Sure I use gcc & ecgs because they're free, but they're hardly start-of-the-art as far as optimizing goes. If borland released a cheap optimizing C++ compiler for Linux, I'd sure be interested.
NT beat Linux (on these benchmarks, at least), even when Linux was optimally configured. We whined for a retest with an optimally configured Linux, and got it - hey, we even got a lot of changes in the test setup - UP as well as SMP, different RAID controller, NT as well as Win 95 clients... BUT, we STILL lost.
Rather than whining, I'd say there's a lot to be grateful for - mainly the fact that the testing has shown some prior bottleneck assumptions to be wrong, and has exposed the real problems which can now be addressed. Given the results, the press coverage has also been rather kind to Linux.
For software developers, this is still the best way to get a job. Companies use recruiters to avoid having to screen applicants themselves, and typically pay 15-20% of the positions salary to the recruiter for there services. If the recruiter thinks your suited for a position, then you're very likely to get an interview.
From my own experience and circle of friends, job search sites do not seem to be a main-stream or common way to find a job - in fact I don't know *anyone* who has got a job this way. Most software jobs in my geographic area (NY/CT) are never advertized in the paper or internet, and are *only* available thru headhunters.
Science is about explaining facts, not denying them.
If a mug suddenly floated off your desk in apparent contradiction to the law of gravity, the correct scientific response would not be to deny it because it is impossible, but rather to explain it. Maybe the guy in the room below you is conducting a gravity blocking experimemnt...
If "cold fusion" research is giving rise to "unaccounted heat", then that should be explained. Maybe fusion is not the explanation, but maybe it is. It would be extremely foolhardly to claim that current scientific knowledge is complete and entirely correct.
AOL have a deal with Bell Atlantic to use their ADSL service, and the there's a bill being pushed by a Virginia (AOL home) congressman to force cable companies to sell service to allow other companies to act as cable ISPs. Look for much more broadband competition in the the Bell Atlantic region soon as they roll out their ADSL service, and the cable companies compete. I work for Bell Atlantic, and have been told this is seen as a critical first-to-market race.
Trick people into thinking they're buying a dog, then secretly load it up with subversive Sony technology! Maybe it's got a Sony "nightshot" vision system, and is beaming images of naked American chicks back to Japan!
Seems like every time someone builds a robot with video cameras that look like eyes, that people suddenly give it more credence. Brooks's Cog seems like a made-for-TV project. It's still bug-level technology in a human form factor. If you want high-level emergent behaviour, you need an architecture that supports it. If you simply put together a bunch of low level behaviours you're not going to suddenly get higher level cognition... you'll just get a cool made-for-TV reactive robot.
ARM originally stood for Acorn Risc Machine - it was designed by Acorn. The acronym was changed to mean Advanced Risc Machines when it was spun off as a separate company with Apple and VLSI technology.
Not only did you have to hold the RAM packs on with rubber bands, but the common way to keep the thing from overheating was a carton old cold milk stood on top of it. Ah good old uncle Clive! I used to work at Acorn, and remember Clive coming to one of our Christamas partys and tryign to hit on our (extremely hot) personel girl! Then there's Chris Curry, one of Acorn's founders, who was found shagging some bimbo on his desk after another party!:)
The ARM is one of the few true RISC designs around, with all 32 bit instructions, and all instructions conditional. This is the reason for the tiny die size and low poer requirements. I used to work for Acorn Computers (the "A" in ARM) in the early 80's, and knew the main designer of the ARM - Steve Furber. As of a couple of years ago he was working at the University of Manchester (UK), where he had fabricated a fully asynchronous ARM CPU that he had been working on for years. This is basically a dataflow design, and (potentially) saves huge amounts of power by removing the need to clock the whole chip. The ARM was partly inspired by the 6502 which was what the early Acorn computers were based on, and was minimal enough to practically be a RISC design itself. Ah, the good old days...
How many net servers in the real world run off SMP boxes? Most ISPs use server farms of uniprocressor machines - much better bang for the buck. No-one's denying that Linux's SMP performance could be improved, but exactly how it compares to NT (which has it's own set of problems) is really unknown to this stage due to lack of fair testing.
The Oracle test I mentioned took one approach to fairness in testing both NT and Linux out of the box with no tuning on either side.
Given how artificial benchmarks are, the real world observations of NT vs Linux performance should probably be given more weight anyway. A quad zeon box is hardly what people are running Linux servers on - many are running on 486's! Try that with NT...
