A method and apparatus for optimizing game coding parameters based on hardware parameters

Abstract

This invention provides a method and apparatus for optimizing game coding parameters based on hardware parameters, belonging to the field of data optimization technology. Specifically, it includes: determining a model control target based on an update method; determining a coding optimization group within the coding control group based on the adjustment data of the game coding parameters of the model control target and the update method of the model control target in the coding control group; and determining the parameter control method for the game coding of the coding optimization group in different model control targets using stuttering detection data and model control target data during the model control process in different coding control groups, thereby improving the stability of the terminal during operation.

Description

Technical Field

[0001] This invention belongs to the field of data optimization technology, and in particular relates to a method and apparatus for optimizing game coding parameters based on hardware parameters. Background Technology

[0002] As game graphics continue to improve, the demands on hardware resources are also increasing. However, the hardware configurations of different user devices vary greatly, and the performance of the same device fluctuates under different usage scenarios (such as charging, high-temperature environments, and background multitasking). Traditional games typically run with fixed encoding parameters (such as fixed resolution and fixed frame rate), and these fixed parameters cannot adapt to the real-time state of the hardware.

[0003] To address the aforementioned technical issues, the invention patent application CN202511698701.1, "A Front-End Development Optimization Method and System," employs a resource loading optimization strategy to execute the incremental update instruction set based on terminal operating environment performance parameters. It also eliminates cross-platform compatibility conflicts through rendering pipeline reorganization technology, outputting consistent content presentation across multiple terminals. However, the following technical problems remain: When optimizing game coding parameters for a terminal using models and hardware parameters, the differences in terminal hardware parameters lead to significant impacts from external factors and game runtime. Consequently, the model may frequently adjust game parameters, making it difficult to meet the required stability of the terminal's operation. Therefore, determining the optimization control method for game coding parameters based on the frequency of model adjustments to ensure the reliability of terminal operation has become an urgent technical problem to be solved.

[0004] Therefore, there is an urgent need for a method and device for optimizing game coding parameters based on hardware parameters. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a method for optimizing game coding parameters based on hardware parameters, which includes: S1 divides terminals into different groups based on terminal model, determines the optimization method for game coding parameters of terminals in the group based on the lag monitoring data of the terminals in the group, determines the coding control group in the group based on the optimization method, and determines the update method for model control target in the coding control group by combining the terminal change data in the coding control group. S2 determines the model control target based on the update method, and determines the coding optimization group in the coding control group based on the adjustment data of the game coding parameters of the model control target and the update method of the model control target in the coding control group. S3 uses stuttering detection data and model control target data from different model control targets in different coding control groups to determine the parameter control method for the game coding of the coding optimization group in different model control targets.

[0006] The beneficial effects of this invention are as follows: Based on the adjustment data of the game coding parameters of the model control target and the update method of the model control target in the coding control group, the coding optimization group in the coding control group is determined. That is, the group that frequently adjusts the game coding parameters during the game operation of the model control target is identified to avoid the occurrence of technical problems such as poor system stability.

[0007] By utilizing stuttering detection data and model control target data from different coding control groups and different model control targets during the model control process, the parameter control method for game coding of the coding optimization group under different model control targets is determined. First, the overall model recognition capability of all coding control groups is evaluated, which is reflected by the average value of the "model recognition matching coefficient". If the overall capability is insufficient, the status quo should be maintained to accumulate more data and avoid data loss due to blind intervention. Then, based on the matching coefficient of the optimization group itself, the number of stuttering risk targets, and the frequency of adjustment values, it is decided whether to adopt fixed parameter operation (to verify whether it is more stable) or to continue to maintain model control. Through this hierarchical decision-making, it is ensured that there is sufficient data to support model iteration. At the same time, targeted tests are conducted on unstable optimization groups to improve the reliability of overall control.

[0008] Furthermore, terminals are divided into different groups based on their model numbers, specifically including: Terminals of the same model are grouped together.

[0009] Furthermore, the lag monitoring data of the terminal includes the lag periods and lag durations of the lag periods during the operation of the game.

[0010] Furthermore, the method for determining the optimization method of the game encoding parameters of the terminals in the group is as follows: Based on the lag monitoring data of the terminals in the group, determine the lag periods of the terminals in the group during the game's operation; By utilizing the distribution data of lag periods in different game runtime segments, the terminal at risk of lag is identified. An optimization method for game encoding parameters of terminals in the group that are at risk of lag.

[0011] Furthermore, the terminal with the risk of lag is a terminal that experiences lag periods in different game runtime stages.

[0012] Furthermore, the method for determining the parameter control method of the game coding for the coding optimization group under different model control objectives is as follows: Using the stuttering detection data of different model control targets in the coded control group during the model control process, identify the model control targets whose number of stuttering periods in the model control process does not meet the requirements, and use them as stuttering risk control targets; Based on the lag risk control target and simulation control target data in the coded control group, the model identification matching coefficient of the coded control group is determined; A parameter control method for determining the game coding parameters of the coding optimization group under different model control objectives is used, based on the model identification matching coefficients of different coding control groups and the frequent adjustment values ​​of the coding optimization group.

[0013] In a second aspect, the present invention provides a computer device, comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described game coding parameter optimization method based on hardware parameters when running the computer program.

[0014] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of a game coding parameter optimization method based on hardware parameters; Figure 2 This is a flowchart illustrating the method for determining the optimization of game encoding parameters for terminals within a group; Figure 3This is a flowchart illustrating the method for determining the update method of the model control objective in the coded control group; Figure 4 This is a flowchart illustrating the method for determining the coding optimization group within the coding control group. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0019] Example 1 like Figure 1 As shown, this application provides a method for optimizing game encoding parameters based on hardware parameters, specifically including: S1 divides terminals into different groups based on terminal model, determines the optimization method for game coding parameters of terminals in the group based on the lag monitoring data of the terminals in the group, determines the coding control group in the group based on the optimization method, and determines the update method for model control target in the coding control group by combining the terminal change data in the coding control group. S2 determines the model control target based on the update method, and determines the coding optimization group in the coding control group based on the adjustment data of the game coding parameters of the model control target and the update method of the model control target in the coding control group. S3 uses stuttering detection data and model control target data from different model control targets in different coding control groups to determine the parameter control method for the game coding of the coding optimization group in different model control targets.

