The guys at Tech Report also has an article in which they dissect parts of the benchmark and provide what both FutureMark and nVidia's comments on the matter.
But what if a missile targeted, tracked and controlled using Microsoft software, aimed at a critical target, the command unit fails (BSOD, GPF, Access Violation...), and hits a hospital that not only contains civilians but possibly other troops on our side?
(BTW - I'm against this warmongering and possible war)
Two of the processors are in there are Itanium 2 processors, the other two are the first Itanium processors. Still, compare the prices of those processors to say a Sun UltraSparc III 900MHz or the newer 1+ GHz processors (from the Sun Store, a 1.05GHz USIIIcu with 8MB cache lists for just under $7000), or even Alphas, PA-RISCs and POWER4/POWER4+... it doesn't look at bad.
For those who don't need a 64-bit processor but could still use 36-bit addressing and PAE, you have the Xeon MP processors that run around $3700 for a 2GHz processor with 2MB L3 cache.
As other posts have stated, The Inquirer's report comes from the Linux Kernel mailing list. It's not an interview like the submitter stated, but rather excerpts of messages that Linus posted and responses to his posts.
There are 32-bit x86 processors that can handle more than 32-bit memory addressing, the Intel Xeon processors come to mind (which can address up to 36-bits)... the only problem is that the application and OS needs to support windowing or PAE (Physical Address Extension) to allow use of > 4GB of memory.
The only problem with windowing and use of PAE is that there is a long delay (from the processor's point of view) to shift the window compared to accessing something within that window. On the bright side, the delay isn't nowhere as bad as having to go to virtual memory and paging files.
The supposed 64-bit extensions for the Intel x86 chips that has been floating around at Intel is called Yamhill (which is named after a county, city, river and Indian tribe in Oregon).
Video cards can pump out high resolutions at fairly good refresh rates, but there are two limiting factors: the monitor and the quality of the video.
Sure, a video card can pump out 1600x1200 at 85Hz, but it doesn't help if a monitor can only do 1280x1024 at 75Hz. Also, the quality of the RAMDAC and the components between it and the monitor are also very critical. Some people have complained about the GeForce's 2D quality, some haven't...
That's why I love my Matrox G450 and dual monitors at work (1600x1200 @ 85Hz each):)
My guess is that in conjunction with the Rights Management Services (RMS) in Windows Server 2003, either the application or the server would read the metadata/DRM flags in the document before the actual operations would take place.
Let's say that a Word document with DRM flags to only allow print by specific execs, is in a share that only allows C-level execs access, and someone somehow got access into the share and attempts to print the data. The server would of course allows the user (who has access via the share) to read the file, but the DRM flags only allow certain people to print, so the user is denied print access by Word.
That's just an example that I was able to pull from this tired mind... but you probably get the gist of it.
I know your being a bit sarcastic or fecitious, but many of the free TrueType or OpenType fonts available on the Internet aren't exactly the best fonts, primarily when printing or used in any high-resolution, anti-aliased, and/or large font size scenarios. It all has to do with how the fonts are hinted, constructed, tweaked and tuned. It's a painful process, even for professionals who spend their work hours producing fonts.
I personally think it's great that they are providing high-quality fonts that can pretty much be free to distribute or hacked... mostly being a free (gratis) replacement for Verdana (and a couple of other fonts Microsoft includes in Windows and Office).
Why not use/dev/random or another pseudo-random number generator instead of/dev/zero, or at least do one round of zero's, one round of random data, and repeat say... 5-10 times?:)
Aw heck, just rip any boy-band of Spheres CD to fill up the hard drive and you've got enough random noise to fill it up. Just don't listen to it while you's ripping it:D
I guess I need to get caught up on the NetBSD side of things:) Last I heard it was being written and tested, but didn't know if it was considered "stable" or not.
I believe that zero copy networking is being developed into NetBSD and FreeBSD's stacks as well. I'm not sure if it's happening with the OpenBSD stack nor the one used in Mac OS X/Darwin.
It may not be Mac OS X or Darwin-specific, but it seems that Intel is working with Wasabi Systems/NetBSD to work on getting iSCSI to run on NetBSD. Some of this may work up to Darwin, then into Mac OS X... but it will take a while, mostly when Apple is focusing on Fibre Channel with their XServe RAID units.
