Replies: 3 comments 2 replies
-
In the Binary Contents compare method, single threading was faster in my environment, so I made it single threaded. |
Beta Was this translation helpful? Give feedback.
-
No, none. |
Beta Was this translation helpful? Give feedback.
-
I repeated the test with a subset of the files from first test (now I have about 2x10 GB of files in two folders that I compare, it was 2x100GB in first test) and now I get different results. Binary Content is fastest:
Strange... |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
When comparing two folders I noticed that the default "Full Contents" method is about twice as fast the "Binary Contents" method, even if the docs say: This method is faster than the Quick contents compare method (which in turn is: This method is faster than Full Contents, because it does not load the files.)
The reason is probably that the former use multiple CPU cores/threads, while the binary mode uses one.
For binary contents method Task manager shows almost no load on my SSD, while about 33% load on the CPU (i7-9850H 6core/12thread) - mostly by MsMpEng.exe
Some quick benchmarks (comparing two big folders, that happen to be identical):
Is there a reason for not using multiple threads for binary contents method? It may not give more speed on HDD, but today we use SSD. Also with a lot of RAM, there is a good chance the files are in cache, even when using a HDD.
PS: I used WinMergePortable-2.16.30 64 bit version
Beta Was this translation helpful? Give feedback.
All reactions