Game resource loading decision-making method and device under comprehensive performance

Abstract

The invention provides a game resource loading decision-making method and device under comprehensive performance, and relates to the technical field of game resource dynamic management, and the method comprises the steps: carrying out the movement route prediction of a game role, and obtaining a plurality of predicted movement routes, a plurality of route prediction probabilities and a plurality of predicted movement speeds; calling resource quality grades, and performing loading quality optimization of a plurality of route map resources by combining a plurality of predicted moving speeds and integrating map loading quality and performance to obtain a plurality of optimal loading quality grades; and pre-loading the plurality of route map resources according to the plurality of optimal loading quality levels. According to the resource allocation method and device, the technical problem that due to the fact that an existing method lacks dynamic adaptability to real-time game scenes and player behaviors, timeliness and accuracy of game resource allocation are insufficient can be solved, resource allocation is carried out according to the predicted resource requirements, scientificity, timeliness and accuracy of resource allocation can be improved, and resource allocation efficiency is improved. The overall resource utilization rate is optimized, and the game experience is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic management of game resources, and in particular to a method and device for making decisions on loading game resources under comprehensive performance. Background Art

[0002] In the existing game development process, resource loading methods usually adopt preloading or static loading strategies to load resources (animation, sound effects, etc.) that may be used in the game into memory in advance to ensure that the required content can be smoothly presented when the player enters a specific area. However, with the complexity of game scenes and the dynamic nature of player interaction, the traditional static resource loading method has gradually revealed its limitations, especially in game environments with large resource demands and high loading requirements. Loading delays and screen freezes are prone to occur. For example, if the player suddenly accelerates to an unloaded area or changes the perspective quickly, the system often experiences screen delays and freezes due to untimely resource loading, which significantly affects the smoothness and immersion of the game.

[0003] In summary, existing resource loading methods lack dynamic adaptability to real-time game scenes and player behaviors, especially in open world or real-time competitive games, where player actions and scenes change frequently. Traditional loading methods are difficult to respond to these rapid changes in a timely manner, resulting in insufficient timeliness and accuracy in resource allocation. Summary of the invention

[0004] The purpose of the present invention is to provide a game resource loading decision method and device under comprehensive performance, so as to solve the technical problem that the existing game resource loading method lacks dynamic adaptability to real-time game scenes and player behaviors, resulting in insufficient timeliness and accuracy of game resource allocation.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a method for making game resource loading decisions under comprehensive performance, which is implemented by a game resource loading decision device under comprehensive performance, including: receiving a real-time position of a game character, predicting the movement route of the game character, obtaining multiple predicted movement routes and multiple route prediction probabilities, classifying movement speeds according to the multiple predicted movement routes, and obtaining multiple predicted movement speeds; according to the multiple route prediction probabilities, retrieving multiple route map resources of the multiple predicted movement routes, wherein the retrieval ratio of the multiple route map resources is the same as the multiple route prediction probabilities; retrieving the resource quality level set by the user in historical time, combining the multiple predicted movement speeds, and optimizing the loading quality of the multiple route map resources for the purpose of comprehensive map loading quality and performance, and obtaining multiple optimal loading quality levels; according to the multiple optimal loading quality levels, pre-loading the multiple route map resources to complete the loading of game resources.

[0007] In a second aspect, the present invention further provides a game resource loading decision device under comprehensive performance, which is used to execute a game resource loading decision method under comprehensive performance as described in the first aspect, including: a moving route prediction module, which is used to receive the real-time position of a game character, predict the moving route of the game character, obtain multiple predicted moving routes and multiple route prediction probabilities, and classify the moving speeds according to the multiple predicted moving routes to obtain multiple predicted moving speeds; a map resource retrieval module, which is used to retrieve multiple route map resources of the multiple predicted moving routes according to the multiple route prediction probabilities, wherein the retrieval ratio of the multiple route map resources is the same as the multiple route prediction probabilities; a loading quality optimization module, which is used to retrieve the resource quality level set by the user in the historical time, and optimize the loading quality of the multiple route map resources in combination with the multiple predicted moving speeds for the purpose of comprehensive map loading quality and performance to obtain multiple optimal loading quality levels; a game resource loading module, which is used to pre-load the multiple route map resources according to the multiple optimal loading quality levels to complete the game resource loading.

[0008] In a third aspect, the present invention further provides an electronic device, comprising:

[0009] At least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor so that the at least one processor can perform the steps of any one of the methods described in the first aspect above.