I think they chose a BSD license variant (which is called the Intel Open Source License) over the LGPL because it was the best business decision for them and RedHat was okay with that. There is a reason why there are so many Open Source licenses because it provides people, developers and companies many different options to choose from. It would suck if there was only one Open Source licenses since that would limit one's choice and make others nervous about releasing their work in such a license.
If you don't like the BSD license, then don't use it as the other poster mentioned. No one is forcing you to use a license that you don't want to use. It all depends on which license is the best option for the situation. Kind of like how you could use a screwdriver to drive in a nail, but wouldn't a hammer or a mallet work better?
W3C calls their CSS and HTML specifications as "Recommendations" rather than "Requirements" or "Standard", per se. Instead, they provide specifications on how user agents (be it browsers or cell phones) are recommended to follow.
Of course, it would be lovely if all browser makers were to forced to follow the recommendations down to the nitty gritty, but even the recommendations don't always provide strict requirements on how a property or class should be rendered.
The fact that Microsoft is pushing out (delibrately) a broken style sheet is just wrong.
The reason for them using dual channel PC2100 (aka DDR266) instead of faster DDR SDRAM is to keep the CPU front side bus and the memory bus as synchronized as possible (4.2GB vs. 4.2GB). Running the memory bus out of sync from the FSB may end up causing additional latency as well as the fact that the CPUs can only saturate up to 4.2GB/s (since Xeons use a shared bus instead of the point-to-point bus used by the Athlons).
The problem with the current Athlons is the fact that both processors get the full FSB bandwidth, but are limited by a single-channel DDR memory controller. That should change with the Athlon 64/Opteron processors... but that's still a bit down the road.
Take a look at dual Xeon motherboards that have the Intel E7505 chipset, which provides support for AGP 8X along with dual-channel DDR SDRAM. The i860 chipset (which uses dual channel Rambus) will only allow up to AGP 4X. Supermicro and Tyan have motherboards available using the E7505 chipset.
Right now, you will not see much, if any, difference in performance between AGP 4X and AGP 8X...
I'm not saying that it's an excuse, just agreeing that installation of the hotfixes is no easy task. I hate it but I patch up servers, even with the most arcane install process, because it's going to be easier than cleaning up the mess that a worm or exploit could cause.
Slashdot Mirror
In AOL's case, it would probably is cheaper for them to stick with pressed CD's instead of using recordable CDs in the first place.
Thanks for the link... I'll add that to my to-read list for tonight.
The guys at Tech Report also has an article in which they dissect parts of the benchmark and provide what both FutureMark and nVidia's comments on the matter.
But what if a missile targeted, tracked and controlled using Microsoft software, aimed at a critical target, the command unit fails (BSOD, GPF, Access Violation...), and hits a hospital that not only contains civilians but possibly other troops on our side?
(BTW - I'm against this warmongering and possible war)
Road hog!
What will we see next, a Harley that can use biodiesel?
For those who don't need a 64-bit processor but could still use 36-bit addressing and PAE, you have the Xeon MP processors that run around $3700 for a 2GHz processor with 2MB L3 cache.
As other posts have stated, The Inquirer's report comes from the Linux Kernel mailing list. It's not an interview like the submitter stated, but rather excerpts of messages that Linus posted and responses to his posts.
There are 32-bit x86 processors that can handle more than 32-bit memory addressing, the Intel Xeon processors come to mind (which can address up to 36-bits)... the only problem is that the application and OS needs to support windowing or PAE (Physical Address Extension) to allow use of > 4GB of memory.
The only problem with windowing and use of PAE is that there is a long delay (from the processor's point of view) to shift the window compared to accessing something within that window. On the bright side, the delay isn't nowhere as bad as having to go to virtual memory and paging files.
- AlphaServer ES47
- AlphaServer ES80
- AlphaServer GS1280
On the workstation side, you have the AlphaStation ES47.The supposed 64-bit extensions for the Intel x86 chips that has been floating around at Intel is called Yamhill (which is named after a county, city, river and Indian tribe in Oregon).
Video cards can pump out high resolutions at fairly good refresh rates, but there are two limiting factors: the monitor and the quality of the video.
