IIRC, IBM is going to deliver a system that's based on POWER and Cell chips for this exact reason - no commodity chip currently beats the Cell in floating-point brute force and POWER is quite good on everything else (including floating-point).
We will see some very interesting machines in the next few years.
As far as killing him of leukemia 6 months from now, this would not be enough to contain any information he migh have - 6 months is time enough to write a biography.
If this is a case of silencing him because he was about to disclose something really nasty, why not run over him with a truck or hit him with a falling brick or, even better, making him vanish without a trace? That's nothing a boat and a pair of cement shoes couldn't achieve. Although it's unclear if anyone was ever murdered with cement shoes, I could bet some money it's more frequent than with Polonium.
If I am ever to have information people would kill me to avoid spreading, I would spread it as fast and wide as I could, leaving my potential killers scrambling for damage control and lower my value as a target.
Of course, they might kill me later, for revenge, but later is better than sooner.
But remaining a high value target is something really dumb to do.
The royalties part may not work, but, certainly, bearing the costs of an invalidation and, eventually, having to pay damages for a bad patent would be quite an interesting change.
BTW, it should be required that the patent applicant should prove it made a reasonable effort to find about prior art and demonstrate it found none.
This patent, obviously, would fail in that regard.
- With a public and standard ISA, you will have Linux-compatible drivers shortly
- With a public and standard ISA, people will have a single standard to code against. Library support should be excellent.
- While your über-FPUs/vector accelerators/stream processors (what GPUs are made of) are not GPU-ing something, they can accelerate SSL, physics processing and any other vector-friendly activity you may have. Playing Flash content, maybe.
- GPUs are memory-hungry. The added memory bandwidth will benefit all software, not only graphics-intensive stuff.
- There is nothing that precludes you from using a stand-alone GPU, provided you have the drivers. But your CPU will have a couple high performance units that can give it a hand. Think asymmetric SLI.
We will see how well the idea performs by watching the Cell processor (a CPU with 8 "GPU"s attached) in the PS3. That's roughly the same idea.
they hand assembled a couple units from the parts that will be used in the production lines to make sure everything works and that the production process does what is expected from it.
After they say "yes, this is right" they will start the assembly lines that will mass-produce the computers.
I am not sure what this could imply for the current US administration.
As far as I get, they are responsible for invading two countries, one of them without much of a reason, after bombing it back to stone-age (the other was pretty fucked up before they got there) and establishing a network of illegal prisons and deportations that allow prisoners to be tortured and held without right to the slightest semblance of due process.
It's sad we came to this. Saddam should not have never been supported. The same goes to the Shah and Khomeini (who wasn't supported by the US, anyway) and a very long list of small dictators.
Today's verdict shows that no former ruler is immune to a sufficiently motivated tribunal.
That's why we need cheap, expendable vehicles with cheap, expendable crews.
Now seriously, wouldn't it be incredibly useful to have a transfer vehicle docked to the ISS at all times, able to pick up a stranded shuttle crew and carry them back to the ISS?
It could even be a simple, Progress-like vehicle. How much fuel would be needed for a couple orbit transfers? And in potential rescue missions like this, it could use an ion engine to take it to the pick-up point several weeks after undocking from the ISS. It would be viable for an unmanned vehicle to take months to do an orbit change. Chemical rockets would only be used to carry the crew back to the ISS.
Even better would be something that could attach to the now unmanned shuttle and take it to the ISS for further damage assessment and possible repairs. Ion engines could be useful here too - there would be no hurry to take the damaged shuttle to the ISS.
"Name fiftie countries with higher free speech standards than the us."
According to the list at http://www.rsf.org/, the US is #53, so with some effort, I could write down (copy-paste, really) at least 52 of them. According to the ranking, the US shares 53th place with Botswana, Croatia and Tonga.
Copy-paste wasn't so hard. Enjoy. Had to jump a couple so slashdot would allow me to post the list.
