CN104658041A - Dynamic scheduling method for entity models in distributed three-dimensional virtual environment - Google Patents
Dynamic scheduling method for entity models in distributed three-dimensional virtual environment Download PDFInfo
- Publication number
- CN104658041A CN104658041A CN201510077170.4A CN201510077170A CN104658041A CN 104658041 A CN104658041 A CN 104658041A CN 201510077170 A CN201510077170 A CN 201510077170A CN 104658041 A CN104658041 A CN 104658041A
- Authority
- CN
- China
- Prior art keywords
- model
- solid model
- solid
- template
- dimensional virtual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention provides a dynamic scheduling method for entity models in a distributed three-dimensional virtual environment. The method comprises steps as follows: step 1, a layering data structure of each entity model in the distributed three-dimensional virtual environment is established; step 2, a mapping relation of an entity model status information data packet sent by each simulation node and a model type template is established; step 3, a buffer pool M1 and a storage pool M2 are established; step 4, visibility of the entity models is judged according to the entity model status information data packet, model components required by the entity models are loaded according to the mapping relation of the step 2, and the entity models are subjected to regrouping operation with adoption of layering data structures established in the step 1; step 5, the variety and the number of the models in the buffer pool M1 established in the step 3 are dynamically adjusted in real time. On the basis that the dynamic entity fidelity in the three-dimensional virtual environment is not reduced, internal and external memory data interaction frequency of the entity models is substantially reduced, so that the dynamic scheduling process of the whole scene resources is rapid and smooth.
Description
Technical field
The invention belongs to 3-D Visualization Simulation field, particularly relate to the solid model dynamic dispatching method in this kind of large-scale complex virtual environment of such as Virtual Battlefield.
Background technology
Distributed Three-dimensional virtual environment is a kind of real-time network interactive environment based on virtual reality technology and network technology, it is by the communication between the computing node that is distributed in various places, share local or the global data information of virtual three-dimensional environment, work in coordination with being distributed on many computing nodes of task, provide virtual three-dimensional space that is a kind of entirety, real, that can immerse to user.Distributed Three-dimensional Virtual Environment is due to its favorable expandability, and the features such as wide adaptability, are widely used in Military Simulation, network amusement etc.
Along with the development of computer graphics techniques, the graphical effect of Distributed Three-dimensional virtual environment is more and more true to nature, bring the sharply increase of memory space and operand thereupon, for this problem, usually adopt out-of-core technology contextual data dynamically to be called in as required at present and recall internal memory, but because external memory data acquisition speed is comparatively slow, inside and outside deposit data will inevitably cause system delay alternately frequently, makes to play up frame per second and becomes very unstable.
For the static model (model that whole render process spatial location and state remain unchanged) in virtual environment, mechanism is it is predicted by call number on the basis of data dispatch, namely prejudge and obtain and the closely-related data set of tomorrow requirement, the scheduling quantum that may increase suddenly on average arrives multiple interframe and completes, and ensures that render smooth is continuous.But for the dynamic model in virtual environment, owing to there is the solid model of a large amount of complex structures in Distributed Three-dimensional virtual environment, and these model space positions and state are among random dynamic change always, the call number mechanism of it is predicted cannot predict, existing technology thus cannot be adopted to realize dynamic dispatching.
Summary of the invention
For solving the problem, the invention provides a kind of solid model dynamic dispatching method of Distributed Three-dimensional virtual environment, it significantly reduces deposit data frequency of interaction inside and outside solid model on the basis not reducing scenario entities fidelity in three-dimensional virtual environment, thus makes whole scenario resources dynamic dispatching process rapidly, smoothly.
