DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 20 rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter because claim 20 is directed to a signal per se.
Claim 20 recites a computer-readable medium. The broadest reasonable interpretation of a claim drawn to a computer readable medium (also called machine readable medium and other such variations) typically covers forms of non-transitory tangible media and transitory propagating signals per se in view of the ordinary and customary meaning of computer readable media. See MPEP 2111.01. When the broadest reasonable interpretation of a claim covers a signal per se, the claim must be rejected under 35 U.S.C. 101 as covering non-statutory subject matter. The USPTO recognizes that applicants may have claims directed to computer readable media that cover signals per se, which the USPTO must reject under 35 U.S.C. 101 as covering both non- statutory subject matter and statutory subject matter. A claim drawn to such a computer readable medium that covers both transitory and non-transitory embodiments may be amended to narrow the claim to cover only statutory embodiments to avoid a rejection under 35 U.S.C. $ 101 by adding the limitation "non-transitory" to the claim. Such an amendment would typically not raise the issue of new matter because the broadest reasonable interpretation relies on the ordinary and customary meaning that includes signals per se. Applicant’s specification in Paragraph 120 recites “In this manner, computer-readable media generally may correspond to (1) tangible computer-readable storage media, which is non-transitory or (2) a communication medium such as a signal or carrier wave.” Applicant’s disclosure states that a “computer-readable medium” could either correspond to a non-transitory computer-readable medium or a signal or carrier wave. Therefore, claim 20 must be rejected under 35 U.S.C. 101 as it potentially corresponds to a computer readable medium that cover signal per se.
As an additional note, a non-transitory computer readable medium having executable programming instructions stored thereon is considered statutory as non-transitory computer readable media excludes transitory data signals.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 8-10, 16 and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather et al (US 7633506 B1) and Wong et al (US 20230102063 A1), hereinafter Leather and Wong respectively.
Regarding claim 1, Leather teaches an apparatus for graphics processing (“The present invention relates to a parallel pipeline graphics system” - Abstract), comprising: at least one memory (memory 815) ; the at least one processor (processor 813), individually or in any combination, is configured to: assign a first subset of the set of workloads to a first graphics pipeline and a second subset of the set of workloads to a second graphics pipeline (“When geometry data is sent to the back-end, it is divided up and provided to one of the parallel pipelines” – Abstract [NOTE: geometry data processing is a form of workload]); and render the first subset of the set of workloads in the first graphics pipeline and the second subset of the set of workloads in the second graphics pipeline (“use appropriate pipelines to operate on the geometry” – Fig 2, Step 230 . [NOTE: geometry data goes through the rendering process by their respective pipelines to then be placed in a frame buffer for display.]). Leather does not teach at least one processor coupled to the at least one memory and, based at least in part on information stored in the at least one memory, the at least one processor, individually or in any combination, is configured to: obtain an indication of a set of workloads. However, Wong teaches at least one processor coupled to the at least one memory (“The apparatus includes a computer processor, a computer memory operatively coupled to the computer processor” – Par 17, Lines 3-4) and, based at least in part on information stored in the at least one memory, the at least one processor, individually or in any combination, is configured to: obtain an indication of a set of workloads (“cause the apparatus to receive a request that includes a description of a workload from a workload initiator” – Par 17, Lines 7-9. [NOTE: the processor obtains a request of a workload as indication]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather to incorporate the teachings of Wong to configure the processor to obtain an indication of a set of workloads. It is known in the art that by implementing some form of indication to communicate that there are workloads present will allow the system to efficiently distribute the workloads to the graphics pipeline and allocate the necessary resources.
Regarding claim 19, the claim describes a method that performs the function of claim 1. Therefore, method claim 19 corresponds to the apparatus disclosed in claim 1 and is rejected for the same reasons of obviousness as used above.
Regarding claim 20, the claim describes a computer readable medium (CRM) that performs the function of claim 1. Therefore, CRM claim 20 corresponds to the apparatus disclosed in claim 1 and is rejected for the same reasons of obviousness as used above.
