CN103544729A - Animation data processing method and system - Google Patents
Animation data processing method and system Download PDFInfo
- Publication number
- CN103544729A CN103544729A CN201310437547.3A CN201310437547A CN103544729A CN 103544729 A CN103544729 A CN 103544729A CN 201310437547 A CN201310437547 A CN 201310437547A CN 103544729 A CN103544729 A CN 103544729A
- Authority
- CN
- China
- Prior art keywords
- animation
- data
- input value
- time schedule
- gpu
- 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
Images
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention discloses an animation data processing method and system. The method includes: defining to-be-calculated animation data into a uniform data structure, by a CPU (central processing unit), calculating time schedule according to the input value in the data structure, and storing the calculated time schedule in the data structure; by a GPU (graphics processing unit), parallelly calculating the animation data to obtain animation schedule according to the time schedule; after GPU's calculation, by the CPU, calculating the final values of attributes of the current object according to the animation schedule calculated by the GPU and the input value in the data structure, and using a rendering system to render the calculation results. Due to the fact that the calculation, with the largest calculation quantity, of the animation schedule is performed by the GPU, calculation efficiency can be well increased, and operation efficiency of the whole system is increased.
Description
Technical field
The present invention relates to animation field, relate in particular to a kind of for calculating fast animation data disposal route and the system of animation data.
Background technology
Rise due to smart mobile phone and intelligent television, 3DLauncher(3D desktop) the 3D application such as by large-scale application in smart machine, but be similar to the application such as 3D animation also cannot large-scale application in smart machine, its main cause is: for animation data, still adopt at present the mode of CPU cycle calculations, its flow process is as follows: S1, according to time schedule, by CPU cycle calculations animation data; After S2, animation calculate, all object state is determined, now uses GPU to play up; S3 plays up after a complete frame, recalculates next frame animation data.Animation data computation process is wherein carried out by CPU circulation, this still can carry out for less animations of animation data amount such as plane animations, if but animation data amount is large, especially for 3D animation, its data processing amount scale is very huge, so if adopt traditional CPU cycle calculations mode, its counting yield is very low, to greatly drag slow whole system travelling speed, affect whole system performance.
Therefore, prior art has yet to be improved and developed.
Summary of the invention
In view of above-mentioned the deficiencies in the prior art, the object of the present invention is to provide a kind of animation data disposal route and system, be intended to solve the low problem of existing animation data disposal route counting yield.
Technical scheme of the present invention is as follows:
A disposal route, wherein, comprises step:
A, animation data to be calculated is defined as to unified data structure, CPU calculates time schedule according to the input value in data structure, and the time schedule calculating is kept in data structure;
B, GPU, according to time schedule, carry out parallel computation to animation data and obtain animation progress;
After C, GPU calculate, the animation progress that CPU calculates according to GPU and the input value in data structure calculate the end value of current each attribute of object, and utilize graph rendering system to play up result of calculation.
Described animation data disposal route, wherein, described step B specifically comprises:
B1, use graph rendering system create for sharing the data buffer of data in GPU video memory;
B2, the animation data that includes time schedule result of calculation is written to data buffer;
B3, from data buffer, obtain each input value in computation process, and calculate animation progress according to time schedule and input value;
B4, the animation progress calculating is written to the relevant position in data buffer.
Described animation data disposal route, wherein, described step B3 specifically comprises:
B31, obtain for the treatment of every part of parallel thread ID that animation data is corresponding in data buffer;
B32, according to parallel thread ID, obtain data buffer and operate accordingly first address;
B33, according to operation first address obtain each input value in corresponding animation data computation process.
Described animation data disposal route, wherein, step C specifically comprises:
C1, CPU be each input value in reading out data structure and animation progress from data buffer;
C2, CPU calculate the end value of current each attribute of object according to each input value and animation humidometer;
C3, utilize graph rendering system to carry out animation to play up, with by animate to screen.
Described animation data disposal route, wherein, described data buffer is created by OpenGL or direct3d.
