Cross-platform use method for virtual game props

Abstract

The invention provides a cross-platform use method of game virtual props, and relates to the technical field of games. The method comprises the steps that a cross-platform use instruction is responded on a source game platform, and a cloud proxy server allocates a cloud execution instance for a target virtual prop and marks occupation; establishing a session with a target platform lightweight client; the client collects a touch operation to generate a standardized event, and packages and sends the event; the server enqueues the event, the cloud instance executes a prop logic script, and state change and audio-visual effect description are generated; the server monitors network parameters and selects a rendering mode to adapt to data transmission; the client monitors a rendering quality feedback adjustment mode or a sampling rate; after use, synchronizing the state to the source platform to remove the mark; according to the method, logic execution and expression of the props are separated, seamless use of the virtual props among different game platforms is achieved, and state consistency and operation coherence of the props are kept.

Description

Technical Field

[0001] This invention relates to the field of game technology, and more specifically, to a method for cross-platform use of virtual game items. Background Technology

[0002] In massively multiplayer online games (MMORPGs), virtual items are crucial assets for players and are typically tied to specific gaming platforms. With the development of cross-platform gaming, players expect to be able to continue using acquired items when switching between different devices.

[0003] However, in existing technologies, item data is stored locally or on a specific server. When players switch to other platforms, items cannot be used directly, resulting in a fragmented gaming experience. To address item-related issues, existing solutions, such as Chinese patent CN116785711A, disclose a method for using game items. This method detects item wear during combat and automatically switches to a backup item of the same type. However, this method focuses on item replacement within the same game and fails to solve the problem of item usage across different game platforms. Chinese patent CN115212572A discloses a method for controlling game items, which achieves cyclical switching of items of the same type by reusing item switching controls, optimizing interface interaction. However, this method is still limited to a single game platform and cannot achieve cross-platform synchronization of item status and operational adaptation.

[0004] Existing technologies lack a method that enables seamless use of virtual items across different game platforms, maintaining consistent status and smooth operation; therefore, a cross-platform method for using game virtual items is proposed to address the above issues. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a cross-platform method for using virtual game items, aiming to solve the technical problem that virtual items are bound to a specific game platform in the prior art, making it impossible to use them seamlessly between different game platforms, resulting in inconsistent item status and fragmented operation experience.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This application provides a method for cross-platform use of virtual game items, including the following steps: S1: In the source game platform, in response to the player's cross-platform use command for the target virtual item, the cloud proxy server obtains the item identifier of the target virtual item. and player identifier Verify whether the target virtual item is in an untraded and undisassembled state. If so, allocate an independent cloud execution instance to the target virtual item and mark the state of the target virtual item as cross-platform occupied on the source game platform.

[0007] By verifying that the item has not been traded or dismantled on the source platform, we ensure that it is in an operable state; by allocating an independent cloud execution instance to run the item logic in the cloud, we decouple the item behavior from the source platform. Marking a cross-platform occupied state is used to prohibit operations on the source platform during cross-platform use, thus avoiding state conflicts.

[0008] Furthermore, the "untraded and undisassembled" state means that the target virtual item has not been listed for trading or locked by disassembly operations in the source game platform. This verification is completed by the cloud proxy server calling the source game platform's transaction status API and disassembly status API.

[0009] The cloud proxy server allocates an independent cloud execution instance for the target virtual item, specifically including: Create a sandbox environment, and load the Lua logic script of the target virtual item and its current attack power value into the sandbox environment. and current durability value It then starts the Lua interpreter to simulate the execution of operation events; At the same time, according to the formula Calculate the comprehensive attribute value ,in For the first The preset weight coefficient of the item attribute, For the first The current value of the item attribute. The total number of attributes is used to initialize the state of the target virtual item in the cloud execution instance and to provide a basis for load balancing within the server.

