They aren't cooling anything, rather they are preventing being heated by the sun. Big difference.
+5 Insightful? What are you mods smoking?
We're talking about the inside of a PC case. They're moving out the heat that was generated by electrical power dissipation in the CPU (and other components). IOW, they're cooling the CPU. The sun is about 0.65 W/sq. inch. If your PC is in the direct sun (and mine's not), that's about 1% compared to a Pentium.
Oh, I get it: NASA, Shuttle. It's supposed to be +1 Funny. No offense to nB, but I'd vote -1 WTF?!
I went to a science and engineering school, which explains the lack of Macs.
Science and engineering was traditionally one of the Mac's strong markets. When I was an EE undergrad (87-91), there were far more Macs than PCs in the labs, and some software development classes were done only on Macs. I got to learn Unix, too. As other posters have mentioned, free PC donations from Dell and Intel (with the requirement to run Windows) quickly replaced anything else. It's hard to compete with FREE. Now the fresh grads are all Windows users, and surprised there's anything else.
When Mathworks cancelled MATLAB for the Mac, there was enough outcry that they eventually brought it back. Wolfram had enough sense to keep Mathematica on the Mac all along. There are plenty of people in science, engineering, research, and higher education that are still using Macs. I think MacOS X has begun a Mac resurgence with lots of geeks. It's nice to have Unix and productivity apps on the same box. LOTS of OSS programs just run on MacOS X.
there really are not a lot of Macs in use compared to Windows machines
And that means what, exactly? Linux Insider has definitive proof that scientists should use Macs.:) Are you a lemming, or a scientist?:)
My first computer was a Leading Edge Model-D with a 7.16 MHz 8088 and a switch that let you run at 4.77 MHz! It came with a "Magnavox Professional RGB Monitor 80", that does CGA and has a composite video input.
The box is long dead, but I still use the monitor at least weekly -- it's the video for my Shuttle Athlon XP 2400+ PVR-250 MythTV Linux box. It needs a periodic whack to fix the colors, but generally works fine. Manufactured in January 1987!
My G4(2x867MHz,MDD) has exactly this same problem. I've reset the PMU -- it makes no difference. I've even taken it to the shop twice. Of course, it's always working by the time it gets there.:) It seems that by the time it cools off, it works fine again.
I also have the new (supposedly quieter) power supply for the MDDs. I haven't installed it yet because I'm afraid to add another variable to the equation.
Their advice in the shop: buy AppleCare -- we can't find it, but it'll probably be back!
Re:What about Transputers?
on
Grid Processing
·
· Score: 2, Informative
no one in America noticed them We used transputers on quite a large number of projects right here at the University of Texas.
the NIH principle Actually, the problem was that they were slow and complicated. They went so long between family upgrades that eventually we could replace a large array of transputers with a few regular CPUs. Not to mention that we can also get a handy little thing like an OS on general purpose CPUs.
programming languages designed for parallelism Did I mention complicated? Occam was part of the problem. The scientific world wants to program in C or Fortran, or some extension of them, or some library called by them. That's why MPI is so popular.
not all problems can be done faster by doing more of it at once I'm not sure I agree. Having more capability at each compute node means less need for partitioning. (The part you say is hard.)
Obviously there's a lot of work to be done in parallel processing. You can hardly blame Inmos's problems on geography (or America for Inmos's problems). They looked very promising for awhile, but just didn't keep up.
Graduate schools are also largely populated with students that are not U.S. citizens, and are not permitted to work on these types of projects. That has profs looking for other sources of funding.
I think there were about 12 books, but only a few are available at PG. ERB's work was some of the first titles there. Pretty cool (and very imaginative) old-school Sci-Fi. You may have heard of his other work -- Tarzan.
As is (almost) every AltiVec FFT benchmarked on the site -- FFTW is the slowest one listed. I included FFTW only because it is a "well known" scalar point of reference -- it is obviously very well known to djb. The topic of discussion is AltiVec, which djbfft clearly does not use. No matter how fast of a scalar implementation djbfft is, it cannot hope to compete with vBigDSP.
I mean absolutely no disrespect to Dr. Bernstein. On his site he makes statements such as this: How many other high-performance libraries have been excluded from the FFTW benchmarks, or slowed down by the FFTW authors? Not good.
I would suggest that he show head-to-head results in an easily digestable format. The FFTW people have made it very easy for him to do so. He could fix the compile flags on benchFFT that he is complaining about, and then post the results. As in "put your money where your mouth is."
by using interleaved data, you are benchmarking a scalar vDSP against your other scalar algorithms
This is clearly not the case. On an 800 MHz G4, the peak scalar performance is 1600 MFLOPS.