A disposal system, wherein, comprising:
Time schedule computing module, for animation data to be calculated is defined as to unified data structure, CPU calculates time schedule according to the input value in data structure, and the time schedule calculating is kept in data structure;
Parallel computation module,, carries out parallel computation to animation data and obtains animation progress according to time schedule for GPU;
Rendering module, after calculating for GPU, the animation progress that CPU calculates according to GPU and the input value in data structure calculate the end value of current each attribute of object, and utilize graph rendering system to play up result of calculation.
Described animation data disposal system, wherein, described parallel computation module comprises:
Data buffer creating unit, for being used graph rendering system to create for sharing the data buffer of data at GPU video memory;
Time schedule writing unit, for being written to data buffer by the animation data that includes time schedule result of calculation;
Animation progress computing unit, for obtain each input value computation process from data buffer, and calculates animation progress according to time schedule and input value;
Animation progress writing unit, for being written to the animation progress calculating the relevant position of data buffer.
Described animation data disposal system, wherein, described animation progress computing unit specifically comprises:
Parallel thread ID obtains subelement, for obtaining for the treatment of parallel thread ID corresponding to every part of data buffer animation data;
Operation first address obtains subelement, for obtaining data buffer according to parallel thread ID, operates accordingly first address;
Input value is obtained subelement, for obtain each input value of corresponding animation data computation process according to operation first address.
Described animation data disposal system, wherein, described rendering module comprises:
Data-reading unit, for each input value from the reading out data structure of data buffer and animation progress;
End value computing unit, for calculating the end value of current each attribute of object according to each input value and animation humidometer;
Rendering unit, plays up for utilizing graph rendering system to carry out animation, with by animate to screen.
Beneficial effect: the present invention is first defined as unified data structure by animation data to be calculated, and calculate time schedule according to the input value in data structure by CPU, then by GPU according to time schedule, animation data is carried out to parallel computation and obtains animation progress; After GPU calculates, CPU calculates the end value of current each attribute of object according to animation progress and input value, and utilizes graph rendering system to play up result of calculation.Method of the present invention is due to the animation progress calculating section of animation calculated amount maximum is transferred to GPU parallel computation, thereby can be good at improving counting yield, improves the operational efficiency of whole system.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of animation data disposal route of the present invention preferred embodiment.
Fig. 2 is the particular flow sheet of step S102 in method shown in Fig. 1.
Fig. 3 is the graph of a relation of animation progress and time schedule in linear gradient animation.
Fig. 4 is the graph of a relation of animation progress and time schedule in curve gradual change animation.
Fig. 5 is the particular flow sheet of step S203 in flow process shown in Fig. 2.
Fig. 6 is the corresponding relation figure of parallel thread and animation data.
Fig. 7 is the particular flow sheet of step S103 in method shown in Fig. 1.
Fig. 8 is the structured flowchart of animation data disposal system of the present invention preferred embodiment.
Fig. 9 is the concrete structure block diagram of parallel computation module in system shown in Figure 8.
Figure 10 is the concrete structure block diagram of animation progress computing unit in module shown in Fig. 9.
Figure 11 is the concrete structure block diagram of rendering module in system shown in Figure 8.
Embodiment
The invention provides a kind of animation data disposal route and system, for making object of the present invention, technical scheme and effect clearer, clear and definite, below the present invention is described in more detail.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
Refer to Fig. 1, Fig. 1 is the process flow diagram of a kind of animation data disposal route of the present invention preferred embodiment, and as shown in the figure, it comprises step:
S101, animation data to be calculated is defined as to unified data structure, CPU calculates time schedule according to the input value in data structure, and the time schedule calculating is kept in data structure;
S102, GPU, according to time schedule, carry out parallel computation to animation data and obtain animation progress;
After S103, GPU calculate, the animation progress that CPU calculates according to GPU and the input value in data structure calculate the end value of current each attribute of object, and utilize graph rendering system to play up result of calculation.