[0020] Furthermore, terminals are divided into different groups based on their model numbers, specifically including: Terminals of the same model are grouped together.

[0021] Furthermore, the lag monitoring data of the terminal includes the lag periods and lag durations of the lag periods during the operation of the game.

[0022] Specifically, such as Figure 2 As shown, the method for determining the optimization method of game encoding parameters for terminals in the group is as follows: The core objective of this technical solution is to develop a dynamic game encoding parameter optimization method for groups of terminals of the same model, effectively suppressing stuttering during game operation. Even terminals of the same model can exhibit performance differences due to hardware status, operating environment, and other factors, resulting in varying stuttering characteristics during gameplay. By analyzing stuttering monitoring data from all terminals within the group, "stuttering-risk terminals" that experience stuttering across multiple game runtime stages are identified. Based on the proportion of these terminals, the suitability of the current game encoding parameters for the group is determined. If there are too many stuttering-risk terminals, it indicates that the current parameter settings are too high and cannot meet the performance requirements of most terminals in the group. In this case, the overall game encoding parameters need to be controlled at a lower level, and fine-tuned in conjunction with hardware parameters to reduce the stuttering rate and improve user experience.

[0023] First, based on the stuttering monitoring data of each terminal within the group, the stuttering periods for each terminal during game execution are extracted. Second, by comparing the stuttering distribution across different game execution periods, terminals experiencing stuttering in multiple time periods are identified. These terminals are the most performance-sensitive and are key indicators for evaluating the rationality of encoding parameters. Finally, the number of such terminals is counted. If it exceeds a preset threshold, it is determined that the current encoding parameters cannot effectively suppress stuttering. The group needs to be marked as an "encoding control group," and a minimum encoding parameter strategy should be adopted. Simultaneously, dynamic control should be implemented using a hardware parameter model to minimize stuttering while ensuring basic smoothness.

[0024] S11 uses the lag monitoring data of the terminals in the group to determine the lag periods of the terminals in the group during the game's operation; Grouping: This refers to grouping all terminals of the same model into the same group to ensure that the hardware is similar and facilitates comparative analysis.

[0025] Lag monitoring data includes the periods of lag recorded by the terminal during game operation (i.e., the time intervals in which lag occurred) and the duration of each lag period. This data is usually collected by the terminal's built-in performance monitoring module.

[0026] Lag period: refers to the continuous period of time during game operation when the terminal experiences obvious lag (such as a sudden drop in frame rate or screen freezing), such as the 5-second interval from 10:05:30 to 10:05:35.

[0027] The significance of defining a "group" is to classify terminals with the same hardware into one category, eliminate the impact of hardware differences on stuttering analysis, and make subsequent optimization strategies more targeted.

[0028] The reason for collecting "lag periods" rather than just the number of lags is that lag duration can more accurately reflect the severity of lag and provide a quantitative basis for subsequent risk identification.

[0029] A game operation platform grouped all devices with the "GamePhone X" model into a single group, containing 1000 devices. During the past week of game operation, each device recorded its own lag events. For example, device A experienced two lag events during gameplay: the first during the login phase (18:30:10 to 18:30:15), lasting 5 seconds; and the second during the match phase (18:45:22 to 18:45:25), lasting 3 seconds. This information was aggregated into lag period data.

[0030] S12 uses the distribution data of lag periods in different game runtime segments to determine the terminal with lag risk in the terminal. Game runtime period: refers to a complete game running process, during which the terminal is constantly in the game running process.

[0031] Distribution data of lag periods: refers to statistical information on whether each terminal experienced lag during game runtime, and the duration of the lag.

[0032] Lag-prone terminals: These are defined as terminals that experience lag periods across different game runtime stages. Such terminals exhibit lag across different time periods, indicating that their overall performance may be insufficient, rather than being sporadic in specific scenarios.

[0033] The concept of "game runtime segment" is introduced to differentiate the context of stuttering and avoid misjudging occasional stuttering as a risk. If a terminal stutters only during a specific period (such as resource loading), it may be due to excessive resource demand during that period; however, if stuttering occurs in multiple periods, it indicates that the terminal's performance may not meet the basic requirements of the game, which is the real risk.

[0034] The significance of defining "stuttering risk terminals" lies in focusing on those terminals that most need optimization, as their existence is a key basis for deciding whether to adjust encoding parameters.

[0035] Terminal A in the group experiences lag periods during every run, thus classifying it as a high-risk terminal for lag. Terminal B does not experience lag periods during every run, therefore it is not considered a high-risk terminal for lag. By analyzing the lag periods of all terminals, we can determine the specific time periods during which each terminal experienced lag, ultimately identifying a list of all high-risk terminals in the group.

[0036] S13 determines an optimization method for the game encoding parameters of terminals in the group that are at risk of lag.

[0037] Furthermore, the terminal with the risk of lag is a terminal that experiences lag periods in different game runtime stages.

[0038] It is understandable that if the number of terminals at risk of lag in the group does not meet the requirements, it means that the terminals in the group are unable to effectively suppress lag under the current game encoding parameters. Therefore, the optimization method for the game encoding parameters of the terminals in the group is to control the game encoding parameters to the minimum, that is, to use the group as an encoding control group, and to combine the hardware parameters and models of the terminals to control the game encoding parameters, thereby reducing lag during operation.

[0039] Game encoding parameters: These are configurable parameters that affect game image quality and smoothness, such as resolution, frame rate limit, texture quality, and shadow effects. The levels of these parameters directly affect the load on the terminal.

[0040] Optimization Method: In this solution, the number of terminals at risk of lag determines whether a unified adjustment to the game encoding parameters for the entire group is necessary. If adjustment is required, the "minimum state" parameters are used, and the group is marked as the "encoding control group." Subsequent dynamic control is then implemented based on the terminal's hardware parameters and model.

[0041] Minimum settings: This refers to setting the game's encoding parameters to the minimum configuration supported by the game, such as minimum resolution, 30 frames per second maximum, and turning off special effects, in order to minimize the load on the terminal.