[0010] The beneficial effects of the present invention are:

[0011] By predicting the moving route according to the real-time position of the game character, multiple predicted moving routes and multiple route prediction probabilities are obtained, and the moving speeds are classified according to the multiple predicted moving routes to obtain multiple predicted moving speeds; then, according to the multiple route prediction probabilities, multiple route map resources of the multiple predicted moving routes are retrieved, wherein the retrieval ratio of the multiple route map resources is the same as the multiple route prediction probabilities; then, the resource quality level set by the user in the historical time is retrieved, and combined with the multiple predicted moving speeds, the loading quality of the multiple route map resources is optimized for the purpose of comprehensive map loading quality and performance, and multiple optimal loading quality levels are obtained; finally, the multiple route map resources are pre-loaded according to the multiple optimal loading quality levels to complete the loading of game resources. In other words, by accurately predicting the game resource requirements according to the actual game scene, and pre-configuring resources according to the predicted resource requirements, the scientificity, timeliness and accuracy of resource allocation can be improved, the overall resource utilization rate can be optimized, and the user's gaming experience can be significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A flowchart of a method for making decisions on game resource loading under comprehensive performance provided by the present invention;

[0013] Figure 2 A schematic diagram of the structure of a game resource loading decision system with comprehensive performance provided by the present invention;

[0014] Figure 3 This is a schematic structural diagram of an electronic device provided by the present invention.

[0015] In the accompanying drawings, the components represented by the reference numerals are described as follows:

[0016] Moving route prediction module 11, map resource retrieval module 12, loading quality optimization module 13, game resource loading module 14, electronic device 500, memory 510, processor 520, first computer program 511. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0018] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0019] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0020] For example, please refer to the attached Figure 1 The present invention provides a method for making a decision on game resource loading under comprehensive performance, which is applied to a device for making a decision on game resource loading under comprehensive performance, and specifically comprises the following steps:

[0021] S100: receiving the real-time position of the game character, predicting the moving route of the game character, obtaining multiple predicted moving routes and multiple route prediction probabilities, classifying the moving speeds according to the multiple predicted moving routes, and obtaining multiple predicted moving speeds.

[0022] In one embodiment, step S100 of the present invention further includes:

[0023] S110: Based on the game character operation data records within the user's historical time, a sample character position set is collected, and sample movement routes under different sample character positions are collected to obtain multiple sample movement route sets, and multiple sample route probability sets are labeled based on the proportion of different sample movement routes in each sample movement route set; S120: The sample character position set, multiple sample movement route sets and multiple sample route probability sets are used as supervised training data to train a movement route predictor; S130: The real-time position of the game character is received, input into the movement route predictor, and multiple predicted movement routes and multiple route prediction probabilities are predicted.

[0024] Specifically, the game character operation data records within the user's historical time are obtained, wherein the game is a racing game, a multiplayer competitive shooting game, etc., which can be set according to the actual game type; the historical time can be set in combination with the user's game scene, for example, within the past 3 months; the game character operation data records include the character position, the character position trajectory (moving route, etc.), the character movement speed and mode, etc. Then, the sample character position is collected according to the game character operation data record to obtain a sample character position set; then, the sample movement routes under different sample character positions in the sample character position set are collected to obtain multiple sample movement route sets, wherein the sample character position and the sample movement route set correspond one to one. Then, the proportion of different sample movement routes in each sample movement route set is counted, that is, the number of occurrences of each route is divided by the total number of routes of the position sample, to obtain the proportion of each route, and then the proportion of each sample movement route is used as its probability, recorded in the sample route probability set, indicating the possibility of the character choosing the route when starting from the position, and multiple sample route probability sets are obtained. For example, in the game, players starting from a certain building may have three common routes (such as going to resource point A, entering safe zone B, and avoiding combat zone C). By counting the frequency of players' choices in historical data, the probability of each route can be marked.

[0025] Then, a mobile route predictor is constructed based on the BP neural network. The mobile route predictor is a BP neural network model that can be iteratively optimized in machine learning and is obtained through supervised training. The mobile route predictor includes an input layer, multiple hidden layers and an output layer. The input data of the input layer is the character position, and the output data of the output layer is multiple mobile routes and multiple route probabilities. Then, the sample character position is used as input, and the sample mobile route set and the sample route probability set are used as output. The sample character position set, the multiple sample mobile route sets and the multiple sample route probability sets are used as supervised training data to perform supervised training on the mobile route predictor. First, the character position is input into the input layer of the predictor, and the data is transmitted layer by layer through the hidden layers until the output layer, and multiple mobile routes are output. and its probability; then use the loss function (such as cross entropy loss or mean square error) to calculate the difference between the movement route and probability predicted by the model and the true label (the set of sample movement routes and the set of sample route probabilities). The smaller the loss value, the higher the accuracy of the model prediction; further use the back propagation algorithm to calculate the gradient of the loss with respect to each layer of parameters, and pass the gradient back to each layer of the network to gradually update the network parameters; then use the optimization algorithm (such as SGD, Adam, etc.) to adjust the weights and bias parameters of the model so that the model gradually reduces the prediction error; finally, repeat the process of forward propagation, loss calculation, back propagation and parameter update until the loss of the model on the training set converges to a lower value, indicating that the model has learned to predict the movement route and probability from the input position, and then output the trained movement route predictor.