N Country Score 1 Finland 0,50 - Iceland 0,50 - Ireland 0,50 - Netherlands 0,50 5 Czech Republic 0,75 6 Estonia 2,00 - Norway 2,00 8 Slovakia 2,50 - Switzerland 2,50 10 Hungary 3,00 - Latvia 3,00 - Portugal 3,00 - Slovenia 3,00 (...) 26 Namibia 6,00 27 Lithuania 6,50 - United Kingdom 6,50 29 Costa Rica 6,67 30 Cyprus 7,50 31 South Korea 7,75 32 Greece 8,00 - Mauritius 8,00 34 Ghana 8,50 35 Australia 9,00 - Bulgaria 9,00 - France 9,00 - Mali 9,00 39 Panama 9,50 40 Italy 9,90 41 El Salvador 10,00 - Spain 10,00 43 Taiwan 10,50 44 South Africa 11,25 45 Cape Verde 11,50 - Macedonia 11,50 - Mozambique 11,50 - Serbia and Montenegro 11,50 49 Chile 11,63 50 Israel 12,00 51 Japan 12,50 52 Dominican Republic 12,75 53 Botswana 13,00 - Croatia 13,00 - Tonga 13,00 - United States of America 13,00
I must confess I was a bit lost when, after a couple Ubuntu releases, I installed Fedora Core 6 for a spin. It seems everything is in subtly different places.
I don't know if anyone noticed that, but the Pixel vehicle seems somewhat unstable with its single engine.
You can see it here http://media.armadilloaerospace.com/2006_09_23/Pix elQualification.wmv, later in the hover, when it starts to oscilate the engine thrust vector in order to stabilize the craft. The oscilation seems to be increasing, but it's hard to tell since the hover itself is too short.
I know they _are_ rocket scientists and, no doubt, know about this. But it seems to be a major obstacle to achieve the 180-second hover goal.
I wonder what can be done to improve it without compromising the single-engine simplicity.
Of course, doing fancy things on the 8086/8088 was horrible, but the fact that it was source-compatible with the 8085 (I think) made it possible to directly port many CP/M programs and have some titles ready to run very early.
Of course, they wouldn't be able to address more than 64K of memory, but it was not a big issue since the original PC came in configurations with as little as 16K of RAM.
And when compared to the crazy (and brilliant) things computer designers were doing with 8-bit computers to use more than 64K of memory (Apple II, Apple III, various 8085/Z-80 designs), the 8086 is easy to program. I did a lot of bank-switching in my Apple II days and I can tell you the PC was rather boring and unchallenging in comparison.
Again, the 68000 was so much nicer.
When time travel becomes available we will be able to directly test these ideas.
The PC had a fortunate spy incident in which IBM OS basics were stolen before the PC came out.
What?! Can you explain that?
The 8086 was very similar to the 8085, so it was trivial to translate ASM code that ran on 8085 to 8086 (if not the other way around). MD-DOS was, more or less very similar to CP/M, and porting software was easy.
There is also IBM's misjudgement that the BIOS alone could stop clone makers, even if the PC was made with off-the-shelf parts. This, and the non-exclusive agreement on MS-DOS between IBM and MS made the clone industry possible.
Not being known to be risky is a long way from not being risky. While a new design can be a huge improvement over an old one, specially one we know very well how it failed, it still may fail in unforeseen ways. Any engineering process is a mostly continuous spiral towards "good enough" with some mishaps along the way.
Slashdot Mirror
IIRC, IBM is going to deliver a system that's based on POWER and Cell chips for this exact reason - no commodity chip currently beats the Cell in floating-point brute force and POWER is quite good on everything else (including floating-point).
We will see some very interesting machines in the next few years.
As far as killing him of leukemia 6 months from now, this would not be enough to contain any information he migh have - 6 months is time enough to write a biography.
If this is a case of silencing him because he was about to disclose something really nasty, why not run over him with a truck or hit him with a falling brick or, even better, making him vanish without a trace? That's nothing a boat and a pair of cement shoes couldn't achieve. Although it's unclear if anyone was ever murdered with cement shoes, I could bet some money it's more frequent than with Polonium.