[0042] Combining hardware parameters and models: This means that after setting the parameters to the minimum, further personalized fine-tuning is needed based on the hardware performance of each terminal (such as CPU, GPU model, memory size) and the preset performance model. For example, for terminals with strong performance, some parameters can be appropriately increased, but the overall level is still controlled at a low level.

[0043] The number of terminals at risk of stuttering reflects the overall adaptability of the group under the current encoding parameters. If there are many such terminals, it means that the current parameters are too high for most terminals, causing widespread stuttering. Therefore, it is necessary to reduce the parameters to the lowest possible level to ensure basic smoothness.

[0044] To optimize the smoothness of its popular mobile games across different devices, a game company grouped user devices and analyzed lag performance. Specifically, all devices designated as "flagship models" were grouped into one group, containing 1000 devices. During gameplay, the system automatically collected lag data for each device, including the duration of lag and the corresponding game scenario.

[0045] Step S11: Determine the period of lag. For example, if terminal #001 experiences a lag during the game and it lasts for 5 seconds, the lag data of all terminals is collected into the database.

[0046] Step S12: Identify terminals at risk of lag. The system counts whether each terminal has experienced at least one lag during all operations. If a terminal has lag records in all operations, it is marked as a terminal at risk of lag. After counting, a total of 250 terminals in the group meet this condition, meaning there are 250 terminals at risk of lag.

[0047] Step S13: Determine the optimization method. The company's preset threshold for the percentage of terminals at risk of stuttering is 20%. The percentage of terminals at risk of stuttering in the group is 250 / 1000=25%, which exceeds the threshold. Therefore, it is determined that the current game encoding parameters (default medium-high quality, 60 frames per second) are insufficient to effectively suppress stuttering, and overall optimization of this group is required.

[0048] The optimization method was determined as follows: All terminals in this group were forced to set their game encoding parameters to the lowest possible level, i.e., resolution reduced to 720p, frame rate locked at 30fps, and effects such as shadows and anti-aliasing disabled. Simultaneously, this group was designated as the "Encoding Control Group," and these parameters will be dynamically adjusted based on each terminal's hardware parameters (such as CPU frequency and memory usage) and preset performance models. For example, for terminals with better performance, some effects could be enabled at the lowest possible level, but the overall focus remained on low load to ensure minimal stuttering.

[0049] Specifically, such as Figure 3 As shown, the method for determining the update method of the model control target in the coding control group is as follows: For terminal groups already marked as "encoding control groups," a dynamic "model control target" update method is developed to determine which terminals require further fine-tuning of game encoding parameters through performance models. Encoding control groups refer to those with a high proportion of terminals at risk of lag, whose game encoding parameters have been forcibly set to the lowest possible level. Therefore, it is necessary to use models to adjust game encoding parameters so that they can run beyond the lowest possible state and maintain image quality to a certain extent. However, since the applicability of the model cannot be determined, it is necessary to select the terminals most in need of model intervention as "model control targets" based on factors such as group size and terminal changes. This solution uses multi-level judgment to prioritize terminals in groups with severe lag, large group changes, and large scale to be included in model control, thereby maximizing lag suppression effects when the optimization effect is difficult to determine.

[0050] First, assess the proportion of all coded control groups. If the proportion is too high, it indicates widespread lag issues, and a strict update strategy is applied to all groups (i.e., lag duration exceeding the limit is immediately included in model control). Otherwise, further analyze the terminal changes in each coded control group (the number of new terminals joining the group daily). If the average change is large, it indicates group instability, and a strict strategy is also applied. If the change is small, a lenient strategy is applied to the remaining groups based on the proportion of strictly controlled groups (both lag duration exceeding the limit and lag occurring during specific time periods must be met). Finally, if a decision is still undecided, prioritize groups by size, with larger groups receiving priority for a strict strategy. This hierarchical approach ensures that model resources are allocated to the groups with the most prominent problems and the widest impact.

[0051] S21 determines the number of coding control groups in the group based on the coding control groups in the group; It is understood that, based on the number of coding control groups in the group, the proportion of the number of coding control groups in the group is determined, and it is determined whether the proportion of the number of coding control groups in the group is greater than the preset control group proportion threshold. If so, the update method for the model control target in all coding control groups is as follows: if the total duration of the lag period in the most recent preset time period does not meet the requirements, the terminal is taken as the model control target, and it is determined whether the terminal can reliably run under the control of the model by adjusting the game coding parameters. If not, proceed to step S22.

[0052] Group: refers to a collection of terminals divided by terminal model, with each model corresponding to a group.

[0053] Encoding control groups: These are groups whose game encoding parameters have been forced to be set to the lowest level due to a high proportion of terminals at risk of lag. These groups require close monitoring.

[0054] Preset control group number percentage threshold: A pre-set percentage value used to determine whether there are too many groups that need to be strictly controlled.

[0055] By analyzing the percentage of groups controlled by the coding system, the severity of the overall lag issue can be assessed. If the vast majority of groups are already under coding control, it indicates widespread performance insufficiency. In this case, the strictest model control target selection strategy should be applied to all coding control groups indiscriminately; that is, any terminal experiencing lag exceeding the limit should be included in the model to alleviate lag as quickly as possible.

[0056] If the percentage is not high, it indicates that the problem is mainly concentrated in a few groups, requiring further refined analysis to avoid wasting resources.

[0057] A gaming platform has 20 terminal model groups, 12 of which are marked as encoding control groups. Therefore, the proportion of encoding control groups is 12 / 20 = 60%. Assuming a preset threshold of 50% for the proportion of control groups, 60% > 50%, so the platform directly enters the "Yes" branch. A strict update method is applied to all 12 encoding control groups: for each terminal, if the total duration of lag periods in the last 7 days exceeds 20 minutes (i.e., does not meet the requirement), then that terminal is used as the model control target. Subsequently, its game encoding parameters are dynamically adjusted through the model, and its reliable operation is verified. If the proportion is not greater than the threshold, the process proceeds to S22.

