A testing method and system for graphics card

Abstract

The present invention relates to the field related to graphics card test management, and discloses a testing method and system for graphics cards. By performing a staged balance test on the graphics card, the correlation between the power consumption of the graphics card and the heat dissipation of a fan is evaluated, and then the correlation can be compared with the factory scheduling to determine the heat dissipation capacity deviation of the graphics card due to individual differences in the graphics card and the specific usage environment of the current user. Then, separate scheduling management can be performed on separate graphics cards, so that different users in different usage environments can always keep the graphics card under the optimal scheduling plan, ensure the management stability of the graphics card temperature, and avoid the occurrence of situations such as overheating of the graphics card due to the large difference between the environment and individual factors and the experimental environment of the factory scheduling plan.

Description

Technical Field

[0001] The present invention relates to the field related to graphics card test management, and in particular to a testing method and system for graphics cards. Background Art

[0002] A graphics card is a hardware device specifically used to process graphic data, including rendering and decoding of 2D and 3D images and videos, as well as general computing tasks. A graphics card usually consists of two parts: a computing body and a heat dissipation structure. The heat dissipation part is used to dissipate the heat of the main body and is also an important structure that is essential for the operation of the graphics card.

[0003] The graphics card heat dissipation scheduling scheme in the prior art is usually fixed based on the manufacturer's factory policy. The factory policy is experimented and formulated according to the standard specifications in the graphics card manufacturing, and has no correlation with the individual differences of the graphics card itself and the differences in the user's usage environment. However, in actual use, different graphics cards used by different users have individual differences generated during production, resulting in the scheduling scheme based on the factory policy cannot provide the best heat dissipation management service for the graphics card, and the use effect is poor. Summary of the invention

[0004] The object of the present invention is to provide a testing method and system for a graphics card to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A testing method for a graphics card, comprising:

[0007] Control the maximum heat dissipation output, and perform a power consumption upper limit test on the graphics card based on a preset load program to obtain the maximum power consumption of the current graphics card. The load program is used to schedule the graphics card core and video memory based on the test requirements and control them to run at a certain occupancy rate;

[0008] The maximum power consumption is tested and selected at a preset power consumption test interval to obtain a power consumption test ladder including a plurality of power consumption test nodes, the power consumption test ladder is tested step by step based on a load program, and the temperature of the graphics card is kept in a balanced state through fan control to obtain the balanced state test data corresponding to the array;

[0009] Performing individual evaluation based on the equilibrium test data and the corresponding test power consumption to obtain the offset coefficient of the current graphics card in the current environment, wherein the offset coefficient is used to characterize the staged value of the effectiveness of the heat dissipation capability of the specific graphics card in the specific environment;

[0010] The preset factory scheduling is offset-adjusted according to the offset coefficient to obtain a personalized scheduling solution, where the personalized scheduling solution is used to characterize a heat dissipation scheduling solution for a graphics card under different power consumption.

[0011] As a further solution of the present invention: the step of controlling the temperature of the graphics card to be in a balanced state by fan control to obtain the balanced state test data corresponding to the array specifically includes:

[0012] Based on the preset temperature control threshold, if the test temperature of the current power consumption test node is less than the temperature control threshold, wait for the temperature to reach the temperature control threshold, and control the fan speed to make the temperature constant at the temperature control threshold, and record the fan speed and graphics card power consumption in the equilibrium state, and record them as equilibrium state test data;

[0013] If the test temperature of the current power consumption test node is greater than the temperature control threshold, the residual temperature difference between the upper limit temperature wall of the current graphics card and the temperature control threshold is calculated, and the residual temperature difference is divided based on the remaining number of power consumption test nodes to obtain multiple balance nodes; and based on the balance node corresponding to the current power consumption test node, the fan speed is controlled to make the temperature constant, and the fan speed and graphics card power consumption in the equilibrium state are recorded, and recorded as equilibrium test data.

