CN101446908A - Method for improving special effect processing efficiency based on multi-CPU platform - Google Patents
Method for improving special effect processing efficiency based on multi-CPU platform Download PDFInfo
- Publication number
- CN101446908A CN101446908A CNA2007101780566A CN200710178056A CN101446908A CN 101446908 A CN101446908 A CN 101446908A CN A2007101780566 A CNA2007101780566 A CN A2007101780566A CN 200710178056 A CN200710178056 A CN 200710178056A CN 101446908 A CN101446908 A CN 101446908A
- Authority
- CN
- China
- Prior art keywords
- cpu
- special effect
- effect processing
- processing task
- platform
- 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
Abstract
The invention discloses a method for improving special effect processing efficiency based on a multi-CPU platform. The method comprises the following steps: creating a special effect processing task; assigning the special effect processing task to various CPUs for execution according to number of the CPUs of the multi-CPU platform; and gathering and submitting execution results of various CPUs. In the embodiment of the invention, the special effect processing task is assigned to various CPUs for execution, which improves service efficiency of the CPUs and enhances the special effect processing efficiency of the multi-CPU platform. Furthermore, in the embodiment of the invention, a CPU special effect processing exclusive call mode can ensure coherence of the special effect processing task operations, reduce switching among tasks, and further improve the special effect production process performance of a non-linear editing system, thus acquiring more real-time special effects in programming, clipping higher-quality programs and improving the satisfaction degree of the audience.
Description
Technical field
The present invention relates to the programme production technique field, particularly relate to a kind of method that improves special effect processing efficiency based on multi-CPU platform.
Background technology
Continuous progress along with computer technology and integrated circuit technique, linear editing system (system that is used for video-tape and film) is extraordinary obsolescence, is replaced based on computer hardware platforms, the nonlinear editing system of looking Audio Processing card and large scale digital memory carrier at present.
Nonlinear editing system is the product that computing machine and digital to television technology combine, and it is added by computing machine basically looks the Audio Processing card and be equipped with corresponding non-linear editing software and form.The non-line editing technique of numeral is applied in characteristics such as to have editting function in the video program production process powerful, convenient to operation, therefore is widely used in the making of TV programme.Wherein, the non-linear ＂ of the ＂ in the nonlinear editing system is meant be not workflow with the linear mode, and promptly it has broken through in the past that the operator must successively be the operator scheme of order with Pgmtime under the linear work environment.Under the nonlinear operation mode, the operator can freely select target, needn't be order with the time order and function.Simply say, need not be as edit tape when being exactly montage F.F. or to bringing the search picture to and fro.
Under the conventional linear edit mode, need the combination of multiple devices such as videocorder, cut bank, animation stand, sound console, film titler usually, finish program later stage montage, Series Code, stunt making, subtitle superposition, look work such as audio mix.And under non-linear editing mode, only need 1 nonlinear editing system can finish that several equipment, multidigit technician cooperate the work that just can finish under above-mentioned traditional edit mode based on digital technology.In conventional linear program making process,, can only rely on to duplicate sometimes and achieve the goal because linear editing can't be made amendment quickly and easily.Because index word is big, so influenced picture quality to a great extent, has also prolonged the time of program making.And non-linear editing system biggest advantage just is its non-linear editing, it can carry out montage to program quickly and easily, and nonlinear editing system is under the computer disk recording mode, each section material all is that the form with a computer documents is stored on the computer disk, by software for editing, can freely make up the editing that these files are finished whole program.Revise fragment if desired in the middle of completed program, lengthening or shortening only need to revise relevant file sequence and get final product.
Because there is above-mentioned huge advantage in nonlinear editing system, so very fast having obtained in the program making of TV station used widely.Usually for the workflow of non-linear editing, can regard steps such as input, editor, output simply as, it uses flow process mainly to be divided into as follows:
1, material collection and input, collection are exactly to utilize nonlinear editing system to convert analog video, sound signal to digital signal to store in the computing machine, and perhaps the digital video with the outside stores in the computing machine, becomes manageable material.Input mainly is image that other software processes is crossed, sound or the like, imports in the nonlinear editing system.
2, material editor, material editor are provided with going into a little and going out a little of material, to select only part, and different materials of winding in chronological order then.
3, stunt is handled, and for video material, stunt is handled and comprised transition, special efficacy, synthetic stack; For audio material, stunt is handled and is comprised transition, special efficacy; Shocking picture effect is exactly to produce in this course.Therefore the software and hardware Capability Requirement to nonlinear editing system is very high in stunt is handled, and not only processing speed is very slow if can not reach requirement, and often the real-time of special effects can not reach requirement.