:)
Sure, a video card can pump out 1600x1200 at 85Hz, but it doesn't help if a monitor can only do 1280x1024 at 75Hz. Also, the quality of the RAMDAC and the components between it and the monitor are also very critical. Some people have complained about the GeForce's 2D quality, some haven't...
That's why I love my Matrox G450 and dual monitors at work (1600x1200 @ 85Hz each)
My guess is that in conjunction with the Rights Management Services (RMS) in Windows Server 2003, either the application or the server would read the metadata/DRM flags in the document before the actual operations would take place.
Let's say that a Word document with DRM flags to only allow print by specific execs, is in a share that only allows C-level execs access, and someone somehow got access into the share and attempts to print the data. The server would of course allows the user (who has access via the share) to read the file, but the DRM flags only allow certain people to print, so the user is denied print access by Word.
That's just an example that I was able to pull from this tired mind... but you probably get the gist of it.
http://news.com.com/2100-1001-985496.html
If some people thought that GNU/RMS is bad enough, now we have MS-RMS!
Thanks for clarifying and correcting me :)
Sorry... I guess I meant to say that they spend all of their work days tweaking and tuning fonts.
I know your being a bit sarcastic or fecitious, but many of the free TrueType or OpenType fonts available on the Internet aren't exactly the best fonts, primarily when printing or used in any high-resolution, anti-aliased, and/or large font size scenarios. It all has to do with how the fonts are hinted, constructed, tweaked and tuned. It's a painful process, even for professionals who spend their work hours producing fonts.
I personally think it's great that they are providing high-quality fonts that can pretty much be free to distribute or hacked... mostly being a free (gratis) replacement for Verdana (and a couple of other fonts Microsoft includes in Windows and Office).
Why not use /dev/random or another pseudo-random number generator instead of /dev/zero, or at least do one round of zero's, one round of random data, and repeat say... 5-10 times? :)
:D
Aw heck, just rip any boy-band of Spheres CD to fill up the hard drive and you've got enough random noise to fill it up. Just don't listen to it while you's ripping it
I guess I need to get caught up on the NetBSD side of things :) Last I heard it was being written and tested, but didn't know if it was considered "stable" or not.
I believe that zero copy networking is being developed into NetBSD and FreeBSD's stacks as well. I'm not sure if it's happening with the OpenBSD stack nor the one used in Mac OS X/Darwin.
It may not be Mac OS X or Darwin-specific, but it seems that Intel is working with Wasabi Systems/NetBSD to work on getting iSCSI to run on NetBSD. Some of this may work up to Darwin, then into Mac OS X... but it will take a while, mostly when Apple is focusing on Fibre Channel with their XServe RAID units.
If you don't like the BSD license, then don't use it as the other poster mentioned. No one is forcing you to use a license that you don't want to use. It all depends on which license is the best option for the situation. Kind of like how you could use a screwdriver to drive in a nail, but wouldn't a hammer or a mallet work better?
W3C calls their CSS and HTML specifications as "Recommendations" rather than "Requirements" or "Standard", per se. Instead, they provide specifications on how user agents (be it browsers or cell phones) are recommended to follow.
Of course, it would be lovely if all browser makers were to forced to follow the recommendations down to the nitty gritty, but even the recommendations don't always provide strict requirements on how a property or class should be rendered.
The fact that Microsoft is pushing out (delibrately) a broken style sheet is just wrong.
The reason for them using dual channel PC2100 (aka DDR266) instead of faster DDR SDRAM is to keep the CPU front side bus and the memory bus as synchronized as possible (4.2GB vs. 4.2GB). Running the memory bus out of sync from the FSB may end up causing additional latency as well as the fact that the CPUs can only saturate up to 4.2GB/s (since Xeons use a shared bus instead of the point-to-point bus used by the Athlons).
The problem with the current Athlons is the fact that both processors get the full FSB bandwidth, but are limited by a single-channel DDR memory controller. That should change with the Athlon 64/Opteron processors... but that's still a bit down the road.
Right now, you will not see much, if any, difference in performance between AGP 4X and AGP 8X...
I'm not saying that it's an excuse, just agreeing that installation of the hotfixes is no easy task. I hate it but I patch up servers, even with the most arcane install process, because it's going to be easier than cleaning up the mess that a worm or exploit could cause.