[0026] Finally, the real-time position of the game character (such as coordinates and orientation in the game) is received, and the real-time position is input into the mobile route predictor for route prediction, and multiple predicted mobile routes and multiple route prediction probabilities are output. By constructing a mobile route predictor, multiple possible mobile routes and probabilities can be effectively predicted based on the real-time position of the character in a real-time environment, providing precise support for subsequent dynamic resource loading and game performance optimization.

[0027] In one embodiment, step S100 of the present invention further includes:

[0028] S140: Based on the game character operation data records within the user's historical time, a sample movement route set is collected, and the average movement speeds of different sample movement routes are collected to obtain a sample movement speed set; S150: A mapping relationship between the sample movement route set and the sample movement speed set is constructed to obtain a movement speed predictor; S160: The multiple predicted movement routes are respectively input into the movement speed predictor, the sample movement speeds of the sample movement routes that are closest are associated, and a plurality of predicted movement speeds are obtained by classification.

[0029] Specifically, a sample movement route set is collected based on the game character operation data records within the user's historical time; then multiple historical movement speeds of each sample movement route are obtained, and the average of the multiple historical movement speeds is calculated to obtain the average movement speed, and a sample movement speed set is obtained, for example, the sample movement speed is that the game character moves 1m per second. Then, each route in the sample movement route set is mapped one-to-one with its corresponding average movement speed, and all sample movement routes and their corresponding average speeds are integrated to generate a sample route-average speed data set; then, based on the classification decision principle, the sample route is used as a child node, and the average speed is used as a leaf node of the child node, and a movement speed predictor is constructed according to the sample route-average speed data set.

[0030] Then, the multiple predicted movement routes are respectively input into the movement speed predictor for speed prediction, and in the movement speed predictor, a distance measurement method (such as Euclidean distance, cosine similarity, etc.) is used to calculate the similarity between each predicted route and each route in the sample movement route set. For each predicted movement route, the most similar sample movement route is found, that is, the distance or similarity between the two is the highest. This closest sample route can best reflect the movement characteristics of the current predicted route. Then the sample movement speed of the most similar sample movement route is obtained, and the sample movement speed is used as the predicted movement speed of the current predicted movement route to obtain multiple predicted movement speeds, in which the predicted movement speeds correspond to the predicted movement routes one by one. By constructing a movement speed predictor, the movement speed of the character under different routes can be efficiently predicted, providing support for the dynamic resource loading strategy in the game.

[0031] S200: Retrieving a plurality of route map resources of the plurality of predicted moving routes according to the plurality of route prediction probabilities, wherein a ratio of retrieving the plurality of route map resources is the same as the plurality of route prediction probabilities.

[0032] In one embodiment, step S200 of the present invention further includes:

[0033] S210: In the game resource library, all route map resources of the multiple predicted moving routes are retrieved; S220: In the multiple route map resources, multiple resource retrieval ratios are generated according to the sizes of the multiple route prediction probabilities, and the map resources closest to the multiple predicted moving routes in the multiple route map resources are respectively extracted to obtain multiple route map resources.

[0034] Specifically, all route map resources of the multiple predicted moving routes are retrieved in the game resource library, including terrain resources, building and structure resources, environmental objects, sound and special effect resources, etc., to obtain multiple route map resources. Then, multiple resource retrieval ratios are generated according to the size of the predicted probabilities of the multiple routes. First, the probability values ​​of all predicted routes are added to obtain the sum of the predicted probabilities. Then, the ratio of the predicted probability of each route to the sum of the predicted probabilities is used as the resource extraction ratio to obtain multiple resource retrieval ratios. For example, assuming that there is a first predicted route (predicted probability 20%), a second predicted route (predicted probability 30%) and a third predicted route (predicted probability 50%), the sum of the predicted probabilities is 1, and the resource extraction ratio of the first predicted route is 20%.

