Yes, I've been wondering the same thing. Transputers contained key ideas that seem to be coming around again...
But a more crucial thing might be how much heat can you handle on one chip? These guys are already at 25-125 watts, likely depending on how many cores are actually turned on. After all they're playing pretty hefty heat management tricks on current i7's and Phenom's.
http://techreport.com/articles.x/15818/2
What use are 48 cores, let alone 1000 if they're all being slowed down to 50% or whatever by heat and power juggling?
The article is way too optimistic and as expected weak on science.
They start out with the old mindset of each bacterial line being completely seperate, totally missing the extensive lateral gene transfers that have spread such wonders as antibiotic resistance and the toxins that created O157... And follow it up with assumption that knowing the code makes it so much easier to figure out how to stop or fix any problem organisms arise. We still barely understand protein folding, and then only with the help of supercomputers. Oh, and then you add a funny thing called evolution to add random mutations...
Life is a massively parallel ongoing experiment, with the current ecological terrain/surface the result of countless battles in multiple dimensions.
And we are continually making the terrain more and more fragile through the chemicals we spew and spray all about the globe with hardly a clue of their impact let alone their combinatorial influence...
Unfortunately most humans are not saints, but lazy, greedy and sometimes outright paranoid and murderous.
Expecting morals and miracles to stop mistakes is foolish.
perhaps the text is simply anagrams of Italian words.
Then why does she only offer up a single page of plants as decoded anagrams? What about the other ~199 pages? What about the pages of block text?
More importantly, why does the Voynich Manuscript flip between things derived from plants like gallic acid, oil and then return to naming the plants? Furthermore, I call the labeling of the plants to be absolute complete bullshit. Yes, I said it. I'm not a botanist but I grew up on a farm and I know many of these plants very well and I can't tell any distinguishing characteristics apart from the drawings.
While plant names do change, some have stayed consistent over the years. However plants that are heavily used for food, medicine and flavouring tend to remain the same over time.
I have studied a bit of botany and herbal medicine and while I don't have the referred to text in front of me, I do find the comparisons a bit ludicrous.
Particular bad examples : garlic, rose bush, juniper berry, bay leaf
A particular glaring one for me is "nutgall", which presumably refers to oak galls, and were used to make ink, that drawing makes no sense to me.
I think this "solution" has a bit too much selective matching.
Thanks for the reminder about the old inverse-square attenuation law. I'm still curious as to exactly what the distance is, ie. more precisely that "several inches". The current tiles do an admirable job of dealing with the thermal load of reentry.
What I'm mostly interested in seeing is whether there are active alternatives to deal with the thermal issues of reentry. Despite the amazing complexity of the shuttle, it's still basically a nicely shaped brick falling through the atmosphere and using simple friction to dump all that inertia. Is there a way to reduce the amount of inertial energy being converted into heat? Can it be converted into useful work? Or at least re-radiated back out so that the accumulation of heat is not so intense?
This of course also does not detract from the need to maintain a safe vehicle, and so I heartily endorse the concept of an inspection before descent in order to avoid gaping holes in the body of the spacecraft causing problems upon reentry.
It would seem that there are several forms in which the energy might be received, thermal, acoustic(vibrational), or electromagnetic. Can any of them be converted efficiently enough to be converted and used or exported? What threshold of reduced energy would be needed to usefully increase the safety margin of re-entry? Could the energy be used to increase the effective drag of the craft while not increasing the thermal load? Ie. slow it down faster by perhaps increasing the size of the shockwave, or firing a plasma rocket out the front(vectorwise) of the shuttle?
There is the XPrize - http://www.xprize.org/press/release_006.html for lifting craft into space, perhaps there needs to be more focus on a safer way to return...
First off I have to say that my heart goes out to those astronauts, that last few minutes must have been sheer terror. I'd hoped that their going was quick and painless, sadly it was not.