[0058] S22 uses the terminal change data in the coding control group to determine the number of updated terminals in the coding control group on different dates; The above steps include the following: S221 determines the average number of terminal updates in the coding control group on different dates based on the number of terminal updates in the coding control group on different dates, and determines whether the average number of terminal updates in the coding control group on different dates is greater than a preset terminal number threshold. If so, the update method for the model control target in the coding control group is determined as follows: if the total duration of the lag period in the most recent preset time period does not meet the requirements, the terminal is taken as the model control target, and it is determined whether the adjustment of game coding parameters under the control of the model can reliably run. If not, proceed to step S222. Average number of updates: Calculate the average number of new terminals added each day over a recent period (e.g., one week).

[0059] Preset terminal number threshold: A pre-set number value used to determine whether group changes are significant.

[0060] The total duration of the lag period does not meet the requirements: This means that the cumulative lag duration of the terminal in the most recent preset time period exceeds a certain threshold, indicating that the lag is severe.

[0061] A high average number of updates indicates that the group is expanding rapidly and the problem is spreading. At this point, it's necessary to immediately and rigorously screen the devices within the group, bringing those experiencing severe lag into the model's control to curb the problem's spread. If the average number is not high, it indicates that the group is relatively stable, and a more detailed assessment can proceed.

[0062] S222 updates the model control target by determining whether the total duration of the lag period within the most recent preset time period does not meet the requirements. Specifically, if the terminal is used as the model control target's encoding control group, it is treated as a strict control group. The proportion of the strict control group in the encoding control group is determined to be greater than a preset control group proportion threshold. If so, the remaining encoding control groups are then updated by determining whether the total duration of the lag period within the most recent preset time period does not meet the requirements, and there are lag periods with unmet durations. Specifically, if the terminal is used as the model control target, it is determined whether the terminal can reliably operate under the model's control by adjusting game encoding parameters. If not, the process proceeds to step S23.

[0063] Strictly controlled groups: These are groups that have been determined to use the strict update method through S21 or S221.

[0064] Preset control group percentage threshold: A pre-set percentage used to determine whether the strictly controlled group has become the majority.

[0065] There are stuttering periods that do not meet the requirements: This is a stricter condition than simply not meeting the total duration requirement. It may require that the duration of at least one stuttering period exceeds a certain threshold in order to exclude the interference of multiple short stuttering periods.

[0066] If the proportion of strictly controlled groups is already high, it means that most of the problem groups have been covered. The remaining few groups can be screened with relatively lenient conditions, that is, not only should the total duration exceed the limit, but there should also be obvious long lag events, to ensure that model resources are really used where they are most needed. If the proportion is not high, it is necessary to further distinguish them through other indicators (such as group size).

[0067] S23 uses the number of coded control groups in the group and the number of updates of terminals in the coded control groups on different dates to determine the update method of the model control target in the coded control group.

[0068] In the above steps, based on the number of terminals in the coding control group, the sorting result of the coding control group in different coding control groups is determined. If the sorting result of the coding control group in different coding control groups is before the target position, the update method for the model control target in the coding control group is as follows: if the total duration of the lag period in the most recent preset time period does not meet the requirements, the terminal is taken as the model control target, and it is determined whether the terminal can reliably operate under the control of the model by adjusting the game coding parameters. If the sorting result of the coding control group in different coding control groups is not before the target position, the update method for the model control target in the remaining coding control groups is as follows: if the total duration of the lag period in the most recent preset time period does not meet the requirements, and there are lag periods with durations that do not meet the requirements, the terminal is taken as the model control target, and it is determined whether the terminal can reliably operate under the control of the model by adjusting the game coding parameters.

[0069] It should be noted that the sorting results are obtained by sorting the terminals in the coding control group from high to low.

[0070] Sorting results: Sorted according to the number of terminals in each coding control group (i.e., group size), with larger groups ranked first.

[0071] Target position: A pre-defined sorting position threshold, such as the top 30% of positions.

[0072] If none of the above conditions are met, the final decision will be based on the group size. Larger groups have a wider impact and should be given priority for strict strategies; smaller groups should be given more lenient strategies to balance resource allocation.

[0073] To optimize the smoothness of its various games, a game company manages terminal model groups. Currently, there are 20 terminal model groups, of which 12 are marked as encoding control groups due to a high percentage of terminals at risk of lag. Step S21: Calculate the percentage of terminals in the encoding control groups. 12 / 20 = 60% > 50%, therefore, a strict update method is applied to all 12 encoding control groups: Within each group, if a terminal's cumulative lag time exceeds 20 minutes in the past 7 days, that terminal is designated as a model control target, and its encoding parameters are dynamically adjusted by the model thereafter.

[0074] Assume that after a period of time, the problem is alleviated and the number of coded control groups decreases. Now, reassess, assuming there are only 5 coded control groups at this point, the total number of groups remains 20, accounting for 25% (<50%), and proceed to S22.

[0075] Step S22: Analyze terminal change data. For these 5 coding control groups, count the number of terminals newly added to the group each day for the past 7 days and calculate the average: Group A: Average 50 units / day, Group B: Average 10 units / day, Group C: Average 5 units / day, Group D: Average 2 units / day, Group E: Average 0 units / day The preset threshold for the number of terminals is 20. Group A has an average value of 50, which is greater than 20. Therefore, Group A is determined to be subject to a strict update method (i.e., if the lag time is greater than 20 minutes in the last 7 days, it will be included in the model control) and becomes a "strictly controlled group". The average value of the remaining groups is ≤20, so proceed to S222.

[0076] Step S222: Calculate the proportion of the strictly controlled group: Currently, only group A is under strict control, accounting for 1 / 5 = 20% < 30%, therefore it does not meet the conditions and will be transferred to S23.

[0077] Step S23: Sort by group size: The remaining four groups (B, C, D, and E) have the following number of terminals: Group B: 2000, Group C: 1500, Group D: 800, and Group E: 500. Sorted from largest to smallest, the top 30% has a ratio of 4 * 0.3 = 1.2, so the first group is selected, which is Group B. Therefore, Group B uses a strict update method (inclusion if there is a lag of >20 minutes). Groups C, D, and E use a more lenient update method: terminals must simultaneously meet the condition of having a lag of >20 minutes in the last 7 days and at least one instance of a lag of >5 minutes to be included in the model control.