[0014] As a further solution of the present invention: the step of performing individual evaluation based on the equilibrium test data and the corresponding test power consumption to obtain the offset coefficient of the current graphics card in the current environment specifically includes:

[0015] The heat output of the fan per unit time is evaluated based on the equilibrium test data of multiple power consumption test nodes. The heat output is determined based on the temperature difference and is proportional to the heat generation power. The temperature difference is characterized by:

[0016] ;

[0017] in, Used to characterize heat generation power, That is, the heat accumulation time, Characterizes the mass of an object, Characterizes the specific heat capacity of an object;

[0018] The corresponding airflow velocity of the fan is calculated based on the equilibrium test data of multiple power consumption test nodes. The airflow velocity is characterized by the ratio of the airflow volume to the effective area of ​​the fan outlet, that is:

[0019] ;

[0020] in, is a constant factor, Indicates the fan speed. Indicates the fan diameter, Indicates the effective area of ​​the fan outlet;

[0021] Determine the fan viscosity of the current graphics card in the current environment, characterized by:

[0022] ;

[0023] Among them, the , , They are used to characterize the temperature difference and airflow velocity of adjacent test nodes respectively;

[0024] Function fitting is performed on the viscosity of several power consumption test nodes to obtain the offset coefficient fitting function of the graphics card individual in the current environment.

[0025] As a further solution of the present invention: it also includes a basic personality judgment step of an individual, specifically including:

[0026] The offset coefficient fitting function is judged to obtain a constant function segment of the offset coefficient fitting function, wherein the constant function segment is used to characterize the fan viscosity when the fan is at a low speed and the heat conduction effect is not affected by the environment and the structure on the airflow, that is, the fan can increase in proportion to the power consumption of the graphics card and completely conduct the additional heat generated by the increased power in the same amount;

[0027] The fan speed of the constant function segment is compared and calculated with the factory scheduling to obtain the basic personality of the current graphics card, and the basic personality is used to characterize the machining accuracy deviation value of the physical structure of the graphics card.

[0028] As a further solution of the present invention: it also includes the steps of:

[0029] When it is detected that the initial temperature of the graphics card has dropped, the fan is scheduled to dive based on a personalized scheduling plan. The scheduling dive is used to indicate that when the fan reaches the scheduling speed, the fan speed is slowly reduced so that the graphics card temperature reaches the set value under the corresponding power consumption in the personalized scheduling plan.

[0030] An embodiment of the present invention is to provide a testing system for a graphics card, comprising:

[0031] An individual limit test module is used to control the maximum output of heat dissipation and perform a power consumption upper limit test on the graphics card based on a preset load program to obtain the maximum power consumption of the current graphics card. The load program is used to schedule the graphics card core and video memory based on the test requirements and control them to run at a certain occupancy rate;

[0032] A balanced grouping test module is used to test the maximum power consumption at a preset power consumption test interval, obtain a power consumption test ladder including a plurality of power consumption test nodes, test the power consumption test ladder step by step based on a load program, and make the temperature of the graphics card in a balanced state through fan control, so as to obtain the balanced state test data corresponding to the array;

[0033] An individual offset evaluation module, used to perform individual evaluation based on the equilibrium test data and the corresponding test power consumption, so as to obtain an offset coefficient of the current graphics card in the current environment, wherein the offset coefficient is used to characterize a staged value of the effectiveness of the heat dissipation capability of a specific graphics card in a specific environment;

[0034] The personalized scheduling management module is used to perform an offset adjustment on the preset factory scheduling according to the offset coefficient to obtain a personalized scheduling solution, wherein the personalized scheduling solution is used to characterize the heat dissipation scheduling solution of the graphics card under different power consumption.

[0035] As a further solution of the present invention: the balanced grouping test module includes:

[0036] The heat dissipation priority test unit is used to make a judgment based on a preset temperature control threshold. If the test temperature of the current power consumption test node is less than the temperature control threshold, the unit waits for the temperature to reach the temperature control threshold, controls the fan speed to make the temperature constant at the temperature control threshold, and records the fan speed and graphics card power consumption in the equilibrium state, which are recorded as equilibrium state test data.

[0037] The temperature priority test unit is used to calculate the residual temperature difference between the upper temperature wall of the current graphics card and the temperature control threshold if the test temperature of the current power consumption test node is greater than the temperature control threshold, and divide the residual temperature difference based on the remaining number of power consumption test nodes to obtain multiple balance nodes; and control the fan speed to make the temperature constant based on the balance node corresponding to the current power consumption test node, record the fan speed and graphics card power consumption in the equilibrium state, and record them as equilibrium state test data.