Specifically, in step S101, first calculative animation data is defined as to unified data structure, for animation data is managed and unifiedly calculated, the typical displacement animation of take is example, animation data can be defined as to the structure in table one:
Table one
In above-mentioned data structure, comprise two large divisions: input value and result of calculation, input value is wherein raw animation data, result of calculation is for to calculate according to raw animation data.Result of calculation wherein comprises three parts: time schedule, animation progress, object current location (being the end value of each attribute of object), its computation process is expressed as computation process 1, computation process 2, computation process 3.
Wherein, computation process 1: time schedule=(Dang front Shi Jian Dian – time starting point)/(time finishes some – time starting point)
Computation process 2: animation progress=function(time schedule, animation types, animation factor of influence ...)
Computation process 3: object current location=displacement reference position+(displacement finishes position – displacement reference position) * animation progress
Above-mentioned each computing formula is applicable to time-based animation, it is according to Time Calculation, to go out the time of having spent to account for full-time ratio be time schedule, then according to time schedule and animation factor of influence, calculate animation progress, finally calculate the end value of current each attribute of object, such as object current location etc.In above-mentioned computing formula, computation process 1 and computation process 3 are all fixed, its computation process is relatively simple, main calculated amount need to calculated the computation process 2 of animation progress, so the present embodiment is transferred to GPU by computation process 2, by GPU parallel computation, to improve counting yield, for example, when an animation system has a plurality of independent animation, when each animation needs the complicated calculations such as use curve parsing, adopt the data structure in the present invention, it can greatly reduce computing time by parallel computation, make the front time shorten of animate each time, and then increase the drafting number of times of p.s..
In the present embodiment, computation process 1 and computation process 3 or calculated by CPU, this is because computation process 1 and computation process 3 multiplication and division for once all, this does not have much burdens for CPU cycle calculations, can not increase computing time.In addition, computation process 1 and computation process 3 nonjoinders are calculated to GPU, reason is also: different animations, although algorithm is similar, but data type is but also different, that for example displacement animation calculates is space vector type vector3, the speed float that speed animation calculates, the calculative data type of dissimilar animation is different, and committed memory is also different.And computation process 1 and computation process 3 have all related to the calculating of concrete data type, if also all transfer to GPU to process computation process 1 and computation process 3, because different types of data committed memory is in different size, GPU just can not guarantee to use the data block of formed objects to carry out cutting for all animation nodes so, can not use unified data type to calculate, can greatly increase like this complexity of computation process 1 and computation process 3.
2 differences of computation process, although its calculated amount is maximum, its computation process has nothing to do with concrete data type, so when parallel computation mass data, also can allow GPU use unified data structure size, unified data type is calculated, and lowers structure complexity simultaneously.
Because the present embodiment transfers to GPU to carry out computation process 2, and computation process 1 and computation process 3 are still carried out by CPU, so, in data structure, need be by computation process 1, computation process 2, the corresponding separately storage of variable of computation process 3, first CPU can carry out computation process 1 like this, after being finished, computation process 1 fills in computation process 1 result, and computation process 1 result is passed to GPU, by GPU, carry out computation process 2, and filling in computation process 2 results, last CPU carries out computation process 3 and fills in net result.
In above-mentioned steps, animation data is all preserved with unified data structure, and for example system has 10,000 animations needs to calculate, and so the data that construct 10,000 parts of animation progresses calculating needs is calculated for GPU.
Step S102 is GPU and carries out computation process 2, is according to time schedule, animation data is carried out to parallel computation and obtain animation progress; Specifically, as shown in Figure 2, this step comprises:
S201, use graph rendering system create for sharing the data buffer of data in GPU video memory;
This data buffer can be created by various rendering systems, for example OpenGL or direct3d, data sharing (for example reading data or data writing) can be carried out in this data buffer, for example, when parallel computation is realized by OpenCL, can for OpenCL, share by this interface of clCreateFromGLBuffer, the mode of this data buffer can improve the data utilization ratio between graph rendering system and parallel computation greatly, has avoided extra data transfer overhead.
S202, the animation data that includes time schedule result of calculation is written to data buffer;
In abovementioned steps, by CPU, calculate time schedule, this time schedule result of calculation can be written to data buffer so, data are ready like this, just can carry out the calculating of data by GPU.
