Method for detecting overclocking operational performance of GPU (graphic processing unit) and then grading GPU

Abstract

The invention discloses a method for detecting the overclocking operational performance of a GPU (graphic processing unit) and then grading the GPU, and the method is used for detecting the overclocking operational performance of the GPU and then grading the GPU, wherein the GPU has an overclocking detection frequency band, and the overclocking detection frequency band contains a plurality of overclocking detection frequencies. The method comprises the following steps: firstly, inputting an operating voltage to the GPU, and operating the GPU with a plurality of overclocking detection frequencies so as to obtain a frequency-standard power consumption value curve; then, inputting a boosting voltage to the GPU, and operating the GPU with a plurality of overclocking detection frequencies so as to obtain a frequency-boost power consumption value curve; and finally, carrying out comparison on the slopes of the frequency-standard power consumption value curve and the frequency-boost power consumption value curve, then grading the GPU according to the obtained comparison result.

Description

technical field [0001] The invention relates to a method for detecting and grading the overfrequency performance of a graphics processor, in particular to a method for detecting and grading the overfrequency performance of a graphics processor by means of boosting voltage. Background technique [0002] Since the circuit structure of the Graphic Processing Unit (GPU) is quite complicated, and many uncontrollable factors, such as temperature, dust, and pressure, often occur during the manufacturing process of the GPU, the manufacturer cannot determine the performance of the GPU. The final operating frequency can only provide an overclocking detection frequency band corresponding to the graphics processor. Therefore, the manufacturer must classify the graphics processor according to the performance of the graphics processor. [0003] In the prior art, the overclocking performance of graphics processors is classified by overclocking the graphics processors for several hours, tha...

AI Technical Summary

Problems solved by technology

It can be seen that when the overclocking detection frequency band of the graphics processor is larger, it will take more time for detection, which will increase the production cost of the manufacturer.

[0004] In addition, some graphics processors may be normal during the detection process for several hours, but if the detection is carried out for tens of hours, abnormalities may occur

It can be seen that in the conventional technology, the use of long-term detection methods cannot provide a stable and objective detection result, which in turn affects the goodwill

Images