[0078] Final update method summary: Groups A and B: Strict updates; Groups C, D, and E: Relaxed updates. Through this hierarchical strategy, model resources are prioritized for the most problematic groups (Groups A and B) and those terminals experiencing severe lag, achieving efficient management.

[0079] Specifically, such as Figure 4 As shown, the method for determining the coding optimization group in the coding control group is as follows: The core decision-making objective of this technical solution is to further identify which groups belong to the "coding optimization groups" requiring special attention from the "coding control groups" that have already implemented model dynamic control. These are groups that frequently adjust game coding parameters during the game operation of the model-controlled target, in order to verify and optimize the model's control effect. Terminals in the coding control group dynamically adjust game coding parameters through the model to suppress lag. However, the adjustment frequency and effectiveness vary among different groups: some groups adjust frequently but stably, while others may over-adjust or have poor results. By analyzing the parameter adjustment behavior of the model control target (i.e., the number of adjustments in different game runtime segments), terminals that adjust too frequently or are unstable (parameter adjustment targets) can be identified, thereby evaluating the stability and effectiveness of the entire group's adjustments. Combined with the group's original model control target update method (strict or lenient), it is ultimately determined whether the group needs to be focused on and optimized as a coding optimization group. If the group itself uses a strict update method, there is already sufficient data to verify the model's reliability, and no additional optimization is needed; if a lenient method is used, it is necessary to determine whether it should be included in the optimization scope based on the proportion and matching degree of the adjusted targets, in order to balance the sufficiency of model verification and the stability of system operation.

[0080] First, based on the number of parameter adjustments made by each model control terminal during different game runtime periods, a "frequent adjustment period" is defined, and "parameter adjustment targets" that are frequently adjusted during game execution are selected. Then, for each coding control group, it is categorized according to the update method of its model control targets: if it is a strict update, it is directly determined not to belong to the coding optimization group; if it is a lenient update, the proportion of parameter adjustment targets within the group (i.e., the frequent adjustment value) is calculated. If this value is low, it indicates stable adjustment and it does not belong to the optimization group; if it is high, the proportion of parameter adjustment targets to the total number of terminals in the group (adjustment matching coefficient) is further calculated. If this coefficient exceeds a threshold, it is determined to be a coding optimization group; otherwise, it does not. This hierarchical selection ensures that optimization resources are focused on groups with prominent adjustment problems that may affect model reliability.

[0081] S31 uses the adjustment data of the game coding parameters of the model control target to determine the number of times the game coding parameters of the model control target are adjusted in different game runtime segments, and uses the number of times the game coding parameters are adjusted to determine the frequent adjustment periods in the game runtime segment; It should be noted that the frequent adjustment period in the game runtime is the game runtime period in which the number of times the game encoding parameters are adjusted is greater than a preset adjustment number threshold.

[0082] Model control targets: These are terminals selected within the coding control group whose game coding parameters are dynamically adjusted by the model. These terminals are typically included in model control due to severe lag.

[0083] Game runtime period: refers to a complete game running process, during which the terminal is constantly in the game running process.

[0084] Number of times game encoding parameters are adjusted: This refers to the number of times the model adjusts the terminal's encoding parameters (such as resolution and frame rate) during each game runtime. These adjustments are proactively made based on performance predictions.

[0085] Frequent adjustment period: Defined as a game runtime period in which the number of adjustments exceeds a preset adjustment threshold. This threshold is used to distinguish between normal adjustments and excessive adjustments.

[0086] By tracking the number of adjustments over different time periods, we can reveal the performance sensitivity of the device during game execution. An excessive number of adjustments indicates frequent model intervention, potentially suggesting that the device is struggling to operate reliably and stably under the current model.

[0087] The significance of defining "frequent adjustment periods" lies in quantifying the intensity of adjustments, providing a basis for subsequent identification of terminals with abnormal adjustments.

[0088] S32 determines the parameter adjustment target in the model control target based on the frequent adjustment period of the model control target in the coded control group; It should be noted that the parameter adjustment target is a model control target in which the proportion of frequently adjusted time periods in the game runtime exceeds a preset proportion threshold.

[0089] Parameter adjustment targets: These refer to model control targets whose frequency of adjustment exceeds a preset threshold across all game runtime periods. Terminals exhibiting frequent adjustments across different time periods indicate abnormal parameter adjustment behavior and require close monitoring.

[0090] Frequent adjustments during a single game's cloud creation process may be due to the characteristics of that time period, but frequent adjustments across multiple processes indicate that the model is overly adjusting parameters for the terminal. By using a percentage threshold for filtering, we can focus on terminals that are generally over-adjusted. The significance of defining a "parameter adjustment target" is to identify individual terminals that require further analysis or optimization.

[0091] S33 determines whether the coding control group belongs to the coding optimization group based on the parameter adjustment target in the coding control group and the update method of the model control target in the coding control group.

[0092] Furthermore, if the update method for the model control target in the coding control group is to take the terminal as the model control target when the total duration of the lag period in the most recent preset time period does not meet the requirements, then the coding control group can provide more verification data for whether the model can reliably control the game coding parameters, thereby reducing lag. Therefore, it is determined that the coding control group does not belong to the coding optimization group.

[0093] Determine whether the model control target update method for this coding control group is "if the total duration of the lag period in the most recent preset time period does not meet the requirements, then the terminal is taken as the model control target" (i.e., strict update). If so, then it is determined that this group does not belong to the coding optimization group.

[0094] Additionally, it is understood that if the update method for the model control target in the coding control group does not fall under the category of "if the total duration of the lag period in the most recent preset time period does not meet the requirements," i.e., when the terminal is used as the model control target, the following situations apply: Case 1: Based on the proportion of parameter adjustment targets in the model control targets of the coding control group, determine the frequent adjustment value of the coding control group. If the frequent adjustment value of the coding control group is less than the preset adjustment threshold, it is determined that the parameter adjustment stability of the model control targets in the coding control group is relatively high, and therefore it does not belong to the coding optimization group. Frequent adjustment value: refers to the proportion of the number of parameter adjustment targets within a group to the total number of control targets of all models in that group.