If I am ever to have information people would kill me to avoid spreading, I would spread it as fast and wide as I could, leaving my potential killers scrambling for damage control and lower my value as a target.
Of course, they might kill me later, for revenge, but later is better than sooner.
But remaining a high value target is something really dumb to do.
The royalties part may not work, but, certainly, bearing the costs of an invalidation and, eventually, having to pay damages for a bad patent would be quite an interesting change.
BTW, it should be required that the patent applicant should prove it made a reasonable effort to find about prior art and demonstrate it found none.
This patent, obviously, would fail in that regard.
As added benefits:
- With a public and standard ISA, you will have Linux-compatible drivers shortly
- With a public and standard ISA, people will have a single standard to code against. Library support should be excellent.
- While your über-FPUs/vector accelerators/stream processors (what GPUs are made of) are not GPU-ing something, they can accelerate SSL, physics processing and any other vector-friendly activity you may have. Playing Flash content, maybe.
- GPUs are memory-hungry. The added memory bandwidth will benefit all software, not only graphics-intensive stuff.
- There is nothing that precludes you from using a stand-alone GPU, provided you have the drivers. But your CPU will have a couple high performance units that can give it a hand. Think asymmetric SLI.
We will see how well the idea performs by watching the Cell processor (a CPU with 8 "GPU"s attached) in the PS3. That's roughly the same idea.
In the meantime, I bet it will work just fine.
AMD is making the x86 more Cell-like.
What a moron...
Idiot.
they hand assembled a couple units from the parts that will be used in the production lines to make sure everything works and that the production process does what is expected from it.
After they say "yes, this is right" they will start the assembly lines that will mass-produce the computers.
Chances are your phone was crippled by your cell-phone company. Most phones are friendlier.
Buildings that use it - and similar technologies - could get tax reductions or other compensation for doing somthing for the public good.
If done this way, it could work.
I am not sure what this could imply for the current US administration.
As far as I get, they are responsible for invading two countries, one of them without much of a reason, after bombing it back to stone-age (the other was pretty fucked up before they got there) and establishing a network of illegal prisons and deportations that allow prisoners to be tortured and held without right to the slightest semblance of due process.
It's sad we came to this. Saddam should not have never been supported. The same goes to the Shah and Khomeini (who wasn't supported by the US, anyway) and a very long list of small dictators.
Today's verdict shows that no former ruler is immune to a sufficiently motivated tribunal.
The ones you buy a couple years from now.
Writing this on a notebook that outpaces the US$ 100K workstation of a couple years back.
You know... It would be fun to have a Theo x RMS deathmatch.
It would be better to have a good enough telescope in a serviceable orbit than having a super-fancy telescope in an utterly difficult to get at orbit.
Ease of maintenance is directly related to operational cost and, therefore, to the lifespan of the equipment.
And, being easily serviceable would mean new instruments could be attached and old instruments removed with much more ease.
As someone else said, there must be something escaping me. After all, there must be a good reason I am not building spaceships myself.
BTW, why not a Moon-based one?
That's why we need cheap, expendable vehicles with cheap, expendable crews.
Now seriously, wouldn't it be incredibly useful to have a transfer vehicle docked to the ISS at all times, able to pick up a stranded shuttle crew and carry them back to the ISS?
It could even be a simple, Progress-like vehicle. How much fuel would be needed for a couple orbit transfers? And in potential rescue missions like this, it could use an ion engine to take it to the pick-up point several weeks after undocking from the ISS. It would be viable for an unmanned vehicle to take months to do an orbit change. Chemical rockets would only be used to carry the crew back to the ISS.
Even better would be something that could attach to the now unmanned shuttle and take it to the ISS for further damage assessment and possible repairs. Ion engines could be useful here too - there would be no hurry to take the damaged shuttle to the ISS.
"Name fiftie countries with higher free speech standards than the us."
