Their database server is crashing _too_ fast and _too_ much. The memory appears to be the culprit. AFAIK it's under a support contract from Penguin Computing.
But: Will someone be definitely there to maintain the damn things? Thats where some of the money to buy PC's, servers, *cough*windows*cough* comes from. It's called a Support contract. If this thing blows in 2 years, will I get a replacement under warranty?
Wikipedia is one of the best resources out there. I did a school project on Stars and I found that Wikipedia simply blows other resources away. (I've never seen "Oh Boy, an F grade kills me" in any other encyclopedia).
Being a) a minor b) in australia c) without credit card unfortunately makes it hard for me to donate to them at the moment. I'll have to see what I can do (any Australians here willing to forward donations?)
Perhaps some dummy RC5-72 blocks generated by distributed.net's public sourcecode version is a good benchmark (distributed.net's public src release IS NOT compatible with it's closed source buffers)
Not true at all. The closed portion of the nVidia drivers is only the X driver. The kernel module is and has always been open source, and you can compile it for whatever version of the kernel you wish (assuming the kernel module interface hasn't changed drastically, of course).
AFAIK According to the README file in the driver source distro, The OpenGL implementation is (C) SGI.
AFAIK Utah GLX is working on an Riva 128/TNT driver. Good luck to them. I was using my TNT with the NV binary drivers until the start of this year. Still put out playable fps at 1280x1024 in Quake 3(Linux)
If this serves right, then when you power your PC up after a major refit, your graphics card will split in half while trying to play a game for the 1st time on your refitted rig:(
AFAIK Utah GLX is working on an Open Source driver for RIVA TNT's.
Last time I heard they got quake running on it:)
nVIDIA would prefer not to release any details of there drivers under NDA. But why waste your time trying to grab info about a card, when you can reverse engineer the driver?
RivaTuner, a overclocking application includes several user contributed driver patches for both NVIDIA and ATI.
The most notable of these is SoftQuadro and NVStrap. People found a way to make a plain old GeForce a Quadro by simply changing the PCI device ID - until NVIDIA stopped them by blocking this in the driver. So people took out their disassemblers and created patches against the driver.
So, 1 reason NOT to release source code is so that people can't softmod their GeForce cards into Quadro ones.
The next reason is all the proprietary NV_ OpenGL extensions which allow you to do a whole bunch of stuff in hardware. While some of these have made their way into the OpenGL Standard specification, NVIDIA doesn't want either ATI, XGI or S3 to get hold of sourcecode to these extensions.
Also, remember that SGI owns the OpenGL implementation in the NVIDIA driver - not NVIDIA.
One of the best things about NVIDIA cards is that it's guaranteed to run anything. NVIDIA doesn't want people downloading custom drivers and complaining to them - infact they have already screamed down the throat of OmegaDrive because people downloaded his driver and submitted complaints to NVIDIA - not OmegaDrive.
This is also what the 'The Way That It's Meant to be Played' program is for. 'TWTIMTBP' (abbreviation, dumbass) is not a performance enhacing program. It is to cut down the number of bugs that developers have in their code with NVIDIA GPU's. Electronic Arts is already happy with it - they released a game (don't remember the name) and out of the ~175 bug reports, only ~6 were to do with NVIDIA GPU's. Thats impressive. If only Mozilla's Envaglism program had the same sort of success:(
I completely disagree with you. No offence, but you appear to be exibiting 'The Microsoft Effect', where customers waited to upgrade to Windows X Service Pack X before upgrading to X Windows version. Microsoft itself tried to stem this behaviour by declaring Windows Server 2003 Beta 2 == RTM and Windows Server 2003 Final Release == Service Pack 1.
I've been using 2.6.0-test* kernels on my boxes/boxen for some time now, and the performance improvement is great. If your a Linux gamer, upgrade now. I gained 20fps in tuxracer alone by upgrading:).
There are other changes. Kernel Modularisation has changed a lot, and you probably won't find yourself modprob'ing away everything like you did before.