Regarding claim 8, Leather in view of Wong teaches the apparatus of claim 1. Leather does not teach wherein the at least one processor, individually or in any combination, is further configured to: allocate a first resource for the first subset of the set of workloads and a second resource for the second subset of the set of workloads; and store first data associated with the first subset of the set of workloads in the first resource and second data associated with the second subset of the set of workloads in the second resource. However, Wong further teaches wherein the at least one processor, individually or in any combination, is further configured to: allocate a first resource for the first subset of the set of workloads and a second resource for the second subset of the set of workloads (“The resource manager determines, based on the utilization metrics and one or more policies, a workload allocation recommendation for the workload” – Abstract. [NOTE: resource manager as disclosed by Wong will consider the characteristics associated with the workload before assigning the resource. When combined with Leather’s disclosure, one of ordinary skill could use the resource manager to allocate resources for both each subset of the set of workloads.]); and store first data associated with the first subset of the set of workloads in the first resource and second data associated with the second subset of the set of workloads in the second resource (“Thus, the workload initiator can determine whether placing a workload on a particular processing resource is preferable based on runtime behavior of the system and policies established of the workload.” – Abstract. [NOTE: Necessary data associated with the first and second subset of the set of workloads would need to be stored when the workload is placed in the respective processing resource.]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather to incorporate the teachings of Wong to allocate a first and second resource to the first and second subset of workloads respectively and store the data associated with the subset of workloads to their respective resources. Allocating resources to the workloads is a standard application of graphics processing as they are needed to execute the workload. Storing the data associated with the workloads makes the change of assignment from one pipeline to another more manageable since it will contain information regarding the workload.
Regarding claim 9, Leather in view of Wong teaches the apparatus of claim 8. Leather does not teach wherein the first data is distinct from the second data, and wherein the allocation of the first resource and the second resource is based on the first data being distinct from the second data. However, Wong further teaches wherein the first data is distinct from the second data, and wherein the allocation of the first resource and the second resource is based on the first data being distinct from the second data (“The resource manager determines, based on the utilization metrics and one or more policies, a workload allocation recommendation for the workload” – Abstract.) [NOTE: Wong discloses that the resource manager assigns the workloads to different resources based on the data and characteristics associated with the respective workloads. This implies that the workloads are assigned resources depending on the specific data associated to the workload. Otherwise, the workloads would be potentially assigned the same resource if the workloads were not distinct.]. It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather to incorporate the teachings of Wong to allocate the first and second resource based on the first data being distinct from the second data. By taking into consideration the data associated with the workloads, the resources can efficiently be allocated to the workload that would make best use of it. This design choice will also aid in the reassigning of the workloads to different graphics pipeline since the distinct data will make it easier to identify the workloads.
Regarding claim 10, Leather teaches the apparatus of claim 8. Leather does not teach wherein to store the first data associated with the first subset of the set of workloads in the first resource and the second data associated with the second subset of the set of workloads in the second resource, the at least one processor, individually or in any combination, is configured to: concurrently store the first data associated with the first subset of the set of workloads in the first resource and the second data associated with the second subset of the set of workloads in the second resource. However, Wong further teaches wherein to store the first data associated with the first subset of the set of workloads in the first resource and the second data associated with the second subset of the set of workloads in the second resource, the at least one processor, individually or in any combination, is configured to: concurrently store the first data associated with the first subset of the set of workloads in the first resource and the second data associated with the second subset of the set of workloads in the second resource (In some cases, the workload can include work items that can be executed on the integrated GPU and other work items that can be executed on the discrete GPU concurrently.” – Par 31, Lines 21-24. [NOTE: Wong discloses that the workload work items can be executed concurrently. The execution of workloads would require that the associated data be concurrently stored to their respective allocated resources.]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather to incorporate the teachings of Wong to configure the processor to configure the first and second subsets of the set of workloads to concurrently store the first and second data respectively. This strategy makes best use of the multiple pipelines to create the predicted outcome of reduced latency and parallelization.
Regarding claim 16, Leather and Wong teach the apparatus of claim 1. Leather in view of Wong further teaches wherein to assign the first subset of the set of workloads to the first graphics pipeline and the second subset of the set of workloads to the second graphics pipeline, the at least one processor, individually or in any combination, is configured to: map the first subset of the set of workloads to the first graphics pipeline and the second subset of the set of workloads to the second graphics pipeline (“When geometry data is sent to the back-end, it is divided up and provided to one of the parallel pipelines” - Abstract). [NOTE: One of ordinary skill would recognize that Leather shows subsets of a set of workloads, in the form of geometry data to be processed, being assigned to their respective graphics pipeline. The term “map” does not limit how the workloads are assigned to the graphics, therefore the same reasoning for rejection of claim 1 can be used for claim 16.]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to modify Leather in view of Wong to further incorporate Leather’s teaching of distributing the geometry data to parallel pipelines. Assigning the workloads to graphics pipeline is a known technique in the art of graphics processing to perform rendering operations.