The solid model dynamic dispatching method of Distributed Three-dimensional virtual environment of the present invention, its method comprises the following steps:
Step 1, sets up the hierarchical data structure of each solid model in described Distributed Three-dimensional virtual scene, and parts each in described solid model are saved as a model resource file respectively;
Wherein, described hierarchical data structure is: the hierarchical structure of each parts, each component names, the local coordinate system of each parts and the anchor point of each parts in solid model;
Step 2, set up the mapping relations of solid model state information data packet PDU that in described Distributed Three-dimensional virtual scene, each simulation node sends and model class template in the mode of external file, described mapping relations are: the model class name template of solid model ID, serviceable condition state and model class template, derive from mode code mod one_to_one corresponding;
Wherein, described solid model state information data packet PDU comprises: the three dimensional space coordinate of solid model ID, solid model and serviceable condition state; The content of described model class template comprises model class name template, basic model and Derived model; Described basic model comprises: model resource file, basic components identification code that the step 1 in the external memory storage that component layer level structure, basic components are corresponding is preserved; Described Derived model comprises: derive from mode code mod, derive from assembling side-play amount and the assembling relative direction of model resource file, derivation component identification code and derivation parts that the step 1 in external memory storage corresponding to parts is preserved;
Step 3, creates and reclaims the Buffer Pool M1 of invisible solid model and deposit the storage pool M2 of all visible solid model indexes three-dimensional virtual environment from three-dimensional virtual environment;
Step 4: obtain present frame three-dimensional scenic view information, and the visibility judge of solid model is carried out in conjunction with described solid model state information data packet, the sightless solid model be present in scene is put into Buffer Pool M1, and its model is set to invisible, the index of its model is deleted from storage pool M2; To not existing in scene or existed but the visible solid model that changes of serviceable condition, according to there is corresponding model class model instance situation in the mapping relations in step 2 and Buffer Pool M1, or carry out solid model and reuse operation or carry out solid model reorganization operation or carry out solid model reconstructed operation.
Further, described step 4 comprises:
Step 40, obtains the PDU that present frame three-dimensional scenic view information and all simulation nodes send, and wherein three-dimensional scenic viewpoint data comprise the position of viewpoint, direction, horizontal field of view angle and visibility;
Step 41, according to the three dimensional space coordinate of solid model and three-dimensional scenic viewpoint data calculate each solid model viewing angle and look eye-distance from, and Ergodic judgement is carried out to all PDU received from network: if viewing angle and depending on eye-distance within the range of observation being all in setting and field angle, then solid model is visible solid model, performs step 44; Otherwise solid model is invisible solid model, perform step 42;
Step 42, with the solid model index of solid model ID for its correspondence in keyword search M2, if searched, then illustrates that this solid model exists in three-dimensional virtual scene, then perform step 43; If do not searched, then illustrate that this entity does not exist in scene, do not need to process;
Step 43, adds Buffer Pool M1 by the index of solid model and solid model, template and mod, and is set to invisible by this solid model; The index of this solid model is deleted from M2 simultaneously;
Step 44: with the solid model index of solid model ID for its correspondence in keyword search M2:
If searched, then illustrate that this solid model exists in three-dimensional virtual scene, judge whether the state of this solid model changes further, if changed, first perform step 43, then perform step 45; If do not changed, illustrate that this solid model does not need to upgrade, not operation dispatching process;
If do not searched, illustrate that this model does not also join three-dimensional virtual scene, perform step 45;
Step 45, according to template and mod obtaining model class template in the mapping relations that solid model ID and state sets up from step 2, and retrieve in M1, if the solid model that the value retrieving template and mod is all mated, then this model is performed to the model reusability operation of step 46; If only have the value of template to mate, then enter the model reorganization operation of step 47; If template and mod does not mate, then enter the model reconstruction operation of step 48;
Step 46, model reuse operation: the index of solid model is added M2, and is set to visible, the index of model, template and mod are deleted from Buffer Pool M1 simultaneously;
Step 47, the reorganization operation of model: search template identical in Buffer Pool M1, the model that mod absolute difference is minimum, is called master pattern by this model, is called object module by needing the model of restructuring; The derivation mode finding derivation mode corresponding to object module mod to describe T and master pattern mod corresponding in the template resource file that template is corresponding describes O, first component identification codes all in T is traveled through, parts corresponding in master pattern are replaced with the parts that T specifies; Then travel through all component identification codes in O, first whether decision means identification code exists in T, if exist, does not then carry out any operation, if do not exist, then parts corresponding in master pattern is replaced back parts corresponding to basic model; Master pattern is performed to the model reusability operation of step 46;
Step 48, Construction of A Model operates: according to the basic model component layer level structure in the template resource file that template is corresponding, load the model resource file of basic components in external memory storage successively, construct basic model B according to top-down order; If mod is not equal to 0, according to step 47 pair basic model B execution model reorganization operation; Model index after restructuring is added M2, and is set to visible.
