Sign in Join

Live chat for Free Pascal, Lazarus, Delphi and every Pascal programmer.

Mailing lists / fpc-devel / 2020-07

just for information

4 messages · 4 people ·

Reply by email Subscribe to this list Your subscription

Read-only archive. Use j and k to move between messages.

Samuel Herzog

Hello List-Members,
first of all thank you all for the FPC 3.2 release! I am the maintainer of sourceforge.net/projects/scimarkfordelphi/

|
|
| |
Scimark for Delphi Lazarus Free Pascal

Download Scimark for Delphi Lazarus Free Pascal for free. Scimark for Delphi/Lazarus/Free Pascal. Benchmark for ... |

|

|

I have updated with latest versions FPC 3.2 and Delphi 10.4 Sydney. The results you can either verify yourself (by downloading sources or executables)or look at the attached screenshot. "Composite score" ist the average of the 5 different "benchmarks". I am a little irritated about the scores of "sparse matmult" and "LU". Have a nice sunday.
Sam

Reply by email

Jonas Maebe
to Samuel Herzog
Samuel Herzog wrote 7 lines
The results you can either verify yourself (by downloading sources or executables) or look at the attached screenshot. "Composite score" ist the average of the 5 different "benchmarks". I am a little irritated about the scores of "sparse matmult" and "LU".

While the results are certainly interesting, keep in mind that SciMark (and LU in particular; I don't know about sparse matmult) is generally not a very good benchmark in terms of being representative for real world programs, especially as far as non-JIT compilers are concerned. See e.g.
https://www.phoronix.com/forums/forum/software/programming-compilers/870805-gcc-6-1-vs-llvm-clang-3-9-compiler-performance?p=871540#post871540 (last post on the page) by Honza Hubička, a gcc compiler developer (*).

Of course, compiler developers still tend to focus on it because it is used a lot to compare compiler performance and nobody likes to look bad. And there are of course still optimisations that will improve both those benchmarks and real world programs.

Jonas

(*) who has a very interesting blog on compiler optimisations, btw:
http://hubicka.blogspot.com

Reply by email

J. Gareth Moreton
to Jonas Maebe

I'm always up for finding improvements in the built binaries so long as the adjustments to the compiler are reasonable.  I don't have Delphi 10.4 Sydney or anything close to that version unfortunately, so I cannot compare the disassembly of each binary produced, but I'll see what I can do someday.

Gareth aka. Kit

Jonas Maebe wrote 32 lines
On 2020-07-12 09:37, Samuel Herzog via fpc-devel wrote: > The results you can either verify yourself (by downloading sources or > executables) > or look at the attached screenshot. > > "Composite score" ist the average of the 5 different "benchmarks". > > I am a little irritated about the scores of "sparse matmult" and "LU". While the results are certainly interesting, keep in mind that SciMark (and LU in particular; I don't know about sparse matmult) is generally not a very good benchmark in terms of being representative for real world programs, especially as far as non-JIT compilers are concerned. See e.g. https://www.phoronix.com/forums/forum/software/programming-compilers/870805-gcc-6-1-vs-llvm-clang-3-9-compiler-performance?p=871540#post871540 (last post on the page) by Honza Hubička, a gcc compiler developer (*). Of course, compiler developers still tend to focus on it because it is used a lot to compare compiler performance and nobody likes to look bad. And there are of course still optimisations that will improve both those benchmarks and real world programs. Jonas (*) who has a very interesting blog on compiler optimisations, btw: http://hubicka.blogspot.com _______________________________________________ fpc-devel maillist  -  https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus

Reply by email

Sven Barth
to Samuel Herzog: first of all thank you all for the FPC 3.2 release! I am the maintainer of…
Samuel Herzog wrote 31 lines
Hello List-Members, first of all thank you all for the FPC 3.2 release! I am the maintainer of sourceforge.net/projects/scimarkfordelphi/ <https://sourceforge.net/projects/scimarkfordelphi/> Scimark for Delphi Lazarus Free Pascal Download Scimark for Delphi Lazarus Free Pascal for free. Scimark for Delphi/Lazarus/Free Pascal. Benchmark for ... <https://sourceforge.net/projects/scimarkfordelphi/> I have updated with latest versions FPC 3.2 and Delphi 10.4 Sydney. The results you can either verify yourself (by downloading sources or executables) or look at the attached screenshot. "Composite score" ist the average of the 5 different "benchmarks". I am a little irritated about the scores of "sparse matmult" and "LU". Have a nice sunday.

Do you also have a comparison between 3.2.0 and 3.0.4? Just to see what changed.

Also do you know whether the calculated results are *correct*? Just to know whether e.g. the "Sparse matmult" is only fast, because the generated code leaves out essential steps...

Regards,
Sven

Reply by email