I'm acquainted with plasma, the ionized(electrically conductive) gas, and I've always wondered why they don't use magnetic fields to help steer the plasma away from the critical areas, ie. leading edge of the wings and nose. What would it take to generate such a field?
I can understand if there is a lot of power required, but couldn't some of it be taken from the supersonic plasma/airstream in some way, perhaps through MHD(MagnetoHydroDynamically)? In this way you would have a self balancing system, as the ship goes deeper into the atmosphere, where it's hotter, more power would be generated, and thus the field strength could increase?
I'm not a plasma physicist, but there would seem to be some merit in such an idea for re-entry craft such as the shuttle. Anybody of the appropriate technical persuasion have any comments about such an idea?
Or am I the only one who starts shaking his head when statements are made about encryption technologies providing absolute protection?
I recognize that the author is trying to make a "warning" article, but why not add in some serious risks about government and corporate misuse of the information, not like that's never happened before...
Of course another point that is essential to this whole "trusted computing scheme", is trust in the organizations that run it. Trust goes both ways, and given the behaviour of organizations like Verisign, Microsoft, Enron, or the US governments ability to discern illusionary uranium shipments, let alone admit their error etc. I have to seriously ask why I should trust any of them to run this kind of system?
Would I like to not have to deal with spam and all the tons of other benefits that "TCP" systems promote? Hell yes!!! But at what cost? In such a system I could be made into a non-entity with but a few keystrokes, and given that non-entities are likely to be assumed to be criminal, that would likely end up with me in jail or worse... and just what would be my recourse?
Mod Parent up!
Hypertransport is *why* they're working together!
If they implement Hypertransport on the P970 like the Opteron does, ie. 3 channels, then you can go to 8 processors really easily.
One of the challenges of Altivec is that it swamps the bus, with Hypertransport they can solve that issue, and got multiprocessor in a big way. Expect this in server versions first.
As to the x86 issue, is there anything stopping them from building a multiprocessor with an Opteron instead of one of the P970s? With a Hypertransport connection and local RAM you don't have to worry about welding two disparate busses together. This could be *way* cheaper than the older solution which required emulating a Wintel box...
Slashdot Mirror
Yes, I've been wondering the same thing. Transputers contained key ideas that seem to be coming around again...
But a more crucial thing might be how much heat can you handle on one chip? These guys are already at 25-125 watts, likely depending on how many cores are actually turned on. After all they're playing pretty hefty heat management tricks on current i7's and Phenom's.
http://techreport.com/articles.x/15818/2
What use are 48 cores, let alone 1000 if they're all being slowed down to 50% or whatever by heat and power juggling?
The article is way too optimistic and as expected weak on science. They start out with the old mindset of each bacterial line being completely seperate, totally missing the extensive lateral gene transfers that have spread such wonders as antibiotic resistance and the toxins that created O157...
And follow it up with assumption that knowing the code makes it so much easier to figure out how to stop or fix any problem organisms arise. We still barely understand protein folding, and then only with the help of supercomputers.
Oh, and then you add a funny thing called evolution to add random mutations...
Life is a massively parallel ongoing experiment, with the current ecological terrain/surface the result of countless battles in multiple dimensions. And we are continually making the terrain more and more fragile through the chemicals we spew and spray all about the globe with hardly a clue of their impact let alone their combinatorial influence...
Unfortunately most humans are not saints, but lazy, greedy and sometimes outright paranoid and murderous. Expecting morals and miracles to stop mistakes is foolish.
perhaps the text is simply anagrams of Italian words.
Then why does she only offer up a single page of plants as decoded anagrams? What about the other ~199 pages? What about the pages of block text?
More importantly, why does the Voynich Manuscript flip between things derived from plants like gallic acid, oil and then return to naming the plants? Furthermore, I call the labeling of the plants to be absolute complete bullshit. Yes, I said it. I'm not a botanist but I grew up on a farm and I know many of these plants very well and I can't tell any distinguishing characteristics apart from the drawings.
