| IDL 5.3 Performance ? [message #18838] |
Tue, 08 February 2000 00:00  |
Richard Tyc
Messages: 69
Registered: June 1999
|
Member |
|
|
IDL speed gurus:.
I just received my new Dell machine last week. It's a top of the line dual
processor Pentium which should be blistering fast. I promptly began to do
some speed tests using the idlspec2 from JD Smith at Cornell (results have
been sent...) and also some app specific tests using my medical image
application for which we bought the machine. Needless to say, I am not
impressed with the performance so far but am confused at what the problem
is. I am leaning toward saying its IDL 5.3 if this is possible.
The current machine in question is a Dell Precision Workstation 420, dual
Pentium III 733 MHz, 512 Mb Rambus RDRAM Memory on NT4 SP5. It uses a
relatively low-end graphics card, a Matrox G400 Max. I tend to think this
may be where the problem lies. It had IDL 5.3 installed
I compared the performance with a previous machine I got which is now in the
hands of our mechanical engineers running Autodesks Mechanical Desktop.
It was a Dell Precision Workstation 410, dual 700 MHz Pentium III, 1024 Mb
SDRAM, with a screaming fast Wildcat 4000 graphics card. It had IDL 5.2.1
installed.
Anyway, the tests in question should really be exploiting the CPU
performance so I thought it was irrelevant the older Dell had the high end
graphics card. I noticed the TIME_TEST3 performance was alot worse. For
example running an empty for loop 2000000 times took 0.07799 units on the
700 MHz vs 0.172 on the new 733 MHz Dell.
My app also takes almost twice as long on the current Dell and most of the
work is number crunching and displaying rendered volumes (IDLgrvolume) which
should not take advantage of high end graphics cards like the wildcat but
rather CPU performance because it uses a software ray tracing technique.
So, the main difference seems to be the old machine had IDL5.2.1 and the new
IDL5.3 and I know from SPEC benchmarks the new Dell using the 733 Mhz
Pentium, the Rambus memory etc IS faster. So, is it possible IDL 5.3 may
run applications/benchmarks slower ? It's hard to believe and I think other
factors are at play but its odd even the CPU speed tested in idlspec2 is
slower.
Q. for JD Smith: is the is2_53.sav test program alot different than the
version for IDL 5.2 ?
Any comments/ideas ?
Rich
|
|
|
|
| Re: IDL 5.3 Performance ? [message #18908 is a reply to message #18838] |
Thu, 10 February 2000 00:00  |
Richard Tyc
Messages: 69
Registered: June 1999
|
Member |
|
|
I did install IDL 5.2.1 and it did show a major improvement in some areas:
Platform: Dell 420 Dual Pentium III 733 MHz, 512 Mb RDRAM, Matrox G400 Max
IDL5.3 IDL5.2.1
TIME_TEST3 6.405 4.235
GRAPHICS_TIMES3 4.593 2.625 ? not sure why-same
card ??
IO test 0.578 0.594
Some significant differences
Invert 192^2 random matrix 0.266 0.07799
Generate 1000000 random nos 0.141 0.063
Smooth 512x512 float array 0.109 0.047
My own application ran 20% faster
Rich
David McClain <dmcclain@azstarnet.com> wrote in message
news:sa1qmsfchse80@corp.supernews.com...
> The only fair way to make this comparison is to install the old IDL 5.2x
on
> your new machine and run the same code. There are so many hardware
> variations with respect to bus width to memory, how many processors, how
> large and what mapping the secondary and primary caches offer, etc, etc. I
> would be interested to hear your results...
>
> David McClain, Sr. Scientist
> Raytheon Systems Co.
> Tucson, AZ
>
> Richard Tyc <richt@sbrc.umanitoba.ca> wrote in message
> news:87q5b0$g0k$1@canopus.cc.umanitoba.ca...
>> IDL speed gurus:.
>>
>> I just received my new Dell machine last week. It's a top of the line
dual
>> processor Pentium which should be blistering fast. I promptly began to
do
>> some speed tests using the idlspec2 from JD Smith at Cornell (results
have
>> been sent...) and also some app specific tests using my medical image
>> application for which we bought the machine. Needless to say, I am not
>> impressed with the performance so far but am confused at what the
problem
>> is. I am leaning toward saying its IDL 5.3 if this is possible.
>>
>> The current machine in question is a Dell Precision Workstation 420,
dual
>> Pentium III 733 MHz, 512 Mb Rambus RDRAM Memory on NT4 SP5. It uses a
>> relatively low-end graphics card, a Matrox G400 Max. I tend to think
this
>> may be where the problem lies. It had IDL 5.3 installed
>>
>> I compared the performance with a previous machine I got which is now in
> the
>> hands of our mechanical engineers running Autodesks Mechanical Desktop.
>> It was a Dell Precision Workstation 410, dual 700 MHz Pentium III, 1024
Mb
>> SDRAM, with a screaming fast Wildcat 4000 graphics card. It had IDL
5.2.1
>> installed.
>>
>> Anyway, the tests in question should really be exploiting the CPU
>> performance so I thought it was irrelevant the older Dell had the high
end
>> graphics card. I noticed the TIME_TEST3 performance was alot worse.
For
>> example running an empty for loop 2000000 times took 0.07799 units on
the
>> 700 MHz vs 0.172 on the new 733 MHz Dell.