Further, the method also comprises the step of the model instance quantitative proportion of each model class in dynamic conditioning Buffer Pool M1: recall all model class in section sometime and add up stored in the model instance quantity stored in the model instance quantity of Buffer Pool and Current buffer pond, if stored in quantity much larger than recalling quantity, then the model instance quantity of this model class in Buffer Pool is reduced by half; If recalling quantity is greater than stored in quantity, then by the model instance of this model class in Buffer Pool by copying, its quantity increase is recalled with stored in two times of difference.
Further, the method also comprises the step of dynamic conditioning Buffer Pool M1 size: add up the model instance total quantity of model class all in Buffer Pool in section sometime, if be greater than a certain threshold value, within this time period, the variance yields of the model instance quantity of different model class is as priority, deletes the model instance of model class.
Beneficial effect:
Solid model dynamic dispatching method of the present invention builds based on the model data structures of the stratification of model assembly-model class-model entity by xml template, the component-level achieving solid model in dynamic dispatching process is reused, minimizing model geometric data and data texturing are to the consumption of memory source, and reduce the frequency of interaction of interior external storage, guarantee that virtual scene is played up real-time and level and smooth.
The present invention is directed to transaction quantity in Large-scale Distributed Virtual Environment many, kind relatively less and distribution random feature, adopt real-time scheduling's the reusing and recombinating by model assembly based on Buffer Pool, to greatest extent the model data in internal memory is reused, and according to ought for the previous period in call in the quantity of different model class example in the classification that recalls Buffer Pool model and Number dynamics adjustment Buffer Pool, to improve the efficiency of model reusability and reconstruct, dynamic conditioning is carried out to Buffer Pool inner model total quantity simultaneously, with controlled entity model resource to the consumption of memory source.
Accompanying drawing explanation
Fig. 1 is solid model structural representation in the solid model dynamic dispatching method of Distributed Three-dimensional virtual scene of the present invention;
Fig. 2 is Stryker model class schematic diagram of illustrating in the solid model dynamic dispatching method of Distributed Three-dimensional virtual scene of the present invention;
Fig. 3 is model class template schematic diagram in the solid model dynamic dispatching method of Distributed Three-dimensional virtual scene of the present invention;
Fig. 4 is the mapping relations schematic diagram of solid model and model class template in the solid model dynamic dispatching method of Distributed Three-dimensional virtual scene of the present invention;
Fig. 5 is solid model visibility judge schematic diagram in the solid model dynamic dispatching method of Distributed Three-dimensional virtual scene of the present invention;
Fig. 6 is the solid model dynamic dispatching method process flow diagram of Distributed Three-dimensional virtual environment of the present invention.
Embodiment
As shown in Figure 6, the solid model dynamic dispatching method of Distributed Three-dimensional virtual environment of the present invention comprises the following steps:
Step 1: the hierarchical data structure setting up solid model in Distributed Three-dimensional virtual scene.Concrete method is:
Step 11: the hierarchical data structure of solid model comprises the hierarchical structure of each parts in solid model as shown in Figure 1, each component names, the local coordinate system of each parts and the anchor point (solid model carries out reference point during parts assembling) of each parts, wherein the anchor point of same class parts (the main weapon as armored vehicle) only has one, translation change is carried out by arranging side-play amount during assembling, relative angle is set and carries out rotational transform, adjust the relative space relation between the local coordinate system of each parts and anchor point, finally each parts are saved as a model resource file respectively.
