1 | <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
|
---|
2 | <HTML>
|
---|
3 | <HEAD>
|
---|
4 | <META http-equiv="Content-Type" content="text/html; charset=Windows-1252">
|
---|
5 | <TITLE>b (Benchmark) command</TITLE>
|
---|
6 | <LINK href="style.css" rel="stylesheet" type="text/css">
|
---|
7 | </HEAD>
|
---|
8 |
|
---|
9 | <BODY>
|
---|
10 |
|
---|
11 | <H1>b (Benchmark) command</H1>
|
---|
12 |
|
---|
13 | <P>Measures speed of the CPU and checks RAM for errors.</P>
|
---|
14 |
|
---|
15 | <H4>Syntax</H4>
|
---|
16 |
|
---|
17 | <PRE class="syntax">
|
---|
18 | b [number_of_iterations] [-mmt{N}] [-md{N}] [-mm={Method}]
|
---|
19 | </PRE>
|
---|
20 |
|
---|
21 | <P>There are two tests:<P>
|
---|
22 | <OL>
|
---|
23 | <LI>Compressing with LZMA method
|
---|
24 | <LI>Decompressing with LZMA method
|
---|
25 | </OL>
|
---|
26 |
|
---|
27 | <P>The benchmark shows a rating in MIPS (million instructions per second).
|
---|
28 | The rating value is calculated from the measured CPU speed and it
|
---|
29 | is normalized with results of Intel Core 2 CPU with multi-threading option
|
---|
30 | switched off. So if you have Intel Core 2 Duo,
|
---|
31 | rating values must be close to real CPU frequency.</P>
|
---|
32 |
|
---|
33 | <P>You can change the upper dictionary size to increase memory usage by -md{N} switch.
|
---|
34 | Also, you can change the number of threads by -mmt{N} switch.</P>
|
---|
35 |
|
---|
36 | <P>The <B>Dict</B> column shows dictionary size. For example, 21 means 2^21 = 2 MB.</P>
|
---|
37 |
|
---|
38 | <P>The <B>Usage</B> column shows the percentage of time the processor is working.
|
---|
39 | It's normalized for a one-thread load. For example, 180% CPU Usage for 2 threads
|
---|
40 | can mean that average CPU usage is about 90% for each thread.</P>
|
---|
41 |
|
---|
42 | <P>The <B>R / U</B> column shows the rating normalized for 100% of CPU usage.
|
---|
43 | That column shows the performance of one average CPU thread.</P>
|
---|
44 |
|
---|
45 | <P><B>Avr</B> shows averages for different dictionary sizes.</P>
|
---|
46 | <P><B>Tot</B> shows averages of the compression and decompression ratings.</P>
|
---|
47 |
|
---|
48 | <P>Compression speed and rating strongly depend on memory (RAM) latency.
|
---|
49 |
|
---|
50 | <P>Decompression speed and rating strongly depend on the integer performance of the CPU.
|
---|
51 | For example, the Intel Pentium 4 has big branch
|
---|
52 | misprediction penalty (which is an effect of its long pipeline) and pretty slow
|
---|
53 | multiply and shift operations. So, the Pentium 4 has pretty low decompressing ratings.</P>
|
---|
54 |
|
---|
55 | <P>You can run a CRC calculation benchmark by specifying -mm=crc.
|
---|
56 | That test shows the speed of CRC calculation in MB/s. The first column shows the size of the block.
|
---|
57 | The next column shows the speed of CRC calculation for one thread. The other columns are results
|
---|
58 | for multi-threaded CRC calculation.</P>
|
---|
59 |
|
---|
60 |
|
---|
61 | <H4>Examples</H4>
|
---|
62 |
|
---|
63 | <PRE class="example">
|
---|
64 | 7z b
|
---|
65 | </PRE>
|
---|
66 | runs benchmarking.
|
---|
67 |
|
---|
68 | <PRE class="example">
|
---|
69 | 7z b -mmt1 -md26
|
---|
70 | </PRE>
|
---|
71 | runs benchmarking with one thread and 64 MB dictionary.
|
---|
72 |
|
---|
73 | <PRE class="example">
|
---|
74 | 7z b 30
|
---|
75 | </PRE>
|
---|
76 | <P>runs benchmarking with default settings for 30 iterations.</P>
|
---|
77 |
|
---|
78 | </BODY>
|
---|
79 | </HTML>
|
---|