[0095] Preset adjustment threshold: A pre-set ratio value used to determine the prevalence of abnormal terminal adjustments within a group.

[0096] If the proportion of parameter adjustment targets is low, it indicates that most models have stable control target adjustments, the overall group condition is good, and no optimization is needed.

[0097] Case 2: If the frequent adjustment value of the coding control group is not less than the preset adjustment threshold, obtain the number of parameter adjustment targets in the coding control group, determine the adjustment matching coefficient of the coding control group based on the proportion of the number of parameter adjustment targets in the terminal, and determine whether the adjustment matching coefficient of the coding control group is greater than the preset matching coefficient threshold. If yes, the coding control group is determined to belong to the coding optimization group; otherwise, the coding control group is determined not to belong to the coding optimization group.

[0098] Adjustment matching coefficient: This refers to the proportion of the number of parameter adjustment targets out of all terminals in the group (not just model control targets). This reflects the prevalence of adjustment anomalies across the entire group.

[0099] Preset matching coefficient threshold: A pre-set ratio value.

[0100] When parameter adjustment objectives constitute a significant portion of the model's control objectives, it's necessary to further examine their proportion within the entire group. A high proportion indicates that adjustment anomalies have been effectively validated under a large number of parameter adjustment objectives, suggesting a higher risk of frequent model adjustments. In such cases, the model should be treated as an optimization group.

[0101] In one possible embodiment: Group X: Update method: Strict update (if the lag duration exceeds the limit, it will be included in the model control); Total number of terminals: 200; Number of target units controlled by the model: 50 (included due to severe lag); Parameter adjustment target: Through the analysis of adjustment data of these 50 terminals, it was found that 10 terminals meet the condition of "frequent adjustment period ≥ 50%", and thus become the parameter adjustment target.

[0102] Since group X uses strict updates, according to step S33, it is directly determined that it does not belong to the coding optimization group.

[0103] Group Y: Update method: lenient update (only those that simultaneously meet the criteria of exceeding the limit for lag duration and having long lag periods are included in the model control), total number of terminals: 150, target number of models for control: 30; Parameter adjustment target: Among these 30 units, 12 are parameter adjustment targets.

[0104] Calculate the frequent adjustment value = 12 / 30 = 40% ≥ 30%, then proceed to case 2.

[0105] The matching coefficient was adjusted to 12 / 150 = 8% < 15%, therefore it was determined that it did not belong to the coding optimization group.

[0106] Specifically, the stuttering detection data of the model control target during the model control process is determined based on the stuttering detection period of the model control target during the model control process.

[0107] This technical solution provides a coding optimization group identification method based on model-controlled target adjustment behavior. By analyzing the adjustment frequency of terminals in different time periods, it identifies terminals with abnormal adjustments and then evaluates the adjustment stability of the entire group. It focuses optimization resources on the groups that truly need attention. For groups using strict update methods, the model effect can be directly verified due to the abundance of data, without the need for additional optimization. For lenient update groups, the matching coefficient is adjusted to avoid misjudgment due to insufficient data.

[0108] By using a two-level threshold (frequent adjustment value and adjustment matching coefficient) for screening, we ensured that groups with prominent adjustment problems were included in the optimization, while preventing the entire group from being frequently optimized due to problems with a few terminals, thus maintaining the stability of system operation.

[0109] Furthermore, the method for determining the parameter control method of the game coding for the coding optimization group under different model control objectives is as follows: The core decision-making objective of this technical solution is to develop a set of game coding parameter control methods for different "model control objectives" within terminal groups identified as "coding optimization groups." Coding optimization groups refer to terminal groups identified through steps S31 to S33 that exhibit frequent parameter adjustment behavior and unstable operation under the current model control. The existence of such groups indicates that the model's control over its internal terminals is ineffective and may require adjustment to determine whether it can operate reliably under stable coding parameters. However, whether to immediately adopt new control methods (such as fixed parameter operation) for these groups depends on several factors: First, it is necessary to evaluate the overall model recognition capability of all coding control groups, reflected by the average "model recognition matching coefficient." If the overall capability is insufficient, the status quo should be maintained to accumulate more data, avoiding data loss due to blind intervention. Second, it is necessary to examine the proportion of groups with matching deviations in other groups besides the current optimization group. If the proportion is high, it indicates that the model generally has recognition problems, and data accumulation is also needed rather than intervention. When all the above conditions are met, the decision is made based on the matching coefficient of the optimization group itself, the number of targets with stall risk, and the frequency of adjustments. The decision is then made to either run with fixed parameters (to verify if it is more stable) or continue maintaining model control. This tiered decision-making ensures that model iterations have sufficient data support, while targeted testing is conducted on unstable optimization groups to improve the overall reliability of control.

[0110] S41 uses the stuttering detection data of different model control targets in the coded control group during the model control process to determine the model control targets whose number of stuttering periods in the model control process does not meet the requirements, and uses them as stuttering risk control targets. Encoding control group: refers to the terminal model group that has been marked as encoding control (i.e., the game encoding parameters are set to the lowest state).

[0111] Model control target: refers to the terminal selected in the coding control group whose game coding parameters are dynamically adjusted by the model.

[0112] Stuttering detection data: This includes the stuttering periods recorded by the model control target during model intervention, as well as the number of stuttering periods.

[0113] Stuttering risk control target: Defined as the model control target where the number of stuttering periods exceeds a preset threshold (e.g., 3 times / week) during the model control process.

[0114] S42 determines the model identification matching coefficient of the coded control group based on the lag risk control target and simulation control target data in the coded control group; Model identification matching coefficient: defined as the difference between the proportion of normal control targets and the proportion of lag risk control targets, i.e., (total number of model control targets - number of lag risk control targets) / total number of model control targets. This coefficient reflects the model's effectiveness in identifying and controlling lag; a higher coefficient indicates a more effective model.

[0115] The success rate of model control can be intuitively expressed by calculating the difference, and the larger the total number of targets, the higher the statistical reliability of the coefficient.