Step 12: the model class template adopting all solid models in xml document definition Distributed Three-dimensional virtual scene, the solid model that the corresponding class formation of each model class template is close.
As shown in Figure 3, the content of described model class template comprises model class name template, basic model (being similar to base class) and Derived model (being similar to derived class).
Wherein basic model comprises: model resource file, basic components identification code that the step 11 in the external memory storage that component layer level structure, basic components are corresponding is preserved;
Derived model comprises: derive from mode code mod, derive from assembling side-play amount and the assembling relative direction of model resource file, derivation component identification code and derivation parts that the step 1 in external memory storage corresponding to parts is preserved;
Above-mentioned derivation parts and basic components are mated by identification code; The assembling side-play amount and the assembling relative direction that derive from parts refer to that the local coordinate system deriving from parts is for the relative position of anchor point and relative angle.Represent basic model when derivation mode code is 0, derive from the degree of approximation of the value reflection derivation mode of mode code, the absolute difference namely deriving from mode code is less, represents that the difference between Derived model is less, can be recombinated by fewer operation of components.
The structure of basic model and composition are directly derived by modeling tool, wherein implies anchor point information and hierarchical relationship.The all parts local coordinate system of basic model is consistent with the anchor point of these parts, namely do not need to carry out translation change and rotational transform during all basic model parts assemblings, side-play amount and relative direction are all (0,0,0,), Stryker (Stryker) the serial battlebus of us army as shown in Figure 2 may be defined as a model class.
Step 2: the mapping relations setting up solid model PDU that in Distributed Three-dimensional virtual scene, each simulation node sends and model class template in the mode of external file, the solid model PDU (state information data packet) that wherein each simulation node sends comprises a solid model ID, the three dimensional space coordinate of solid model and serviceable condition state (intact, light damage, middle damage, heavily damage); Mapping relations are: the model class name template of solid model ID, serviceable condition state and model class template and the mode of derivation code mod one_to_one corresponding, as shown in Figure 4.
Step 3: virtual scene initialization.Two continuous print storage spaces are created, respectively as storing the Buffer Pool M1 of the invisible solid model resource reclaimed from three-dimensional virtual environment and depositing the storage pool M2 of all visible solid model indexes three-dimensional virtual environment in internal memory.
Step 4: adopt the real-time scheduling of Buffer Pool to dispatch solid model;
Step 40: contextual data and distributed simulation node data obtain.Play up frame at each to start, obtain the solid model PDU that present frame three-dimensional scenic view information and all simulation nodes send, wherein three-dimensional scenic viewpoint data comprise the position of viewpoint, direction, horizontal field of view angle and visibility;
Step 41: Solid simulation model visibility judge.The solid model PDU that all simulation nodes of present frame send is traveled through, according to the three dimensional space coordinate of solid model and three-dimensional scenic viewpoint data calculate each solid model viewing angle and look eye-distance from, as shown in Figure 5, if viewing angle and depending on eye-distance within the range of observation being all in setting and field angle, then solid model is visible solid model, performs step 44; Otherwise solid model is invisible solid model, perform step 42.
Step 42: the process of invisible solid model.For sightless solid model, first with the solid model index of solid model ID for its correspondence in keyword search storage pool M2, if searched, then illustrate that this solid model exists in three-dimensional virtual scene, then perform step 43; If do not searched, then illustrate that this entity does not exist in scene, do not need to process.
Step 43: Buffer Pool is stored in operation.This solid model (comprising the index of this solid model, model class name template and derivation mode code mod) is added Buffer Pool M1, and this solid model is set to invisible; The index of this solid model is deleted from storage pool M2 simultaneously.
Step 44: the process of visible solid model.For visible solid model, first with the solid model index of solid model ID for its correspondence in keyword search storage pool M2:
If searched, then illustrate that this solid model exists in three-dimensional virtual scene, judge whether the serviceable condition state of this solid model changes further, if changed, first perform step 43, then perform step 45; If do not changed, illustrate that this solid model does not need to upgrade, not operation dispatching process; If do not searched, illustrate that this model does not also join three-dimensional virtual scene, perform step 45.