Well, Linux currently has 17% of the server market, and is estimated to be growing at 25% a year for the next few years...
If you check http://www.netcraft.net/survey/ you'll see that Apache massively dominates the web server market with around a 60% share. Being open source rather than commercially backed obviously hasn't stopped it from putting a huge dent in Microsoft's sales.
I'm sure Microsoft wishes that Linux _was_ a traditional single company commercial vendor, since that would give them a target to shoot at.
My point was that it's a uniprocessor machine, and that benchmarking (and real-world deployment) has shown Linux to pretty conclusively outperform NT on uniprocessor machines. Quite likely FreeBSD would outperform Linux in some server benchmarks, but that's beside the point.
As Mindcraft's web site says (paraphrasing) "you identify your goals, we do the testing to satisfy them". Given that the paying customer was identified as Microsoft, it should come as no surprise that the goal was to show NT being faster then Linux. Bear in mind though, that all _independent_ testing has shown exactly the opposite to be true, certainly for uniprocessor machines such as the ftp.cdrom.com server.
There have yet to be any standard SMP benchmarks (TPC-D, SPECWeb96 etc) published, although an unofficial Oracle benchmark indicated Linux to beat NT there also.
Also bear in mind that the "Mindcraft" testing has since been shown to have been performed in a Microsoft lab (the "Mindcraft" e-mails originated from a Microsoft domain)...
Ultimately, all the "Mindcraft" tests really proved is that Microsoft is starting to take Linux as a _very_ serious threat to NT - not surprising given the Linux server marketshare and growth numbers.
Microsoft is attempting to recover from the PR nightmare resulting from this testing by redoing the tests with "unimpeachable" Linux configuration expertise supplied by Linus and Alan Cox... but as those two have indicated, this is a complete farce, and you can expect the "retest" results to be as information free as the first ones.
Everyone's actions are driven by satisfaction - just that different people enjoy different things. You may enjoy getting the guts of the worlds 513th CD player to work, someone else may enjoy addign the chrome, and someone else writing the documentation.
Note also that pleasure is a complicated many faceted thing... Helping others may appear to be altruistic, but people only do it because it makes them feel good. Finishing the last 10% if a project may be a grind, but many people do it because of the satisfaction of completion/whatever - maybe the source of pleasure is different from then first 90%, but it's still pleasure driven.
Certainly open source projects may seem more haphazard than things we are forced to do professionally, but the level of effort/completion put into them is really globally optimized!... If the 70% complete status of a project sufficiently annoys someone, then it will rise to the level where completing it is becomes the most satisfying thing they cad do - and they will do it.
BTW, do you _really_ think a robot could replace you in having sex with your wife? Sounds like you need to use a little imagination or whipped cream!;-)
RedHat is working with Precision Insight to add OpenGL direct rendering to XFree86. This will allow clients that are on the same machine as the server to render directly into the framebuffer. SGI already has this.
XFree also supports the DGA and MIT_SHM X extensions than can be used to speed up local clients.
Remember that X is just the low level graphics library - it's what's on top of it (the low level and desktop environment widgets and dialogs) that contribute to an app's look and feel. The currently popular widget sets and environments - qt/KDE and gtk/GNOME are still very young, but improving very fast. Remember that Windows itself was essentially useless until 3.0, and the UI was unusable until Windows '95...
Remember that the current Linux user base is only in the 12-15 million range, but growing _very_ fast. Mexican schools are adopting Linux/GNOME, and China is likely to be predominantly Linux too for cost reasons (it already has a pretty large Linux user base, and there's only just been released a Chinese version of Linux!)... as the desktop user base grows, there's going to be some very rapid advances...
.. a single zeon makes a kick-ass server, but in the mindshaft testing lab you need a quad zeon! Hmmm.. of course it _was_ optimized for NT - probably the logic was that with 4 cpus running NT, then at any time ONE of them might actually be working!
Slashdot Mirror
The flight recorder was retrieved and had recordings of them screaming after the explosion - pretty conclusive evidence they were alive I'd say.
Sure I use gcc & ecgs because they're free, but they're hardly start-of-the-art as far as optimizing goes. If borland released a cheap optimizing C++ compiler for Linux, I'd sure be interested.
Forget the applications they are currently anticipating - the fundamental physics here is the exciting thing - very wierd!
.cx domain ?
BTW, where/what's the
Hmmm.. maybe that's the whole point of this - just to get some free/cheap publicity for their product? Maybe I'm just too cynical?