[0038] As a further solution of the present invention: the individual offset assessment module includes:

[0039] The thermal evaluation unit is used to evaluate the heat output of the fan per unit time based on the equilibrium test data of multiple power consumption test nodes. The heat output is determined based on the temperature difference and is proportional to the heat generation power. The temperature difference is characterized by:

[0040] ;

[0041] in, Used to characterize heat generation power, That is, the heat accumulation time, Characterizes the mass of an object, Characterizes the specific heat capacity of an object;

[0042] A dynamic evaluation unit is used to calculate the corresponding airflow velocity of the fan based on the equilibrium test data of multiple power consumption test nodes, where the airflow velocity is characterized by the ratio of the airflow volume to the effective area of ​​the fan outlet, that is:

[0043] ;

[0044] in, is a constant factor, Indicates the fan speed. Indicates the fan diameter, Indicates the effective area of ​​the fan outlet;

[0045] The characterization unit is used to determine the fan viscosity of the current graphics card in the current environment, characterized by:

[0046] ;

[0047] Among them, the , , They are used to characterize the temperature difference and airflow velocity of adjacent test nodes respectively;

[0048] The personalized evaluation unit is used to perform function fitting on the viscosity of several power consumption test nodes, and obtain the offset coefficient fitting function of the individual graphics card in the current environment.

[0049] As a further solution of the present invention: it also includes a basic evaluation module, specifically including:

[0050] a controllable stage judgment unit, used to judge the offset coefficient fitting function, and obtain a constant function segment of the offset coefficient fitting function, wherein the constant function segment is used to characterize the fan viscosity when the fan is at a low speed and the heat conduction effect is not affected by the environment and structure on the airflow, that is, the fan can increase in proportion to the power consumption of the graphics card and completely conduct the additional heat generated by the increased power in the same amount;

[0051] The basic personality judgment unit is used to compare and calculate the factory scheduling based on the fan speed of the constant function segment to obtain the basic personality of the current graphics card, and the basic personality is used to characterize the processing accuracy deviation value of the physical structure of the graphics card.

[0052] As a further solution of the present invention: it also includes a diving control module;

[0053] The dive control module is used to schedule the fan to dive based on a personalized scheduling plan when it is detected that the initial temperature of the graphics card has dropped. The scheduled dive is used to indicate that when the fan reaches the scheduled speed, the fan speed is slowly reduced so that the graphics card temperature reaches the set value under the corresponding power consumption in the personalized scheduling plan.

[0054] Compared with the prior art, the beneficial effects of the present invention are as follows: by performing a staged balance test on the graphics card, the correlation between the graphics card power consumption and the heat dissipation of the fan can be evaluated, and then compared with the factory scheduling, to determine the deviation in the heat dissipation capacity of the graphics card due to individual differences in the graphics card and the specific usage environment of the current user, and then a separate scheduling management can be performed on a separate graphics card, so that different users in different usage environments can always keep the graphics card under the optimal scheduling plan, ensuring the stability of the graphics card temperature management, and avoiding the occurrence of graphics card overheating due to the large gap between the environment and individual factors and the experimental environment of the factory scheduling plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 The present invention is a flowchart of a testing method for a graphics card.

[0056] Figure 2 The present invention is a specific flow chart of the balanced state test data steps of a test method for a graphics card.

[0057] Figure 3 The present invention is a block diagram of a test system for a graphics card. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0059] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0060] like Figure 1 The present invention provides a method for testing a graphics card, which includes the following steps:

[0061] S10, controlling the maximum output of heat dissipation, and performing a power consumption upper limit test on the graphics card based on a preset load program to obtain the maximum power consumption of the current graphics card, wherein the load program is used to schedule the graphics card core and the graphics memory based on the test requirements and control them to run at a certain occupancy rate;

[0062] S20, testing points of the maximum power consumption at a preset power consumption test interval, obtaining a power consumption test ladder including a plurality of power consumption test nodes, testing the power consumption test ladder step by step based on a load program, and making the temperature of the graphics card in a balanced state through fan control, so as to obtain balanced state test data corresponding to the array;

[0063] S30, performing individual evaluation based on the equilibrium test data and the corresponding test power consumption to obtain a deviation coefficient of the current graphics card in the current environment, wherein the deviation coefficient is used to characterize a phased value of the effectiveness of the heat dissipation capability of the specific graphics card in the specific environment;

[0064] S40, performing an offset adjustment on the preset factory scheduling according to the offset coefficient to obtain a personalized scheduling solution, wherein the personalized scheduling solution is used to characterize a heat dissipation scheduling solution of a graphics card under different power consumptions.