Claim(s) 2 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, and Liang et al (US 20200020067 A1), hereinafter Liang.
Regarding claim 2, Leather in view of Wong teaches the apparatus of claim 1. Leather in view of Wong does not teach wherein the first graphics pipeline is a binning render (BR) pipeline and the second graphics pipeline is a binning visibility (BV) pipeline. However, Liang teaches wherein the first graphics pipeline is a binning render (BR) pipeline (“configured to divide a buffer into which a frame is to be rendered into a plurality bins” – Par 39, Lines 12-14) and the second graphics pipeline is a binning visibility (BV) pipeline (“configured to generate visibility information for each bin of the plurality of bins during the binning pass” – Par 39, Lines 15-17. [NOTE: visibility information is further disclosed by Liang as any information that indicates whether primitives are visible or not.]) or wherein the first graphics pipeline is the BV pipeline and the second graphics pipeline is the BR pipeline. It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong by combining it with the teachings of Liang to have the first graphics pipeline be a binning render pipeline (BR) and the second graphics pipeline be a binning visibility pipeline (BV). Although it is not explicitly disclosed by Liang that the BR and BV pipelines define a first and second pipeline, one of ordinary skill in the art would recognize BR and BV pipelines as known concepts in the art. Binning render pipelines are known to handle the rasterization and shading of primitives while binning visibility pipelines are known to make the determination between visible and invisible primitives. Implementing BR and BV pipelines as separate graphics pipeline benefits the apparatus by improving workload distribution (load -balancing) efficiency.
Claim(s) 3-6 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, and Palmer et al (US 10217183 B2).
Regarding claim 3, Leather in view of Wong teaches the apparatus of claim 1. Leather in view of Wong does not teach wherein the at least one processor, individually or in any combination, is further configured to: change an assignment of at least one of (1) a portion of the first subset of the set of workloads to the second graphics pipeline or (2) a portion of the second subset of the set of workloads to the first graphics pipeline. However, Palmer teaches wherein the at least one processor, individually or in any combination, is further configured to: change an assignment of at least one of (1) a portion of the first subset of the set of workloads (“allocating a plurality of processing units to process tasks associated with a graphics pipeline” - Abstract) to the second graphics pipeline (“reallocating the at least one processing unit to process tasks associated with the compute pipeline.” - Abstract) or (2) a portion of the second subset of the set of workloads to the first graphics pipeline. [NOTE: Palmer discloses a plurality of processing units to process tasks/workloads that were initially assigned in a “graphics pipeline” which are then reallocated to a “compute pipeline” after receiving a request for the reallocation. This concept can be combined with Leather to substitute with the two graphics pipelines instead.] It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to incorporate the teachings of Palmer to configure the processor to change the assignment of at least one of a portion of the set of workloads from one graphics pipeline to the other. Reassigning/migrating tasks from one pipeline to another is a common technique done to prevent one pipeline from staying idle. One of ordinary skill in the art could make this combination in order to produce the predictable result of a system that has dynamic load balancing and less idle time when one pipeline is overloaded.
Regarding claim 4, Leather in view of Wong and Palmer teaches the method of claim 3. Leather in view of Wong does not teach wherein the at least one processor, individually or in any combination, is further configured to: determine that at least one of (1) the portion of the first subset of the set of workloads is not to be rendered by the first graphics pipeline or (2) the portion of the second subset of the set of workloads is not to be rendered by the second graphics pipeline, wherein the change of assignment is based on the determination. However, Palmer further teaches wherein the at least one processor, individually or in any combination, is further configured to: determine that at least one of (1) the portion of the first subset of the set of workloads is not to be rendered by the first graphics pipeline or (2) the portion of the second subset of the set of workloads is not to be rendered by the second graphics pipeline (the parallel processor may be configured to drain any pending graphics work from at least a portion of the graphics pipeline prior to reallocating the at least one processing unit to process tasks associated with the compute pipeline” – Col 3, Lines 61-65. [NOTE: Palmer’s disclosure states that when a decision to reallocate the tasks from the graphics pipeline to the compute pipeline is made, the processor is configured to drain any pending work which implies that no further rendering in the initial pipeline can be done.]), wherein the change of assignment is based on the determination (After the graphics WFI command has executed, the processor may reallocate processor resources from the graphics pipeline to the compute pipeline in order to process the synchronous compute tasks” – Col 14, Lines 26-29 [NOTE: The wait-for-idle (WFI) command waits for the graphics pipeline to be drained from its tasks before the system is allowed to move the workload from the first pipeline to the second)]. It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to further incorporate the teachings of Palmer to prevent the first and second subset of the workloads from rendering in their respective graphics pipelines. One of ordinary skill would opt to implement this known technique in graphics processing to avoid unnecessary rendering which takes up a lot of resources to execute.