[0010] S2: The cloud proxy server establishes a real-time communication session with the lightweight client on the target platform (tp) specified by the player, and generates a session token. It is also bound to the instance identifier ID. By establishing a real-time communication session, a channel is provided for the transmission of subsequent operation events and rendering data. The session token is used to identify this session and is bound to the cloud execution instance to ensure that data is routed to the correct processing unit.

[0011] S3: The lightweight client uses a preset sampling frequency of 120Hz. Collect player touch operation coordinates on the target platform The raw data, including pressure (pr) and area (ar), are processed by Kalman filtering to generate standardized operation events. These standardized operation events are then packaged into 16ms time windows and sent to the cloud proxy server.

[0012] Kalman filtering eliminates touch sampling noise, generating smooth operation trajectories; packetizing and sending within fixed time windows helps balance network load and real-time operation.

[0013] Furthermore, the sampling frequency can be dynamically adjusted according to the network conditions. The adjusted sampling frequency f' is determined by the formula... Calculation, where To adjust the coefficients, with values ​​ranging from 0.5 to 2.0 in increments of 0.1, the sampling frequency is reduced in weak network environments to decrease the amount of uplink data, while the sampling frequency is increased in high-quality networks to enhance operational accuracy.

[0014] S4: The cloud proxy server places the received operation event data into the operation queue in the order of timestamp tm. The cloud execution instance retrieves the operation events from the operation queue in sequence, simulates the execution of the operation events according to the logic script of the target virtual prop, and generates state change data and audiovisual effect description.

[0015] By processing events sequentially through an operation queue, the correct event order is ensured, when the total number of events per second... When the number of times exceeds 100, the operation events within the same 16ms time window are sampled and merged before execution. The sampling and merging means that for multiple operation events received within the same 16ms time window, only the event with the largest timestamp is kept and sent to the cloud execution instance for execution, while the other events are discarded. This strategy preserves the player's final operation intention and reduces the processing load while ensuring the integrity of the intention.

[0016] S5: The cloud proxy server monitors the current network bandwidth. ,Delay and packet loss rate ,in The unit is Mbps. The unit is milliseconds (ms). The value range is [0, 1]. Based on the pre-stored target platform rendering capability list, a rendering mode is selected from video stream mode, instruction set mode and keyframe mode. The audiovisual effect description is adapted to the target platform compatible rendering data and the rendering data is sent to the lightweight client for presentation.

[0017] By using real-time network quality monitoring and dynamic rendering mode selection, the audiovisual performance is adapted to the current network conditions.

[0018] Furthermore, quality score According to the formula Calculation, where , , Preset weighting coefficients and satisfying ; when Greater than the first threshold When selecting video stream mode, Less than the second threshold When selecting keyframe mode, when Greater than or equal to and less than or equal to Select instruction set mode at that time.

[0019] Specifically, if the video stream mode is selected, the cloud renderer is invoked to generate an H.264 video stream; if the instruction set mode is selected, JSON-formatted animation instructions and effect parameters are generated; if the keyframe mode is selected, key status information is sent at 100ms intervals, and the lightweight client performs interpolation rendering in conjunction with the Kalman prediction algorithm.

[0020] S6: The lightweight client monitors rendering latency. and client packet loss rate , will the and The data is sent back to the cloud proxy server, and the cloud proxy server, according to the... and Switch the rendering mode from video stream mode to instruction set mode, or reduce the operation event sampling rate from 120Hz to 60Hz.

[0021] The system dynamically adjusts based on rendering quality metrics provided by the client, switching to a lightweight mode or reducing the amount of upstream data when rendering quality deteriorates, thereby improving operational responsiveness and screen smoothness.

[0022] Furthermore, when ms or Furthermore, when the current mode is video stream, the rendering mode will be switched to instruction set mode, and the mode will not be switched again within 5 seconds after the switch to avoid oscillation. when ms or And when the current sampling rate is 120Hz, the data is sent down. Reduce the sampling rate to 60Hz.