You're lucky to get 800 MFLOPS, just like the (very good) scalar FFTW. That's exactly why we have SIMD/AltiVec.
I'm using Apple's ctoz() and ztoc() to convert from interleaved to split and back, and getting "about the same as vBigDSP" -- that's 1-2 GFLOPS. Without these conversions, that curve peaks at about 4 GFLOPS.
You also could have checked your incorrect assumption if you had looked at the provided source code.
The djbfft website clearly emphasizes the Pentium, and makes no mention of AltiVec. I have personally found FFTW to be relatively fast on most (non-vector) platforms. However, the AltiVec FFTs smoke it.
You have to give the FFTW people credit: they have provided their results, and made it easy for you to reproduce them. I guess that's why they're at MIT, and djbfft is at.... Where is it again?
I don't really care who wrote my FFT, I just want it to be FAST and CORRECT.
The djbfft website sounds like a lot of whining. If they think they've been wronged, they should provide a version of benchFFT that shows how it is better than FFTW, and provide a way for people to verify that.
It's important to point out that the "traditional way" to store complex numbers (and do an FFT) is with interleaved data. If your application uses interleaved data (and most do), then this library is for you. Split data is being used more because of the obvious performance improvements, but you have to change your code (and your mindset) to use it.
I've updated it to say: as much as twice as fast as vBigDSP on split data (but about the same on interleaved data).
I'd like to get an open-source split-data version to add to the site, so Linux users get good split-data FFTs, too.
A DAC or a pot would be a tremendous waste. What you want (as several people have mentioned) is Pulse Width Modulation. You can adjust the brightness by adjusting the duty cycle. I'd recommend an PIC or some such.
A similar bill has already been passed in California. This bill could have helped to prevent my daughter's abduction to Mexico. She has been missing since last April. FindSabrina.org for details.
Let me tell you, your definition of "stuff that matters" changes when your child has been abducted.
If the police asked you about the crack house next door, shouldn't you tell them everything you know? Get the crooks out of your neighborhood! Online communities are no different, and eBay wants a community of trust. It works largely on the honor system!
Believe me, as someone relying on law enforcement (see my.sig), they have better things to do than invade your privacy. They typically have large caseloads, and can't dedicate enough resources to the ones they have.
People want to "stay out of it" but then complain because society is going to hell! I applaud eBay's policy.
I've used LinuxPPC for years. I've ditched it on my desktops -- I don't even dual-boot anymore. (Of course, my myriad embedded systems still use it). MacOS X has all the Unixy goodness I need, plus commercial apps and great multimedia support. With Orobor and Fink, I've got it all!
Have you been reading/. lately? All the Alpha Geeks and nuttin up on Jaguar. Somehow, it just works!
To stay on topic: The iApps are pretty darn cool and worth paying for if you want upgrades.
I'm still searching for my missing daughter -- she was abducted over 8 months ago. I was on America's Most Wanted in November, but she still hasn't been found.
Please help to give a geek dad the only thing he wants for Christmas -- Sabrina back at home. Please mod this up, and please visit FindSabrina.org. Please tell your friends to look at her picture to see if they've seen her.
This Christmas has really sucked for me and my family. (Sorry about the repeat -- that's my.sig)
I selected FeaturePrice.com, the top rated site. They do more for less than my old service. They have several packages, depending on the service you want. My only complaint was paying annually up-front.
While you're at it, please visit my site (in my sig), and look for my abducted daugher. We'll give it the slashdot test -- I have "unlimited" bandwidth.:)
---
How secure are Internet transmissions with Canopy?
Radio communication, by it's very nature, is secure.
What the... ?!?!
Canopy systems offer additional advances security with
over-the-air-DES encrypted communication. And, because
Canopy systems provide 128 bit authentication, only
authorized canopy customers can gain access to your
Canopy system.
Ahhhh, marketing. When will they learn? And right next to the "hacking sattelites" article.:)
Slashdot Mirror
Your == possessive. You're == you are.
You're awfully inflamatory. The LED is not 99% efficient -- the reflector inside the LED is. Typical white LEDs are well under 40%.
They aren't cooling anything, rather they are preventing being heated by the sun. Big difference.
+5 Insightful? What are you mods smoking?
We're talking about the inside of a PC case. They're moving out the heat that was generated by electrical power dissipation in the CPU (and other components). IOW, they're cooling the CPU. The sun is about 0.65 W/sq. inch. If your PC is in the direct sun (and mine's not), that's about 1% compared to a Pentium.
Oh, I get it: NASA, Shuttle. It's supposed to be +1 Funny. No offense to nB, but I'd vote -1 WTF?!
I went to a science and engineering school, which explains the lack of Macs.