[0116] S43 determines the parameter control method for game coding of the coding optimization group in different model control objectives by using the model identification matching coefficients in different coding control groups and the frequent adjustment values ​​of the coding optimization group.

[0117] Furthermore, the model identification matching coefficient of the coded control group is determined based on the difference between the proportion of simulated control targets and the proportion of lag risk control targets in the coded control group.

[0118] Furthermore, a parameter control method for determining the game coding of the coding optimization group under different model control objectives is specifically included, based on the model identification matching coefficients of different coding control groups and the frequent adjustment values ​​of the coding optimization group. Scenario 1: If the average matching coefficient of the model recognition of different coding control groups is less than the preset matching coefficient threshold, it is difficult to effectively determine whether the model can reliably solve the lag problem. Therefore, in order to effectively provide more data for model iteration, it is determined that the parameter control method of the game coding of the coding optimization group in different model control targets will not perform parameter control processing, and will still perform parameter control processing according to the original model.

[0119] First, calculate the average of the model recognition matching coefficients for all coded control groups, denoted as avg_match.

[0120] Scenario 1 (Data Accumulation Strategy When Overall Recognition Ability is Insufficient): If avg_match is less than the preset matching coefficient threshold, it indicates that the model's overall recognition ability is insufficient. In this case, several pre-optimization groups will interrupt the accumulation of data controlled by the original model, and running with fixed parameters cannot provide dynamic comparison information. Therefore, to ensure the reliability of subsequent model updates, all encoding optimization groups will not undergo parameter control processing, but will continue to accumulate data according to the original model's parameter control.

[0121] Scenario 2: If the average model recognition matching coefficient of different coding control groups is not less than the preset matching coefficient threshold, the coding control groups whose model recognition matching coefficient is less than the preset matching coefficient threshold are regarded as matching deviation groups. If the proportion of matching deviation groups in the coding control groups other than the coding optimization group does not meet the requirements, in order to effectively provide more data for model iteration, the parameter control methods of the game coding of the coding optimization group in different model control targets are determined not to perform parameter control processing, and parameter control processing is still performed according to the original model.

[0122] Scenario 2 (Data Accumulation Strategy When There Are Too Many Overall Deviation Groups): If `avg_match` is not less than a preset matching coefficient threshold, groups with matching coefficients less than the preset threshold are considered as deviation groups. For each coding optimization group, the proportion of deviation groups in other coding control groups (excluding that group) is calculated. If this proportion does not meet the requirements (i.e., exceeds the preset proportion threshold), it indicates that the model as a whole still has a large number of deviation groups. In this case, intervention on a single optimization group cannot solve the general problem and may even result in data loss. Therefore, this optimization group is not subject to parameter control and is still controlled according to the original model.

[0123] Scenario 3: If the proportion of matching deviation groups in the coding control group excluding the coding optimization group meets the requirements, and if there are no matching deviation groups in the coding control group excluding the coding optimization group, then the parameter control method for the game coding of the coding optimization group in different model control targets is determined to be the preset control method. That is, in model control targets where the number of stuttering periods within the most recent preset duration is less than the preset duration threshold, the game coding parameter with the longest runtime in the game coding parameters of the coding optimization group in different model control targets is used for parameter control processing to determine whether stuttering can be effectively suppressed and frequent adjustments avoided.

[0124] Scenario 3 (Fixed parameter optimization strategy for groups with good overall performance and no bias): If the proportion of matching biased groups in other groups besides the optimized group meets the requirements, and there are no biased groups, it indicates that the model performs well in other groups. For this frequently adjusted optimized group, a preset control method (basic version) is adopted, that is, among the model control targets where the number of stuttering periods within the most recent preset duration is less than a threshold, the game coding parameters with the longest runtime in this group are fixed to verify whether fixed parameters can improve stability.

[0125] Case 4: If there is a matching deviation group in the coding control group other than the coding optimization group, determine whether the model recognition matching coefficient of the coding optimization group is above the preset coefficient threshold. If not, it will not provide enough verification data for adjusting the deviation for model iteration. Therefore, the parameter control method of the game coding of the coding optimization group in different model control targets is determined to be the preset control method. That is, in the model control targets where the number of stuttering periods in the most recent preset duration is less than the preset duration threshold, the game coding parameter with the longest runtime in the game coding parameters of the coding optimization group in different model control targets is used for parameter control processing to determine whether stuttering can be effectively suppressed and frequent adjustments can be avoided. If so, proceed to the next step. Scenario 4 (Differentiation Strategy When a Deviation Group Exists and the Optimized Group's Own Control Effect is Poor): If a deviation group exists in other groups besides the optimized group, and the matching coefficient of that optimized group is lower than the preset coefficient threshold, it indicates that its own control effect is poor. In this case, it is necessary to make a comprehensive judgment based on the number of model-controlled targets and the number of stall risk targets within that group: if the number of targets is large and the number of risk targets is small, proceed to the next step; otherwise, the data is insufficient, and the preset control method (basic version) needs to be tested to determine whether it can operate reliably.

[0126] If the frequency adjustment value of the coding optimization group is greater than a preset frequent adjustment threshold, then the parameter control method for the game coding of the coding optimization group in different model control objectives is determined to be another control method. That is, in model control objectives where the number of stuttering periods in the most recent preset duration is less than the preset duration threshold, and the proportion of game runtimes with stuttering periods in history meets the requirements, the game coding parameter with the longest runtime in the game coding parameters of the coding optimization group in different model control objectives is used for parameter control processing to determine whether stuttering can be effectively suppressed and frequent adjustments can be avoided. If not, in order to effectively provide more data for model iteration, it is determined that the parameter control method for the game coding of the coding optimization group in different model control objectives will not be subject to parameter control processing, and the parameter control processing will still be performed according to the original model.