[0065] In this embodiment, a test method for a graphics card is provided. By performing a staged balance test on the graphics card, the correlation between the power consumption of the graphics card and the heat dissipation of the fan is evaluated, and then the graphics card can be compared with the factory scheduling to determine the heat dissipation capacity deviation due to individual differences in the graphics card and the specific usage environment of the graphics card under the current user. Then, a separate scheduling management can be performed on a separate graphics card, so that different users in different usage environments can always put the graphics card under the optimal scheduling plan, ensure the stability of the management of the graphics card temperature, and avoid the occurrence of graphics card overheating due to the large difference between the environment and individual factors and the experimental environment of the factory scheduling plan; the graphics card heat dissipation scheduling plan in the prior art is usually fixedly set based on the factory strategy of the manufacturer, and the factory strategy is experimented and formulated according to the standard specifications in the production and manufacturing of the graphics card, and has no correlation with the individual differences of the graphics card itself and the user's usage environment differences. However, in actual use, different graphics cards used by different users have individual differences generated in production (the influencing factors of individual differences include the quality of the chip, the packaging difference of the heat dissipation duct, etc.), which may be overclocked to release higher power, or the power usage caused by instability. and frequency limitation, as well as the user's chassis environment (affecting the airflow exchange during heat dissipation) and the external environment are different, which will lead to the poor effect of factory scheduling in actual use. Therefore, this embodiment provides a test adjustment scheme based on the user's individual graphics card for self-assessment and scheduling adjustment: Based on the individual differences that may be caused by packaging and other processes in the production and manufacturing of individual graphics cards, it is first necessary to test the power consumption upper limit (the power consumption upper limit here refers to the power consumption upper limit that can be stably operated and output, and the task is built and judged through the load program), obtain the power consumption upper limit, and then the power consumption interval can be split based on the power consumption upper limit to obtain several power consumption test nodes. For example, within (0-100), 50 nodes are selected for testing at each interval of a value, and at each power consumption test node, the fan and temperature limit are adjusted to make the temperature in a balanced state, and the data such as the fan and graphics card power consumption are recorded, and then the heat dissipation capacity of the fan for the graphics card in this equilibrium state can be calculated, and it is fitted based on several test nodes to judge the change in the heat dissipation capacity of the graphics card in the current environment, and by associating it with the factory scheduling, the setting of a personalized scheduling scheme is realized.

[0066] like Figure 2 As shown, as another preferred embodiment of the present invention, the step of making the graphics card temperature in a balanced state by controlling the fan to obtain the balanced state test data corresponding to the array specifically includes:

[0067] S21, judging based on a preset temperature control threshold, if the test temperature of the current power consumption test node is less than the temperature control threshold, waiting for the temperature to reach the temperature control threshold, and controlling the fan speed to make the temperature constant at the temperature control threshold, and recording the fan speed and graphics card power consumption in the equilibrium state, and recording them as equilibrium state test data;

[0068] S22, if the test temperature of the current power consumption test node is greater than the temperature control threshold, the residual temperature difference between the upper temperature wall of the current graphics card and the temperature control threshold is calculated, and the residual temperature difference is divided based on the remaining number of power consumption test nodes to obtain multiple balance nodes; and based on the balance node corresponding to the current power consumption test node, the fan speed is controlled to make the temperature constant, and the fan speed and graphics card power consumption in the equilibrium state are recorded, and recorded as equilibrium state test data.

[0069] In this embodiment, the steps for making the graphics card in a balanced state are specifically described. In scheduling, when the power consumption of the graphics card is low, its heat generation rate is low. Therefore, the heat dissipation demand of the fan can be reduced by increasing its temperature to a certain extent, and the fan speed can be reduced to reduce noise. Therefore, a temperature control threshold is set here. When the power of the graphics card is low, the graphics card is heated up with the temperature control threshold as the boundary, and after reaching the temperature control threshold, the fan is regulated and reaches a balanced state; when the temperature of the graphics card is high, the heat is generated faster, and it is necessary to increase the heat dissipation by increasing the fan speed. However, in order not to excessively keep the fan at the highest speed all the time, it is also necessary to allow the temperature to rise in stages according to the temperature upper limit, so that the temperature and fan speed are increased synchronously to achieve the purpose of coordinated control.

