Slashdot Mirror

Karma whore alert! Please mod down, and mod this anonymous post up!

archive.progeny.com (US or Canada)
ftp.twoguys.org (US or Canada)
ftp3.sourceforge.net (US or Canada)
ftp.rpmfind.net (US or Canada)
ftp.sourceforge.net (US or Canada)
ftp.cse.buffalo.edu (US or Canada)
ftp.yggdrasil.com (US or Canada)
planetmirror.com (Australia)
ftp.sunet.se (Europe)
ftp.dataplus.se (Europe)
ftp.easynet.nl (Europe)
ftp.unina.it (Europe)
ftp.belnet.be (Europe)
ftp.codefactory.se (Europe)
ftp.tr.gnome.org (Europe)
fr.rpmfind.net (Europe)
ftp.acc.umu.se (Europe)
ftp.no.gnome.org (Europe)
ftp.dit.upm.es (Europe)
fr2.rpmfind.net (Europe)
linux.cem.itesm.mx (South America)

archive.progeny.com (US or Canada)
ftp.twoguys.org (US or Canada)
ftp3.sourceforge.net (US or Canada)
ftp.rpmfind.net (US or Canada)
ftp.sourceforge.net (US or Canada)
ftp.cse.buffalo.edu (US or Canada)
ftp.yggdrasil.com (US or Canada)
planetmirror.com (Australia)
ftp.sunet.se (Europe)
ftp.dataplus.se (Europe)
ftp.easynet.nl (Europe)
ftp.unina.it (Europe)
ftp.belnet.be (Europe)
ftp.codefactory.se (Europe)
ftp.tr.gnome.org (Europe)
fr.rpmfind.net (Europe)
ftp.acc.umu.se (Europe)
ftp.no.gnome.org (Europe)
ftp.dit.upm.es (Europe)
fr2.rpmfind.net (Europe)
linux.cem.itesm.mx (South America)

Optimized number crunching aps for Mac smoke (most) anything else out there, and certainly anything on the x86 side of computing.

Sorry, but this just isn't true. As far as I know, there is one single application (Genentech BLAST) where a heavily altivec-optimized G4 version beats a completely unoptimized x86 version. What Apple don't say is that there are several commercial x86-optimized BLAST versions that beat the Genentech one.

The true story is that anything using double precision (and most scientific/numbercrunching code does!) will run a factor two faster on x86 than G4, simply because Altivec only does single precision. There are also several applications optimized for double precision SSE2, while there is no such alternative on the G4.

It is also a myth that Altivec should be so much better than SSE. Sure, it does better per clock, but the clock in a P4 is more than twice that of a G4. Just check out e.g. ATLAS or GROMACS. Both these programs use Altivec on G4 and SSE/SSE2 on x86, and the x86 performance is significantly better even in single precision, not to mention the hardware price.

Finally, this assumes you don't consider the cost of porting the code to Altivec, or that writing it in Altivec from the beginning makes it unportable. This might not be important to photoshop users, but it is for scientific use: there isn't a single Altivec-capable machine among the world's 500 fastest supercomputers. If you just take a portable source and compile it, code generated by the Intel compilers on x86 will run about twice as fast as anything generated on G4.

No, I'm not a 'mac basher'. But you don't serve the Mac community by inflating benchmark results (Apple "flops" vs. real double precision flops, comparing hand-optimized code vs. nonoptimized code, not giving any details on how benchmarks where performed, not publishing SPECbench results like ever other vendor does).

All this will do is make people p*ssed off when they compile their first code on a G4 and realize it's a factor 2 slower.

For a comparision, have a look at a recent AMD cluster installation at

HPC2N. They could have used SSE/3DNow instructions to increase a bogus "flopcount" by a factor 4, but they don't. Instead they only mention the standard double precision flops (800 Gflop peak), and in practice they really achieve 481 Gflops on LINPACK.

Old, but here it goes:

100 Mbps FTTH in my home.

I also remember reading about these folks who made their own neighborhood 100 Mbps fiber network. The screenshot of the FTP download speed is just plain silly. The creator of the page even mentions "and 25-40 Mbps is possible most of the time - that means it is their single hard disk limiting the speed!". :)

Driving around in a station wagon "full" of equipment sounds like ancient times. I took a course in "mobility and wireless networks" at my University (Swedish)), where a representative from Ericsson was invited and demo:ed a device that looked exactly like a normal cell phone.

He hooked it up to a laptop via serial and USB and got all kinds of data from the GSM/GPRS network.

It was also very easy to inject malformatted data in the GSM/GPRS network and simply make it crash, he told us. The networks in the area where they develop it were prone to be down... ;)

Here you have a link to the official website (http://www.ericsson.com/tems/gsm/pocket-gsm.shtml ) along with a picture of it. He told us it cost about US$20.000 to get hold of one of them, and many phone operators in Europe use them.

It would be great to see how much throughputdisks SCSI and IDE support. I worked at a place where they had a SCSI disk controller (a hub) where a lot of machines could connect to the array and see the disks as local disks. I have seen some cases where there is a fiber channel controller going into another fiber channel controller and the actual disks were simple scsi disks.

