Lahey Support
08-15-2003, 01:33 AM
Hi Michael,
thanks for your tests.
> Perfect agreement (to the last digit) with approved answers on both
> machines.
With a Monte Carlo code I would expect that there at least COULD be
differences, not only between runs on different hardware systems but also
between two runs on the same system, because you are working with
random numbers. Very fine that this is not the case.
> if it is possible to wrap up the executable and the offending input file
> into a simple package that I could try on some of the test machines I have
> here, I'd be very interested, and would try and get trial access to a P-4
> for a test.
Thanks for your offer to run my code on your systems. My problem is, that
I've hardly the time to prepare the code and input files or to do my own
tests
within the next weeks (several deadlines for deliverables and project
reports
in September...). I suggest to do this in October, if you agree.
> Otherwise, could you please keep us informed - I'd like to know what you
> find, since I am on the verge of buying my first AMD machine, but with
some
> trepidation.
I've been running an AMD Athlon 800 MHz without any problems
for 1 1/2 years now. In January I bought a new Athlon 1200 MHz systems
which runs well for four month. Then it "started to hang" in intervals
between
some minutes and some hours - worthless for simulation runs over
some days! Some repairs and hardware upgrades - and now I have a
system with 1333 MHz (133 MHz FSB, AK12 board) which still "hangs"
twice a day. I decided to give it back to the trader and to change to the
good old Intel system.
By the way, benchmark tests in several journals say that Athlon/1333MHz
and P4/1700MHz systems are nearly identical in speed. May be this is the
case for office applications, games, and so on. My experience with
my simulation model (FORTRAN code, Lf95 compiler) is that the
P4/1800MHz system performs by about a factor 2 better than the
Athlon/1200 MHz system:
System CPU time
Athlon 800MHz 100 %
Athlon 1200 MHz ~ 70%
P4 1800 MHz ~ 32%
It seems that the standard benchmarks say nothing about the performance
of our typical applications. So I'm really lucky that I decided to give the
Athlon system back to the trader and to try Intel again!
Though I have to find out the reason for the differences in the simulation
results...
Klaus
METCON Umweltmeteorologische Beratung Dr. Klaus Bigalke
Jappopweg 9h, D-25421 Pinneberg
Tel. 04101/6938-56 Fax: 04101/6938-57 Mobil: 0171/6249543
Email: [address removed] Web: www.metcon-umb.de
----------------------------------------------------------
To unsubscribe, send to [address removed] the following
as the first and only line of the message body:
unsubscribe fortran
----------------------------------------------------------
thanks for your tests.
> Perfect agreement (to the last digit) with approved answers on both
> machines.
With a Monte Carlo code I would expect that there at least COULD be
differences, not only between runs on different hardware systems but also
between two runs on the same system, because you are working with
random numbers. Very fine that this is not the case.
> if it is possible to wrap up the executable and the offending input file
> into a simple package that I could try on some of the test machines I have
> here, I'd be very interested, and would try and get trial access to a P-4
> for a test.
Thanks for your offer to run my code on your systems. My problem is, that
I've hardly the time to prepare the code and input files or to do my own
tests
within the next weeks (several deadlines for deliverables and project
reports
in September...). I suggest to do this in October, if you agree.
> Otherwise, could you please keep us informed - I'd like to know what you
> find, since I am on the verge of buying my first AMD machine, but with
some
> trepidation.
I've been running an AMD Athlon 800 MHz without any problems
for 1 1/2 years now. In January I bought a new Athlon 1200 MHz systems
which runs well for four month. Then it "started to hang" in intervals
between
some minutes and some hours - worthless for simulation runs over
some days! Some repairs and hardware upgrades - and now I have a
system with 1333 MHz (133 MHz FSB, AK12 board) which still "hangs"
twice a day. I decided to give it back to the trader and to change to the
good old Intel system.
By the way, benchmark tests in several journals say that Athlon/1333MHz
and P4/1700MHz systems are nearly identical in speed. May be this is the
case for office applications, games, and so on. My experience with
my simulation model (FORTRAN code, Lf95 compiler) is that the
P4/1800MHz system performs by about a factor 2 better than the
Athlon/1200 MHz system:
System CPU time
Athlon 800MHz 100 %
Athlon 1200 MHz ~ 70%
P4 1800 MHz ~ 32%
It seems that the standard benchmarks say nothing about the performance
of our typical applications. So I'm really lucky that I decided to give the
Athlon system back to the trader and to try Intel again!
Though I have to find out the reason for the differences in the simulation
results...
Klaus
METCON Umweltmeteorologische Beratung Dr. Klaus Bigalke
Jappopweg 9h, D-25421 Pinneberg
Tel. 04101/6938-56 Fax: 04101/6938-57 Mobil: 0171/6249543
Email: [address removed] Web: www.metcon-umb.de
----------------------------------------------------------
To unsubscribe, send to [address removed] the following
as the first and only line of the message body:
unsubscribe fortran
----------------------------------------------------------