S203, from data buffer, obtain each input value in computation process, and calculate animation progress according to time schedule and input value;
In this step, GPU can carry out concrete calculating, and for different animations, its calculated amount is also different.For example, for typical linear gradient animation, animation progress and time schedule present linear relationship as shown in Figure 3, in computation process, do not need extra animation factor of influence, the concrete formula of computation process 2 is: f(t)=t, wherein, f(t) represent animation progress, t represents time schedule, is f(t) linear relationship with t, calculates animation progress here by computation process 2, only need time schedule parameter, so computation process is very simple, calculated amount is also relatively little, can not embody the advantage that adopts GPU parallel computation; For more complicated animation system, for example curve gradual change animation (as described in Figure 4, it shows the curve gradual change relation that a kind of animation progress and time schedule present), curve motion can be divided into number of different types, such as adopting the function superposed simulations such as simple sin, cos, tg, can directly adopt so several fixing curve types to be combined to form, if and adopt self-defined Bezier reference mark to customize, need to partly add extra reference mark data at animation factor of influence so, for example typical Bezier cubic curve formula is:
Constant p wherein
0, p
1, p
2, p
3can be used as animation factor of influence is saved in data structure, in compute classes during like the gradual change animation of above-mentioned Bezier cubic curve, the calculated amount of computation process 2 is very large, and computation process is complicated, adopts the GPU parallel computation in the present invention can greatly improve counting yield.
In carrying out parallel computation process, need to carry out corresponding parallel computation according to actual calculated amount, specifically, as shown in Figure 5, it comprises step:
S301, obtain for the treatment of every part of parallel thread ID that animation data is corresponding in data buffer;
First this step needs to create parallel thread, if there are 10,000 parts of animation datas (every piece of data size is identical) to need to calculate, need so to create 10,000 parallel threads, as shown in Figure 6, each parallel thread is independently carried out the task of oneself separately, first need to obtain parallel thread ID, this parallel thread ID correspondence the animation data of need processing.
S302, according to parallel thread ID, obtain data buffer and operate accordingly first address;
According to parallel thread ID, can from data buffer, get corresponding operation first address, because data structure is consistent, so operation first address can simple computation obtain.
S303, according to operation first address obtain each input value in corresponding animation data computation process.
By operation first address, can correspond to corresponding animation data, thereby from data structure, get each input value, for example reference position, time starting point, time end point etc., also can obtain system time, the total size in memory block etc. the data that all parallel thread can be shared herein as required.
For example, according to time schedule, each input value (animation factor of influence, animation types), can calculate by the formula of computation process 2 animation progress.
S204, the animation progress calculating is written to the relevant position in data buffer.
This step is that result of calculation is written to the corresponding result the unknown in data buffer, i.e. animation progress field in data structure.
In step S103, after GPU calculates, the animation progress that CPU can calculate according to GPU and the input value in data structure calculate the end value of current each attribute of object, then utilize graph rendering system to play up result of calculation.As shown in Figure 7, it specifically comprises the steps:
S401, CPU be each input value in reading out data structure and animation progress from data buffer;
S402, CPU calculate the end value of current each attribute of object according to each input value and animation humidometer;
This step is aforesaid computation process 3, can calculate the end value of each attribute by animation progress and each input value.
S403, utilize graph rendering system to carry out animation to play up, with by animate to screen.
Based on said method, the present invention also provides a kind of animation data disposal system, and as shown in Figure 8, it comprises:
Time schedule computing module 100, for animation data to be calculated is defined as to unified data structure, CPU calculates time schedule according to the input value in data structure, and the time schedule calculating is kept in data structure;
Further, as shown in Figure 9, described parallel computation module 200 comprises:
Data buffer creating unit 210, for being used graph rendering system to create for sharing the data buffer of data at GPU video memory;
Time schedule writing unit 220, for being written to data buffer by the animation data that includes time schedule result of calculation;
Animation progress computing unit 230, for obtain each input value computation process from data buffer, and calculates animation progress according to time schedule and input value;
Animation progress writing unit 240, for being written to the animation progress calculating the relevant position of data buffer.About ins and outs existing detailed description in detail in method above of above-mentioned modular unit, therefore repeat no more.