I tend to agree. The reason I come here is:
40% Interesting news stories gathered in one place
60% Informed/popular opinions - keeping track of what's going on...
Slashdot is really just as much, if not more, a specialized bulletin board as a news site.
NT beat Linux (on these benchmarks, at least), even when Linux was optimally configured. We whined for a retest with an optimally configured Linux, and got it - hey, we even got a lot of changes in the test setup - UP as well as SMP, different RAID controller, NT as well as Win 95 clients... BUT, we STILL lost.
Rather than whining, I'd say there's a lot to be grateful for - mainly the fact that the testing has shown some prior bottleneck assumptions to be wrong, and has exposed the real problems which can now be addressed. Given the results, the press coverage has also been rather kind to Linux.
For software developers, this is still the best way to get a job. Companies use recruiters to avoid having to screen applicants themselves, and typically pay 15-20% of the positions salary to the recruiter for there services. If the recruiter thinks your suited for a position, then you're very likely to get an interview.
From my own experience and circle of friends, job search sites do not seem to be a main-stream or common way to find a job - in fact I don't know *anyone* who has got a job this way. Most software jobs in my geographic area (NY/CT) are never advertized in the paper or internet, and are *only* available thru headhunters.
Science is about explaining facts, not denying them.
If a mug suddenly floated off your desk in apparent contradiction to the law of gravity, the correct scientific response would not be to deny it because it is impossible, but rather to explain it. Maybe the guy in the room below you is conducting a gravity blocking experimemnt...
If "cold fusion" research is giving rise to "unaccounted heat", then that should be explained. Maybe fusion is not the explanation, but maybe it is. It would be extremely foolhardly to claim that current scientific knowledge is complete and entirely correct.
Ben
AOL have a deal with Bell Atlantic to use their ADSL service, and the there's a bill being pushed by a Virginia (AOL home) congressman to force cable companies to sell service to allow other companies to act as cable ISPs. Look for much more broadband competition in the the Bell Atlantic region soon as they roll out their ADSL service, and the cable companies compete. I work for Bell Atlantic, and have been told this is seen as a critical first-to-market race.
Trick people into thinking they're buying a dog, then secretly load it up with subversive Sony technology! Maybe it's got a Sony "nightshot" vision system, and is beaming images of naked American chicks back to Japan!
Seems like every time someone builds a robot with video cameras that look like eyes, that people suddenly give it more credence. Brooks's Cog seems like a made-for-TV project. It's still bug-level technology in a human form factor. If you want high-level emergent behaviour, you need an architecture that supports it. If you simply put together a bunch of low level behaviours you're not going to suddenly get higher level cognition... you'll just get a cool made-for-TV reactive robot.
ARM originally stood for Acorn Risc Machine - it was designed by Acorn. The acronym was changed to mean Advanced Risc Machines when it was spun off as a separate company with Apple and VLSI technology.
Not only did you have to hold the RAM packs on with rubber bands, but the common way to keep the thing from overheating was a carton old cold milk stood on top of it. Ah good old uncle Clive! I used to work at Acorn, and remember Clive coming to one of our Christamas partys and tryign to hit on our (extremely hot) personel girl! Then there's Chris Curry, one of Acorn's founders, who was found shagging some bimbo on his desk after another party! :)
The ARM is one of the few true RISC designs around, with all 32 bit instructions, and all instructions conditional. This is the reason for the tiny die size and low poer requirements. I used to work for Acorn Computers (the "A" in ARM) in the early 80's, and knew the main designer of the ARM - Steve Furber. As of a couple of years ago he was working at the University of Manchester (UK), where he had fabricated a fully asynchronous ARM CPU that he had been working on for years. This is basically a dataflow design, and (potentially) saves huge amounts of power by removing the need to clock the whole chip. The ARM was partly inspired by the 6502 which was what the early Acorn computers were based on, and was minimal enough to practically be a RISC design itself. Ah, the good old days...
You can use this site:
http://www.netcraft.com/cgi-bin/Survey/whats
to find out what server and OS are being used by a given domain name. Try egg.microsoft.com !
This works by recognising the characteristic signatures of the different OS's TCP/IP stacks as they respond to a bunch of wierd packets.
How many net servers in the real world run off SMP boxes? Most ISPs use server farms of uniprocressor machines - much better bang for the buck. No-one's denying that Linux's SMP performance could be improved, but exactly how it compares to NT (which has it's own set of problems) is really unknown to this stage due to lack of fair testing.