Regarding claim 5, Leather in view of Wong and Palmer teach the apparatus of claim 4. Leather in view of Wong does not teach wherein the determination is based on at least one of the first graphics pipeline being idle or the second graphics pipeline being idle. However, Palmer further teaches wherein the determination is based on at least one of the first graphics pipeline being idle or the second graphics pipeline being idle (“Once the TPCs 320 associated with the graphics pipeline are idle, then the TPC Manager 710 can allocate all of the TPCs 320 to the compute pipeline” – Col 19, Lines 31-33). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to further incorporate the teachings of Palmer to make the determination that the portion of the first and second subsets should not be rendered based on whether or not the graphics pipeline is idle. It is known in the art that when a graphics pipeline is idle, then it no longer performs a rendering operation. This determination yields the predictable result of the system being more efficient in communicating what workloads should refrain from rendering.
Regarding claim 6, Leather in view of Wong and Palmer teach the apparatus of claim 3. Leather in view of Wong does not teach wherein the at least one processor, individually or in any combination, is further configured to: output an indication of the change of assignment of at least one of (1) the portion of the first subset of the set of workloads to the second graphics pipeline or (2) the portion of the second subset of the set of workloads to the first graphics pipeline. However, Palmer further teaches wherein the at least one processor, individually or in any combination, is further configured to: output an indication of the change of assignment of at least one of (1) the portion of the first subset of the set of workloads to the second graphics pipeline or (2) the portion of the second subset of the set of workloads to the first graphics pipeline (“Once the TPC Manager 710 has received the acknowledge message, the TPC Manager 710 transmits a message to the compute work distribution unit 225 that indicates that the TPCs 320 have been allocated to the compute pipeline.” – Col 17, Lines 64-67 and Col 18, Line 1. [NOTE: the acknowledgement message tells the Texture Processing Clusters (TPC) Manager that the graphics workload (first pipeline) can safely be reallocated to the compute pipeline 9second pipeline) and another message is transmitted when to change of assignment occurs]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to further incorporate the teachings of Palmer to output an indication of a change of assignment of at least one of the subsets of the set of workloads from the initial graphics pipeline to the other. Having some form of indication to signal the change of assignment of the workloads is a common practice in the art that has the known result of improving system communication to prevent latency. Wong’s use of a form of message to indicate the change of assignment would also benefit Leather’s graphics system by allowing the user to keep track of the assignment changes based on the indications.
Claim(s) 7 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, Palmer, and Alla et al (US 11176734 B1), hereinafter Alla.
Regarding claim 7, Leather in view of Wong and Palmer teaches the apparatus of claim 6. Leather in view of Wong and Palmer, does not teach wherein to output the indication of the change of assignment, the at least one processor, individually or in any combination, is configured to: output the indication of the change of assignment via a visibility stream in a graphics pipeline. However, Alla teaches wherein to output the indication of the change of assignment, the at least one processor, individually or in any combination, is configured to: output the indication of the change of assignment via a visibility stream in a graphics pipeline (“In some aspects, the visibility stream can be analyzed to determine which primitives, or vertices of primitives, are visible or not visible.” – Col 10, Lines 24-26. [NOTE: Alla only discloses the use of visibility streams to determine the visible and invisible primitives. After the combination, outputting an indication of the change of assignment as taught by Palmer can use the visibility streams as taught by Alla to teach wherein to output the indication of the change of assignment, the at least one processor, individually or in any combination, is configured to: output the indication of the change of assignment via a visibility stream in a graphics pipeline]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong and Palmer to incorporate the teachings of Alla to use a visibility stream to output the indication of the change of assignment. It is known in the art that visibility streams contain information concerning what primitives in the workload are visible or not. This plays a critical role in determining how rendering should be executed since it unnecessary to take up resources to render invisible primitives. One of ordinary skill could combine the visibility stream with the indication of a workload’s change in assignment since it would be natural for the visibility stream to convey the state of a workload such as when it is being reassigned to a different graphics pipeline.
Claim(s) 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, and Shao et al (US 6982713 B2), hereinafter Shao.