[0070] As another preferred embodiment of the present invention, the step of performing individual evaluation based on the equilibrium test data and the corresponding test power consumption to obtain the offset coefficient of the current graphics card in the current environment specifically includes:

[0071] The heat output of the fan per unit time is evaluated based on the equilibrium test data of multiple power consumption test nodes. The heat output is determined based on the temperature difference and is proportional to the heat generation power. The temperature difference is characterized by:

[0072] ;

[0073] in, Used to characterize heat generation power, That is, the heat accumulation time, Characterizes the mass of an object, Characterizes the specific heat capacity of an object;

[0074] The corresponding airflow velocity of the fan is calculated based on the equilibrium test data of multiple power consumption test nodes. The airflow velocity is characterized by the ratio of the airflow volume to the effective area of ​​the fan outlet, that is:

[0075] ;

[0076] in, is a constant factor, Indicates the fan speed. Indicates the fan diameter, Indicates the effective area of ​​the fan outlet;

[0077] Determine the fan viscosity of the current graphics card in the current environment, characterized by:

[0078] ;

[0079] Among them, the , , They are used to characterize the temperature difference and airflow velocity of adjacent test nodes respectively;

[0080] Function fitting is performed on the viscosity of several power consumption test nodes to obtain the offset coefficient fitting function of the graphics card individual in the current environment.

[0081] In this embodiment, the steps of performing individual evaluation based on the equilibrium test data and the corresponding test power consumption to obtain the offset coefficient of the current graphics card in the current environment are specifically described. The fan viscosity here is mainly determined based on the correlation between the airflow velocity required to carry away a certain amount of heat and the fan speed. Because when the amount of heat that needs to be carried away per unit time increases, the heat dissipation efficiency needs to be increased by the airflow velocity, and the airflow velocity is related to the fan speed. However, due to the influence of the distribution of the front and rear heat dissipation structures of the fan and the airflow structure of the chassis, the airflow velocity will not be completely increased based on the increase in fan speed. Therefore, the influence of external factors such as the environment and chassis on the viscosity of the graphics card fan can be judged based on this (in an ideal situation, the airflow velocity obtained based on the rotation speed should be directly proportional to the increase in heat, but the rotation speed used here is based on the actual test and has been affected by viscosity, so the viscosity judgment can be realized), and the offset coefficient fitting function can be obtained.

[0082] As another preferred embodiment of the present invention, it also includes a basic personality judgment step of an individual, specifically including:

[0083] The offset coefficient fitting function is judged to obtain a constant function segment of the offset coefficient fitting function, wherein the constant function segment is used to characterize the fan viscosity when the fan is at a low speed and the heat conduction effect is not affected by the environment and the structure on the airflow, that is, the fan can increase in proportion to the power consumption of the graphics card and completely conduct the additional heat generated by the increased power in the same amount;

[0084] The fan speed of the constant function segment is compared and calculated with the factory scheduling to obtain the basic personality of the current graphics card, and the basic personality is used to characterize the machining accuracy deviation value of the physical structure of the graphics card.

[0085] In this embodiment, a basic personality judgment step is added. This process mainly aims to determine from the viscosity the deviation caused by problems such as the graphics card's own process. For example, factors such as the efficiency of the heat pipe caused by packaging problems will cause different graphics cards to have different efficiencies in heat conduction from the graphics card core to the heat sink in actual use, which in turn leads to different heat dissipation efficiencies.

[0086] As another preferred embodiment of the present invention, the steps are also included:

[0087] When it is detected that the initial temperature of the graphics card has dropped, the fan is scheduled to dive based on a personalized scheduling plan. The scheduling dive is used to indicate that when the fan reaches the scheduling speed, the fan speed is slowly reduced so that the graphics card temperature reaches the set value under the corresponding power consumption in the personalized scheduling plan.

[0088] In this embodiment, an automatic dive function based on ambient temperature is also included. For example, when the user uses air conditioning, or the ambient temperature changes due to daily external climate factors, the fan speed does not reach the specified personalized scheduling plan, but the temperature has dropped to the specified value in the personalized scheduling plan. Therefore, the fan speed can be further reduced.