[0035] Then, among the multiple all-route map resources, the map resources closest to the multiple predicted moving routes are extracted according to the multiple resource extraction ratios. First, among the all-route map resources corresponding to each predicted moving route, the resources are sorted from near to far according to the distance to the predicted route; then, according to the resource retrieval ratio, a certain proportion of resources are extracted from the resources closest to the predicted route. For example, the map resources on both sides of the route can be divided into multiple parts according to the distance from the route, such as 1 part every 5% of the distance, and then retrieved according to the resource retrieval ratio. For example, if the resource retrieval ratio is 10%, the 10% map resources closest to the predicted route are loaded, that is, two nearest map resources are extracted; and multiple route map resources are obtained, wherein the route map resources correspond to the predicted moving route one by one. By using the strategy of extracting the nearest resources in proportion, resources can be effectively concentrated on high-probability paths, ensuring that users have a smooth gaming experience on the main paths, while saving resource overhead on low-probability paths.

[0036] S300: Retrieve the resource quality level set by the user in the historical time, combine the multiple predicted moving speeds, optimize the loading quality of the multiple route map resources for the purpose of comprehensive map loading quality and performance, and obtain multiple optimal loading quality levels.

[0037] In one embodiment, step S300 of the present invention further includes:

[0038] S310: Retrieve multiple historical resource quality levels of the user within a preset historical time range; S320: Calculate the average of the multiple historical resource quality levels to obtain the resource quality level.

[0039] Specifically, multiple resource quality levels recorded within a preset historical time range are retrieved from the user's historical game data. Each resource quality level represents the user's image quality preference in different time periods. For example, the level range is 1 to 5, where 5 represents the highest image quality and 1 represents the lowest image quality. Then, the average of the retrieved multiple historical resource quality levels is calculated, and the average calculation result is used as the resource quality level. By obtaining the resource quality level, the current resource quality level can be dynamically adjusted based on the user's historical preferences, ensuring that resource loading meets the user's expectations and optimizes system performance.

[0040] In one embodiment, step S300 of the present invention further includes:

[0041] S330: Obtain the median resource quality level, use the ratio of the resource quality level and the median resource quality level to perform a product correction calculation on the preset quality weight to obtain the quality weight, and use 1 minus the quality weight to obtain the performance weight, wherein the preset quality weight is 0.5; S340: Obtain the resource loading quality space, and randomly assign multiple first loading quality levels to the multiple route map resources within the resource loading quality space.

[0042] Specifically, the median resource quality level is obtained, and the median resource quality level is the median of the user's historical resource quality level range. For example, if the historical resource quality level ranges from 1 to 5, the median resource quality level is 3. Then, the preset quality weight is obtained, where the preset quality weight is 0.5, and then the ratio of the average resource quality level to the median resource quality level is used as the correction coefficient to perform a product correction calculation on the preset quality weight, that is, the correction coefficient is multiplied by the preset quality weight, and the product of the two is used as the corrected quality weight. For example, if the resource quality level is 2.7, the median resource quality level is 3, and the preset quality weight is 0.5, then the corrected quality weight is 0.9 times 0.5, which is 0.45; further, 1 is subtracted from the quality weight, and the difference between the two is used as the performance weight. For example, assuming that the quality weight is 0.45, the performance weight is 0.55.

[0043] Obtain a resource loading quality space, where the resource loading quality space is a resource quality loading range, which can be set according to actual scenarios, for example, the resource loading quality space is set to level 1 to level 8; then, randomly assign multiple first loading quality levels to the multiple route map resources in the resource loading quality space, that is, randomly select a loading quality value in the resource loading quality space, set it as the loading quality level of the route map resource, and obtain multiple first loading quality levels, wherein the first loading quality levels correspond to the route map resources one by one.

[0044] S350: performing loading speed analysis according to the multiple first loading quality levels and the multiple route map resources to obtain multiple first loading speed information;

[0045] In one embodiment, step S350 of the present invention further includes:

[0046] S351: According to the game resource loading history data of the device, collect a sample loading quality level set and a sample route map resource set, and collect the loading speed information under different sample loading quality levels and sample route map resources to obtain a sample loading speed information set; S352: Use the sample loading quality level set, sample route map resource set and sample loading speed information set as supervised training data to train a loading speed predictor; S353: Combine the multiple first loading quality levels and multiple route map resources respectively, input them into the loading speed predictor, and obtain multiple first loading speed information through predictive analysis.

[0047] Specifically, the game resource loading history data of the device is obtained, and the sample loading quality level and sample route map resources are collected based on the game resource loading history data, that is, the records of loading quality level and the route map resources used in the historical loading process are extracted from the historical data of the device to obtain a sample loading quality level set and a sample route map resource set. Then, the loading speed information under different sample loading quality levels and sample route map resources is collected. The loading speed can be expressed by the amount of resources loaded per unit time. For example, 5m of route map resources in the game map are loaded per second to obtain a sample loading speed information set.