Regarding claim 11, Leather in view of Wong teaches the apparatus of claim 8. Leather in view of Wong, does not teach wherein the first resource is a first depth buffer and the second resource is a second depth buffer, and wherein the first depth buffer and the second depth buffer are cleared prior to storage of the first data associated with the first subset of the set of workloads in the first depth buffer and the second data associated with the second subset of the set of workloads in the second depth buffer. However, Shao teaches wherein the first resource is a first depth buffer and the second resource is a second depth buffer, and wherein the first depth buffer and the second depth buffer are cleared prior to storage of the first data associated with the first subset of the set of workloads in the first depth buffer and the second data associated with the second subset of the set of workloads in the second depth buffer (“At the beginning of a frame, the depth buffer has to be cleared to a value specified by the application so that the depth engine can process pixels correctly” – Col 1, Lines 66-67 and Col 2, Line 1. [NOTE: Shao teaches the use of depth buffers and how they need to be cleared before any processing occurs. After the combination, Leather’s graphics system, which includes a first and second graphics pipeline, can incorporate Wong’s resource manager to allocate resources to the workloads in the first and second graphics pipeline and Shao’s use of the depth buffer to teach wherein the first resource is a first depth buffer and the second resource is a second depth buffer, and wherein the first depth buffer and the second depth buffer are cleared prior to storage of the first data associated with the first subset of the set of workloads in the first depth buffer and the second data associated with the second subset of the set of workloads in the second depth buffer). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to incorporate the teachings of Shao to have the first resource be a first depth buffer and second resource be a second depth buffer. It is common in the art to use depth buffers as conventional resources for storing data related to depth associated with workloads. Allocating a first and second depth buffer to the first and second subset of the set of workloads will allow for independent depth testing.
Claim(s) 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, M. Leather (US 8933945 B2), and Akenine-Moller et al (US 20180082464 A1), hereinafter M. Leather and Akenine-Moller respectively.
Regarding claim 12, Leather in view of Wong teaches the apparatus of claim 1. Leather in view of Wong does not teach, wherein a first portion of the first graphics pipeline and a second portion of the second graphics pipeline overlap at a shared graphics pipeline, wherein the shared graphics pipeline is a pixel pipeline. However, M. Leather teaches wherein a first portion of the first graphics pipeline and a second portion of the second graphics pipeline overlap at a shared graphics pipeline, wherein the shared graphics pipeline is a pixel pipeline (“the first graphics pipeline 101 and the back end circuitry 42 of the second graphics pipeline 102 share the front end circuitry 35, in that the first and second graphics pipelines 101 and 102 receive the same pixel data 36 provided by the front end circuitry 35” – Col 4, Lines 18-22. [NOTE: M. Leather discloses that the first and second graphics pipeline share the front-end circuitry which receives primitive data and handles processing to generate pixel data.]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to incorporate the teachings of M. Leather to have a portion of the first and second graphics pipelines overlap at a shared pixel graphics pipeline. It is a known technique to have a portion of the pipelines share hardware/resources to prevent duplicating hardware and allows multiple workload streams to utilize common processing stages. Therefore, this design choice based on the combination would yield predictable results such has dynamic balancing of workloads and efficient utilization of hardware/resources. Leather in view of Wong and M. Leather still does not teach wherein the set of workloads is a set of depth-only workloads. However, Akenine-Moller teaches wherein the set of workloads is a set of depth-only workloads (“At 1500, the state of the graphics pipeline is initially set to “depth only” to perform the Z-prepass operation.” - Par 130, Lines 6-8.). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong and M. Leather to incorporate the teachings of Akenine-Moller to have the set of workloads be a set of depth-only workloads. Depth-only workloads are known in the art for determining visibility information without using the resources for full shading and color rendering. Having the set of workloads be a set of depth-only workloads is a common form of optimization.
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, and Alla.