What metrics do people use? Price surely favors IDE, and proabably scalability -- it is very easy to install a dozen scsi controllers and connect to several hundreds of disks because of the several channels provided.

A quote here:

Connectivity: The ATA interface can only address two devices while SCSI can address eight devices (Narrow SCSI), 16 devices (Wide SCSI), 32 (Very Wide SCSI) or 126 (FireWire). There are also many peripherals available to SCSI only and not ATA.


The page I got this quote from is at http://www.acc.umu.se/~sagge/scsi_ide/

S

http://www.acc.umu.se/~tfytbk/mattgrand has a description of our real 100 Mbps Internet connected network (access also at 100 Mbps!)...

We have this kind of limits already in our 100 Mbps Internet access@home.

There also is a destinction between the "internal" network (citywide) and the rest of the world.

See my page (already hit by /. a few months ago, 62 000 visitors) at http://www.acc.umu.se/~tfytbk/mattgrand.

It would also be nice, now, if the government could help facilitate future growth via funded expansion of shared infrastructure. Broadband access is apparently become less readily available and more costly, right at a time when demand for access is increasing.

As long as you are talking about extremely local governments, or even stuff smaller than governments at the community level, that's fine. But you don't really need government at all. Government is the lazy man's solution, an attempt to get someone else to pay for it. Look at that neighborhood fiber network in Sweden or the Coop in Colorado and you'll see that the problem can be solved just by having people get off their asses and doing something about it. And if we're not willing to spend the money directly and efficiently in our own communities, how dare we ask that it be done indirectly with federal taxes by unaccountable fund-siphoning middlemen beaurocratic grifters? (Oops, I think my anti-fed bias is showing. ;-)

We don't need no stinking federal network. The only problems are with the "last mile". At higher levels above that, the private sector can very easily handle the job.

Wireless is cute, but I still think that wherever possible, cheap fibre optic cables should be used.

3M's Volition Series is an excellent example -- we used it to get 100 Mbps Internet in our block.

Combine fast connectivity with a local DMZ to which it is for free to connect and all data traffic within the DMZ is free, and you've got a success on your hands.

Use wireless only where it is impossible to install fibre optics, but you will be amazed over how many tubes/pipes already present under roads et.c. which can be used with a permit!

As I have written before here on /., we already have a MAN in Umeå, Sweden. We are currently enjoying 100 Mbps full duplex per fibre optics to our homes.

On http://www.acc.umu.se/~tfytbk/mattgrand [57.000 visitors], I describe the fibre optical network we ourselves installed in my parent's block and it is connected to the city's MAN.

The smart thing is that the hub in the entire MAN is a "DMZ" to which it is FREE to connect for anyone (you have to either provide your own router or rent a port in someone elses router).

ALL traffic within the DMZ is for FREE and the only thing that cost is capacity out to the rest of the world (since someone has to pay for the connection to the closest Internet Exchange Point).

The concept is excellent! They have upgraded the backbone in the DMZ to 10 Gbps and lots of equipment is 1 Gbps, amongst others the ISP (Bonet) that we are using. We certainly enjoy high speed, especially within our city (100.000 inhabitants), where I always get ~4000 kB/s sustained speed! (for example from my appartment, 6 km (4 miles) from my parent's house, to my server which I left back home when I moved here.)

I really think that this concept should be considered to boost the public's interest in connecting to the MAN - here in Umeå we have around 70% (!) of all households connected to this fabulous network - without doubt I live in the 'most connected' city in the world!

The fibre optics were dug during the 90s by our community owned power company, that were smart enough to put down single mode fibres when they upgraded the hot water pipelines that provides the city with heat.

Abstract

We consider the use of graphical passwords, passwords based on pictures and drawings rather than on text, from a broad perspective. Results from memory research are used as a basis for a discussion of what properties a password system should have to provide passwords that are hard to guess but easy to remember. These ideas are used to guide the design of a large number of prototype systems, which are described in detail. The security properties of the different systems are evaluated, and the most promising candidates are singled out.

We also discuss how the technique of fuzzy commitment can be used to create secure fuzzy password systems --- systems that allow the user to make small errors when entering her password --- and show that such systems must be carefully designed to avoid leaking information to the attacker.

We evaluate the systems through user testing and show that while visual passwords are not a miracle cure for all problems with passwords, they still might be a useful complement to traditional password systems.

http://www.cs.umu.se/~niklasf/exjobb/

I recommend you check out my classmate's Master Thesis on http://www.cs.umu.se/~niklasf/exjobb/.

He is currently working at RSA Labs, so I imagine he knows what he is talking about! :)

I can't resist to once more point out that we are accessing the Internet per 100 Mbps full duplex fibre link to our home.

I have a very detailed description on this page on how we installed a very high-tech network in our entire block.

The page have been slashdotted once before, so the visitor counter have passed 52 000!

Best regards, Tomas

My five cents: Stick to Cat5(e) in the house, but leave separate conduits with nothing but a wire in them (very hard to pull additional cables when other cables are present).

