|Message: Re: speed differences between 4.4.0 and 4.3.2 with examples...||Not Logged In (login)|
Click on the Forum title, e.g. on the "Forums by Category" page, to read a sequence of postings to the Forum and its threads all in one page. If you are only interested in one thread or the thread following a specific posting, click the thread or the posting, which takes you to a smaller page, which contains only the part you are interested in and may be easier to navigate.
Messages are "chained" if there are only replies at the first level, i.e. 1/1.html, 1/1/1.html etc. In case of "chained" messages the message number is replaced by the icon and there is no indentation.
Inline: Display the subject line only or also the text of the posting(s); for the choice "All" the "Outline" choices are switched off.
|1||0||1||no text / full text of posting|
|2||1||All||text for level 1 only / text for All postings|
Outline: Choose the depth of the posting thread, successive toggle controls provide increasing detail.
|1||2||1||2 levels / 1 level (original posting)|
|2||3||2||3 levels / 2 levels|
|3||3||All||3 levels / all levels (all postings)|
Hi Frank, several reasons explain the performance degradation you noticed from release 3.2 to release 4.0. Definitely, one thing that will already help and you should do, is to apply patch-01 to release 4.0 that you can download from the web. However, there're other issues which make 4.0 slower than 3.2 and we are tracing them all: - 4.0 uses CLHEP-1.7, we experienced a serious performance degradation in handling 3/4-vectors in this new version of CLHEP. These issues have been reported to CLHEP editors and they're working to fix this; - if you're testing on a Linux box, the setup for Linux-g++ has changed since 3.2 to adopt ANSI C++ setup. We noticed a slight penalty for this as well; - 4.0 includes a new mechanism for handling touchables in the geometry, which are now reference-counted. We know a 5% penalty (in the worst cases) is due to this; - 4.0 EM physics includes a more precise algorithm for multiple scattering, which in some cases may lead up to 10% performance loss; - a new mechanism for handling cuts in range is implemented in 4.0, that we identified is not yet optimally tuned. We're currently working for cleaning this penalty and even optimising the overall performance. We plan to distribute a second patch by end of February which also will deal with most of these issues. Cheers, Gabriele
|Inline Depth:||Outline Depth:||Add message:|