[0089] like Figure 3 As shown, the present invention also provides a testing system for a graphics card, which comprises:

[0090] The individual limit test module 100 is used to control the maximum output of heat dissipation and perform a power consumption upper limit test on the graphics card based on a preset load program to obtain the maximum power consumption of the current graphics card. The load program is used to schedule the graphics card core and video memory based on the test requirements and control them to run at a certain occupancy rate;

[0091] The balanced grouping test module 200 is used to test the maximum power consumption at a preset power consumption test interval, obtain a power consumption test ladder including a plurality of power consumption test nodes, test the power consumption test ladder step by step based on a load program, and make the temperature of the graphics card in a balanced state through fan control, so as to obtain the balanced state test data corresponding to the array;

[0092] An individual offset evaluation module 300 is used to perform an individual evaluation based on the equilibrium test data and the corresponding test power consumption to obtain an offset coefficient of the current graphics card in the current environment, wherein the offset coefficient is used to characterize a staged value of the effectiveness of the heat dissipation capability of a specific graphics card in a specific environment;

[0093] The personalized scheduling management module 400 is used to perform an offset adjustment on the preset factory scheduling according to the offset coefficient to obtain a personalized scheduling solution, wherein the personalized scheduling solution is used to characterize the heat dissipation scheduling solution of the graphics card under different power consumption.

[0094] As another preferred embodiment of the present invention, the balanced grouping test module includes:

[0095] The heat dissipation priority test unit is used to make a judgment based on a preset temperature control threshold. If the test temperature of the current power consumption test node is less than the temperature control threshold, the unit waits for the temperature to reach the temperature control threshold, controls the fan speed to make the temperature constant at the temperature control threshold, and records the fan speed and graphics card power consumption in the equilibrium state, which are recorded as equilibrium state test data.

[0096] The temperature priority test unit is used to calculate the residual temperature difference between the upper temperature wall of the current graphics card and the temperature control threshold if the test temperature of the current power consumption test node is greater than the temperature control threshold, and divide the residual temperature difference based on the remaining number of power consumption test nodes to obtain multiple balance nodes; and control the fan speed to make the temperature constant based on the balance node corresponding to the current power consumption test node, record the fan speed and graphics card power consumption in the equilibrium state, and record them as equilibrium state test data.

[0097] As another preferred embodiment of the present invention, the individual offset assessment module includes:

[0098] The thermal evaluation unit is used to evaluate the heat output of the fan per unit time based on the equilibrium test data of multiple power consumption test nodes. The heat output is determined based on the temperature difference and is proportional to the heat generation power. The temperature difference is characterized by:

[0099] ;

[0100] in, Used to characterize heat generation power, That is, the heat accumulation time, Characterizes the mass of an object, Characterizes the specific heat capacity of an object;

[0101] A dynamic evaluation unit is used to calculate the corresponding airflow velocity of the fan based on the equilibrium test data of multiple power consumption test nodes, where the airflow velocity is characterized by the ratio of the airflow volume to the effective area of ​​the fan outlet, that is:

[0102] ;

[0103] in, is a constant factor, Indicates the fan speed. Indicates the fan diameter, Indicates the effective area of ​​the fan outlet;

[0104] The characterization unit is used to determine the fan viscosity of the current graphics card in the current environment, characterized by:

[0105] ;

[0106] Among them, the , , They are used to characterize the temperature difference and airflow velocity of adjacent test nodes respectively;

[0107] The personalized evaluation unit is used to perform function fitting on the viscosity of several power consumption test nodes, and obtain the offset coefficient fitting function of the individual graphics card in the current environment.

[0108] As another preferred embodiment of the present invention, it also includes a basic evaluation module, specifically including:

[0109] a controllable stage judgment unit, used to judge the offset coefficient fitting function, and obtain a constant function segment of the offset coefficient fitting function, wherein the constant function segment is used to characterize the fan viscosity when the fan is at a low speed and the heat conduction effect is not affected by the environment and structure on the airflow, that is, the fan can increase in proportion to the power consumption of the graphics card and completely conduct the additional heat generated by the increased power in the same amount;

[0110] The basic personality judgment unit is used to compare and calculate the factory scheduling based on the fan speed of the constant function segment to obtain the basic personality of the current graphics card, and the basic personality is used to characterize the processing accuracy deviation value of the physical structure of the graphics card.

[0111] As another preferred embodiment of the present invention, it also includes a diving control module;

[0112] The dive control module is used to schedule the fan to dive based on a personalized scheduling plan when it is detected that the initial temperature of the graphics card has dropped. The scheduled dive is used to indicate that when the fan reaches the scheduled speed, the fan speed is slowly reduced so that the graphics card temperature reaches the set value under the corresponding power consumption in the personalized scheduling plan.

