Slow performance on large text files

[glenndrayton]

In the past we have used UltraCompare for two-way text comparison. We upgraded to BeyondCompare because it is nicely integrated into Code Co-op and it performs really well on small source code files. But one other use we have is comparing large text files e.g. 30-200 MB in size. I am disappointed in the slow performance of BC on such cases. For example, comparing two 30MB text files takes 5 seconds on UC, but takes so long in BC we have to kill the program.

Is there some advice I can get on how to improve BC on such large cases? There seems to be a fundamental difference in the way BC and UC do comparisions, with UC being slow on small files compared to BC, but BC failing to handle large files. Is there some kind of "large file mode" for BC or is this a known limitation i.e. do we have to keep using two comparison tools?

[chrroe]

Hi Glenn!

From my experiance also during the beta phase of BC3 I know that the syntax-highlightning feature can slow down the file handling in text compares.
What type are your files exactly? If the filename-extensions match one of the definitions unfer the "Tools -> File Formats..." dialog you could try to disable the corresponding checkbox.
Then retest the comparision of the large files. Does that improve speed?

Another approach could be to modify the parameters for the Line Alignment algorithm. In the session settings for text compares there is a dialogbox called "Alignment". Here you can modify the parameters for line alignement from fast to slow and vice versa.

Hope this helps ...


Bye
Christoph

[Aaron]

Hello,

Does BC successfully complete if you let it sit for several minutes on a 30 meg file? Is it loading the file and you can see the text or pre-processing?

What types of files are you loading?

What is the command line you are using/how you are launching BC?