12,065,060 members (46,070 online)
Rate this:
See more:
please give me some parameter which is used to evaluate the performance of compression algorithm

tell me how it is used to valuate...

thanks
Posted 11-May-12 4:24am
PKriyshnA1.7K

Rate this:

## Solution 2

Given a set of suitable input data, there are essentially these paramenters: the `compression ratio` (size_after_compression/size_before_compression) and both the `compression` and `decompression speed` (bytes/sec).
v3
SAKryukov 11-May-12 18:02pm

Important note: in this answer, last '/' character is not a division operator, because otherwise 1) slow decompression speed would mean good algorithm; 2) infinitely slow decompression (thread never completes, due to a programming bug, zero decompression speed) would cause division by zero; in floating-point arithmetic it would produce quality equal to +Infinity. :-)
My 5. :-)
--SA
CPallini 12-May-12 5:00am

It was indeed confusing, I've changed that.
BTW Thank you.
SAKryukov 14-May-12 15:31pm

You are welcome.
I did not actually think it was confusing; I just tried to joke about it, picturing some stupid reader who could possibly think that '/' could be a division operator :-)
However, this joke was inspired by the real-life experience with some inquirers here at CodeProject... :-)

Cheers,
--SA
Rate this:

## Solution 1

When it comes to loss-less compression (i.e. compressing a file for storage), then the main point is compressed file size divided by original file size, followed by the time required for compression.
Sometimes, a lossy compression is used, e.g. jpg for images, or mp3 for audio. Here, a measurement for loss is required, but that is harder to describe, because in the end the result is evaluated by a human user who has to see a nice picture or listen to "high-fidelity" music.

Top Experts
Last 24hrsThis month
 OriginalGriff 190 RDBurmon 100 ppolymorphe 50 CPallini 50 nv3 40
 Dave Kreskowiak 1,857 Richard MacCutchan 1,250 d@nish 980 OriginalGriff 974 CPallini 790