[0048] Then, a loading speed predictor is constructed based on a BP neural network. The loading speed predictor includes an input layer, multiple hidden layers and an output layer. The input data of the input layer is the loading quality level and the route map resource, and the output data of the output layer is the loading speed information. Then, the sample loading quality level and the sample route map resource are used as training input data, and the sample loading speed information is used as training supervision data. The sample loading quality level set, the sample route map resource set and the sample loading speed information set are used as training data to supervise the training of the loading speed predictor. First, the sample loading quality level and the sample route map resource features are input into the network, and the predicted value of the loading speed is generated through the input layer, the hidden layer and the output layer; then, the mean square error is used as the loss function to calculate the error between the predicted loading speed and the actual loading speed, and the gradient of the loss function to each parameter is calculated, and the error is transmitted from the output layer back to the hidden layer and the input layer layer by layer, and the weight and bias of each layer are updated; the process of forward propagation, loss calculation, back propagation and parameter update is repeated until the value of the loss function converges on the training set, that is, the set error threshold is reached or the predetermined number of iterations is completed, and a trained loading speed predictor is obtained.

[0049] Then, the multiple first loading quality levels and multiple route map resources are combined respectively, that is, each route map resource corresponds to a first loading quality level, and the combined data is input into the loading speed predictor for prediction, and multiple first loading speed information is output, wherein the larger the loading quality level, the more route map resources, and the lower the loading speed.

[0050] S360: Calculate and obtain first loading fitness of the multiple first loading quality levels according to the quality weight, the performance weight, the multiple first loading speed information, the multiple predicted moving speeds, and the multiple first loading quality levels;

[0051] In one embodiment, step S360 of the present invention further includes:

[0052] S361: Among them, LOAD FIT is the loading fitness, M is the number of multiple route map resources, and w z is the quality weight, w x is the performance weight, ω i is the resource weight set according to the route prediction probability corresponding to the i-th route map resource, Z i is the loading quality level of the i-th route map resource, Z mid is the median resource quality level, V i is the loading speed information of the i-th route map resource, V yi is the predicted moving speed corresponding to the i-th route map resource.

[0053] Specifically, a loading fitness calculation formula is constructed. In the loading fitness calculation formula, LOAD FIT is the loading fitness. The larger the loading fitness, the better the overall loading effect. M is the number of multiple route map resources. z is the quality weight, for example, 0.45, w x is the performance weight, for example, 0.55, ω i is the resource weight set according to the route prediction probability corresponding to the i-th route map resource, where ω i is the ratio of the route prediction probability corresponding to the i-th route map resource to the sum of the M route prediction probabilities corresponding to the M route map resources; Z i is the loading quality level of the i-th route map resource, Z mid is the median resource quality level, V i is the loading speed information of the i-th route map resource, V yi is the predicted moving speed corresponding to the predicted moving route of the i-th route map resource.

[0054] Then, a loading fitness calculation formula is used to perform loading fitness calculation according to the quality weight, performance weight, multiple first loading speed information, multiple predicted moving speeds, and multiple first loading quality levels to obtain the first loading fitness of the multiple first loading quality levels. By constructing a loading fitness calculation formula, the accuracy, reliability, and efficiency of loading fitness calculation can be improved, thereby improving the efficiency and accuracy of loading quality optimization.

[0055] S370: Continue to optimize the loading quality of the multiple route map resources until the optimization converges, and output multiple optimal loading quality levels with the largest loading fitness.

[0056] Specifically, multiple second loading quality levels are randomly assigned to the multiple route map resources in the resource loading quality space, wherein the second loading quality level of each route map resource is different from the first loading quality level; and the same method for calculating the first loading fitness is used to calculate multiple second loading fitnesses of the multiple second loading quality levels. Then, the loading quality of the multiple route map resources is continuously optimized using the same method until a predetermined number of iterative optimizations is met, and multiple loading fitnesses are obtained; finally, multiple loading quality levels with the largest loading fitness are output and set as multiple optimal loading quality levels. By randomly generating a combination of loading quality levels in the resource loading quality space, calculating the loading fitness and iterating the optimization multiple times, the optimal combination of loading quality levels can be found to achieve a balance between loading speed and resource quality, thereby optimizing the loading performance and user experience of the game.

[0057] S400: Pre-loading the multiple route map resources according to the multiple optimal loading quality levels to complete the loading of game resources.