A word of warning: Sort out your Processor cooling arrangements before upgrading. 2.6 kernels don't like inadequately cooled processors. If your box hasn't had any sort of cooling upgrade in ~1-2 years, take off the cover right now (if warranty permits) and clean out the heatsink/dried thermal grease/frigged thermal tape.
Yeah, but how many hits do you still get from "Mozilla/4.0 {MSIE 5.5; Windows 98}" on a non-public server?
I get a lot on my box, and my box doesn't even serve a production website.
Don't always blame M$. Blame ignorant admins/users. At my school, about everyone, with the exception of/.'ers like me was around 25+ unpatched Windows vunrebilities. We use MS SQL for one of our apps, and they didn't patch the Slammer worm until it started appearing on the network. ditto for Blaster.
For me, ext3 doesn't offer much data redundancy over plain old ext2. It still suffers from ext2's dataloss problems. Infact, with ext3, I had the horror of i/o errors. Once I had a bad powerdown, and I came close to reformatting just because it wouldn't let me into php.conf.
Personally, anyone looking for a data-redundancy fix should use ReiserFS (which we have had for a looong time), JFS or XFS.
Slashdot Mirror
read the post by Brion above.
Their database server is crashing _too_ fast and _too_ much. The memory appears to be the culprit. AFAIK it's under a support contract from Penguin Computing.
It seems like a good idea.
But:
Will someone be definitely there to maintain the damn things? Thats where some of the money to buy PC's, servers, *cough*windows*cough* comes from. It's called a Support contract. If this thing blows in 2 years, will I get a replacement under warranty?
Yeah I know that :) The only thing that looks good in the photo is the Qube.
Keep in mind that this ad was from PC Magazine Australia in 1999.
Guys, they deserve it.
Wikipedia is one of the best resources out there. I did a school project on Stars and I found that Wikipedia simply blows other resources away. (I've never seen "Oh Boy, an F grade kills me" in any other encyclopedia).
Being a) a minor b) in australia c) without credit card unfortunately makes it hard for me to donate to them at the moment. I'll have to see what I can do (any Australians here willing to forward donations?)
Come on.. even a single buck can help anybody.
The Qube 3 and RaQ 550 Source code was released to the Cobalt Users Group of Japan under a BSD-link license.
Since their server is down, this is the google cache
Did you know that Cobalt has the biggest market share of on-line Linux servers after Redhat?
The Qube 3 sourcecode was released to the Cobalt Users Group of Japan at open.cobaltqube.org (down at the moment) :(
What a sad ending. I am still drooling over this sexy Cobalt Qube 2 advertisment
Perhaps some dummy RC5-72 blocks generated by distributed.net's public sourcecode version is a good benchmark (distributed.net's public src release IS NOT compatible with it's closed source buffers)
Thank god for gnupg/pgp.
"But someone else has the password to this file, honest!"
"Oops! I just pressed secure wipe!"
AFAIK along with the hl2 leak, I thought that the Half Life src was released for modders?
Slashdot needs a new mod point:
"+1, Unfortunate" and "+1, Segfualted"
Thank god their ipv6 implementation is not so crap.
But their IPP implementation is an example of M$'es non-respect for standards.
Not true at all. The closed portion of the nVidia drivers is only the X driver. The kernel module is and has always been open source, and you can compile it for whatever version of the kernel you wish (assuming the kernel module interface hasn't changed drastically, of course).
AFAIK According to the README file in the driver source distro, The OpenGL implementation is (C) SGI.
AFAIK Utah GLX is working on an Riva 128/TNT driver. Good luck to them. I was using my TNT with the NV binary drivers until the start of this year. Still put out playable fps at 1280x1024 in Quake 3(Linux)
Wasn't it's codename 'Columbia'
:(
If this serves right, then when you power your PC up after a major refit, your graphics card will split in half while trying to play a game for the 1st time on your refitted rig
AFAIK Utah GLX is working on an Open Source driver for RIVA TNT's.
:)
:(
Last time I heard they got quake running on it
nVIDIA would prefer not to release any details of there drivers under NDA. But why waste your time trying to grab info about a card, when you can reverse engineer the driver?
RivaTuner, a overclocking application includes several user contributed driver patches for both NVIDIA and ATI.
