it need a library, you are more likely than not stopped right there.
Also, I've never heard of Compiz until this story.
One major reason that the newer notebooks might consume less power on average is because the desknote type system has been marginalized. Those systems were just monsters in weight and power consumption, I've seen some that used desktop Pentium 4 chips. One guy I knew used carry-on type luggage to haul it around rather than a notebook bag, because it was too big to fit in any existing notebook bag, being 2" thick, 17" screen and requiring two or three honking fans.
It is silly, but I think it would be nice if you could pick a time period. I know that the photos are updated in a patchwork manner, but if there was a way to "peel back" the history, I think it would be very interesting.
True. Actually, there were very, very few Firewire enclosures that could power a 3.5" drive. Actually, the only one I've seen was from Weibetech and they discontinued it. Those enclosures were too expensive. The 2.5" drives are too slow and low capacity to justify using just to save a power brick in a desktop situation, for mobile needs, a 2.5" drive is almost necessary, the 3.5" drives were usually too big.
Maybe if you are designing a budget car, but I don't think that money is a valid excuse for a BMW's user interface. They can afford to use industrial grade encoders and slides.
The ISS probably has significant heating and cooling systems, but I doubt a fan would be that hard to use. Astronauts in microgravity are supposed to exercise often, so I think they probably have that sort of problem taken care of.
I think the core execution engine is supposed to be improved, so this problem might be fixed. One way to know is to try the alpha build and see what happens. It's probably fine, I've used the nightlies without too much issue. Memory isn't a problem for me anyway, my system has more memory than I've ever seen it use.
When it comes to GameStop vs. anyone else, I think my biases would lean to whoever that other party is. Let's just say that GameStop's retail practices aren't totally upstanding.
It's old enough that you shouldn't trust. Remember "don't trust anyone over 30"?
Now they need a "large print" version of the DS. None of this "micro" stuff.
In the US, abolishing copyright would probably mean amending the constitution.
I make a distinction between personal use and redistribution. Copying what you bought, for personal use, is fine with me, but I don't think redistribution is acceptable. I think that used to be well within line of what was granted through copyright, before it was turned into a monster.
This ease of copying thing is one reason I really don't bother making software. I can't afford to fund myself, I don't think anyone else will, and how I make it worth my time is with the users paying for the software. If a person that would otherwise pay for the software can just download the software, then that doesn't support my work. The type of software I would make is specialized enough that tip-ware won't work, but then, I'm not convinced that tip-ware funding would make it worth my time developing software, there's quite a bit of risk there.
That's why I just make my money with physical objects. My products are hard to duplicate without expensive equipment or a lot of time, and I try to price it such that it's not worth trying to copy it. I suppose that might go away if FAB quality goes up and it's affordable to buy for home use, but that is a long ways away, all forms of rapid prototyping produce crude objects, especially the stuff made with the FAB project.
I think using breeder reactors is a very good idea, but I do understand the aversion to allowing this technology to proliferate because it's easy to change to make weapons grade plutonium too.
I'm not convinced symmetrical multithreading is a gimmick. The designers of the Alpha chip were puting in SMT onto the EV8 core and generally those designers weren't prone to including gimmicks into their chips. I think the difference was that Alphas had six functional units rather than three. Most of those same designers went on to work for AMD after they left Compaq.
However, it's worth noting, that these are clearly AMD ideas.
* On die memory controller - AMD's idea - and it's been in use for quite a while now
* Embedded GPU - a rip off of the AMD fusion idea, announced shortly after the aquisition of AMD.
Neither on-die memory controler, nor embedded GPU were AMD's idea, they came from elsewhere. For example, the DEC Alpha 21066 chip had an on-die memory controller, introduced in 1994. I'm sure someone else can submit where there was a CPU with an on-die graphics controller, maybe some ARM chips did that.