4, captions are made, and captions are very important parts in the program, and it comprises literal and two aspects of figure.It is very convenient that nonlinear editing system is made captions, almost do not have the effect that can't realize, and also have a large amount of templates to select.
5, output and generation after editing saving is finished, just can be exported back and record on the video-tape; Also video file be can generate, online, imprinting VCD and DVD or the like are published to.
Along with the continuous increase that spectators require TV programme, TV station needs a large amount of the use to look audio effect to reach the effect of playing up material when programming.Therefore along with the continuous development of computer technology, under the traction of this demand, increasing nonlinear editing system uses multi-CPU platform and multinuclear and Hyper-Threading, promotes the hardware capabilities of nonlinear editing system.The processing speed of Effect Algorithm has been accelerated in the application of above-mentioned technology in nonlinear editing system, can increase the track of more multi-layered number, improves the real-time of special effects.
Yet the inventor is when realization is of the present invention, find to exist at least in the prior art following technological deficiency: though in existing nonlinear editing system, adopt the platforms of multi-CPU platforms and multinuclear and Hyper-Threading in a large number, but existing stunt processing procedure still relies on one of them single CPU to handle its Effect Algorithm, make that the performance of multi-CPU platform can not be fully played, and causing the idle of other CPU, is the significant wastage to existing nonlinear editing system resource.
Summary of the invention
The problem that the embodiment of the invention will solve provides a kind of method that improves special effect processing efficiency based on multi-CPU platform, be used for that special effect processing task is assigned to each CPU and go up execution, increase the utilization rate of CPU to greatest extent, thus the technological deficiency that the performance of multi-CPU platform can not be not fully exerted in the customer service prior art.
For achieving the above object, the embodiment of the invention proposes a kind of method based on multi-CPU platform raising special effect processing efficiency on the one hand, may further comprise the steps: create special effect processing task; CPU quantity according to described multi-CPU platform is assigned to upward execution of each CPU with described special effect processing task; The execution result and the submission that gather each CPU.
Wherein, described CPU quantity according to multi-CPU platform is assigned to each CPU with described special effect processing task and goes up execution and specifically comprise: according to the CPU quantity of described multi-CPU platform described special effect processing task mean allocation is gone up to each CPU and carry out.
Wherein, comprise also that when described each CPU carried out the special effect processing task of distributing, other special effect processing task were set to blocked state.
Wherein, at the described execution result that gathers each CPU and after submitting to, further comprising the steps of: the described special effect processing task of mark is completion status.
Wherein, after described special effect processing task is labeled as completion status, carry out other special effect processing task.
Wherein, realize that by Windows API described special effect processing task is assigned to each CPU goes up execution.
Wherein, described CPU quantity according to multi-CPU platform specifically comprises described special effect processing task mean allocation to the last execution of each CPU: the image averaging that will load stunt is assigned to each CPU and goes up execution.
The embodiment of the invention also proposes a kind of computer program, comprises that some instructions are in order to carry out the aforementioned method that improves special effect processing efficiency based on multi-CPU platform.
The embodiment of the invention also proposes a kind of storage medium, stores above-mentioned computer program.
The embodiment of the invention also proposes a kind of computer equipment, comprises in order to carry out the aforementioned software that improves the method for special effect processing efficiency based on multi-CPU platform reaching the hardware that cooperates with software.
The technical scheme of the embodiment of the invention has the following advantages, and the embodiment of the invention goes up execution by special effect processing task being assigned to each CPU, has increased the service efficiency of CPU, has improved the special effect processing efficiency of multi-CPU platform.And the formula method of calling is monopolized in the stunt processing of CPU by the embodiment of the invention, can guarantee the continuity of special effect processing task operation, switching between the minimizing task, utilize the CACHE among the CPU better, further improved the performance of nonlinear editing system stunt making flow process, thereby can obtain how real-time special effects in program making, montage goes out higher-quality program, improves spectators' satisfaction.
Description of drawings
Fig. 1 be the embodiment of the invention one improve the process flow diagram of the method for special effect processing efficiency based on multi-CPU platform;
Fig. 2 is that many CPU of embodiment of the invention stunt is carried out synoptic diagram;
Fig. 3 be the embodiment of the invention two improve the process flow diagram of the method for special effect processing efficiency based on multi-CPU platform;
Fig. 4 monopolizes formula stunt call flow chart for the embodiment of the invention.
