Intel(R) Memory Latency Checker - v3.11a

https://www.intel.cn/content/www/cn/zh/download/736633/intel-memory-latency-checker-intel-mlc.html

I7 8550U DDR4 （2400）
Intel(R) Memory Latency Checker - v3.11a
Measuring idle latencies for random access (in ns)...
                Numa node
Numa node            0
       0          99.9

Measuring Peak Injection Memory Bandwidths for the system
Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
Using all the threads from each core if Hyper-threading is enabled
Using traffic with the following read-write ratios
ALL Reads        :      18165.2
3:1 Reads-Writes :      19081.1
2:1 Reads-Writes :      19650.7
1:1 Reads-Writes :      21504.7
Stream-triad like:      18978.5

Measuring Memory Bandwidths between nodes within system
Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
Using all the threads from each core if Hyper-threading is enabled
Using Read-only traffic type
                Numa node
Numa node            0
       0        18398.0

Measuring Loaded Latencies for the system
Using all the threads from each core if Hyper-threading is enabled
Using Read-only traffic type
Inject  Latency Bandwidth
Delay   (ns)    MB/sec
==========================
 00000  210.20    18534.5
 00002  208.25    18752.5
 00008  181.10    20607.4
 00015  160.30    21556.4
 00050  139.76    21687.0
 00100   99.64    13378.6
 00200   91.80     8465.0
 00300   90.19     6143.8
 00400   88.64     4916.3
 00500   89.36     4136.9
 00700   87.92     3216.8
 01000   96.32     2439.4
 01300   96.44     2039.4
 01700   97.12     1719.4
 02500   96.47     1386.8
 03500   98.33     1170.2
 05000   96.55     1029.8
 09000   94.58      881.5
 20000   96.54      755.3

Measuring cache-to-cache transfer latency (in ns)...
Using small pages for allocating buffers
Local Socket L2->L2 HIT  latency        27.6
Local Socket L2->L2 HITM latency        32.0
PS C:\Users\qqk\Downloads\mlc_v3.11a\Windows>

I9 13900K DDR5 （4800） 节能模式
Intel(R) Memory Latency Checker - v3.11a
Measuring idle latencies for random access (in ns)...
                Numa node
Numa node            0
       0         121.0

Measuring Peak Injection Memory Bandwidths for the system
Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
Using all the threads from each core if Hyper-threading is enabled
Using traffic with the following read-write ratios
ALL Reads        :      61092.1
3:1 Reads-Writes :      63507.5
2:1 Reads-Writes :      62615.4
1:1 Reads-Writes :      61941.9
Stream-triad like:      64548.7

Measuring Memory Bandwidths between nodes within system
Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
Using all the threads from each core if Hyper-threading is enabled
Using Read-only traffic type
                Numa node
Numa node            0
       0        60765.4

Measuring Loaded Latencies for the system
Using all the threads from each core if Hyper-threading is enabled
Using Read-only traffic type
Inject  Latency Bandwidth
Delay   (ns)    MB/sec
==========================
 00000  236.68    60852.5
 00002  236.90    60928.9
 00008  233.37    60785.1
 00015  227.29    60682.0
 00050  225.22    60198.1
 00100  212.51    59559.3
 00200  155.69    50617.5
 00300  139.98    35770.7
 00400  133.78    28268.3
 00500  131.54    23443.8
 00700  129.80    17317.7
 01000  128.57    12756.4
 01300  127.17    10078.1
 01700  126.15     7965.9
 02500  124.79     5627.1
 03500  123.01     4241.3
 05000  122.65     3105.3
 09000  121.40     1991.6
 20000  120.87     1194.4

Measuring cache-to-cache transfer latency (in ns)...
Using small pages for allocating buffers
Local Socket L2->L2 HIT  latency        94.5
Local Socket L2->L2 HITM latency        95.7
PS C:\Users\Administrator\Desktop\mlc_v3.11a\Windows>

I9 13900K DDR5 （4800） 正常模式

Intel(R) Memory Latency Checker - v3.11a
Measuring idle latencies for random access (in ns)...
                Numa node
Numa node            0
       0          92.3

Measuring Peak Injection Memory Bandwidths for the system
Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
Using all the threads from each core if Hyper-threading is enabled
Using traffic with the following read-write ratios
ALL Reads        :      70005.1
3:1 Reads-Writes :      63722.5
2:1 Reads-Writes :      63148.3
1:1 Reads-Writes :      62593.9
Stream-triad like:      64811.4

Measuring Memory Bandwidths between nodes within system
Bandwidths are in MB/sec (1 MB/sec = 1,000,000 Bytes/sec)
Using all the threads from each core if Hyper-threading is enabled
Using Read-only traffic type
                Numa node
Numa node            0
       0        70346.1

Measuring Loaded Latencies for the system
Using all the threads from each core if Hyper-threading is enabled
Using Read-only traffic type
Inject  Latency Bandwidth
Delay   (ns)    MB/sec
==========================
 00000  343.26    70592.5
 00002  332.65    70518.2
 00008  321.12    70561.9
 00015  297.88    70745.1
 00050  277.32    70512.0
 00100  236.15    70515.1
 00200  232.57    70448.7
 00300  183.51    68394.1
 00400  127.50    57496.2
 00500  114.50    48865.4
 00700  103.32    37238.6
 01000  100.13    27009.7
 01300   97.86    21401.0
 01700   97.10    16765.1
 02500   95.44    11805.4
 03500   93.88     8709.0
 05000   92.58     6350.9
 09000   91.23     3875.3
 20000   90.59     2139.9

Measuring cache-to-cache transfer latency (in ns)...
Using small pages for allocating buffers
Local Socket L2->L2 HIT  latency        33.1
Local Socket L2->L2 HITM latency        33.8
PS C:\Users\Administrator\Desktop\mlc_v3.11a\Windows>