>>
>> My app also takes almost twice as long on the current Dell and most of
the
>> work is number crunching and displaying rendered volumes (IDLgrvolume)
> which
>> should not take advantage of high end graphics cards like the wildcat
but
>> rather CPU performance because it uses a software ray tracing technique.
>>
>> So, the main difference seems to be the old machine had IDL5.2.1 and the
> new
>> IDL5.3 and I know from SPEC benchmarks the new Dell using the 733 Mhz
>> Pentium, the Rambus memory etc IS faster. So, is it possible IDL 5.3
may
>> run applications/benchmarks slower ? It's hard to believe and I think
> other
>> factors are at play but its odd even the CPU speed tested in idlspec2 is
>> slower.
>>
>> Q. for JD Smith: is the is2_53.sav test program alot different than the
>> version for IDL 5.2 ?
>>
>>
>>
>> Any comments/ideas ?
>>
>> Rich
>>
>>
>
>
|
|
|
|
| Re: IDL 5.3 Performance ? [message #18918 is a reply to message #18838] |
Wed, 09 February 2000 00:00 |
Karri Kaksonen
Messages: 27
Registered: October 1999
|
Junior Member |
|
|
On Wed, 9 Feb 2000, Roy E. Hansen wrote:
> In article <87q5b0$g0k$1@canopus.cc.umanitoba.ca>, "Richard Tyc" <richt@sbrc.umanitoba.ca> wrote:
>> IDL speed gurus:.
>
> Hence, the 1D FFT is a factor of 2 slower in 5.3 than in 5.2.1
This sounds _very_ bad. Especially since they announced that this
was the last release for the Alpha processor running Windows NT.
It kind of takes away all the advantage of having a fast CPU in
the first place...
--
Karri
|
|
|
|
| Re: IDL 5.3 Performance ? [message #18919 is a reply to message #18838] |
Wed, 09 February 2000 00:00 |
John-David T. Smith
Messages: 384
Registered: January 2000
|
Senior Member |
|
|
Richard Tyc wrote:
> Q. for JD Smith: is the is2_53.sav test program alot different than the
> version for IDL 5.2 ?
>
This is exactly the same program, just "recompiled" for 5.3. It is very basic,
and does little more than a bunch of housekeeping and stuff to auto-email the
results. The actual testing is being done with:
time_test3,/NOFILEIO
graphics_times3
and the extracted IO test from time_test3. When I compare time_test.pro from
IDL5.3 to IDL5.2.1, the *only* differences are the addition of the new compiler
directives:
COMPILE_OPT hidden
to hide the compilation. The file time_test3.pro differs only in the version
number listed at the top (1.1 vs. 1.1.2.1), but otherwise has *no* differences.
The story is the same with graphics_times*.pro.
I rechecked the excerpted IO test, and it still checks out.
> Any comments/ideas ?
The limited 5.3 testing I've done does indicate a slowdown, although these tests
were confined to a few machines only. For instance, on the chart on the
IDLSpecII page, you can see I've entered my machine (Dell Dimension 400Mhz PII)
twice, and the IO was hardest hit (by far). To really see what the slow down
is, or if it is general, we need people to recontribute 5.3 idlspec2 entries
from machines they had sent in before, making sure to use exactly the same
descriptions and text so that i can sort those out together. Just take a look
at http://www.astro.cornell.edu/idlspec/is2_sorted.html to find your old entry.
In the interest of a little more certainty, I did a side by side time_test3 with
v5.2.1 and v5.3. I got:
5.2.1 5.3
1 0.21400 1 0.20214
2 0.23642 2 0.23430
3 0.25929 3 0.26889
4 0.24083 4 0.25496
5 0.14772 5 0.19550
6 1.0328 6 1.0214
7 0.50235 7 0.50162
8 0.46114 8 0.46131
9 0.44494 9 0.44587
10 0.87304 10 0.87744
11 0.67848 11 0.67851
12 0.33605 12 0.24401
13 0.40520 13 0.21952
14 0.27451 14 0.085009
15 0.26790 15 0.25382
16 0.36092 16 0.41647
17 1.4182 17 1.4306
18 0.27534 18 0.26124
19 0.45899 19 0.46114
20 0.46100 20 0.45048
21 0.32254 21 0.32184
22 0.13546 22 0.13159
23 0.90981 23 0.91101
10.7171=Total 10.3288=Total
So this doesn't really support a slowdown. Test 5 "Mult 512 by 512 byte by
constant and store, 30 times", is marginally slower. But tests 12,13, and
especially 14 (LU decomposition) are faster. Even I/O seems similar, which
probably means I was comparing apples and oranges, since my original result
(around .45 seconds) was obtained using a different version of Linux and the C
libraries (I was forced to upgrade in order to use 5.3). When compared side by
side in this way, results really are quite similar. The geometric means were
0.35871513 (v5.3) and 0.38800894 (v5.2.1), indicating the *slight* speedup of
5.3 on specific tests, most notably LU decomp.
IDLSpecMark's were 0.873758 (v5.3) versus 0.891246 (v5.2.1), but had a good deal
of variability on subsequent invocations, and were certainly consistent with
similar performance. My suggestion is to look to the operating system, or try
installing 5.2.1 on your new machine and comparing that way.
JD
--
J.D. Smith |*| WORK: (607) 255-5842
Cornell University Dept. of Astronomy |*| (607) 255-6263
304 Space Sciences Bldg. |*| FAX: (607) 255-5875
Ithaca, NY 14853 |*|
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