Further, as shown in figure 10, described animation progress computing unit 230 specifically comprises:
Parallel thread ID obtains subelement 231, for obtaining for the treatment of parallel thread ID corresponding to every part of data buffer animation data;
Operation first address obtains subelement 232, for obtaining data buffer according to parallel thread ID, operates accordingly first address;
Input value is obtained subelement 233, for obtain each input value of corresponding animation data computation process according to operation first address.About ins and outs existing detailed description in detail in method above of above-mentioned modular unit, therefore repeat no more.
Further, as shown in figure 11, described rendering module 300 comprises:
Data-reading unit 310, for each input value from the reading out data structure of data buffer and animation progress;
End value computing unit 320, for calculating the end value of current each attribute of object according to each input value and animation humidometer;
In sum, the present invention is first defined as unified data structure by animation data to be calculated, and according to the input value in data structure, calculates time schedule by CPU, then by GPU according to time schedule, animation data is carried out to parallel computation and obtains animation progress; After GPU calculates, CPU calculates the end value of current each attribute of object according to animation progress and input value, and utilizes graph rendering system to play up result of calculation.Method of the present invention is due to the animation progress calculating section of animation calculated amount maximum is transferred to GPU parallel computation, thereby can be good at improving counting yield, improves the operational efficiency of whole system.
Should be understood that, application of the present invention is not limited to above-mentioned giving an example, and for those of ordinary skills, can be improved according to the above description or convert, and all these improvement and conversion all should belong to the protection domain of claims of the present invention.
Claims (9)
1. an animation data disposal route, is characterized in that, comprises step:
A, animation data to be calculated is defined as to unified data structure, CPU calculates time schedule according to the input value in data structure, and the time schedule calculating is kept in data structure;
B, GPU, according to time schedule, carry out parallel computation to animation data and obtain animation progress;
After C, GPU calculate, the animation progress that CPU calculates according to GPU and the input value in data structure calculate the end value of current each attribute of object, and utilize graph rendering system to play up result of calculation.
2. animation data disposal route according to claim 1, is characterized in that, described step B specifically comprises:
B1, use graph rendering system create for sharing the data buffer of data in GPU video memory;
B2, the animation data that includes time schedule result of calculation is written to data buffer;
B3, from data buffer, obtain each input value in computation process, and calculate animation progress according to time schedule and input value;
B4, the animation progress calculating is written to the relevant position in data buffer.
3. animation data disposal route according to claim 2, is characterized in that, described step B3 specifically comprises:
B31, obtain for the treatment of every part of parallel thread ID that animation data is corresponding in data buffer;
B32, according to parallel thread ID, obtain data buffer and operate accordingly first address;
B33, according to operation first address obtain each input value in corresponding animation data computation process.
4. animation data disposal route according to claim 1, is characterized in that, step C specifically comprises:
C1, CPU be each input value in reading out data structure and animation progress from data buffer;
C2, CPU calculate the end value of current each attribute of object according to each input value and animation humidometer;
C3, utilize graph rendering system to carry out animation to play up, with by animate to screen.
5. animation data disposal route according to claim 2, is characterized in that, described data buffer is created by OpenGL or direct3d.
6. an animation data disposal system, is characterized in that, comprising:
Time schedule computing module, for animation data to be calculated is defined as to unified data structure, CPU calculates time schedule according to the input value in data structure, and the time schedule calculating is kept in data structure;
Parallel computation module,, carries out parallel computation to animation data and obtains animation progress according to time schedule for GPU;
Rendering module, after calculating for GPU, the animation progress that CPU calculates according to GPU and the input value in data structure calculate the end value of current each attribute of object, and utilize graph rendering system to play up result of calculation.
7. animation data disposal system according to claim 6, is characterized in that, described parallel computation module comprises:
Data buffer creating unit, for being used graph rendering system to create for sharing the data buffer of data at GPU video memory;
Time schedule writing unit, for being written to data buffer by the animation data that includes time schedule result of calculation;
Animation progress computing unit, for obtain each input value computation process from data buffer, and calculates animation progress according to time schedule and input value;
Animation progress writing unit, for being written to the animation progress calculating the relevant position of data buffer.