Embodiment
The inventor is at the technological deficiency of the nonlinear editing system of available technology adopting multi-CPU platform, can be when carrying out special effect processing task, this special effect processing task is assigned to each CPU goes up execution, the resource of the nonlinear editing system of existing multi-CPU platform not only can be made full use of, the processing speed of whole special effect processing task can also be improved by division of tasks.And the embodiment of the invention has also proposed a kind of formula of monopolizing stunt method of calling to CPU, can fully guarantee the continuity that CPU uses, the minimizing task is switched the influence to the CPU processing speed, can utilize the CACHE among the CPU better, further improve the performance of nonlinear editing system stunt making flow process.
Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail:
As shown in Figure 1, for the embodiment of the invention one improve the process flow diagram of the method for special effect processing efficiency based on multi-CPU platform, can improve the performance of nonlinear editing system by this embodiment, this embodiment may further comprise the steps:
Step S101 creates special effect processing task.Wherein special effect processing task need be carried out a large amount of Effect Algorithm, is the processing procedure of an algorithm intensity.
Step S102, the stunt execution module is transferred above-mentioned special effect processing task, and according to the CPU quantity of multi-CPU platform in the nonlinear editing system this special effect processing task is assigned to upward execution of each CPU.The stunt processing speed not only can be improved like this, and other cpu resources can be made full use of.As shown in Figure 2, for many CPU of embodiment of the invention stunt is carried out synoptic diagram, in this embodiment with the stunt division of tasks, be assigned to these four CPU of CPU0, CPU1, CPU2 and CPU3 and go up execution, carry out a special effect processing task simultaneously by four CPU, improve the speed that stunt is handled, satisfy the real-time requirement of special effects.Wherein, as a preferred embodiment, if the inventor finds the special effect processing task difference that each CPU is assigned to, have have more lack, will occur being assigned to the time that the CPU of more Processing tasks handles so will be longer, and the CPU task of distributing less task also needs the more CPU of wait task after executing, and therefore still can have certain wasting of resources problem.Therefore the embodiment of the invention proposes the special effect processing task mean allocation is gone up execution to each CPU, also need not mutual wait like this between the CPU, further improves the service efficiency of nonlinear editing system.For example being a two field picture when loading stunt, this image laterally or vertically can be divided into four parts, four CPU that transfer to respectively among Fig. 2 carry out respectively, the time basically identical carried out of these four CPU like this, thus guaranteed processing speed the most efficiently.Wherein, the embodiment of the invention also proposes a kind of mode that realizes special effect processing task is assigned to the last execution of each CPU by Windows API; yet this mode is not the unique mode that can carry out the special effect processing task distribution; therefore realize the distribution of special effect processing task by other modes, also all should be embodiment of the invention protection domain and contain.
Step S103 after each CPU finishes the work, gathers execution result and the submission of each CPU.
As shown in Figure 3, improve the process flow diagram of the method for special effect processing efficiency for the embodiment of the invention two based on multi-CPU platform, this embodiment on the basis of embodiment one by the formula of monopolizing stunt invocation pattern to CPU, can guarantee the continuity that CPU uses, further improve the performance of nonlinear editing system.This embodiment may further comprise the steps:
Step S301 creates special effect processing task.Wherein special effect processing task need be carried out a large amount of Effect Algorithm, is the processing procedure of an algorithm intensity.
Step S302, the stunt execution module is transferred above-mentioned special effect processing task, and according to the CPU quantity of multi-CPU platform in the nonlinear editing system this special effect processing task is assigned to upward execution of each CPU.The stunt processing speed not only can be improved like this, and other cpu resources can be made full use of.As shown in Figure 2, for many CPU of embodiment of the invention stunt is carried out synoptic diagram, in this embodiment with the stunt division of tasks, be assigned to these four CPU of CPU0, CPU1, CPU2 and CPU3 and go up execution, carry out a special effect processing task simultaneously by four CPU, improve the speed that stunt is handled, satisfy the real-time requirement of special effects.Wherein, as a preferred embodiment, if the inventor finds the special effect processing task difference that each CPU is assigned to, have have more lack, will occur being assigned to the time that the CPU of more Processing tasks handles so will be longer, and the CPU task of distributing less task also needs the more CPU of wait task after executing, and therefore still can have certain wasting of resources problem.Therefore the embodiment of the invention proposes the special effect processing task mean allocation is gone up execution to each CPU, also need not mutual wait like this between the CPU, further improves the service efficiency of nonlinear editing system.For example being a two field picture when loading stunt, this image laterally or vertically can be divided into four parts, four CPU that transfer to respectively among Fig. 2 carry out respectively, the time basically identical carried out of these four CPU like this, thus guaranteed processing speed the most efficiently.Wherein, the embodiment of the invention also proposes a kind of mode that realizes special effect processing task is assigned to the last execution of each CPU by Windows API; yet this mode is not the unique mode that can carry out the special effect processing task distribution; therefore realize the distribution of special effect processing task by other modes, also all should be embodiment of the invention protection domain and contain.