The most notable of these is SoftQuadro and NVStrap. People found a way to make a plain old GeForce a Quadro by simply changing the PCI device ID - until NVIDIA stopped them by blocking this in the driver. So people took out their disassemblers and created patches against the driver.
So, 1 reason NOT to release source code is so that people can't softmod their GeForce cards into Quadro ones.
The next reason is all the proprietary NV_ OpenGL extensions which allow you to do a whole bunch of stuff in hardware. While some of these have made their way into the OpenGL Standard specification, NVIDIA doesn't want either ATI, XGI or S3 to get hold of sourcecode to these extensions.
Also, remember that SGI owns the OpenGL implementation in the NVIDIA driver - not NVIDIA.
One of the best things about NVIDIA cards is that it's guaranteed to run anything.
NVIDIA doesn't want people downloading custom drivers and complaining to them - infact they have already screamed down the throat of OmegaDrive because people downloaded his driver and submitted complaints to NVIDIA - not OmegaDrive.
This is also what the 'The Way That It's Meant to be Played' program is for. 'TWTIMTBP' (abbreviation, dumbass) is not a performance enhacing program. It is to cut down the number of bugs that developers have in their code with NVIDIA GPU's. Electronic Arts is already happy with it - they released a game (don't remember the name) and out of the ~175 bug reports, only ~6 were to do with NVIDIA GPU's. Thats impressive. If only Mozilla's Envaglism program had the same sort of success
Care to give me your IP. I'll have fun playing with your udp port 1434 :) Assuming that you don't have the Slammer patches installed :(
I completely disagree with you. No offence, but you appear to be exibiting 'The Microsoft Effect', where customers waited to upgrade to Windows X Service Pack X before upgrading to X Windows version. Microsoft itself tried to stem this behaviour by declaring Windows Server 2003 Beta 2 == RTM and Windows Server 2003 Final Release == Service Pack 1.
:) .
I've been using 2.6.0-test* kernels on my boxes/boxen for some time now, and the performance improvement is great. If your a Linux gamer, upgrade now. I gained 20fps in tuxracer alone by upgrading
There are other changes. Kernel Modularisation has changed a lot, and you probably won't find yourself modprob'ing away everything like you did before.
A word of warning: Sort out your Processor cooling arrangements before upgrading. 2.6 kernels don't like inadequately cooled processors. If your box hasn't had any sort of cooling upgrade in ~1-2 years, take off the cover right now (if warranty permits) and clean out the heatsink/dried thermal grease/frigged thermal tape.
yes, I got that wrong for some reason, but it suprises me how mnay people can't learn to 'patch' a link :(
Akamai NOC tour
Wired article about Akamai's 'gods-eye' view of the Slammer virus
You might want to have a look at Akamai's NOC at http://www.akamai.com//en/html/about/nocc_tour.htm l
Pictures of Akamai's NOC also were in the Wired article about the Slammer Virus a few months ago.
Yeah, but how many hits do you still get from "Mozilla/4.0 {MSIE 5.5; Windows 98}" on a non-public server?
/.'ers like me was around 25+ unpatched Windows vunrebilities. We use MS SQL for one of our apps, and they didn't patch the Slammer worm until it started appearing on the network. ditto for Blaster.
I get a lot on my box, and my box doesn't even serve a production website.
Don't always blame M$. Blame ignorant admins/users. At my school, about everyone, with the exception of
I've had no real problems with it myself.
Well, I'm going put my own server up soon, I've had enough of static html, and my servlet skills are getting better.
well, ext3 is built over ext2, and still suffers ext2's dataloss problems.
I wouldn't even dare put ext3 on a production system. It isn't worth the advantage over ext2 (counting my own experiances)
yeah, just hold the poweroff button :)
For me, ext3 doesn't offer much data redundancy over plain old ext2. It still suffers from ext2's dataloss problems. Infact, with ext3, I had the horror of i/o errors. Once I had a bad powerdown, and I came close to reformatting just because it wouldn't let me into php.conf.
Personally, anyone looking for a data-redundancy fix should use ReiserFS (which we have had for a looong time), JFS or XFS.
I can confirm that no TV channel uses those allocations.