Step 45: Buffer Pool recalls operation.According to the value of mapped file acquisition template and mod that the value of ID and state is set up in step 2, and retrieve in Buffer Pool M1 according to this, if the solid model that the value retrieving template and mod is all mated, then this model is performed to the model reusability operation of step 46; If only have the value of template to mate, then enter the model reorganization operation of step 47; If do not mated, then enter the model reconstruction operation of step 48;
Step 46: model reuse operation.The index of model is added M2, and is set to visible, the index of model, the value of template and mod are deleted from M1 simultaneously;
Step 47: the reorganization operation of model.In M1, searching template identical, the model that the absolute value of mod difference is minimum, this model is called master pattern, being called object module by needing the model of restructuring; The derivation mode finding derivation mode corresponding to object module mod value to describe T and master pattern mod value correspondence in the template file that template value is corresponding describes O, first component identification codes all in T is traveled through, parts corresponding in master pattern are replaced with the parts that T specifies; Then travel through all component identification codes in O, first whether decision means identification code exists in T, if exist, does not then carry out any operation, if do not exist, then parts corresponding in master pattern is replaced back parts corresponding to basic model; Master pattern is performed to the model reusability operation of step 46.
Step 48: Construction of A Model operates.According to the basic model component layer level structure in the template file that template is corresponding, load the model resource file of basic components in external memory storage according to top-down order successively, and set up level restriction relation, construct basic model B; If mod is not equal to 0, according to step 47 pair basic model B execution model reorganization operation; Model index after restructuring is added M2, and is set to visible,
It is mutual that this step belongs to interior external memory, bring into operation the stage in scene, due to model negligible amounts in Buffer Pool M1, the frequency that this step performs is comparatively frequent, real-time and the flatness of scene rendering may be affected, in order to avoid the appearance of this problem, can at scene initial phase, be the solid model stored in certain kind and quantity in Buffer Pool M1.
Step 5: the size of Buffer Pool in dynamic adjustment step 3;
Concrete method is:
Step 51: the ratio of the model instance of model class each in Buffer Pool M1 is adjusted.In section sometime (as 1s), all model class are recalled and stored in store in the model instance quantity of Buffer Pool and Current buffer pond model instance quantity add up, if stored in quantity much larger than recalling process, then the model instance quantity of this model class in Buffer Pool is reduced by half; If recalling quantity is greater than stored in quantity, then by the model instance of this model class in Buffer Pool by copying, its quantity increase is recalled with stored in two times of difference.
Step 52: the overall size of mould Buffer Pool is adjusted.If the total quantity of all model instances is greater than a certain threshold value (configuring setting according to the complexity of scene and computer hardware by user) in Buffer Pool, with the variance of the model instance quantity of different model class in section sometime (sometime the variance of the model instance quantity of section inner model class represent this model class stored in the frequency recalled, variance larger explanation scheduling frequency is larger, this model class at present in range of observation locus and serviceable condition change greatly; Variance is less, illustrate that scheduling frequency is less, this model class at present in range of observation locus and serviceable condition change less) value is as priority, the model instance of model class is deleted, control in certain scope to realize the consumption of transaction model resource to memory source in scene, do not affect dispatching efficiency simultaneously.
Certainly; the present invention also can have other various embodiments; the direct restructuring of such as pattern number word mark (such as license number) is (because figure notation quantity is too many; cannot enumerate; can not by deriving from the definition of mode in template class; and directly assemble in a program); when not deviating from the present invention's spirit and essence thereof; those of ordinary skill in the art are when making various corresponding change and distortion according to the present invention, but these change accordingly and are out of shape the protection domain that all should belong to the claim appended by the present invention.