:) Are you a lemming, or a scientist? :)
Science and engineering was traditionally one of the Mac's strong markets. When I was an EE undergrad (87-91), there were far more Macs than PCs in the labs, and some software development classes were done only on Macs. I got to learn Unix, too. As other posters have mentioned, free PC donations from Dell and Intel (with the requirement to run Windows) quickly replaced anything else. It's hard to compete with FREE. Now the fresh grads are all Windows users, and surprised there's anything else.
When Mathworks cancelled MATLAB for the Mac, there was enough outcry that they eventually brought it back. Wolfram had enough sense to keep Mathematica on the Mac all along. There are plenty of people in science, engineering, research, and higher education that are still using Macs. I think MacOS X has begun a Mac resurgence with lots of geeks. It's nice to have Unix and productivity apps on the same box. LOTS of OSS programs just run on MacOS X.
there really are not a lot of Macs in use compared to Windows machines
And that means what, exactly? Linux Insider has definitive proof that scientists should use Macs.
My first computer was a Leading Edge Model-D with a 7.16 MHz 8088 and a switch that let you run at 4.77 MHz! It came with a "Magnavox Professional RGB Monitor 80", that does CGA and has a composite video input.
The box is long dead, but I still use the monitor at least weekly -- it's the video for my Shuttle Athlon XP 2400+ PVR-250 MythTV Linux box. It needs a periodic whack to fix the colors, but generally works fine. Manufactured in January 1987!
My G4(2x867MHz,MDD) has exactly this same problem. I've reset the PMU -- it makes no difference. I've even taken it to the shop twice. Of course, it's always working by the time it gets there. :) It seems that by the time it cools off, it works fine again.
I also have the new (supposedly quieter) power supply for the MDDs. I haven't installed it yet because I'm afraid to add another variable to the equation.
Their advice in the shop: buy AppleCare -- we can't find it, but it'll probably be back!
no one in America noticed them
We used transputers on quite a large number of projects right here at the University of Texas.
the NIH principle
Actually, the problem was that they were slow and complicated. They went so long between family upgrades that eventually we could replace a large array of transputers with a few regular CPUs. Not to mention that we can also get a handy little thing like an OS on general purpose CPUs.
programming languages designed for parallelism
Did I mention complicated? Occam was part of the problem. The scientific world wants to program in C or Fortran, or some extension of them, or some library called by them. That's why MPI is so popular.
not all problems can be done faster by doing more of it at once
I'm not sure I agree. Having more capability at each compute node means less need for partitioning. (The part you say is hard.)
Obviously there's a lot of work to be done in parallel processing. You can hardly blame Inmos's problems on geography (or America for Inmos's problems). They looked very promising for awhile, but just didn't keep up.
Graduate schools are also largely populated with students that are not U.S. citizens, and are not permitted to work on these types of projects. That has profs looking for other sources of funding.
Remember Edgar Rice Burrough's Mars books?
Some of them are available at Project Gutenberg
#1 - A Princess of Mars
#2 - Gods Of Mars
#2 - Warlord of Mars
I think there were about 12 books, but only a few are available at PG. ERB's work was some of the first titles there. Pretty cool (and very imaginative) old-school Sci-Fi. You may have heard of his other work -- Tarzan.
Then why aren't we using metric?
:)
Duh! Because it's French.
I mean absolutely no disrespect to Dr. Bernstein. On his site he makes statements such as this: How many other high-performance libraries have been excluded from the FFTW benchmarks, or slowed down by the FFTW authors? Not good.
I would suggest that he show head-to-head results in an easily digestable format. The FFTW people have made it very easy for him to do so. He could fix the compile flags on benchFFT that he is complaining about, and then post the results. As in "put your money where your mouth is."
by using interleaved data, you are benchmarking a scalar vDSP against your other scalar algorithms
This is clearly not the case. On an 800 MHz G4, the peak scalar performance is 1600 MFLOPS. You're lucky to get 800 MFLOPS, just like the (very good) scalar FFTW. That's exactly why we have SIMD/AltiVec.
I'm using Apple's ctoz() and ztoc() to convert from interleaved to split and back, and getting "about the same as vBigDSP" -- that's 1-2 GFLOPS. Without these conversions, that curve peaks at about 4 GFLOPS.
You also could have checked your incorrect assumption if you had looked at the provided source code.
FFTW 3.0 beta ... Anyone want to add it to the benchmark?
Yes! I've run the benchmark, and will add the results shortly. Preliminary indication -- it's no faster than vBigDSP.
The djbfft website clearly emphasizes the Pentium, and makes no mention of AltiVec. I have personally found FFTW to be relatively fast on most (non-vector) platforms. However, the AltiVec FFTs smoke it.