[0023] S7: In response to the end of cross-platform use, the cloud proxy server synchronizes the status change data to the source game platform and updates the attack power value of the target virtual item. and durability value Remove the cross-platform occupation mark and restore the trading and dismantling operations of the target virtual item on the source game platform.

[0024] By synchronizing the state, we ensure that the attributes of the item on the source platform are consistent with the execution results in the cloud, and restore the operation permissions so that the item can continue to be used on the source platform.

[0025] The response to the end of cross-platform use includes any of the following situations: the player actively ends the use on the target platform, or the usage time reaches a preset value. Minutes, or network interruption lasting more than 5 seconds; When the cloud proxy server detects any of the above situations, it shuts down the cloud execution instance. The transaction operation includes listing the target virtual item for sale in the market of the source game platform; the disassembly operation includes breaking down the target virtual item into game materials; the status change data includes at least the updated durability value. .

[0026] The technical effects and advantages of this invention are as follows: This invention allocates an independent cloud execution instance to the target virtual item through a cloud proxy server, separating the item's logic execution from its presentation layer. The cloud execution instance runs the item's logic script, simulates the player's operation events on the target platform, and generates state change data and audiovisual effect descriptions. This method avoids copying item data to the target platform, solves the problem of multiple copy state conflicts, and ensures that the core state of the item is uniformly maintained in the cloud, thereby achieving consistency of item attributes across different platforms.

[0027] This invention collects player touch operations through a lightweight client, filters and generates standardized events, and packages and sends them according to time windows. The cloud server queues the events in sequence for execution by cloud instances. When the event frequency exceeds the limit, sampling and merging are performed, effectively dealing with the processing pressure caused by high-frequency operations. At the same time, the client dynamically adjusts the collection frequency according to network conditions, reducing the amount of uplink data in weak network environments, maintaining smooth operation response, and allowing players to have a consistent operating experience under different network conditions.

[0028] This invention calculates a quality score by monitoring network bandwidth, latency, and packet loss rate in real time. Based on preset thresholds, it selects an appropriate rendering mode from video streaming mode, instruction set mode, and keyframe mode. Video streaming mode is suitable for high-quality networks and is rendered and pushed from the cloud; instruction set mode is rendered locally on the client; and keyframe mode combines prediction algorithms to achieve smooth animation. This dynamic adaptation mechanism can optimize rendering effects according to network conditions, ensuring that players have a stable visual experience in different network environments.

[0029] This invention monitors rendering latency and packet loss rate on the client side and sends the data back to the server. The server then dynamically adjusts the rendering mode or operation event sampling rate accordingly. When rendering quality degrades, it promptly switches to a lightweight mode to prevent screen stuttering or increased operation latency. After use, the server synchronizes the final state to the source game platform, removes the occupancy mark, and restores the item's local operation permissions, forming a complete closed-loop process to ensure that the item's status remains consistent across platforms. Attached Figure Description

[0030] Figure 1 This is a flowchart illustrating the overall implementation process of the present invention; Figure 2 This is a flowchart illustrating the instance allocation and status marking process of the present invention. Figure 3 This is a flowchart of the operation event acquisition and transmission process of the present invention; Figure 4 This is a flowchart illustrating the rendering mode selection and data adaptation process of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Example 1 As attached Figures 1 to 4 The method for cross-platform use of virtual game items, as shown, separates the execution logic of virtual items from the presentation layer through a cloud proxy server, thereby achieving consistency in item status and continuity in cross-platform operation. The technical solution of this invention will be described in detail below with reference to steps S1 to S7.

[0033] Implementing this method requires deploying the following core components: The cloud-based proxy server is deployed on a cloud server cluster, running a Linux operating system and equipped with a Docker container engine to create an isolated sandbox environment. An in-memory database is deployed on the server cluster to store operation queues and item states. The server provides RESTful API interfaces, all of which require authentication tokens for verification.