Where I live, we have 100 Mbps connections to the entire city, and also to the Internet.

It was posted on /. a few weeks ago, but feel free to see how a real network is made on 100 Mbps@home (50 000+ visitors already!).

Actually, it was 3, not 4. And heres where I found the squared-circle quote. That articles title says pi = 4? but then the article talks mostly about Indiana trying to set it to 3.

This is a joke, right?

You've posted at least three times that people complaining about Nautilus' speed (or lack thereof) should ditch their IDE drives and go to SCSI.

You might want to jump a little, I'm gonna throw some basic logic at you.

If the Windows file manager runs as fast on IDE as Nautilus does on SCSI, you can make two conclusions: (1) SCSI is not any faster than IDE, and Nautilus is just slower than the Windows FM. (2) SCSI is faster than IDE, and Nautilus is a lot slower than the Windows FM.

Either way, you're not helping your case.

While we're on the subject, you might want to consider that if only one device on the controller is speaking, SCSI has no real advantage over IDE. That means for most desktop systems, which only have one hard drive, IDE is perfectly sufficient and a hell of a lot cheaper. Do your own research: here's the first link I found at google on the subject.

So drop your ridiculous SCSI fetish and recognize that Nautilus is just slow (even according to your own damned post).

Some people in Sweden rallied together and bought themselves a 100mpbs network for their whole village. They dug up the trenches for the fiber lines themselves. Pretty neat stuff. I don't know if this is exactly what you were talking about when referring to NANs, but this is what came to my mind.

Check it out: http://www.acc.umu.se/~tfytbk/mattgrand/. There's some cool pictures of the equipment, screenshots of 6mb/sec downloads, etc.

Isn't it nice!

I had about 5500-6000 visitors before /. posted the story, mostly due to a posting about six months ago on [Canarie.ca] and links to my page from [Wkmn.com]

I just saw the counter exceed 10 000 hits and our webserver ([ACC.umu.se]) has served you guys with over 1 GB of data already.

I actually received 15 requests for my resume during the storm that followed the posting on the Canarie mailing-list!

One employer from San Francisco, California, actually e-mailed me one day (the first week on my first job) and said: "I'm in Stockholm now and I'm booked for a flight up to Umea (my city).". I was also in Stockholm, but to make a long story short, I chose to stay with my first employer, because it felt so wrong to abandon them after a single week...

I hope he ever will forget me. ;)

I quit that job six months later (june 2001) and now I'm taking a few courses at my old University and I'm thinking about a PhD or I-don't-know-what.

I can't be sure this page will still be up in the afternoon (I'm writing this at 5:30 AM EDT), but do check the community fiber network. And while you're there, scroll all the way to the bottom and click on link that says Visitors (or go directly from here) and check out the people from all over that followed the herd.

Grab it here instead. The source where cut in the post.

OS research has been pursueing these goals for years. There's nothing there that's really very interesting or new. It sounds like they've just browsed the web for a little while and summarized what the various projects are striving for.

One project that's come pretty far is Mosix (I think they're planning to integrate bits into Linux 2.5, but I'm not sure). Then of course there's Plan 9 and Inferno from the fine folks who brought you Unix. And lets not forget Tanenbaum's Amoeba.

Here's a mirror of the PDF documents:

http://www.informatik.umu.se/~meson/oss-eu/

Please use the main site if possible. Don't know if the sysadmin likes this mirror :)

Pick one

I recommend moving to Indiana, where there is a law that has been passed by the state General Assembly declaring that pi == 3. You can read more at http://www.acc.umu.se/~olletg/pi/tribute.html

Bah! Any true artist knows that the one true way to do POV-Ray is with Emacs and pov-mode. ;-)

Here's some I found, it was along the lines of stuff like this

Um, I guess that's "Tanenbaum", if you're referring to the computer scientist (?:) perhaps most known around here for starting the legendary "Linux is obsolete flame fest with Linux and folks, way back when (January 1992, to be specific). I wasn't aware that that quote was due to him, though... Cool! ;^)

I'm interested in replies from anybody who has visited mill's web page and thinks that his is a voice worth listening to on the subject of creating web pages.

I'm willing to say he is. Okay, the colour choices are pretty hideous, but take a look at the page in your fancy-schmancy browser with stylesheets turned on. Now look at it in your fancy-schmancy browser with stylesheets turned off. Now look at it in Lynx. Now look at the source.

Those floating yellow boxes with quotes in them are just <blockquote> elements. No layout is done on a per-quote basis. Compare that to trying to acheive the same thing with tables.

In short, this page rules, and has really made me rethink the viability of doing all of your page layout with CSS.

I visited your web page, and I think you're just bitter because you have one of the ugliest web pages known to man. Perhaps you're taking angry potshots at people who give their pages an attractive design because you have no such aptitude?

I'm interested in replies from anybody who has visited mill's web page and thinks that his is a voice worth listening to on the subject of creating web pages.

Cheers,
ZicoKnows@hotmail.com

He said:

"I still maintain the point that designing a monolithic kernel in 1991 is a fundamental error. Be thankful you are not my student. You would not get a high grade for such a design :-)"