You have to give the FFTW people credit: they have provided their results, and made it easy for you to reproduce them. I guess that's why they're at MIT, and djbfft is at.... Where is it again?
I don't really care who wrote my FFT, I just want it to be FAST and CORRECT.
The djbfft website sounds like a lot of whining. If they think they've been wronged, they should provide a version of benchFFT that shows how it is better than FFTW, and provide a way for people to verify that.
It's important to point out that the "traditional way" to store complex numbers (and do an FFT) is with interleaved data. If your application uses interleaved data (and most do), then this library is for you. Split data is being used more because of the obvious performance improvements, but you have to change your code (and your mindset) to use it.
I've updated it to say: as much as twice as fast as vBigDSP on split data (but about the same on interleaved data).
I'd like to get an open-source split-data version to add to the site, so Linux users get good split-data FFTs, too.
The University of Texas at Austin has set up a website in response to this incident: https://www.utexas.edu/datatheft
A DAC or a pot would be a tremendous waste. What you want (as several people have mentioned) is Pulse Width Modulation. You can adjust the brightness by adjusting the duty cycle. I'd recommend an PIC or some such.
If you do call your state representatives, please also express your support for Texas House Bill 1899 - Prevention of International Parental Child Abduction.
A similar bill has already been passed in California. This bill could have helped to prevent my daughter's abduction to Mexico. She has been missing since last April. FindSabrina.org for details.
Let me tell you, your definition of "stuff that matters" changes when your child has been abducted.
I'd say Tennessee has Texas envy.
:)
top 10 Engineering school
top 15 Education school
top 20 Business school
top 20 Law school
Is Tennessee in any of these lists? They made #50 in one...
How about sports?
Sports Illustrated Top Sports College of 2002
Baseball 2002 National Champs
Currently:
#4 NCAA Men's Basketball
#7 NCAA Football
#7 NCAA Women's Basketball
#9 NCAA Baseball
NFL leading rushers in 2002:
#1 Ricky Williams, Texas grad
#3 Priest Holmes (injured several games!), Texas grad
Hook 'Em Horns! Our Burnt Orange rulez your pale yellow!
(Good thing they don't have a top college IT F**k-up list.
Agreed!
.sig), they have better things to do than invade your privacy. They typically have large caseloads, and can't dedicate enough resources to the ones they have.
If the police asked you about the crack house next door, shouldn't you tell them everything you know? Get the crooks out of your neighborhood! Online communities are no different, and eBay wants a community of trust. It works largely on the honor system!
Believe me, as someone relying on law enforcement (see my
People want to "stay out of it" but then complain because society is going to hell! I applaud eBay's policy.
Go and install ... Linux/PPC
/. lately? All the Alpha Geeks and nuttin up on Jaguar. Somehow, it just works!
Dude, have you used MacOS X Jaguar? It rocks!
I've used LinuxPPC for years. I've ditched it on my desktops -- I don't even dual-boot anymore. (Of course, my myriad embedded systems still use it). MacOS X has all the Unixy goodness I need, plus commercial apps and great multimedia support. With Orobor and Fink, I've got it all!
Have you been reading
To stay on topic: The iApps are pretty darn cool and worth paying for if you want upgrades.
I'm still searching for my missing daughter -- she was abducted over 8 months ago. I was on America's Most Wanted in November, but she still hasn't been found.
.sig)
Please help to give a geek dad the only thing he wants for Christmas -- Sabrina back at home. Please mod this up, and please visit FindSabrina.org. Please tell your friends to look at her picture to see if they've seen her.
This Christmas has really sucked for me and my family.
(Sorry about the repeat -- that's my
I just went to www.10-best-web-site-and-domain-hosting-services.c om and looked at some reviews. I wish I had done that in the first place -- it would have saved lots of money.
:)
I selected FeaturePrice.com, the top rated site. They do more for less than my old service. They have several packages, depending on the service you want. My only complaint was paying annually up-front.
While you're at it, please visit my site (in my sig), and look for my abducted daugher. We'll give it the slashdot test -- I have "unlimited" bandwidth.
---
I'm not trying to sell anything -- my daughter was abducted. I want to spread the word so she can be found.
Newspaper ads are expensive, and my budget is limited. There are so many other ads that you won't notice mine.
If I had the budget of Coca-Cola I could have her on every TV in the world. But I don't.
People spam because it's cheap and easy.
While spam annoys me, too, I intend to enlist it in the search for Sabrina.
Please visit Sabrina's website: FindSabrina.org
Linus passed gas yesterday.
Droves of geeks were seen wafting in his wake, hoping to get a whiff.
Must be a slow day for news.
What the... ?!?!
Ahhhh, marketing. When will they learn? And right next to the "hacking sattelites" article. :)