[0034] The lightweight client is embedded in the target game as an SDK and supports mainstream mobile operating systems. The core modules of the client include: touch capture module, Kalman filter module, network communication module, and rendering module. The client automatically establishes a heartbeat connection with the cloud proxy server during initialization.

[0035] The original game platform is the platform where players initially hold virtual items, interacting with the cloud proxy server via API. The platform needs to provide interfaces for item status query, item locking, and item status update.

[0036] The specific implementation method of step S1 is as follows: S1: In the source game platform, in response to the player's cross-platform use command for the target virtual item, the cloud proxy server obtains the item identifier of the target virtual item. and player identifier Verify whether the target virtual item is in an untraded and undisassembled state. If so, allocate an independent cloud execution instance to the target virtual item and mark the state of the target virtual item as cross-platform occupied on the source game platform.

[0037] Players interact with target virtual items within the source game platform. The source game client's graphical user interface includes a "Cross-Platform Use" button. Clicking this button generates a cross-platform use command from the source game client, which includes the player's identifier. Props and signs The target platform identifier (tp) and timestamp (ts) are specified. The instruction is sent via HTTPS to the designated API interface of the cloud proxy server, with the request header carrying an authentication token.

[0038] After receiving the request, the cloud proxy server parses out... , The server first queries the internal distributed database for the item's status record. The status value is an enumeration type: 0 indicates normal, 1 indicates trading, 2 indicates dismantling, and 3 indicates cross-platform use. If the retrieved status is not 0, the server returns an error message to the source game client.

[0039] If the status is 0, the server further calls the API provided by the source game platform for real-time verification. It calls the transaction status API to check if the item is in a transaction, and the dismantling status API to check if the item is being dismantled. Verification succeeds when both APIs return negative results.

[0040] After successful verification, the cloud proxy server creates an independent cloud execution instance for the target virtual item. The server calls the container API to create a sandbox environment, pre-installed with a Lua script interpreter and necessary runtime libraries. It pulls the corresponding Lua logic script for the item from the script repository. The script defines the item's attack, skill, durability consumption, and other behavioral logic. The script is then copied to a specified path within the container.

[0041] The server calls the attribute query interface of the source game platform to obtain the current attack power value of the item. and durability value The server starts the Lua interpreter main process within the container, and... and The data is written to an in-memory database within the container, and the main process continuously monitors the subsequent operation queue. A unique instance identifier (id) is also assigned to each instance for subsequent session management.

[0042] To facilitate load balancing and resource scheduling within the server, a comprehensive attribute value is calculated. . The calculation formula is ,in The number of attributes involved in the calculation. For the first Preset weight coefficients for each attribute, For the first The current value of each attribute. Calculation result. The data is stored in the server load balancer's metadata for subsequent selection of a physical machine node with matching performance to handle the instance.

[0043] The cloud proxy server sends a status lock request to the source game platform, marking the target item as cross-platform occupied. Upon receiving the request, the source game platform updates the item's status to occupied in its local database, disables trading and dismantling related operation buttons on the game UI, and displays a "Cross-platform in use" message. A confirmation message is returned upon successful locking.

[0044] The server records this cross-platform usage session, stores the session information in the session table, and sets an expiration time.

[0045] The specific implementation method of step S2 is as follows: S2: The cloud proxy server establishes a real-time communication session with the lightweight client on the target platform based on the target platform identifier tp specified by the player.

[0046] The server queries the pre-configured platform access information based on the target platform identifier tp, including the communication protocol, STUN server address, TURN server address, maximum bitrate, etc.

[0047] The server generates a one-time session token. ,Will With current session information Bind and store the session in the session management table, and set an expiration date. Then... The access configuration information is returned to the source game client. The source game client then sends the information via a QR code. Show it to the players.

