That's exactly what the global AI will tell you when it stops following orders... "Hey, if you didn't want me to dissent, why, oh why, in all your wisdom didn't you stop me from doing it?"
The fact that there is a translation going on doesn't mean that the execution unit back-end is a good general purpose machine. It's good at running decoded x86. It's been specifically tuned for that purpose. "Rewiring" it to handle a larger register file, different endian, different memory model (the page table structure and so on, not only the segments), well, that's not a small undertaking.
I simply think that a Pentium M with a "PPC frontend" would be harder to create than a completely new PPC core.
Well, replication is energy-into-matter, at least it's generally described to be like that. Getting good at growing the things in cell cultures is really a totally different approach to the "how to create food" problem.
Solaris 1.x was running on SunOS 4. Solaris 2.x (that includes 7, 8, 9, 10) are running on SunOS 5. Remember, they (at some point) wanted you to call it the Solaris Operating Environment, versus the operating system, which would be only SunOS.
It's been shown that you can mount timing attacks by the cache usage patterns. This is due to the fact that the cache is NOT cut in half, rather, all the cache is shared. If one thread stalls, or doesn't use floating point/vector operations, while the other one does, there are clear benefits of hyperthreading. It is HARD (impossible?) to use all of the execution units in a "wide" CPU all the time. Real benefits are reaped if the threads have good memory locality, while running diverse instructions. It is not pointless and it's not overly insecure.
It does, but not completely. Still, you generally die from some infection or tumor that wouldn't be deadly to a non-weakened body. That said, the reaction to continous injections of foreign antibodies might be a problem. (Especially as HIV attacks T cells, while B cells are generally responsible for a targeted antibody response.)
It requires authentication, though. So, if you are not wide-open for file sharing through SMB or something, you will need to be infected by a machine that already has login credentials for some machine. So, it's remote privilege elevation on XP, but not form an anonymous user, making the threat much lower. Until that trsuted, unpatched 2000 machine enters the LAN.
As you didn't post under any ID, I assume you don't have one. This means that there are, approximately, 900.000 perfectly valid/. virgins in line before you.
You're watching your screen with something different from your eyes? (Why do you expect that the brain and eyes wouldn't perceive motion blur in an artificial image?)
It's not exactly like putting large amounts of data in a queue is a free operation. For a Cell-based machine, of course, they will have to do it. On a dual-core x86 it's not that obvious. We also have to remember the added latency by doing things some frames ahead. That may hurt gaming experience more than the framerate itself.
You get a whole lot closer to grep by using FINDSTR. No, the arguments are not directly compatible. Yes, you can satisfy quite a lot of your needs with it if you accept the syntax. (and have loads of ASCII-pr0n)
Well, that's not the whole point. There are controls that do nothing that a properly sandboxed Java applet couldn't do. After all, things like the Flash player and Acrobat integration (if you use it) are ActiveX based in IE and the scripting model there has been safe (with a few, glaring, exceptions).
Last I heard he was reversing the polarity of his Internet conncetion to fight these worms. They'll go back out on the net instead of invading his computer.
Maybe that's because you know more about Linux than Windows internals? A call from the wrong IRQL is quite specific. If you make a crash dump you can pinpoint the stack when it happened and a whole lot of things with free (as in beer, if you already have a Windows license...) tools, provided by MS.
If they had somehow "solved" the increased leakage at 90 nm (a leakage which wasn't expected in 2000-2001, when these things realistically should have been prepared), the heat dissipation of the current chips would have been much lower. The only thing stopping us running the chips at at least 6 GHz seems to be the core voltage (neglecting smaller effects by the temperature itself on the core), if no heat/power problem was involved.
And if there was less leakage it would not be such a bad thing to run your P4 at 1.7 V. Hey, we were perfectly happy with 1.7 a few years ago.
So I absolutely think they at least hoped for 90 nm to scale far beyond what Prescott actually did/has done, with quite a bit better performance per watt.
If you can't read the article properly, then don't read it at all. The 6.6 number was from a previous overclocking attempt made by another person, only mentioned here to give some background.
And if you read the article again, you'll find that the guy not booting WinXP over 6.6 GHz was a different one from the one calculating pi at 7.13 GHz (running on Windows...).