Step S303, when each CPU carried out the special effect processing task of distributing, other special effect processing task were set to blocked state.Because for multiple task operating system Windows, the same time may have a plurality of tasks in operation, need between the task to switch mutually to carry out, in the process of task switching, need CONTEXT SWITCH like this, the scene that promptly keeps current task is written into the on-the-spot of next task and carries out next task.Therefore task is switched and can be reduced system performance, owing to keep the scene of current task the CACHE performance among the CPU is reduced greatly.Can use CPU as far as possible continuously, needs minimizing CONTEXT SWITCH, better utilization CACHE in order to guarantee special effect processing task.The embodiment of the invention guarantees single in-process by to the control of special effect processing task scheduling, twice independently algorithm operating can not carry out simultaneously.Therefore when each CPU carries out the special effect processing task of distributing, can other special effect processing task be set to blocked state.As shown in Figure 4, for the embodiment of the invention is monopolized formula stunt call flow chart, therefore the stunt execution module is being carried out special effect processing task 1, special effect processing task 2 is set to blocked state, guarantees that therefore special effect processing task 1 can monopolize current stunt execution module.
Step S304 after finishing the work at each CPU, gathers execution result and the submission of each CPU.
Step S305, the described special effect processing task of mark is completion status.
Step S306 after special effect processing task is labeled as completion status, carries out other special effect processing task.In Fig. 4, after finishing, special effect processing task 1 it can be labeled as completion status, and this moment, the stunt execution module can be accepted special effect processing task 2 and execution, also needed other special effect processing task to be set to blocked state when carrying out special effect processing task 2 equally.
The embodiment of the invention goes up execution by special effect processing task being assigned to each CPU, has increased the service efficiency of CPU, has improved the special effect processing efficiency of multi-CPU platform.And the formula method of calling is monopolized in the stunt processing of CPU by the embodiment of the invention, can guarantee the continuity of special effect processing task operation, switching between the minimizing task, utilize the CACHE among the CPU better, further improved the performance of nonlinear editing system stunt making flow process, thereby can obtain how real-time special effects in program making, montage goes out higher-quality program, improves spectators' satisfaction.
Through the above description of the embodiments, those skilled in the art can be well understood to the present invention and can realize by the mode that software adds essential general hardware platform, can certainly pass through hardware, but the former is better embodiment under a lot of situation.Based on such understanding, the part that technical scheme of the present invention contributes to prior art in essence in other words can embody with the form of software product, this computer software product is stored in the storage medium, comprise that some instructions are with so that a computer equipment (can be a personal computer, server, the perhaps network equipment etc.) carry out the described method of each embodiment of the present invention.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (10)
1, a kind of method based on multi-CPU platform raising special effect processing efficiency is characterized in that, may further comprise the steps:
Create special effect processing task;
CPU quantity according to described multi-CPU platform is assigned to upward execution of each CPU with described special effect processing task;
The execution result and the submission that gather each CPU.
2, improve the method for special effect processing efficiency according to claim 1 based on multi-CPU platform, it is characterized in that, described CPU quantity according to multi-CPU platform is assigned to the last execution of each CPU with described special effect processing task and specifically comprises:
CPU quantity according to described multi-CPU platform goes up execution with described special effect processing task mean allocation to each CPU.
3, improve the method for special effect processing efficiency according to claim 1 based on multi-CPU platform, it is characterized in that, also comprise, when described each CPU carried out the special effect processing task of distributing, other special effect processing task were set to blocked state.
4, improve the method for special effect processing efficiency according to claim 1 based on multi-CPU platform, it is characterized in that, after the described execution result that gathers each CPU is also submitted to, further comprising the steps of:
The described special effect processing task of mark is completion status.
5, as improve the method for special effect processing efficiency as described in the claim 3 based on multi-CPU platform, it is characterized in that, after described special effect processing task is labeled as completion status, carry out other special effect processing task.
6, improve the method for special effect processing efficiency according to claim 1 based on multi-CPU platform, it is characterized in that, realize that by Windows API described special effect processing task is assigned to each CPU goes up execution.