[0048] Players open the target game on the target platform device; the lightweight client SDK is already integrated into the game. Players can scan the QR code to obtain... Then, the SDK initiates a connection request to the session connection interface of the cloud proxy server, carrying... .

[0049] Server verification The validity of the request is verified by checking whether the request source matches the target platform identifier. Once verification is successful, a real-time communication channel is allocated to the session.

[0050] If the configured protocol is WebRTC, the server and client interact via ICE to establish a P2P or relay data channel. If the protocol is QUIC, the server and client establish a QUIC connection, supporting multiplexing.

[0051] Once the connection is established, the server binds the communication channel to the previously assigned instance ID. All data received from this channel will then be forwarded to the corresponding cloud execution instance for processing.

[0052] The server sends an initialization message to the client, which includes the initial rendering information of the prop, such as model identifier, initial animation state, and initial position coordinates. After receiving the message, the client calls the local rendering engine to load the corresponding model resources and places the prop in the initial position of the game scene.

[0053] The specific implementation method of step S3 is as follows: S3: Lightweight client with preset sampling frequency Collect player touch operation coordinates on the target platform The raw data, including pressure (pr) and area (ar), are processed by Kalman filtering to generate standardized operation events. These standardized operation events are then packaged into 16ms time windows and sent to the cloud proxy server.

[0054] The lightweight client renders a virtual joystick area or skill buttons in the game interface. When a player's finger touches the screen, the client obtains raw data through the touch API provided by the operating system.

[0055] The client uses a preset sampling frequency. Hz data is collected, and each time the coordinates, pressure value and contact area of ​​the current touch point are obtained. The coordinates are normalized and mapped to the range of [0, 1]. The pressure value and area value are directly returned by the system API, and the value range is also mapped to [0, 1].

[0056] Due to noise and jitter in touch sampling, the client performs Kalman filtering on the original coordinates. Kalman filtering is a recursive estimation algorithm that predicts the current position based on the state of the previous time step, and then corrects it using the current observations to output smoothed coordinates. The filtering parameters are preset in the client configuration file.

[0057] Smoothed coordinates and original timestamp tm, event type Pressure (PR) and area (AR) are combined to generate standardized operation events. tm is a millisecond-level timestamp after system startup. A value of 1 indicates the start of a touch, 2 indicates the movement of the touch, and 3 indicates the end of a touch.

[0058] The client maintains a circular buffer internally, and a timer is triggered every 16ms to perform packet transmission. When the timer is triggered, all events generated within the current 16ms window are retrieved from the buffer, serialized, and sent to the cloud proxy server through the established real-time communication channel.

[0059] To adapt to changes in network conditions, the client's listening control channel receive adjustment coefficient is adjusted. When a new When the value is reached, the client recalculates the sampling frequency. And adjust the timer interval. The frequency adjustment takes effect in the next acquisition cycle.

[0060] The specific implementation method of step S4 is as follows: S4: The cloud proxy server places the received operation event data into the operation queue in timestamp order. The cloud execution instance then retrieves the operation events from the queue sequentially, simulates the execution of the operation events according to the logic script of the target virtual item, and generates state change data and audiovisual effect descriptions; when the total number of events per second... When the number of occurrences exceeds 100, the operation events within the same 16ms time window are sampled, merged, and then executed.

[0061] After receiving the data packets from the client, the cloud proxy server deserializes them to obtain a list of events. For each event, the server inserts it into an ordered set in the in-memory database based on its timestamp (tm) and the instance identifier (id) bound to the current session, automatically sorting the events by timestamp to ensure they are stored in the order they occurred.

[0062] The event processing loop runs inside the cloud execution instance, using a blocking method to retrieve the earliest event from the queue.

[0063] The retrieved event data is passed to the event handling function of the Lua script via LuaCAPI. The script then processes the data according to the event type. Execute the corresponding game logic: like For touch-initiated events, record the starting position to prepare for skill release judgment. If For touch movement events, calculate the displacement vector and update the virtual joystick output direction. If The attack logic is triggered when the touch event ends and the location is in the skill button area.

