Apart Another interesting thing to note is how the bandwidth is affected by the ARM core frequency. If it scaled linearly, we should have an improvement of 62.5% from the increase over Axel Lite in absolute terms800 to 1300 MHz. The average bandwidth at 800 MHz is 5761 MB/s. At 1300 MHz, it is noteworthy that Mito8M exhibits a significant improvement in terms of efficiency too6935 MB/s. Therefore, as shown in the above tablesincrease is 20. This is especially true in 4%. With regard to STREAM benchmark, the case of achieved bandwidth does not scale linearly with ARM core frequency set to 1300 MHz.

Another interesting thing to note is how the bandwidth is affected by the ARM core frequencyPlease see [https://www. If it scaled linearly, we should have an improvement of 62cs.5% from 800 to 1300 MHzvirginia. The average bandwitdh at 800 MHz is 5761 MBedu/s. At 1300 MHz, it is 6935 MBstream/s. Therefore, the increase is 20.4%. With regard to this page] for more details about STREAM benchmark, the achieved bandwidth does not scale linearly with ARM core frequency.

For what regards the memory bandwidth, LMbench provides many results organized in different categories. For the sake of simplicity, the following tables details just a couple of categories. The full results are available for download [http://mirror.dave.eu/axelmito/SBCXMito8M/lmbench-TNMito8M.0-006800MHz.txt here (ARM core frequency set to 800 MHz)] and [http://mirror.dave.eu/mito/Mito8M/lmbench-axellite-iMito8M.MX6Q0-996MHz1300MHz.txt here(ARM core frequency set to 1300 MHz)].

!rowspan="2" |Buffer size!colspan="2" |Bandwitdth

|28611553|2521

|30471567|2546

|30651575|2560

|30771575|2564

|30811577|2564

|30721577|2567

|13091528|2490

|9021531|2494

|7871547|2530

|7751552|2526

|7491514|2518

|6871318|2181

|6421430|2148

|6291420|2108

|6301423|2038

|6321420|2116

|6311365|2117

|6321393|2035

|6331382|2035

|6341372|2050

|6341367|-|1GB|6331998

!rowspan="2" |Buffer size!colspan="2" |Bandwitdth

|37242932|4771

|38483048|4956

|39023100|5046

|39403136|5097

|39583135|5101

|39573150|5120

|39642864|5127

|39673033|5071

|39673093|4886

|39562956|5056

|39473024|5054

|20973075|5092

|21543095|5116

|21143121|5118

|20823137|5120

|20843145|5121

|20853146|5120

|20933146|5125

|20863147|5123

|20893150|5124

|20873144|5125

|20883146|5124

The most There are some interesting results facts to consider stress:* Read and write bandwitdth are those that refer to not effected by the buffer sizes exceeding 1MBsize.* They are significantly affected by the ARM core frequency. For instance, which is the size improvement of the L2 cache. Approximately, read write bandwidth is 630MB/s (7.8about 62% efficiencywhen the buffer is 1GB), while write bandwidth is 2080 MB/s (25.7% efficiency). These numbers are significantly different that practically the ones provided by STREAM. This confirms once again that such results are strongly dependent on the implementation same of the test used to determine the bandwidthincrease in frequency.