[0113] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0114] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the disclosure in the specification and examples. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0115] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A testing method for a graphics card, characterized in that: Include: Control the maximum heat dissipation output, and perform a power consumption upper limit test on the graphics card based on a preset load program to obtain the maximum power consumption of the current graphics card. The load program is used to schedule the graphics card core and video memory based on the test requirements and control them to run at a certain occupancy rate; The maximum power consumption is tested and selected at a preset power consumption test interval to obtain a power consumption test ladder including a plurality of power consumption test nodes, the power consumption test ladder is tested step by step based on a load program, and the temperature of the graphics card is kept in a balanced state through fan control to obtain the balanced state test data corresponding to the array; Performing individual evaluation based on the equilibrium test data and the corresponding test power consumption to obtain the offset coefficient of the current graphics card in the current environment, wherein the offset coefficient is used to characterize the staged value of the effectiveness of the heat dissipation capability of the graphics card in the current environment; Performing an offset adjustment on the preset factory schedule according to the offset coefficient to obtain a personalized schedule, wherein the personalized schedule is used to characterize a heat dissipation schedule of the graphics card under different power consumptions; The step of performing individual evaluation based on the equilibrium state test data and the corresponding test power consumption to obtain the offset coefficient of the current graphics card in the current environment specifically includes: The heat output of the fan per unit time is evaluated based on the equilibrium test data of multiple power consumption test nodes. The heat output is determined based on the temperature difference and is proportional to the heat generation power. The temperature difference is characterized by: ； in, Used to characterize heat generation power, That is, the heat accumulation time, Characterizes the mass of an object, Characterizes the specific heat capacity of an object; The corresponding airflow velocity of the fan is calculated based on the equilibrium test data of multiple power consumption test nodes. The airflow velocity is characterized by the ratio of the airflow volume to the effective area of ​​the fan outlet, that is: ； in, is a constant factor, Indicates the fan speed. Indicates the fan diameter, Indicates the effective area of ​​the fan outlet; Determine the fan viscosity of the current graphics card in the current environment, characterized by: ； Among them, the , , They are used to characterize the temperature difference and airflow velocity of adjacent test nodes respectively; the viscosity of several power consumption test nodes is fitted with functions to obtain the offset coefficient fitting function of the individual graphics card in the current environment.

2. A method for testing a graphics card according to claim 1, characterized in that: The step of controlling the temperature of the graphics card to be in a balanced state by fan control to obtain the balanced state test data corresponding to the array specifically includes: Based on the preset temperature control threshold, if the test temperature of the current power consumption test node is less than the temperature control threshold, wait for the temperature to reach the temperature control threshold, and control the fan speed to make the temperature constant at the temperature control threshold, and record the fan speed and graphics card power consumption in the equilibrium state, and record them as equilibrium state test data; If the test temperature of the current power consumption test node is greater than the temperature control threshold, the residual temperature difference between the upper limit temperature wall of the current graphics card and the temperature control threshold is calculated, and the residual temperature difference is divided based on the remaining number of power consumption test nodes to obtain multiple balance nodes; and based on the balance node corresponding to the current power consumption test node, the fan speed is controlled to make the temperature constant, and the fan speed and graphics card power consumption in the equilibrium state are recorded, and recorded as equilibrium test data.

3. A method for testing a graphics card according to claim 2, characterized in that: Also includes the steps: The offset coefficient fitting function is judged to obtain a constant function segment of the offset coefficient fitting function, wherein the constant function segment is used to characterize the fan viscosity when the fan is at a low speed and the heat conduction effect is not affected by the environment and the structure on the airflow, that is, the fan speed increases in proportion to the power consumption of the graphics card, and the additional heat generated by the increased power is completely and equally conducted; The fan speed of the constant function segment is compared and calculated with the factory scheduling to obtain the basic personality of the current graphics card, and the basic personality is used to characterize the machining accuracy deviation value of the physical structure of the graphics card.

4. A method for testing a graphics card according to claim 2, characterized in that: Also includes the steps: When it is detected that the initial temperature of the graphics card has dropped, the fan is scheduled to dive based on a personalized scheduling plan. The scheduling dive is used to indicate that when the fan reaches the scheduling speed, the fan speed is slowly reduced so that the graphics card temperature reaches the set value under the corresponding power consumption in the personalized scheduling plan.