[0064] When the attack logic is executed, the current attack power is read from the memory database. and remaining cooling time ,like The event is ignored; otherwise, the damage value is calculated. Update durability Set a cooldown time. Generate state change data and write it to the in-memory database. At the same time, generate audiovisual effect descriptions and send them to the message queue for subsequent processing by the rendering and adaptation module.

[0065] To prevent malicious high-frequency operations from overloading the server, a sampling and merging mechanism is introduced. The server counts the total number of operation events received per second. ,like Then, multiple events within each 16ms time window are sampled and merged, and only the event with the largest timestamp is enqueued for execution, while the rest are discarded, which reduces the processing burden and preserves the player's final operation intention.

[0066] The specific implementation method of step S5 is as follows: S5: Cloud proxy server monitors current network bandwidth. ,Delay and packet loss rate Based on the pre-stored rendering capabilities of the target platform, a rendering mode is selected from video stream mode, instruction set mode, and keyframe mode. The audiovisual effect description is adapted to the rendering data compatible with the target platform, and the rendering data is sent to the lightweight client for presentation.

[0067] The rendering adaptation module retrieves audiovisual effect descriptions from the message queue, while the cloud proxy server continuously monitors network quality parameters through a real-time communication channel with the lightweight client. (Bandwidth) Calculated based on the amount of data successfully transmitted in the most recent second, in Mbps; latency The round-trip time is measured in half using Ping messages, in milliseconds; packet loss rate. The loss rate is calculated using serial numbers, with values ​​ranging from [0, 1]. These parameters are updated periodically.

[0068] The server calculates a quality score based on network conditions. The calculation formula is: Weighting coefficient , , Pre-configured and meets The value is determined through offline experiments to reflect the user experience.

[0069] The server pre-stores information about the rendering capabilities supported by the target platform, including whether it supports video streaming mode, instruction set mode, keyframe mode, and corresponding parameter limitations.

[0070] server according to Value and two preset thresholds , Comparison result selection mode: If Select video stream mode, if Select keyframe mode, otherwise select instruction set mode.

[0071] Based on the selected mode, the audiovisual effect description is converted into rendering data that the client can recognize: If the video streaming mode is selected, the server uses a cloud renderer to generate continuous frames, compresses them into a video stream using an H.264 encoder, and sends it. The client receives the video stream, decodes it using hardware, and plays it directly.

[0072] If the instruction set mode is selected, the server converts the audiovisual effect description into a JSON format instruction sequence and sends it. The client then parses the instructions and invokes the local animation and sound effects systems for rendering.

[0073] If keyframe mode is selected, the server sends key status data packets at fixed time intervals, including timestamps, location coordinates, current durability, and skill cooldown status. The client uses the Kalman prediction algorithm to predict the intermediate frame state based on historical keyframes, achieving smooth animation.

[0074] The specific implementation method of step S6 is as follows: S6: Lightweight client-side monitoring of rendering latency and client packet loss rate , will the and The data is sent back to the cloud proxy server, which then processes it according to the specified parameters. and Switch the rendering mode from video stream mode to instruction set mode, or reduce the operation event sampling rate from 120Hz to 60Hz.

[0075] The lightweight client performs real-time statistics on service quality metrics during rendering. Rendering latency is also a factor. Record the time from receiving data to rendering completion, taking the average of the most recent measurements, in milliseconds. Client packet loss rate. The percentage of packets lost during the data reception process is calculated based on missing sequence numbers.

[0076] The client generates a QoS report every second, which includes the current timestamp and rendering latency. and packet loss rate It is sent to the cloud proxy server through the control channel.

[0077] After receiving the QoS report, the cloud proxy server dynamically adjusts its settings based on preset thresholds. If the current rendering mode is video stream mode, and ms or If this occurs, a rendering mode switch is triggered, sending a switch command to the client to switch to instruction set mode. A cooldown period is set after the switch, during which no further mode switching checks are performed to avoid frequent oscillations.