[0058] Specifically, multiple route map resources of multiple predicted moving routes are finally preloaded according to the multiple optimal loading quality levels, that is, each route map resource is loaded in advance. Preloading means that these resources will be loaded into the memory before the player approaches the area, thereby avoiding loading delays or screen freezes when the player actually enters, and completing the loading of game resources.

[0059] In summary, the game resource loading decision method under comprehensive performance provided by the present invention has the following technical effects:

[0060] By predicting the moving route according to the real-time position of the game character, multiple predicted moving routes and multiple route prediction probabilities are obtained, and the moving speeds are classified according to the multiple predicted moving routes to obtain multiple predicted moving speeds; then, according to the multiple route prediction probabilities, multiple route map resources of the multiple predicted moving routes are retrieved, wherein the retrieval ratio of the multiple route map resources is the same as the multiple route prediction probabilities; then, the resource quality level set by the user in the historical time is retrieved, and combined with the multiple predicted moving speeds, the loading quality of the multiple route map resources is optimized for the purpose of comprehensive map loading quality and performance, and multiple optimal loading quality levels are obtained; finally, the multiple route map resources are pre-loaded according to the multiple optimal loading quality levels to complete the loading of game resources. In other words, by accurately predicting the game resource requirements according to the actual game scene, and pre-configuring resources according to the predicted resource requirements, the scientificity, timeliness and accuracy of resource allocation can be improved, the overall resource utilization rate can be optimized, and the user's gaming experience can be significantly improved.

[0061] Embodiment 2: Based on the method for making a decision on game resource loading under comprehensive performance in the above embodiment, the present invention also provides a device for making a decision on game resource loading under comprehensive performance, as shown in the attached Figure 2 ,include:

[0062] A moving route prediction module 11 is used to receive the real-time position of a game character, predict the moving route of the game character, obtain multiple predicted moving routes and multiple route prediction probabilities, and classify the moving speeds according to the multiple predicted moving routes to obtain multiple predicted moving speeds; a map resource retrieval module 12 is used to retrieve multiple route map resources of the multiple predicted moving routes according to the multiple route prediction probabilities, wherein the retrieval ratio of the multiple route map resources is the same as the multiple route prediction probabilities; a loading quality optimization module 13 is used to retrieve the resource quality level set by the user in the historical time, and optimize the loading quality of the multiple route map resources in combination with the multiple predicted moving speeds for the purpose of comprehensive map loading quality and performance to obtain multiple optimal loading quality levels; a game resource loading module 14 is used to pre-load the multiple route map resources according to the multiple optimal loading quality levels to complete the loading of game resources.

[0063] Furthermore, the game resource loading decision device under the comprehensive performance is also used to: collect a sample character position set based on the game character operation data record in the user's historical time, and collect sample movement routes under different sample character positions to obtain multiple sample movement route sets, and according to the proportion of different sample movement routes in each sample movement route set, mark and obtain multiple sample route probability sets; use the sample character position set, multiple sample movement route sets and multiple sample route probability sets as supervised training data to train a movement route predictor; receive the real-time position of the game character, input it into the movement route predictor, and predict and obtain multiple predicted movement routes and multiple route prediction probabilities.

[0064] Furthermore, the game resource loading decision device under the comprehensive performance is also used to: collect a set of sample movement routes based on the game character operation data records within the user's historical time, and collect the average movement speeds of different sample movement routes to obtain a sample movement speed set; construct a mapping relationship between the sample movement route set and the sample movement speed set to obtain a movement speed predictor; input the multiple predicted movement routes into the movement speed predictor respectively, associate the sample movement speeds of the sample movement routes that are closest to each other, and classify to obtain multiple predicted movement speeds.

[0065] Furthermore, the game resource loading decision device under the comprehensive performance is also used to: retrieve all route map resources of the multiple predicted moving routes in the game resource library; generate multiple resource retrieval ratios in the multiple route map resources according to the sizes of the multiple route prediction probabilities, and respectively extract the map resources in the multiple route map resources that are closest to the multiple predicted moving routes to obtain multiple route map resources.

[0066] Furthermore, the game resource loading decision device under the comprehensive performance is also used to: retrieve multiple historical resource quality levels of the user within a preset historical time range; calculate the average of the multiple historical resource quality levels to obtain the resource quality level.

