TI processors can provide a power consumption estimation spreadsheet for corresponding devices, which is based on measured and simulation data to estimate power consumption. Although the operating power consumption of SOC depends on various factors such as electrical parameters, process variations, environmental conditions, and the use cases during processor operation, it can still serve as a rough reference for SOC power consumption. Therefore, software measures such as frequency scaling and reducing loading can be evaluated using this tool as a general reference. The temperature and power consumption of SOC are positively correlated, the higher power consumption leads to faster temperature increases. Hence, the primary goal of software thermal management is to reduce SOC power consumption.

Table 3-1 uses an AM62A processor as an example, and the corresponding tool can be downloaded from here. This method is also applicable to other TI processors. The results of Table 3-1 are statistically analyzed under a junction temperature of 125°C and with all cores operating at maximum frequency.

Results in Table 3-2 are conducted under a junction temperature of 125°C and all cores operating at 80% loading.

From the calculation results, the C7x core in the AM62A has the most significant impact on power consumption. Therefore, when designing the temperature control scheme of the system, after reaching the target temperature, prioritize reducing the frequency and loading of the C7x core.