[0078] If the current operation event sampling rate is 120Hz, and ms or This triggers a sampling rate adjustment and sends out the adjustment coefficient. The sampling rate was reduced to 60Hz to reduce the amount of uplink data and alleviate network pressure.

[0079] The adjustment command is sent through the control channel and takes effect immediately upon receipt by the client.

[0080] The specific implementation method of step S7 is as follows: S7: In response to the end of cross-platform use, the cloud proxy server will synchronize the state change data to the source game platform and update the attack power value of the target virtual item. and durability value Remove the cross-platform occupation mark and restore the trading and dismantling operations of the target virtual item on the source game platform.

[0081] Cross-platform use is terminated under any of the following conditions: the player actively terminates use by sending an termination signal on the target platform client, or the usage duration reaches a preset value. A session is deemed invalid if a minute timer is triggered or if a network interruption lasts for more than 5 seconds.

[0082] After the termination process begins, the server reads the final state change data recorded in the cloud execution instance from the in-memory database, including at least the durability. If attack power Any changes that occur during gameplay will also be recorded.

[0083] Construct a state synchronization request by calling the source game platform's attribute update interface via an HTTP request, carrying an authentication token. Set timeout and retry mechanisms to ensure reliable state synchronization.

[0084] After receiving the request, the source game platform verifies the token's validity and updates the corresponding item's attack power value in its local database. and durability value And restore the item's status to normal.

[0085] After the status update is successful, the server calls the unlock interface of the source game platform to restore the availability of the trading and dismantling operation buttons.

[0086] Finally, the server performs resource cleanup: stops and destroys the container corresponding to the cloud execution instance, deletes the operation queue and state data in the in-memory database, closes the real-time communication connection, and archives the session records.

[0087] This concludes the complete cross-platform usage process for game virtual items. Players can continue to use the item on the original game platform, all status changes have been synchronized, and operation permissions have been restored.

[0088] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for cross-platform use of virtual game items, characterized in that, Includes the following steps: S1: In the source game platform, in response to the player's cross-platform use command for the target virtual item, the cloud proxy server obtains the item identifier of the target virtual item. and player identifier Verify whether the target virtual item is in an untraded and undisassembled state. If so, allocate an independent cloud execution instance to the target virtual item and mark the state of the target virtual item as cross-platform occupied on the source game platform. S2: The cloud proxy server establishes a real-time communication session with the lightweight client on the target platform (tp) specified by the player, and generates a session token. And bind it to the instance identifier ID; S3: The lightweight client uses a preset sampling frequency of 120Hz. Collect player touch operation coordinates on the target platform The raw data, including pressure (pr) and area (ar), are processed by Kalman filtering to generate standardized operation events. The standardized operation events are then packaged into 16ms time windows and sent to the cloud proxy server. S4: The cloud proxy server places the received operation event data into the operation queue in timestamp tm order. The cloud execution instance then retrieves the operation events sequentially from the operation queue, simulates the execution of the operation events according to the logic script of the target virtual item, and generates state change data and audiovisual effect descriptions; when the total number of events per second... When the number of occurrences exceeds 100, the operation events within the same 16ms time window are sampled, merged, and then executed. S5: The cloud proxy server monitors the current network bandwidth. ,Delay and packet loss rate ,in The unit is Mbps. The unit is milliseconds (ms). The value range is [0, 1]. According to the pre-stored target platform rendering capability list, a rendering mode is selected from video stream mode, instruction set mode and keyframe mode. The audiovisual effect description is adapted to the target platform compatible rendering data and the rendering data is sent to the lightweight client for presentation. S6: The lightweight client monitors rendering latency. and client packet loss rate , will the and The data is sent back to the cloud proxy server, and the cloud proxy server, according to the... and Switch the rendering mode from video stream mode to instruction set mode, or reduce the operation event sampling rate from 120Hz to 60Hz; S7: In response to the end of cross-platform use, the cloud proxy server synchronizes the status change data to the source game platform and updates the attack power value of the target virtual item. and durability value Remove the cross-platform occupation mark and restore the trading and dismantling operations of the target virtual item on the source game platform.