Claims (4)
1. a solid model dynamic dispatching method for Distributed Three-dimensional virtual environment, it is characterized in that, the method comprises the following steps:
Step 1, sets up the hierarchical data structure of each solid model in described Distributed Three-dimensional virtual scene, and parts each in described solid model are saved as a model resource file respectively;
Wherein, described hierarchical data structure is: the hierarchical structure of each parts, each component names, the local coordinate system of each parts and the anchor point of each parts in solid model;
Step 2, set up the mapping relations of solid model state information data packet PDU that in described Distributed Three-dimensional virtual scene, each simulation node sends and model class template in the mode of external file, described mapping relations are: the model class name template of solid model ID, serviceable condition state and model class template, derive from mode code mod one_to_one corresponding;
Wherein, described solid model state information data packet PDU comprises: the three dimensional space coordinate of solid model ID, solid model and serviceable condition state; The content of described model class template comprises model class name template, basic model and Derived model; Described basic model comprises: model resource file, basic components identification code that the step 1 in the external memory storage that component layer level structure, basic components are corresponding is preserved; Described Derived model comprises: derive from mode code mod, derive from assembling side-play amount and the assembling relative direction of model resource file, derivation component identification code and derivation parts that the step 1 in external memory storage corresponding to parts is preserved;
Step 3, creates and reclaims the Buffer Pool M1 of invisible solid model and deposit the storage pool M2 of all visible solid model indexes three-dimensional virtual environment from three-dimensional virtual environment;
Step 4: obtain present frame three-dimensional scenic view information, and the visibility judge of solid model is carried out in conjunction with described solid model state information data packet, the sightless solid model be present in scene is put into Buffer Pool M1, and this solid model is set to invisible, the index of this solid model is deleted from storage pool M2 simultaneously; To not existing in scene or existed but the visible solid model that changes of serviceable condition, according to there is corresponding model class model instance situation in the mapping relations in step 2 and Buffer Pool M1, carrying out solid model and reusing operation, reorganization operation or reconstructed operation.
2. the solid model dynamic dispatching method of Distributed Three-dimensional virtual environment as claimed in claim 1, it is characterized in that, described step 4 comprises:
Step 40, obtains the PDU that present frame three-dimensional scenic view information and all simulation nodes send, and wherein three-dimensional scenic viewpoint data comprise the position of viewpoint, direction, horizontal field of view angle and visibility;
Step 41, according to the three dimensional space coordinate of solid model and three-dimensional scenic viewpoint data calculate each solid model viewing angle and look eye-distance from, and Ergodic judgement is carried out to PDU: if viewing angle and depending on eye-distance within the range of observation being all in setting and field angle, then solid model is visible solid model, performs step 44; Otherwise solid model is invisible solid model, perform step 42;
Step 42, with the solid model index of solid model ID for its correspondence in keyword search M2, if searched, then illustrates that this solid model exists in three-dimensional virtual scene, then perform step 43; If do not searched, then illustrate that this entity does not exist in scene, do not need to process;
Step 43, adds Buffer Pool M1 by the index of solid model and solid model, template and mod, and is set to invisible by this solid model; The index of this solid model is deleted from M2 simultaneously;
Step 44: with the solid model index of solid model ID for its correspondence in keyword search M2:
If searched, then illustrate that this solid model exists in three-dimensional virtual scene, judge whether the state of this solid model changes further, if changed, first perform step 43, then perform step 45; If do not changed, illustrate that this solid model does not need to upgrade, not operation dispatching process;
If do not searched, illustrate that this model does not also join three-dimensional virtual scene, perform step 45;
Step 45, according to template and mod obtaining model class template in the mapping relations that solid model ID and state sets up from step 2, and retrieve in M1, if the solid model that the value retrieving template and mod is all mated, then this model is performed to the model reusability operation of step 46; If only have the value of template to mate, then enter the model reorganization operation of step 47; If template and mod does not mate, then enter the model reconstruction operation of step 48;
Step 46, model reuse operation: the index of solid model is added M2, and is set to visible, the index of model, template and mod are deleted from Buffer Pool M1 simultaneously;
Step 47, the reorganization operation of model: search template identical in Buffer Pool M1, the model that mod absolute difference is minimum, is called master pattern by this model, is called object module by needing the model of restructuring; The derivation mode finding derivation mode corresponding to object module mod to describe T and master pattern mod corresponding in the template resource file that template is corresponding describes O, first component identification codes all in T is traveled through, parts corresponding in master pattern are replaced with the parts that T specifies; Then travel through all component identification codes in O, first whether decision means identification code exists in T, if exist, does not then carry out any operation, if do not exist, then parts corresponding in master pattern is replaced back parts corresponding to basic model; Master pattern is performed to the model reusability operation of step 46;
Step 48, Construction of A Model operates: according to the basic model component layer level structure in the template resource file that template is corresponding, load the model resource file of basic components in external memory storage successively, construct basic model B according to top-down order; If mod is not equal to 0, according to step 47 pair basic model B execution model reorganization operation; Model index after restructuring is added M2, and is set to visible.