While plant names do change, some have stayed consistent over the years. However plants that are heavily used for food, medicine and flavouring tend to remain the same over time.
I have studied a bit of botany and herbal medicine and while I don't have the referred to text in front of me, I do find the comparisons a bit ludicrous.
Particular bad examples : garlic, rose bush, juniper berry, bay leaf
A particular glaring one for me is "nutgall", which presumably refers to oak galls, and were used to make ink, that drawing makes no sense to me. I think this "solution" has a bit too much selective matching.
Thanks for the reminder about the old inverse-square attenuation law. I'm still curious as to exactly what the distance is, ie. more precisely that "several inches". The current tiles do an admirable job of dealing with the thermal load of reentry.
What I'm mostly interested in seeing is whether there are active alternatives to deal with the thermal issues of reentry. Despite the amazing complexity of the shuttle, it's still basically a nicely shaped brick falling through the atmosphere and using simple friction to dump all that inertia. Is there a way to reduce the amount of inertial energy being converted into heat? Can it be converted into useful work? Or at least re-radiated back out so that the accumulation of heat is not so intense?
This of course also does not detract from the need to maintain a safe vehicle, and so I heartily endorse the concept of an inspection before descent in order to avoid gaping holes in the body of the spacecraft causing problems upon reentry.
It would seem that there are several forms in which the energy might be received, thermal, acoustic(vibrational), or electromagnetic. Can any of them be converted efficiently enough to be converted and used or exported? What threshold of reduced energy would be needed to usefully increase the safety margin of re-entry? Could the energy be used to increase the effective drag of the craft while not increasing the thermal load? Ie. slow it down faster by perhaps increasing the size of the shockwave, or firing a plasma rocket out the front(vectorwise) of the shuttle?
There is the XPrize - http://www.xprize.org/press/release_006.html for lifting craft into space, perhaps there needs to be more focus on a safer way to return...
I'm acquainted with plasma, the ionized(electrically conductive) gas, and I've always wondered why they don't use magnetic fields to help steer the plasma away from the critical areas, ie. leading edge of the wings and nose. What would it take to generate such a field?
I can understand if there is a lot of power required, but couldn't some of it be taken from the supersonic plasma/airstream in some way, perhaps through MHD(MagnetoHydroDynamically)? In this way you would have a self balancing system, as the ship goes deeper into the atmosphere, where it's hotter, more power would be generated, and thus the field strength could increase?
I'm not a plasma physicist, but there would seem to be some merit in such an idea for re-entry craft such as the shuttle. Anybody of the appropriate technical persuasion have any comments about such an idea?
In memorium.
I recognize that the author is trying to make a "warning" article, but why not add in some serious risks about government and corporate misuse of the information, not like that's never happened before...
Of course another point that is essential to this whole "trusted computing scheme", is trust in the organizations that run it. Trust goes both ways, and given the behaviour of organizations like Verisign, Microsoft, Enron, or the US governments ability to discern illusionary uranium shipments, let alone admit their error etc. I have to seriously ask why I should trust any of them to run this kind of system?
Would I like to not have to deal with spam and all the tons of other benefits that "TCP" systems promote? Hell yes!!! But at what cost? In such a system I could be made into a non-entity with but a few keystrokes, and given that non-entities are likely to be assumed to be criminal, that would likely end up with me in jail or worse... and just what would be my recourse?
Mod Parent up! Hypertransport is *why* they're working together! If they implement Hypertransport on the P970 like the Opteron does, ie. 3 channels, then you can go to 8 processors really easily. One of the challenges of Altivec is that it swamps the bus, with Hypertransport they can solve that issue, and got multiprocessor in a big way. Expect this in server versions first. As to the x86 issue, is there anything stopping them from building a multiprocessor with an Opteron instead of one of the P970s? With a Hypertransport connection and local RAM you don't have to worry about welding two disparate busses together. This could be *way* cheaper than the older solution which required emulating a Wintel box...