[0127] If there are mismatched groups in other groups besides the optimized group, and the matching coefficient of the optimized group itself is not lower than the preset coefficient threshold, then its frequent adjustment value is further examined. If the frequent adjustment value is greater than the preset frequent adjustment threshold, it indicates that the parameter adjustment is too frequent, which may affect stability. Other control methods need to be adopted, that is, the most commonly used parameters are fixed on terminals with fewer recent lags and fewer overall lag periods (terminals that also meet the conditions that the number of lag periods is less than the threshold and the proportion of historical lag periods is small). If the frequent adjustment value is not greater than the threshold, the adjustment behavior is normal, and the original model control is maintained.

[0128] To optimize the smoothness of its mobile game across different devices, a game company implemented coding control for multiple device model groups and identified four coding optimization groups (E, F, G, H). Based on data from all coding control groups, it is necessary to determine which parameter control method to use for the model control objective within each optimization group.

[0129] First, calculate the average matching coefficient for all 8 groups: (0.70 + 0.60 + 0.66 + 0.55 + 0.80 + 0.40 + 0.60 + 0.77) / 8 = (5.08) / 8 = 0.635 ≥ θ1=0.5, proceed to case 2.

[0130] Find the mismatch group (match coefficient < 0.5): there is only one group, F (0.40).

[0131] Now, we will evaluate each optimization group separately: 1. Optimize group E: There are 7 other groups besides E. Among them, the group with mismatch is F, with a proportion of 1 / 7 ≈ 14.3% < β=50%, which meets the requirements.

[0132] Due to the existence of a mismatch group (F), proceed to case 4.

[0133] E has a matching coefficient of 0.80 ≥ θ2=0.5, so we enter case 5.

[0134] Since the frequent adjustment value of E is 35% > γ=30%, other control methods are adopted.

[0135] Specific procedures: Within group E, select model control targets that have experienced fewer than 2 instances of stuttering in the last 7 days, and whose historical stuttering events account for less than 3% of all time periods. Assuming 18 out of 120 targets in group E meet these criteria, run the game with the longest runtime within the group on these 18 devices with fixed encoding parameters (e.g., 720p resolution, 30fps), and observe stuttering changes over a week to verify whether fixed parameters can effectively suppress stuttering and avoid frequent adjustments.

[0136] This technical solution provides a coding optimization group parameter control method based on model recognition matching coefficients and frequently adjusted values. When the overall recognition capability of the model is insufficient or there are many biased groups, the original control is maintained to accumulate data, avoid data loss due to intervention, and ensure that there are enough samples for subsequent model optimization.

[0137] The coding optimization group itself is characterized by frequent adjustments. By conducting targeted tests on fixed parameters, it is possible to verify whether the model is unstable due to over-adjustment, thus providing direction for model improvement. Based on indicators such as the group's own matching coefficient, the number of risk targets, and the frequency of adjustment, different preset control methods with different screening conditions can be used to achieve refined resource allocation.

[0138] For groups that may affect user experience, select terminals with less lag for trial testing to reduce risk; at the same time, ensure basic stability by using the parameter with the longest runtime.

[0139] Example 2 In a second aspect, the present invention provides a computer device, comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described game coding parameter optimization method based on hardware parameters when running the computer program.

[0140] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0141] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0142] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for optimizing game coding parameters based on hardware parameters, characterized in that, Specifically, it includes: Terminals are divided into different groups based on their model. Based on the lag monitoring data of the terminals in the group, an optimization method for the game coding parameters of the terminals in the group is determined. Based on the optimization method, a coding control group is determined in the group. In combination with the terminal change data in the coding control group, an update method for the model control target in the coding control group is determined. Based on the update method, the model control target is determined. According to the adjustment data of the game coding parameters of the model control target, and in combination with the update method of the model control target in the coding control group, the coding optimization group in the coding control group is determined. By utilizing stuttering detection data and model control target data from different model control targets within different coding control groups, the parameter control method for game coding of the coding optimization group in different model control targets is determined.

2. The game encoding parameter optimization method based on hardware parameters as described in claim 1, characterized in that, Terminals are divided into different groups based on their model number, specifically including: Terminals of the same model are grouped together.

3. The game encoding parameter optimization method based on hardware parameters as described in claim 1, characterized in that, The lag monitoring data of the terminal includes the lag periods and the duration of the lag periods during the operation of the game.

4. The game encoding parameter optimization method based on hardware parameters as described in claim 1, characterized in that, The method for determining the optimization method of game encoding parameters for terminals in the group is as follows: Based on the lag monitoring data of the terminals in the group, determine the lag periods of the terminals in the group during the game's operation; By utilizing the distribution data of lag periods in different game runtime segments, the terminal at risk of lag is identified. An optimization method for game encoding parameters of terminals in the group that are at risk of lag.

5. The game coding parameter optimization method based on hardware parameters as described in claim 4, characterized in that, The terminal with the risk of lag is a terminal that experiences lag periods in different game runtime segments.

6. The game encoding parameter optimization method based on hardware parameters as described in claim 4, characterized in that, If the number of terminals at risk of lag in the group does not meet the requirements, the optimization method for the game encoding parameters of the terminals in the group is to control the game encoding parameters to the minimum.

7. The game encoding parameter optimization method based on hardware parameters as described in claim 1, characterized in that, The method for determining the coding optimization group in the coding control group is as follows: Using the adjustment data of the game coding parameters of the model control target, the number of times the game coding parameters of the model control target are adjusted in different game runtime segments is determined, and the number of times the game coding parameters are adjusted is used to determine the frequent adjustment periods in the game runtime segments; Based on the frequent adjustment periods of the model control target in the coded control group, determine the parameter adjustment target in the model control target; Based on the parameter adjustment target and model control target update method in the coding control group, it is determined whether the coding control group belongs to the coding optimization group.

8. The game encoding parameter optimization method based on hardware parameters as described in claim 7, characterized in that, The parameter adjustment target is a model control objective where the proportion of frequently adjusted time periods during game runtime is above a preset threshold.

9. The game coding parameter optimization method based on hardware parameters as described in claim 7, characterized in that, If the update method for the model control target in the coding control group is to determine that the coding control group does not belong to the coding optimization group if the total duration of the lag period in the most recent preset time period does not meet the requirements.

10. A computer device comprising: A memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes a game coding parameter optimization method based on hardware parameters as described in any one of claims 1-9.

Images