[0067] Furthermore, the game resource loading decision device under the comprehensive performance is also used to: obtain the median resource quality level, use the ratio of the resource quality level and the median resource quality level to perform a product correction calculation on the preset quality weight to obtain the quality weight, and use 1 minus the quality weight to obtain the performance weight, wherein the preset quality weight is 0.5; obtain the resource loading quality space, and randomly assign multiple first loading quality levels to the multiple route map resources in the resource loading quality space; perform loading speed analysis based on the multiple first loading quality levels and the multiple route map resources to obtain multiple first loading speed information; calculate the first loading fitness of the multiple first loading quality levels based on the quality weight, performance weight, multiple first loading speed information, multiple predicted moving speeds and multiple first loading quality levels; continue to optimize the loading quality of the multiple route map resources until the optimization converges, and output multiple optimal loading quality levels with the largest loading fitness.

[0068] Furthermore, the game resource loading decision device under the comprehensive performance is also used for: Among them, LOAD FIT is the loading fitness, M is the number of multiple route map resources, and w z is the quality weight, w x is the performance weight, ω i is the resource weight set according to the route prediction probability corresponding to the i-th route map resource, Z i is the loading quality level of the i-th route map resource, Z mid is the median resource quality level, V i is the loading speed information of the i-th route map resource, V yi is the predicted moving speed corresponding to the i-th route map resource.

[0069] Furthermore, the game resource loading decision device under the comprehensive performance is also used to: collect a sample loading quality level set and a sample route map resource set according to the game resource loading history data of the device, and collect loading speed information under different sample loading quality levels and sample route map resources to obtain a sample loading speed information set; use the sample loading quality level set, sample route map resource set and sample loading speed information set as supervised training data to train a loading speed predictor; combine the multiple first loading quality levels and multiple route map resources respectively, input them into the loading speed predictor, and predict and analyze to obtain multiple first loading speed information.

[0070] Example 3: Please refer to Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 500, including a memory 510, a processor 520, and a first computer program 511 stored in the memory 510 and executable on the processor 520, wherein the processor 520 implements the following steps when executing the first computer program 511: receiving a real-time position of a game character, predicting the moving route of the game character, obtaining multiple predicted moving routes and multiple route prediction probabilities, classifying moving speeds according to the multiple predicted moving routes, and obtaining multiple predicted moving speeds; retrieving multiple route map resources of the multiple predicted moving routes according to the multiple route prediction probabilities, wherein the retrieving ratio of the multiple route map resources is the same as the multiple route prediction probabilities; retrieving the resource quality level set by the user in the historical time, combining the multiple predicted moving speeds, optimizing the loading quality of the multiple route map resources for the purpose of comprehensive map loading quality and performance, and obtaining multiple optimal loading quality levels; pre-loading the multiple route map resources according to the multiple optimal loading quality levels to complete the loading of game resources.

[0071] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0072] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0073] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0074] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0075] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0076] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0077] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A game resource loading decision method under comprehensive performance, characterized in that: Methods include: Receiving the real-time position of the game character, predicting the movement route of the game character, obtaining multiple predicted movement routes and multiple route prediction probabilities, and classifying the movement speeds according to the multiple predicted movement routes to obtain multiple predicted movement speeds; Retrieving a plurality of route map resources of the plurality of predicted moving routes according to the plurality of route prediction probabilities, wherein a ratio of retrieving the plurality of route map resources is the same as the plurality of route prediction probabilities; Retrieving the resource quality level set by the user in the historical time, combining the multiple predicted moving speeds, optimizing the loading quality of the multiple route map resources for the purpose of comprehensive map loading quality and performance, and obtaining multiple optimal loading quality levels; According to the multiple optimal loading quality levels, the multiple route map resources are preloaded to complete the loading of game resources.

2. The game resource loading decision method under comprehensive performance according to claim 1, characterized in that: Receiving the real-time position of the game character, predicting the moving route of the game character, and obtaining a plurality of predicted moving routes and a plurality of route prediction probabilities, including: According to the game character operation data records in the user's historical time, a sample character position set is collected, and sample movement routes under different sample character positions are collected to obtain multiple sample movement route sets, and according to the proportion of different sample movement routes in each sample movement route set, multiple sample route probability sets are obtained by marking; Using the sample role position set, the plurality of sample movement route sets and the plurality of sample route probability sets as supervised training data to train a movement route predictor; The real-time position of the game character is received and input into the moving route predictor, and a plurality of predicted moving routes and a plurality of route prediction probabilities are predicted.

3. The game resource loading decision method under comprehensive performance according to claim 1, characterized in that: According to the multiple predicted moving routes, the moving speeds are classified to obtain multiple predicted moving speeds, including: According to the game character operation data records of the user in the historical time, a set of sample movement routes is collected, and the average movement speeds of different sample movement routes are collected to obtain a set of sample movement speeds; Constructing a mapping relationship between the sample moving route set and the sample moving speed set to obtain a moving speed predictor; The plurality of predicted movement routes are respectively input into the movement speed predictor, the sample movement speeds of the sample movement routes closest to them are associated, and a plurality of predicted movement speeds are obtained by classification.