3. the solid model dynamic dispatching method of Distributed Three-dimensional virtual environment as claimed in claim 1 or 2, it is characterized in that, the method also comprises the step of the ratio of the model instance quantity of each model class in dynamic conditioning Buffer Pool M1:
In section sometime, all model class are recalled and add up stored in the model instance quantity stored in the model instance quantity of Buffer Pool and Current buffer pond, if stored in quantity much larger than recalling quantity, then the model instance quantity of this model class in Buffer Pool is reduced by half; If recalling quantity is greater than stored in quantity, then by the model instance of this model class in Buffer Pool by copying, its quantity increase is recalled with stored in two times of difference.
4. the solid model dynamic dispatching method of Distributed Three-dimensional virtual environment as claimed in claim 1 or 2, it is characterized in that, the method also comprises the step of dynamic conditioning Buffer Pool M1 size:
Add up the model instance total quantity of model class all in Buffer Pool in section sometime, if be greater than a certain threshold value, within this time period, the variance yields of the model instance quantity of different model class is as priority, deletes the model instance of model class.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510077170.4A CN104658041A (en) | 2015-02-12 | 2015-02-12 | Dynamic scheduling method for entity models in distributed three-dimensional virtual environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510077170.4A CN104658041A (en) | 2015-02-12 | 2015-02-12 | Dynamic scheduling method for entity models in distributed three-dimensional virtual environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104658041A true CN104658041A (en) | 2015-05-27 |
Family
ID=53249113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510077170.4A Pending CN104658041A (en) | 2015-02-12 | 2015-02-12 | Dynamic scheduling method for entity models in distributed three-dimensional virtual environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104658041A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023315A (en) * | 2016-05-10 | 2016-10-12 | 国网电力科学研究院武汉南瑞有限责任公司 | Three-dimensional entity model fine granularity multi-resolution grading method in distributed virtual reality system |
| CN111324334A (en) * | 2019-11-12 | 2020-06-23 | 天津大学 | Design method for developing virtual reality experience system based on narrative oil painting works |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102044089A (en) * | 2010-09-20 | 2011-05-04 | 董福田 | Method for carrying out self-adaption simplification, gradual transmission and rapid charting on three-dimensional model |
| US20140112149A1 (en) * | 2012-10-22 | 2014-04-24 | Stmicroelectronics (Grenoble 2) Sas | Closed loop end-to-end qos on-chip architecture |
| CN104007971A (en) * | 2014-05-28 | 2014-08-27 | 北京理工大学 | Non-coding fast construction large scale virtual battlefield simulation platform |
-
2015
- 2015-02-12 CN CN201510077170.4A patent/CN104658041A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102044089A (en) * | 2010-09-20 | 2011-05-04 | 董福田 | Method for carrying out self-adaption simplification, gradual transmission and rapid charting on three-dimensional model |
| US20140112149A1 (en) * | 2012-10-22 | 2014-04-24 | Stmicroelectronics (Grenoble 2) Sas | Closed loop end-to-end qos on-chip architecture |
| CN104007971A (en) * | 2014-05-28 | 2014-08-27 | 北京理工大学 | Non-coding fast construction large scale virtual battlefield simulation platform |
Non-Patent Citations (6)
| Title |
|---|
| PAOLO CIGNONI 等: "Adaptive TetraPuzzles: Efficient Out-of-Core Construction and Visualization of Gigantic Multiresolution Polygonal Models", 《ACM TRANSACTIONS ON GRAPHICS》 * |