8. animation data disposal system according to claim 7, is characterized in that, described animation progress computing unit specifically comprises:
Parallel thread ID obtains subelement, for obtaining for the treatment of parallel thread ID corresponding to every part of data buffer animation data;
Operation first address obtains subelement, for obtaining data buffer according to parallel thread ID, operates accordingly first address;
Input value is obtained subelement, for obtain each input value of corresponding animation data computation process according to operation first address.
9. animation data disposal system according to claim 6, is characterized in that, described rendering module comprises:
Data-reading unit, for each input value from the reading out data structure of data buffer and animation progress;
End value computing unit, for calculating the end value of current each attribute of object according to each input value and animation humidometer;
Rendering unit, plays up for utilizing graph rendering system to carry out animation, with by animate to screen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310437547.3A CN103544729A (en) | 2013-09-24 | 2013-09-24 | Animation data processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310437547.3A CN103544729A (en) | 2013-09-24 | 2013-09-24 | Animation data processing method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103544729A true CN103544729A (en) | 2014-01-29 |
Family
ID=49968147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310437547.3A Pending CN103544729A (en) | 2013-09-24 | 2013-09-24 | Animation data processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103544729A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268916A (en) * | 2014-09-10 | 2015-01-07 | 北京金山安全软件有限公司 | View drawing method and device and mobile terminal |
| CN109062620A (en) * | 2018-08-01 | 2018-12-21 | 武汉斗鱼网络科技有限公司 | Android frame animation executes method, apparatus, terminal and readable medium |
| CN109062691A (en) * | 2018-07-19 | 2018-12-21 | 芯视图（常州）微电子有限公司 | A kind of light weight vertex coloring thread generates the method and device of distribution |
| CN109064531A (en) * | 2018-08-01 | 2018-12-21 | 武汉斗鱼网络科技有限公司 | Android animation executes method, apparatus, terminal and readable medium |
| CN111314764A (en) * | 2020-03-04 | 2020-06-19 | 南方电网科学研究院有限责任公司 | Synchronization method of cross-screen animation in distributed rendering environment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080158236A1 (en) * | 2006-12-31 | 2008-07-03 | Reuven Bakalash | Parallel graphics system employing multiple graphics pipelines wtih multiple graphics processing units (GPUs) and supporting the object division mode of parallel graphics rendering using pixel processing resources provided therewithin |
| CN101587583A (en) * | 2009-06-23 | 2009-11-25 | 长春理工大学 | The farm of playing up based on the GPU cluster |
| CN102156703A (en) * | 2011-01-24 | 2011-08-17 | 南开大学 | Low-power consumption high-performance repeating data deleting system |
| US8264503B1 (en) * | 2006-10-24 | 2012-09-11 | Adobe Systems Incorporated | Batch early engagement rules graphics processing |
| CN102722859A (en) * | 2012-05-31 | 2012-10-10 | 北京像素软件科技股份有限公司 | Method for rendering computer simulation scene |
| CN102982506A (en) * | 2012-11-13 | 2013-03-20 | 沈阳信达信息科技有限公司 | Particle system optimization based on graphic processing unit (GPU) |
-
2013
- 2013-09-24 CN CN201310437547.3A patent/CN103544729A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8264503B1 (en) * | 2006-10-24 | 2012-09-11 | Adobe Systems Incorporated | Batch early engagement rules graphics processing |
| US20080158236A1 (en) * | 2006-12-31 | 2008-07-03 | Reuven Bakalash | Parallel graphics system employing multiple graphics pipelines wtih multiple graphics processing units (GPUs) and supporting the object division mode of parallel graphics rendering using pixel processing resources provided therewithin |
| CN101587583A (en) * | 2009-06-23 | 2009-11-25 | 长春理工大学 | The farm of playing up based on the GPU cluster |
