You may just have too much RAM already used by the time you start Eclipse. I generally run Eclipse J2EE + more plugins in about 128-256MB heap space, and as long as that heap space is available on *physical* memory, it's silky smooth even on a 1.2Ghz Duo. You are doing something wrong and it's not Eclipse' fault.
Just play Xenogears end to end and you'll have your fill of mature gaming. Lots of family-friendly concepts like genocide, cannibalism, reincarnation, genome experiments, torture, Freudian psychology, and many more, all pave the way on your noble quest to kill God. It's like every heavy metal album combined. Highly recommended.
"For example, it is known that some vendors optimize their FTL devices for FAT..."
Basically we won't really have full use of our hardware until Windows is forced to evolve.
The same problem led to the creation of WEP. Just because Windows' in-kernel encryption is as useful as piglatin, hardware and drivers have to pick up the slack. Don't say it's because of embedded devices - they've been powerful enough to run Blowfish since before Blowfish itself was specified.
All we need for the flash solution is a revised HID standard that does expose raw blocks. It can still be standard and uniform, just lower level.
Put it like this. If mass storage did not have the HID abstraction and wear levelling circuitry (primitive though it may be), Windows would have absolutely soiled every flash device out there with its uniquely bad IO layer. At least the raw device is slightly protected from Windows by the standard.
That's fair to say, but you can be pretty confident Hibernate is a solid product. Same for Spring. They're many years old, with widespread use in open source and commercial projects. And very encouraging is how they both use unit testing from the ground up, so you can be very confident any given release is exceptionally robust.
I wish I could say the same for the JVM they run on. There were a number of bugs in several official, supported Sun JVM releases which mis-optimised code on amd64. They seem to be fixed, but you'd think they'd test for stuff like that, since even popular projects like Eclipse were crippled by the bugs.
What part of "I don't necessarily agree" don't you understand? I only inferred and rephrased the original poster's suggestion, I didn't lend it any support.
The 99% 1/100 figures don't apply to an X server by any stretch of the imagination. They apply to, say, a video streaming service where the web logic is of minimal computational expense compared to the sheer IO load. I hear it worked for YouTube.
I think he means the administrative overhead of X11 protocol management is not the bottleneck, so it may as well be written in Python. The mathematically intensive and low-level parts, which are only part of the overall code base, can remain C. I don't necessarily agree, as this would increase latency for protocol handling, which really adds up.
This is a common pattern in modern software development - very simple, mechanical C code wrapped by high level, elegant Python (or your scripting language of choice). You get 99% of the performance of pure C with 1/100th the development time. I've done the same in high-performance scientific computing projects with great success.
It's rarely discussed because it's extremely slow. Even on low resolutions it takes an absurd amount of CPU power and latency. On high resolutions it's like a slide show with an awkward guest speaker. There's a reason we have hardware acceleration even for 2D.
Right, because the same kernel binary can boot on ARM mobile phones and SPARC supercomputers. You might just be an idiot or a troll. Of course different machine architectures require different compiled binaries.
On the same architecture, yes, you can boot the same kernel for 4096 CPUs and 1 CPU. That is handled at boot time, not at compile time, but you can optimise further in either direction with some options. Linux seems to do all of this about as well as theoretically possible, including actually patching itself at runtime to optimise for single-processor systems, even if compiled for multi-processor ones.
And why shouldn't it? The mainline Linux kernel runs on realtime embedded devices and 4096-CPU supercomputers, depending on the options used at compile-time. Linux proved it can be done, why can't Microsoft do the same? Do they just not have the engineering skills?
Linux supports NUMA which largely solves that problem, and ccNUMA which solves it even better. It's all about locality once again. Linux has been running on multi-thousand CPU machines for years, and has been optimised and refined by the stakeholders of those projects, so it's not a toy project to show off.
The most recent mainline Linux release has integrated mature patches for 4096 core scalability, that have been developed by high performance computing corporations and tested in the field for years. Previous versions were rated for "only" 1024 cores. That still makes 256 look like a Gameboy.
It must be really hard for Microsoft to compete in the HPC space. I almost feel bad for them. Almost.
You can just boot it off an external USB drive and update it as you would a normal install. But if you're going to do that routinely, you may as well just dual boot.
I don't think it's stupid to allow SSH. They never claimed they were trying to block all politically sensitive content for all people. They only implemented primitive measures to hide sensitive content from the general population, who, just like any general population, don't even know what SSH stands for.
The technically capable population will work out a way around virtually anything. Knowing how to use SSH implies knowing how to change port numbers. There's no point blocking the default SSH port if you're leaving over 65,000 others open.
Slashdot Mirror
You may just have too much RAM already used by the time you start Eclipse. I generally run Eclipse J2EE + more plugins in about 128-256MB heap space, and as long as that heap space is available on *physical* memory, it's silky smooth even on a 1.2Ghz Duo. You are doing something wrong and it's not Eclipse' fault.
Then imagine how good $200,000 headphones would be. They'd include an extra large driver to place on your torso for deep bass.
Just play Xenogears end to end and you'll have your fill of mature gaming. Lots of family-friendly concepts like genocide, cannibalism, reincarnation, genome experiments, torture, Freudian psychology, and many more, all pave the way on your noble quest to kill God. It's like every heavy metal album combined. Highly recommended.