According to the list at http://www.rsf.org/, the US is #53, so with some effort, I could write down (copy-paste, really) at least 52 of them. According to the ranking, the US shares 53th place with Botswana, Croatia and Tonga.
Copy-paste wasn't so hard. Enjoy. Had to jump a couple so slashdot would allow me to post the list.
N Country Score
1 Finland 0,50
- Iceland 0,50
- Ireland 0,50
- Netherlands 0,50
5 Czech Republic 0,75
6 Estonia 2,00
- Norway 2,00
8 Slovakia 2,50
- Switzerland 2,50
10 Hungary 3,00
- Latvia 3,00
- Portugal 3,00
- Slovenia 3,00
(...)
26 Namibia 6,00
27 Lithuania 6,50
- United Kingdom 6,50
29 Costa Rica 6,67
30 Cyprus 7,50
31 South Korea 7,75
32 Greece 8,00
- Mauritius 8,00
34 Ghana 8,50
35 Australia 9,00
- Bulgaria 9,00
- France 9,00
- Mali 9,00
39 Panama 9,50
40 Italy 9,90
41 El Salvador 10,00
- Spain 10,00
43 Taiwan 10,50
44 South Africa 11,25
45 Cape Verde 11,50
- Macedonia 11,50
- Mozambique 11,50
- Serbia and Montenegro 11,50
49 Chile 11,63
50 Israel 12,00
51 Japan 12,50
52 Dominican Republic 12,75
53 Botswana 13,00
- Croatia 13,00
- Tonga 13,00
- United States of America 13,00
I must confess I was a bit lost when, after a couple Ubuntu releases, I installed Fedora Core 6 for a spin. It seems everything is in subtly different places.
They are not racist. They are massist.
Nice to know. I'll try to find a video of that flight.
Sorry. I got the video from a news site. They didn't carry the text of the page
I don't know if anyone noticed that, but the Pixel vehicle seems somewhat unstable with its single engine.
x elQualification.wmv, later in the hover, when it starts to oscilate the engine thrust vector in order to stabilize the craft. The oscilation seems to be increasing, but it's hard to tell since the hover itself is too short.
You can see it here http://media.armadilloaerospace.com/2006_09_23/Pi
I know they _are_ rocket scientists and, no doubt, know about this. But it seems to be a major obstacle to achieve the 180-second hover goal.
I wonder what can be done to improve it without compromising the single-engine simplicity.
Of course, doing fancy things on the 8086/8088 was horrible, but the fact that it was source-compatible with the 8085 (I think) made it possible to directly port many CP/M programs and have some titles ready to run very early.
Of course, they wouldn't be able to address more than 64K of memory, but it was not a big issue since the original PC came in configurations with as little as 16K of RAM.
And when compared to the crazy (and brilliant) things computer designers were doing with 8-bit computers to use more than 64K of memory (Apple II, Apple III, various 8085/Z-80 designs), the 8086 is easy to program. I did a lot of bank-switching in my Apple II days and I can tell you the PC was rather boring and unchallenging in comparison.
Again, the 68000 was so much nicer.
When time travel becomes available we will be able to directly test these ideas.
Sadly, I think it would be a more diverse world.
I think this x86-only stuff is BORING. I miss the 70s and 80s.
What?! Can you explain that?
The 8086 was very similar to the 8085, so it was trivial to translate ASM code that ran on 8085 to 8086 (if not the other way around). MD-DOS was, more or less very similar to CP/M, and porting software was easy.
There is also IBM's misjudgement that the BIOS alone could stop clone makers, even if the PC was made with off-the-shelf parts. This, and the non-exclusive agreement on MS-DOS between IBM and MS made the clone industry possible.
So, what spy incident are you talking about?
Not being known to be risky is a long way from not being risky. While a new design can be a huge improvement over an old one, specially one we know very well how it failed, it still may fail in unforeseen ways. Any engineering process is a mostly continuous spiral towards "good enough" with some mishaps along the way.
The default root password is R1-L2-circle-square-left-R3 and a full circle with the right analog stick.