| Re: Q: IDL benchmarks [message #5805 is a reply to message #5802] |
Fri, 23 February 1996 00:00   |
szoonem
Messages: 16
Registered: June 1995
|
Junior Member |
|
|
In article <4gkdde$9nm@reznor.larc.nasa.gov>, zawodny@arbd0.larc.nasa.gov
(Joseph M Zawodny) wrote:
>
> I agree that the individual test times should be kept.
>
> I have a file of a few systems that I have run TIME_TEST on, ...
> ...
For those who are interested, here are the numbers for a PowerMac 7100/80
with 40MB of RAM running IDL 4.0. Virtual memory is turned off.
1 0.716667 Empty For loop, 1 million times
2 1.63333 Call empty procedure (1 param) 100,000 times
3 0.666667 Add 100,000 integer scalars and store
4 0.766667 25,000 scalar loops each of 5 ops, 2 =, 1 if)
5 0.316667 Mult 512 by 512 byte by constant and store, 10 times
6 0.300000 Shift 512 by 512 byte and store, 10 times
7 0.300000 Add constant to 512 x 512 byte array and store, 10 times
8 0.300000 Add two 512 by 512 byte images and store, 10 times
9 0.466667 Mult 512 by 512 floating by constant and store, 10 times
10 1.10000 Add constant to 512 x 512 floating and store, 10 times
11 0.966667 Add two 512 by 512 floating images and store, 10 times
12 0.183333 Invert a 100 by 100 random matrix
13 0.766667 Transpose 256 x 256 byte, FOR loops
14 0.200000 Transpose 256 x 256 byte, row and column ops
15 0.0666666 Transpose 256 x 256 byte, transpose function
16 1.76667 Log of 100,000 numbers, FOR loop
17 0.350000 Log of 100,000 numbers, vector ops
18 1.70000 Add two 100000 element floating vectors, FOR loop
19 0.0666667 Add two 100000 element floating vectors, vector op
20 0.350000 65536 point real to complex FFT
21 0.216667 Smooth 512 by 512 byte array, 5x5 boxcar
22 0.200000 Smooth 512 by 512 floating array, 5x5 boxcar
23 5.30000 Write and read 10 512 by 512 byte arrays
18.7000=Total Time, 0.46858267=Geometric mean, 23 tests.
I feel I have to add a comment here. Even though I am very happy with
running IDL on my Mac, it has some drawbacks that could be deadly. For
instance, we have to turn virtual memory on if we need to deal with very
large arrays. This makes every thing extremely slow.
------------------------------------------------------------ -----
| Saeid Zoonematkermani | E-Mail: szoonem@astro.sunysb.edu |
| Earth and Space Sciences | Voice: (516)632-8237 |
| State University of New York | Fax: (516)632-8742 |
| Stony Brook, NY 11794-2100 | |
------------------------------------------------------------ -----
| Home Page --> http://ozone.ess.sunysb.edu/ |
------------------------------------------------------------ -----
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