It's even worse than just that.
http://www.linux-mtd.infradead.org/doc/ubifs.html#L_raw_vs_ftl
"For example, it is known that some vendors optimize their FTL devices for FAT..."
Basically we won't really have full use of our hardware until Windows is forced to evolve.
The same problem led to the creation of WEP. Just because Windows' in-kernel encryption is as useful as piglatin, hardware and drivers have to pick up the slack. Don't say it's because of embedded devices - they've been powerful enough to run Blowfish since before Blowfish itself was specified.
All we need for the flash solution is a revised HID standard that does expose raw blocks. It can still be standard and uniform, just lower level.
Put it like this. If mass storage did not have the HID abstraction and wear levelling circuitry (primitive though it may be), Windows would have absolutely soiled every flash device out there with its uniquely bad IO layer. At least the raw device is slightly protected from Windows by the standard.
Except that you're not checking all those passwords at once, like you would with an obviously red paper among white ones.
A lot of the functionality was merged in anyway.
http://www.catb.org/jargon/html/C/cow-orker.html
Actually -Cx gives you context. -C1 gives you one line above and below the matching line, -C2 gives two, etc. -H only gives you the filename.
Firefox bookmarks are backed up automatically, so at least you didn't lose anything.
By the way, run a stable distro with open source drivers. I can't remember the last time anything crashed, let alone something critical like X.Org.
That's fair to say, but you can be pretty confident Hibernate is a solid product. Same for Spring. They're many years old, with widespread use in open source and commercial projects. And very encouraging is how they both use unit testing from the ground up, so you can be very confident any given release is exceptionally robust.
I wish I could say the same for the JVM they run on. There were a number of bugs in several official, supported Sun JVM releases which mis-optimised code on amd64. They seem to be fixed, but you'd think they'd test for stuff like that, since even popular projects like Eclipse were crippled by the bugs.
What part of "I don't necessarily agree" don't you understand? I only inferred and rephrased the original poster's suggestion, I didn't lend it any support.
The 99% 1/100 figures don't apply to an X server by any stretch of the imagination. They apply to, say, a video streaming service where the web logic is of minimal computational expense compared to the sheer IO load. I hear it worked for YouTube.
I think he means the administrative overhead of X11 protocol management is not the bottleneck, so it may as well be written in Python. The mathematically intensive and low-level parts, which are only part of the overall code base, can remain C. I don't necessarily agree, as this would increase latency for protocol handling, which really adds up.
This is a common pattern in modern software development - very simple, mechanical C code wrapped by high level, elegant Python (or your scripting language of choice). You get 99% of the performance of pure C with 1/100th the development time. I've done the same in high-performance scientific computing projects with great success.
It's rarely discussed because it's extremely slow. Even on low resolutions it takes an absurd amount of CPU power and latency. On high resolutions it's like a slide show with an awkward guest speaker. There's a reason we have hardware acceleration even for 2D.
Right, because the same kernel binary can boot on ARM mobile phones and SPARC supercomputers. You might just be an idiot or a troll. Of course different machine architectures require different compiled binaries.
On the same architecture, yes, you can boot the same kernel for 4096 CPUs and 1 CPU. That is handled at boot time, not at compile time, but you can optimise further in either direction with some options. Linux seems to do all of this about as well as theoretically possible, including actually patching itself at runtime to optimise for single-processor systems, even if compiled for multi-processor ones.
There are other architectures besides Intel's. News at 11.
And why shouldn't it? The mainline Linux kernel runs on realtime embedded devices and 4096-CPU supercomputers, depending on the options used at compile-time. Linux proved it can be done, why can't Microsoft do the same? Do they just not have the engineering skills?
Linux supports NUMA which largely solves that problem, and ccNUMA which solves it even better. It's all about locality once again. Linux has been running on multi-thousand CPU machines for years, and has been optimised and refined by the stakeholders of those projects, so it's not a toy project to show off.
The most recent mainline Linux release has integrated mature patches for 4096 core scalability, that have been developed by high performance computing corporations and tested in the field for years. Previous versions were rated for "only" 1024 cores. That still makes 256 look like a Gameboy.
It must be really hard for Microsoft to compete in the HPC space. I almost feel bad for them. Almost.
You can just boot it off an external USB drive and update it as you would a normal install. But if you're going to do that routinely, you may as well just dual boot.
I don't think it's stupid to allow SSH. They never claimed they were trying to block all politically sensitive content for all people. They only implemented primitive measures to hide sensitive content from the general population, who, just like any general population, don't even know what SSH stands for.
The technically capable population will work out a way around virtually anything. Knowing how to use SSH implies knowing how to change port numbers. There's no point blocking the default SSH port if you're leaving over 65,000 others open.
Just copy them into your source tree. BSD license lets you do that, unlike most of the glibc (LGPL) extensions.
Greetings from orbital station.
I can't even tell if you're joking. I can't even tell if that's ironic or just depressing.
I highly doubt the browser uses the streaming units on the Cell, so it's basically just PPC that was available for many years before the Cell.