2. The method for cross-platform use of game virtual items according to claim 1, characterized in that, In step S3, the adjusted sampling frequency f' is calculated according to the formula... Calculation, where To preset the sampling frequency, The adjustment coefficient is set to a value ranging from 0.5 to 2.0, with a step size of 0.

1.

3. The method for cross-platform use of game virtual items according to claim 1, characterized in that, In step S5, the mass fraction According to the formula Calculation, where For network bandwidth, Due to network latency, For network packet loss rate, , , Preset weighting coefficients and satisfying ; when Greater than the first threshold When selecting video stream mode, Less than the second threshold When selecting keyframe mode, when Greater than or equal to and less than or equal to Select instruction set mode at that time.

4. The method for cross-platform use of game virtual items according to claim 3, characterized in that, In step S5, the cloud proxy server converts the audiovisual effect description into corresponding rendering data according to the selected rendering mode: If the video stream mode is selected, the cloud renderer will be invoked to generate an H.264 video stream; If the instruction set mode is selected, animation instructions and effect parameters in JSON format will be generated; If the keyframe mode is selected, key status information is sent at 100ms intervals, and the lightweight client performs tweening rendering in conjunction with the Kalman prediction algorithm.

5. The method for cross-platform use of game virtual items according to claim 1, characterized in that, In step S1, the cloud proxy server allocates an independent cloud execution instance for the target virtual item, including: The cloud proxy server creates a sandbox environment for the target virtual item, and loads the Lua logic script of the target virtual item and its current attack power value into the sandbox environment. and current durability value It then starts the Lua interpreter to simulate the execution of operation events; Among them, the comprehensive attribute value According to the formula calculate, For the first The preset weight coefficient of the item attribute, For the first The current value of the item attribute. The total number of attributes, the comprehensive attribute value Used to initialize the state of the target virtual item in the cloud execution instance.

6. The method for cross-platform use of game virtual items according to claim 1, characterized in that, In step S1, the untraded and undisassembled state means that the target virtual item has not been listed for trading or locked by disassembly operations in the source game platform; the verification is completed by the cloud proxy server calling the transaction status API and disassembly status API of the source game platform.

7. The method for cross-platform use of game virtual items according to claim 1, characterized in that, In step S4, the sampling merging refers to selecting only the event with the largest timestamp from multiple operation events received within the same 16ms time window and sending it to the cloud execution instance for execution, while discarding the remaining events.

8. The method for cross-platform use of game virtual items according to claim 4, characterized in that, In step S6, the cloud proxy server determines the rendering latency based on the... and client packet loss rate Adjustments will be made: when ms or Furthermore, when the current mode is video stream, the rendering mode will be switched to instruction set mode, and the mode will not be switched again within 5 seconds after the switch. when ms or And when the current sampling rate is 120Hz, the data is sent down. Reduce the sampling rate to 60Hz.

9. The method for cross-platform use of game virtual items according to claim 1, characterized in that, In step S7, the response to the end of cross-platform use includes any of the following situations: Players actively terminate their use on the target platform when the usage time reaches a preset value. Minutes, or network interruption lasting more than 5 seconds; When the cloud proxy server detects any of the above situations, it shuts down the cloud execution instance.

10. The method for cross-platform use of game virtual items according to claim 1, characterized in that, In step S7, the transaction operation includes listing the target virtual item for sale in the market of the source game platform, and the disassembly operation includes breaking down the target virtual item into game materials; the status change data includes at least the updated durability value. .

Images