As improve the method for special effect processing efficiency as described in the claim 2 based on multi-CPU platform, it is characterized in that 7, described CPU quantity according to multi-CPU platform goes up execution with described special effect processing task mean allocation to each CPU and specifically comprises:
The image averaging that need is loaded stunt is assigned to upward execution of each CPU.
8, a kind of computer program is characterized in that, comprises that some instructions are in order to carry out the described method that improves special effect processing efficiency based on multi-CPU platform of aforementioned claim 1-7.
9, a kind of storage medium is characterized in that, the described computer program of storage claim 8.
10, a kind of computer equipment is characterized in that, comprises in order to carry out the described software that improves the method for special effect processing efficiency based on multi-CPU platform of aforementioned claim 1-7 reaching the hardware that cooperates with software.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101780566A CN101446908A (en) | 2007-11-26 | 2007-11-26 | Method for improving special effect processing efficiency based on multi-CPU platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007101780566A CN101446908A (en) | 2007-11-26 | 2007-11-26 | Method for improving special effect processing efficiency based on multi-CPU platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101446908A true CN101446908A (en) | 2009-06-03 |
Family
ID=40742598
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007101780566A Pending CN101446908A (en) | 2007-11-26 | 2007-11-26 | Method for improving special effect processing efficiency based on multi-CPU platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101446908A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101980168A (en) * | 2010-11-05 | 2011-02-23 | 北京云快线软件服务有限公司 | Dynamic partitioning transmission method and device |
| CN105007189A (en) * | 2015-07-29 | 2015-10-28 | 无锡天脉聚源传媒科技有限公司 | Video data broadcasting method and device |
-
2007
- 2007-11-26 CN CNA2007101780566A patent/CN101446908A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101980168A (en) * | 2010-11-05 | 2011-02-23 | 北京云快线软件服务有限公司 | Dynamic partitioning transmission method and device |
| CN105007189A (en) * | 2015-07-29 | 2015-10-28 | 无锡天脉聚源传媒科技有限公司 | Video data broadcasting method and device |
| CN105007189B (en) * | 2015-07-29 | 2018-09-25 | 无锡天脉聚源传媒科技有限公司 | A kind of broadcasting method and device of video data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9329745B2 (en) | Rendering in a multi-user video editing system | |
| CN102402458B (en) | Virtual machine and/or multi-level scheduling support on systems with asymmetric processor cores | |
| JP3918951B2 (en) | Method for enabling a multimedia application to run on a hardware platform with various different resource levels, a physical recording medium containing such an application, and a platform system for executing such an application | |
| CN102770841A (en) | Method and apparatus for generating minimum boot image | |
| CN105893126A (en) | Task scheduling method and device | |
| US20110043695A1 (en) | Information Processing Apparatus Enabling an Efficient Parallel Processing | |
| CN102768846B (en) | A kind of audio frequency playing method, device and terminal | |
| CN106033373A (en) | A method and a system for scheduling virtual machine resources in a cloud computing platform | |
| CN103167222A (en) | Nonlinear cloud editing system | |
| CN101365044B (en) | Program information storage and management method for multimedia device and multimedia device | |
| CN102736910B (en) | A kind of general command interface based on plug-in unit defines method | |
| CN103095994A (en) | High-definition medium non-linear editing method used in digital home and device thereof | |
| CN101446908A (en) | Method for improving special effect processing efficiency based on multi-CPU platform | |
| US20150106522A1 (en) | Selecting a target server for a workload with a lowest adjusted cost based on component values | |
| CN102629238A (en) | Method and device for supporting vector condition memory access | |
| US20100049747A1 (en) | Apparatus and method for storing log in a thread oriented logging system | |
| JP4728083B2 (en) | Media processing device | |
| US20170164031A1 (en) | System and Method for Scheduling Clips | |
| ATE221288T1 (en) | SYSTEM FOR MANAGING APPLICATION PROGRAMS USING A VARIABLE MANAGEMENT COMMAND SET | |
| CN103064861A (en) | Method and device for previewing media files and electronic devices | |
| US20230142302A1 (en) | Audio processing | |
| US20230221977A1 (en) | Efficient mass virtual machine (vm) cloning | |
| US20190222882A1 (en) | System and Method for Scheduling Clips | |
| JP2003114677A (en) | Apparatus and method for information processing and storage medium | |
| CN105282509B (en) | A kind of more camera lens editing systems based on Cluster Rendering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20090603 |