The Oracle test I mentioned took one approach to fairness in testing both NT and Linux out of the box with no tuning on either side.
Given how artificial benchmarks are, the real world observations of NT vs Linux performance should probably be given more weight anyway. A quad zeon box is hardly what people are running Linux servers on - many are running on 486's! Try that with NT...
Well, Linux currently has 17% of the server market, and is estimated to be growing at 25% a year for the next few years...
If you check http://www.netcraft.net/survey/ you'll see that Apache massively dominates the web server market with around a 60% share. Being open source rather than commercially backed obviously hasn't stopped it from putting a huge dent in Microsoft's sales.
I'm sure Microsoft wishes that Linux _was_ a traditional single company commercial vendor, since that would give them a target to shoot at.
My point was that it's a uniprocessor machine, and that benchmarking (and real-world deployment) has shown Linux to pretty conclusively outperform NT on uniprocessor machines. Quite likely FreeBSD would outperform Linux in some server benchmarks, but that's beside the point.
As Mindcraft's web site says (paraphrasing) "you identify your goals, we do the testing to satisfy them". Given that the paying customer was identified as Microsoft, it should come as no surprise that the goal was to show NT being faster then Linux. Bear in mind though, that all _independent_ testing has shown exactly the opposite to be true, certainly for uniprocessor machines such as the ftp.cdrom.com server.
... but as those two have indicated, this is a complete farce, and you can expect the "retest" results to be as information free as the first ones.
There have yet to be any standard SMP benchmarks (TPC-D, SPECWeb96 etc) published, although an unofficial Oracle benchmark indicated Linux to beat NT there also.
Also bear in mind that the "Mindcraft" testing has since been shown to have been performed in a Microsoft lab (the "Mindcraft" e-mails originated from a Microsoft domain)...
Ultimately, all the "Mindcraft" tests really proved is that Microsoft is starting to take Linux as a _very_ serious threat to NT - not surprising given the Linux server marketshare and growth numbers.
Microsoft is attempting to recover from the PR nightmare resulting from this testing by redoing the tests with "unimpeachable" Linux configuration expertise supplied by Linus and Alan Cox
Everyone's actions are driven by satisfaction - just that different people enjoy different things. You may enjoy getting the guts of the worlds 513th CD player to work, someone else may enjoy addign the chrome, and someone else writing the documentation.
;-)
Note also that pleasure is a complicated many faceted thing... Helping others may appear to be altruistic, but people only do it because it makes them feel good. Finishing the last 10% if a project may be a grind, but many people do it because of the satisfaction of completion/whatever - maybe the source of pleasure is different from then first 90%, but it's still pleasure driven.
Certainly open source projects may seem more haphazard than things we are forced to do professionally, but the level of effort/completion put into them is really globally optimized!... If the 70% complete status of a project sufficiently annoys someone, then it will rise to the level where completing it is becomes the most satisfying thing they cad do - and they will do it.
BTW, do you _really_ think a robot could replace you in having sex with your wife? Sounds like you need to use a little imagination or whipped cream!
There's also matroxfb which is an accelerated driver for matrox cards. I use it with a Millenium I.
Note that vesafb will work for any card with a VESA 2.0 BIOS and linear frame buffer.
RedHat is working with Precision Insight to add OpenGL direct rendering to XFree86. This will allow clients that are on the same machine as the server to render directly into the framebuffer. SGI already has this.
XFree also supports the DGA and MIT_SHM X extensions than can be used to speed up local clients.
I used to love correcting punch card mistakes using the duplicate button - sounded like a freakin' machine gun! :)
Remember that X is just the low level graphics library - it's what's on top of it (the low level and desktop environment widgets and dialogs) that contribute to an app's look and feel. The currently popular widget sets and environments - qt/KDE and gtk/GNOME are still very young, but improving very fast. Remember that Windows itself was essentially useless until 3.0, and the UI was unusable until Windows '95...
Remember that the current Linux user base is only in the 12-15 million range, but growing _very_ fast. Mexican schools are adopting Linux/GNOME, and China is likely to be predominantly Linux too for cost reasons (it already has a pretty large Linux user base, and there's only just been released a Chinese version of Linux!)... as the desktop user base grows, there's going to be some very rapid advances...
.. a single zeon makes a kick-ass server, but in the mindshaft testing lab you need a quad zeon! Hmmm.. of course it _was_ optimized for NT - probably the logic was that with 4 cpus running NT, then at any time ONE of them might actually be working!