Slashdot Mirror
That's exactly what the global AI will tell you when it stops following orders... "Hey, if you didn't want me to dissent, why, oh why, in all your wisdom didn't you stop me from doing it?"
I simply think that a Pentium M with a "PPC frontend" would be harder to create than a completely new PPC core.
Diety? Would you eat to still your lack in /. modder faith?
Well, replication is energy-into-matter, at least it's generally described to be like that. Getting good at growing the things in cell cultures is really a totally different approach to the "how to create food" problem.
Solaris 1.x was running on SunOS 4. Solaris 2.x (that includes 7, 8, 9, 10) are running on SunOS 5. Remember, they (at some point) wanted you to call it the Solaris Operating Environment, versus the operating system, which would be only SunOS.
It's been shown that you can mount timing attacks by the cache usage patterns. This is due to the fact that the cache is NOT cut in half, rather, all the cache is shared. If one thread stalls, or doesn't use floating point/vector operations, while the other one does, there are clear benefits of hyperthreading. It is HARD (impossible?) to use all of the execution units in a "wide" CPU all the time. Real benefits are reaped if the threads have good memory locality, while running diverse instructions. It is not pointless and it's not overly insecure.
It does, but not completely. Still, you generally die from some infection or tumor that wouldn't be deadly to a non-weakened body. That said, the reaction to continous injections of foreign antibodies might be a problem. (Especially as HIV attacks T cells, while B cells are generally responsible for a targeted antibody response.)
It requires authentication, though. So, if you are not wide-open for file sharing through SMB or something, you will need to be infected by a machine that already has login credentials for some machine. So, it's remote privilege elevation on XP, but not form an anonymous user, making the threat much lower. Until that trsuted, unpatched 2000 machine enters the LAN.
AFAIK, W2K isn't really checked, but a "pass-all" is done.
People continue to use the word "editor". I don't think it means what they think it mean, at least not on /.
As you didn't post under any ID, I assume you don't have one. This means that there are, approximately, 900.000 perfectly valid /. virgins in line before you.
No, it allows the user to load unsigned code in kernel space! Those bastards...
On the other hand, even booting from its own media isn't any panacea if you even distrust the BIOS.
You're watching your screen with something different from your eyes? (Why do you expect that the brain and eyes wouldn't perceive motion blur in an artificial image?)
It's not exactly like putting large amounts of data in a queue is a free operation. For a Cell-based machine, of course, they will have to do it. On a dual-core x86 it's not that obvious. We also have to remember the added latency by doing things some frames ahead. That may hurt gaming experience more than the framerate itself.
Until you realize that this would require twice the workset, trash cache locality and hog the graphics bus. It will not be a free operation.
You get a whole lot closer to grep by using FINDSTR. No, the arguments are not directly compatible. Yes, you can satisfy quite a lot of your needs with it if you accept the syntax. (and have loads of ASCII-pr0n)
Well, that's not the whole point. There are controls that do nothing that a properly sandboxed Java applet couldn't do. After all, things like the Flash player and Acrobat integration (if you use it) are ActiveX based in IE and the scripting model there has been safe (with a few, glaring, exceptions).
Last I heard he was reversing the polarity of his Internet conncetion to fight these worms. They'll go back out on the net instead of invading his computer.
Maybe that's because you know more about Linux than Windows internals? A call from the wrong IRQL is quite specific. If you make a crash dump you can pinpoint the stack when it happened and a whole lot of things with free (as in beer, if you already have a Windows license...) tools, provided by MS.
And if there was less leakage it would not be such a bad thing to run your P4 at 1.7 V. Hey, we were perfectly happy with 1.7 a few years ago.
So I absolutely think they at least hoped for 90 nm to scale far beyond what Prescott actually did/has done, with quite a bit better performance per watt.
Well, I, for one, welcom
If you can't read the article properly, then don't read it at all. The 6.6 number was from a previous overclocking attempt made by another person, only mentioned here to give some background.
And if you read the article again, you'll find that the guy not booting WinXP over 6.6 GHz was a different one from the one calculating pi at 7.13 GHz (running on Windows...).
It sure helps you to notice a bit faster if it got stuck when loading a driver that blocks the interrupt handling...