4. The method for making decisions about game resource loading under comprehensive performance according to claim 1, characterized in that: Retrieving a plurality of route map resources of the plurality of predicted moving routes according to the plurality of route prediction probabilities includes: In the game resource library, retrieve all route map resources of the plurality of predicted movement routes; Among the multiple complete route map resources, multiple resource retrieval ratios are generated according to the sizes of the multiple route prediction probabilities, and the map resources closest to the multiple predicted moving routes among the multiple complete route map resources are respectively extracted to obtain multiple route map resources.

5. The method for making decisions about game resource loading under comprehensive performance according to claim 1, characterized in that: Retrieve the resource quality level set by the user in the historical time, including: Retrieving multiple historical resource quality levels of the user within a preset historical time range; The average of the multiple historical resource quality levels is calculated to obtain the resource quality level.

6. The method for making decisions about game resource loading under comprehensive performance according to claim 1, characterized in that: In combination with the multiple predicted moving speeds, the loading quality of the multiple route map resources is optimized for the purpose of comprehensive map loading quality and performance, and multiple optimal loading quality levels are obtained, including: Obtaining a median resource quality level, using the ratio of the resource quality level to the median resource quality level to perform product correction calculation on a preset quality weight to obtain a quality weight, and subtracting the quality weight from 1 to obtain a performance weight, wherein the preset quality weight is 0.5; Acquire a resource loading quality space, and randomly assign a plurality of first loading quality levels to the plurality of route map resources in the resource loading quality space; Performing loading speed analysis according to the plurality of first loading quality levels and the plurality of route map resources to obtain a plurality of first loading speed information; Calculate and obtain first loading fitness of the multiple first loading quality levels according to the quality weight, the performance weight, the multiple first loading speed information, the multiple predicted moving speeds and the multiple first loading quality levels; The loading quality of the multiple route map resources is continuously optimized until the optimization converges, and multiple optimal loading quality levels with the largest loading fitness are output.

7. The method for making decisions about game resource loading under comprehensive performance according to claim 6, characterized in that: According to the quality weight, the performance weight, the plurality of first loading speed information, the plurality of predicted moving speeds and the plurality of first loading quality levels, the first loading fitness of the plurality of first loading quality levels is calculated and obtained as follows: Among them, LOAD FIT is the loading fitness, M is the number of multiple route map resources, and w z is the quality weight, w x is the performance weight, ω i is the resource weight set according to the route prediction probability corresponding to the i-th route map resource, Z i is the loading quality level of the i-th route map resource, Z mid is the median resource quality level, V i is the loading speed information of the i-th route map resource, V yi is the predicted moving speed corresponding to the i-th route map resource.

8. The method for making decisions about game resource loading under comprehensive performance according to claim 6, characterized in that: Performing a loading speed analysis according to the plurality of first loading quality levels and the plurality of route map resources to obtain a plurality of first loading speed information includes: According to the game resource loading history data of the device, a sample loading quality level set and a sample route map resource set are collected, and loading speed information under different sample loading quality levels and sample route map resources is collected to obtain a sample loading speed information set; The sample loading quality level set, the sample route map resource set and the sample loading speed information set are used as supervised training data to train a loading speed predictor; The multiple first loading quality levels and multiple route map resources are combined respectively, input into the loading speed predictor, and predictive analysis is performed to obtain multiple first loading speed information.

9. A game resource loading decision device under comprehensive performance, characterized in that: The steps for implementing the game resource loading decision method under comprehensive performance as described in any one of claims 1 to 8 include: A moving route prediction module is used to receive the real-time position of the game character, predict the moving route of the game character, obtain multiple predicted moving routes and multiple route prediction probabilities, and classify the moving speeds according to the multiple predicted moving routes to obtain multiple predicted moving speeds; A map resource retrieval module, configured to retrieve a plurality of route map resources of the plurality of predicted moving routes according to the plurality of route prediction probabilities, wherein a retrieval ratio of the plurality of route map resources is the same as the plurality of route prediction probabilities; A loading quality optimization module, used to retrieve the resource quality level set by the user in the historical time, and optimize the loading quality of the multiple route map resources in combination with the multiple predicted moving speeds for the purpose of comprehensive map loading quality and performance, so as to obtain multiple optimal loading quality levels; The game resource loading module is used to preload the multiple route map resources according to the multiple optimal loading quality levels to complete the game resource loading.

10. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the steps of the game resource loading decision method under comprehensive performance as described in any one of claims 1 to 8.

Images