This is very cool but how much is it gonna cost? Correct me if I'm wrong but right now nanotubes are so expensive they can't use 'em to store hydrogen for your hydrogen-powered car. So you're gonna put 'em in paint? This is like the caviar omlette on crack.
i am so busted. i did not understand. everything was working OK because i did not have the two stupid flags on. the two (four in case of the 8.0 version of the compiler) stupid flags (as mentioned above) and in the amd online documentation:
www.amd.com/us-en/assets/content_type/white_papers _and_tech_docs/32035.pdf
are:
for the Intel 7.1 compiler -xK and -xW
Intel 8.0 -xK -xW -xP -xB and -xN
Intel 8.1 -xN and -xP
see table 10 on p. 57
on the 8.0 compiler, -xK slowed down my code (relative to no -xK) and -xP crashed my code:
forrtl: severe (168): Program Exception - illegal instruction
Image PC Routine Line Source
md 0804A124 Unknown Unknown Unknown
md 00B2678A Unknown Unknown Unknown
md 08049FC1 Unknown Unknown Unknown
HOWEVER I would like to mention that -O3 works fine! I never heard of these two (four) foobar flags before today. They are eevil, yes, but the flag combo -axP -xW -ipo -O3 as suggested by the AMD documentation gave me a 0.5% speed up compared to just -O3 on my Opteron (NOTE: I was using the 8.1 compiler where they work fine according to AMD's documentation). I would now like to join the camp of people writing here that says "this is no suprise. it's in the AMD documentation for the love of Mike!"
cool. thanks. i'm having a problem finding the variable that i need to reset. there are about a kagillion posts up there. could you maybe point me toward a particular thread? that would be awesome. thanks.
I work in a computer simulation lab. We have something like 50 AMD processors running code we compiled with the Intel Fortran Compiler and it's been working great for the past two years now. We have a mixture of Athlons and Opterons. Everything is workin' great. No problems. We used to use the Portland Group's fortran compiler until we found that the Intel Fortran Compiler generates faster executables. This was the case for both 32 and 64 bit executables. Started out with Portland Group, switched to Intel...and it works great for us!
Slashdot Mirror
This is very cool but how much is it gonna cost? Correct me if I'm wrong but right now nanotubes are so expensive they can't use 'em to store hydrogen for your hydrogen-powered car. So you're gonna put 'em in paint? This is like the caviar omlette on crack.
you need to figure out how to use it. you can add as many artists as you like. see coments above.
i am so busted. i did not understand. everything was working OK because i did not have the two stupid flags on. the two (four in case of the 8.0 version of the compiler) stupid flags (as mentioned above) and in the amd online documentation: www.amd.com/us-en/assets/content_type/white_papers _and_tech_docs/32035.pdf
are:
for the Intel 7.1 compiler -xK and -xW
Intel 8.0 -xK -xW -xP -xB and -xN
Intel 8.1 -xN and -xP
see table 10 on p. 57
on the 8.0 compiler, -xK slowed down my code (relative to no -xK) and -xP crashed my code:
forrtl: severe (168): Program Exception - illegal instruction
Image PC Routine Line Source
md 0804A124 Unknown Unknown Unknown
md 00B2678A Unknown Unknown Unknown
md 08049FC1 Unknown Unknown Unknown
HOWEVER I would like to mention that -O3 works fine! I never heard of these two (four) foobar flags before today. They are eevil, yes, but the flag combo -axP -xW -ipo -O3 as suggested by the AMD documentation gave me a 0.5% speed up compared to just -O3 on my Opteron (NOTE: I was using the 8.1 compiler where they work fine according to AMD's documentation). I would now like to join the camp of people writing here that says "this is no suprise. it's in the AMD documentation for the love of Mike!"
cool. thanks. i'm having a problem finding the variable that i need to reset. there are about a kagillion posts up there. could you maybe point me toward a particular thread? that would be awesome. thanks.
I work in a computer simulation lab. We have something like 50 AMD processors running code we compiled with the Intel Fortran Compiler and it's been working great for the past two years now. We have a mixture of Athlons and Opterons. Everything is workin' great. No problems. We used to use the Portland Group's fortran compiler until we found that the Intel Fortran Compiler generates faster executables. This was the case for both 32 and 64 bit executables. Started out with Portland Group, switched to Intel...and it works great for us!