| 李胜 等: "超大规模地形场景的高性能漫游", 《软件学报》 * |
| 王建坡: "三维实体模型相关技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王玭茜: "三维空间数据引擎多级缓存管理技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 郑笈: "虚拟战场绘制的数据管理与特效生成", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 钟昀 等: "分布式内存数据库在变电站自动化系统中应用", 《电力自动化设备》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023315A (en) * | 2016-05-10 | 2016-10-12 | 国网电力科学研究院武汉南瑞有限责任公司 | Three-dimensional entity model fine granularity multi-resolution grading method in distributed virtual reality system |
| CN106023315B (en) * | 2016-05-10 | 2019-07-12 | 国网电力科学研究院武汉南瑞有限责任公司 | Three-dimensional entity model fine granularity multiresolution stage division in Distributed Virtual Reality System |
| CN111324334A (en) * | 2019-11-12 | 2020-06-23 | 天津大学 | Design method for developing virtual reality experience system based on narrative oil painting works |
| CN111324334B (en) * | 2019-11-12 | 2023-07-28 | 天津大学 | Design method for developing virtual reality experience system based on narrative oil painting works |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110494895B (en) | Rendering operations using sparse volumetric data | |
| CN106599493B (en) | Visualization implementation method of BIM (building information modeling) model in three-dimensional large scene | |
| US7773087B2 (en) | Dynamically configuring and selecting multiple ray tracing intersection methods | |
| US8284195B2 (en) | Cooperative utilization of spatial indices between application and rendering hardware | |
| US8004518B2 (en) | Combined spatial index for static and dynamic objects within a three-dimensional scene | |
| CN103093499B (en) | A kind of city three-dimensional model data method for organizing being applicable to Internet Transmission | |
| Günther et al. | Ray tracing animated scenes using motion decomposition | |
| CN110399747A (en) | A kind of user right correlating method, querying method and device | |
| CN103559374A (en) | Method for subdividing surface split type curved surfaces on multi-submesh model | |
| Schertler et al. | Towards globally optimal normal orientations for large point clouds | |
| Kim et al. | Multi-level modeling and access control for data sharing in collaborative design | |
| CN104658041A (en) | Dynamic scheduling method for entity models in distributed three-dimensional virtual environment | |
| WO2024002292A1 (en) | Model construction and model data subscription method and apparatus, electronic device and medium | |
| US10163186B2 (en) | Management of a plurality of graphic cards | |
| CN112270083A (en) | Multi-resolution modeling and simulation method and system | |
| JP7161154B2 (en) | Hierarchical bounding volume generation method, computer program and computer device | |
| CN110287351A (en) | A kind of threedimensional model lightweight displaying optimization method | |
| CN105654535B (en) | A kind of model manager system for three-dimensional simulation system | |
| CN102184517A (en) | Fast intersection solving algorithm in dynamic scene | |
| CN104408184A (en) | Two-dimensional visual data display method and device based on data cubes | |
| Garanzha | The use of precomputed triangle clusters for accelerated ray tracing in dynamic scenes | |
| Jiang et al. | View frustum culling algorithm for scene based on adaptive binary tree | |
| Masunaga | The block-world data model for a collaborative virtual environment | |
| He et al. | A novel simplification algorithm based on mls and splats for point models | |
| CN117668966A (en) | Hierarchical index-based BIM model interaction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150527 |
|
| WD01 | Invention patent application deemed withdrawn after publication |