| CN102156703A (en) * | 2011-01-24 | 2011-08-17 | 南开大学 | Low-power consumption high-performance repeating data deleting system |
| CN102722859A (en) * | 2012-05-31 | 2012-10-10 | 北京像素软件科技股份有限公司 | Method for rendering computer simulation scene |
| CN102982506A (en) * | 2012-11-13 | 2013-03-20 | 沈阳信达信息科技有限公司 | Particle system optimization based on graphic processing unit (GPU) |
Non-Patent Citations (1)
| Title |
|---|
| ROBFARBER: ""通过OpenGL映射GPU内存"", 《高性能CUDA应用设计与开发方法与最佳实践》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104268916A (en) * | 2014-09-10 | 2015-01-07 | 北京金山安全软件有限公司 | View drawing method and device and mobile terminal |
| CN109062691A (en) * | 2018-07-19 | 2018-12-21 | 芯视图（常州）微电子有限公司 | A kind of light weight vertex coloring thread generates the method and device of distribution |
| CN109062691B (en) * | 2018-07-19 | 2023-07-04 | 南京军微半导体科技有限公司 | Method for generating and distributing lightweight vertex dyeing threads |
| CN109062620A (en) * | 2018-08-01 | 2018-12-21 | 武汉斗鱼网络科技有限公司 | Android frame animation executes method, apparatus, terminal and readable medium |
| CN109064531A (en) * | 2018-08-01 | 2018-12-21 | 武汉斗鱼网络科技有限公司 | Android animation executes method, apparatus, terminal and readable medium |
| CN109064531B (en) * | 2018-08-01 | 2023-10-20 | 武汉斗鱼网络科技有限公司 | Android animation execution method and device, terminal and readable medium |
| CN111314764A (en) * | 2020-03-04 | 2020-06-19 | 南方电网科学研究院有限责任公司 | Synchronization method of cross-screen animation in distributed rendering environment |
| CN111314764B (en) * | 2020-03-04 | 2022-05-20 | 南方电网科学研究院有限责任公司 | Synchronization method of cross-screen animation in distributed rendering environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107533750B (en) | Virtual image processor, and method and system for processing image data thereon | |
| CN103544729A (en) | Animation data processing method and system | |
| US20140253555A1 (en) | Multiresolution consistent rasterization | |
| CN103761215B (en) | Matrix transpose optimization method based on graphic process unit | |
| CN111445581A (en) | Mesh reconstruction using data-driven priors | |
| CN103617150A (en) | GPU (graphic processing unit) based parallel power flow calculation system and method for large-scale power system | |
| CN102708009B (en) | Method for sharing GPU (graphics processing unit) by multiple tasks based on CUDA (compute unified device architecture) | |
| CN103413343B (en) | A kind of 3D rendering animation engine | |
| CN102591709B (en) | Shapefile master-slave type parallel writing method based on OGR (open geospatial rule) | |
| CN103439743A (en) | Method for drawing seismic data profile graph | |
| CN106201673B (en) | A kind of seismic data processing technique and device | |
| CN104765589A (en) | Grid parallel preprocessing method based on MPI | |
| Steinberger et al. | Parallel generation of architecture on the GPU | |
| CN103345580A (en) | Parallel CFD method based on lattice Boltzmann method | |
| Camp et al. | Parallel stream surface computation for large data sets | |
| US9940689B2 (en) | Latency-resistant sparse simulation technique, system and method | |
| Weber et al. | Efficient parallel extraction of crack-free isosurfaces from adaptive mesh refinement (AMR) data | |
| WO2023160236A1 (en) | Slicing method and apparatus for multi-output neural network, and chip and storage medium | |
| Schreiber et al. | Free-surface lattice-Boltzmann simulation on many-core architectures | |
| CN111091620A (en) | Map dynamic road network processing method and system based on graphics and computer equipment | |
| CN110018812A (en) | A kind of pattern drawing method and device | |
| EP2715564B1 (en) | Mesh sequence file format | |
| CN109492086A (en) | A kind of answer output method, device, electronic equipment and storage medium | |
| CN103177173A (en) | Method and device for selecting network virtual character attiring | |
| CN106504301A (en) | A kind of method for drafting for corresponding vector data by multi-core CPU |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140129 |