5. A test system for a graphics card, characterized in that: Include: An individual limit test module is used to control the maximum output of heat dissipation and perform a power consumption upper limit test on the graphics card based on a preset load program to obtain the maximum power consumption of the current graphics card. The load program is used to schedule the graphics card core and video memory based on the test requirements and control them to run at a certain occupancy rate; A balanced grouping test module is used to test the maximum power consumption at a preset power consumption test interval, obtain a power consumption test ladder including a plurality of power consumption test nodes, test the power consumption test ladder step by step based on a load program, and make the temperature of the graphics card in a balanced state through fan control, so as to obtain the balanced state test data corresponding to the array; An individual offset evaluation module, used to perform individual evaluation based on the equilibrium test data and the corresponding test power consumption, so as to obtain an offset coefficient of the current graphics card in the current environment, wherein the offset coefficient is used to characterize a staged value of the effectiveness of the heat dissipation capability of the graphics card in the current environment; A personalized scheduling management module, used to perform an offset adjustment on the preset factory scheduling according to the offset coefficient to obtain a personalized scheduling scheme, wherein the personalized scheduling scheme is used to characterize a heat dissipation scheduling scheme of a graphics card under different power consumptions; The individual deviation assessment module includes: The thermal evaluation unit is used to evaluate the heat output of the fan per unit time based on the equilibrium test data of multiple power consumption test nodes. The heat output is determined based on the temperature difference and is proportional to the heat generation power. The temperature difference is characterized by: ； in, Used to characterize heat generation power, That is, the heat accumulation time, Characterizes the mass of an object, Characterizes the specific heat capacity of an object; A dynamic evaluation unit is used to calculate the corresponding airflow velocity of the fan based on the equilibrium test data of multiple power consumption test nodes, where the airflow velocity is characterized by the ratio of the airflow volume to the effective area of ​​the fan outlet, that is: ； in, is a constant factor, Indicates the fan speed. Indicates the fan diameter, Indicates the effective area of ​​the fan outlet; The characterization unit is used to determine the fan viscosity of the current graphics card in the current environment, characterized by: ； Among them, the , , They are used to characterize the temperature difference and airflow velocity of adjacent test nodes respectively; The personalized evaluation unit is used to perform function fitting on the viscosity of several power consumption test nodes and obtain the offset coefficient fitting function of the individual graphics card in the current environment.

6. A graphics card testing system according to claim 5, characterized in that: The balanced grouping test module includes: The heat dissipation priority test unit is used to make a judgment based on a preset temperature control threshold. If the test temperature of the current power consumption test node is less than the temperature control threshold, the unit waits for the temperature to reach the temperature control threshold, controls the fan speed to make the temperature constant at the temperature control threshold, and records the fan speed and graphics card power consumption in the equilibrium state, which are recorded as equilibrium state test data. The temperature priority test unit is used to calculate the residual temperature difference between the upper temperature wall of the current graphics card and the temperature control threshold if the test temperature of the current power consumption test node is greater than the temperature control threshold, and divide the residual temperature difference based on the remaining number of power consumption test nodes to obtain multiple balance nodes; and control the fan speed to make the temperature constant based on the balance node corresponding to the current power consumption test node, record the fan speed and graphics card power consumption in the equilibrium state, and record them as equilibrium state test data.

7. A graphics card testing system according to claim 6, characterized in that: It also includes basic assessment modules, including: a controllable stage judgment unit, used to judge the offset coefficient fitting function, and obtain a constant function segment of the offset coefficient fitting function, wherein the constant function segment is used to characterize the fan viscosity when the fan is at a low speed and the heat conduction effect is not affected by the environment and the structure on the airflow, that is, the fan speed increases in proportion to the power consumption of the graphics card, and the additional heat generated by the increased power is completely and equally conducted; The basic personality judgment unit is used to compare and calculate the factory scheduling based on the fan speed of the constant function segment to obtain the basic personality of the current graphics card, and the basic personality is used to characterize the processing accuracy deviation value of the physical structure of the graphics card.

8. A graphics card testing system according to claim 6, characterized in that: Also included is the Dive Control Module; The dive control module is used to schedule the fan to dive based on a personalized scheduling plan when it is detected that the initial temperature of the graphics card has dropped. The scheduled dive is used to indicate that when the fan reaches the scheduled speed, the fan speed is slowly reduced so that the graphics card temperature reaches the set value under the corresponding power consumption in the personalized scheduling plan.

Images