Regarding claim 13, Leather in view of Wong teaches the apparatus of claim 1. Leather in view of Wong does not teach wherein the set of workloads comprises a set of depth pass workloads, and wherein the assignment is based on a change in a depth direction for the set of depth pass workloads. However, Alla teaches wherein the set of workloads comprises a set of depth pass workloads (“FIG. 4 is an example flow diagram illustrating example operations 400 for LRZ depth testing by a GPU. In certain aspects, the operations 400 may be performed by a GPU, such as in the graphics processing pipeline 107 of the processing unit 120 of FIG. 1.” – Col 15, Lines 6-10. [NOTE: GPU performs depth testing related workload/tasks in the graphics pipeline]), and wherein the assignment is based on a change in a depth direction for the set of depth pass workloads (“The GPU may be configured to disable use of the LRZ buffer for depth testing for any of the plurality of commands remaining unprocessed after processing a command of the plurality of commands having a different depth test direction than the corresponding depth test direction stored in the LRZ status buffer.” – Col 2, Lines 4-9. [NOTE: The change in depth test direction causes a change in the behavior of the graphics pipeline which one of skill in the art could combine with Leather’s graphics pipeline to then assign the set of workloads based on the change in depth direction]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to incorporate the teachings of Alla to base the assignment of the set of depth pass workloads to their respective graphics pipeline on the change of depth direction. It is known in the art that assigning depth-only workloads to different graphics pipelines based on a change in depth direction to preserve early depth testing and maintain correct depth processing. Graphics pipelines will maintain the depth test state and reassign workloads configured for the new depth direction when it changes.
Claim(s) 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, and Gruber (US 20200098165 A1)), hereinafter Gruber.
Regarding claim 14, Leather in view of Wong teaches the apparatus of claim 1. Leather in view of Wong does not teach wherein the assignment is based on at least one of: the set of workloads being associated with an occlusion query, a stencil being enabled, a depth test direction change, a memory space size limit, a set of mapping table entries, a depth buffer load, a single primitive mode, or a bin rendering mode. However, Gruber teaches wherein the assignment is based on at least one of: the set of workloads being associated with an occlusion query, a stencil being enabled, a depth test direction change, a memory space size limit, a set of mapping table entries, a depth buffer load (“The first pass of the depth pre-pass results in populating the depth buffer or depth surface 308, such that the depth surface 308 stores information related to the surface” – Par 76, Lines 11-14. [NOTE: Gruber discloses the operation of populating/loading a depth buffer with depth information for further processing. After the combination, the graphics system disclosed in Leather can consider the content loaded in the depth buffer to assign the set of workloads to the graphics pipelines]), a single primitive mode, or a bin rendering mode. It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to incorporate the teachings of Gruber to have the assignment of the set of workloads be based on their association with a depth buffer load. It is standard practice to consider the depth buffer load when assigning workloads to graphics pipelines since it promotes dynamic load balancing between both graphics’ pipelines. One of ordinary skill would be motivated to make this modification for the predicted result of optimizing depth buffer resources and conventional optimization for workload distribution.
Claim(s) 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, and Øygard et al (US 20240169645 A1), hereinafter Øygard.
Regarding claim 15, Leather in view of Wong teaches the apparatus of claim 1. Leather in view of Wong does not teach wherein the at least one processor, individually or in any combination, is further configured to: determine that at least one of (1) a portion of the first subset of the set of workloads has already been rendered in the second graphics pipeline or (2) a portion of the second subset of the set of workloads has already been rendered in the first graphics pipeline; and refrain from rendering at least one of (1) the portion of the first subset of the set of workloads or (2) the portion of the second subset of the workloads in the second graphics pipeline based on the determination. However, Øygard teaches wherein the at least one processor, individually or in any combination, is further configured to: determine that at least one of (1) a portion of the first subset of the set of workloads has already been rendered in the second graphics pipeline or (2) a portion of the second subset of the set of workloads has already been rendered in the first graphics pipeline (“For example, some form of hidden surface removal may be performed before a primitive and/or fragment is sent for rendering, to see if the primitive or fragment etc. will be obscured by a primitive that has already been rendered” – Par 28, Lines 1-4. [NOTE: Øygard discloses that some geometry data such as primitives may be hidden if that particular geometry data has already been rendered and therefore skipped from rendering. This shows that the geometry data within a workload is checked to see if it has already been rendered]); and refrain from rendering at least one of (1) the portion of the first subset of the set of workloads or (2) the portion of the second subset of the workloads in the second graphics pipeline based on the determination (“(in which case the new fragment and/or primitive need not be rendered)” – Par 29, Lines 5-6). [NOTE: After the combination, Leather teaches the first and second graphics pipelines that can determine which portions of the first or second subsets of the sets of workloads have been rendered by incorporating the determination method of Øygard. The determination method also prevents the rendering of the workloads if it has already been done. This combination would then teach wherein the at least one processor, individually or in any combination, is further configured to: determine that at least one of (1) a portion of the first subset of the set of workloads has already been rendered in the second graphics pipeline or (2) a portion of the second subset of the set of workloads has already been rendered in the first graphics pipeline and refrain from rendering at least one of (1) the portion of the first subset of the set of workloads or (2) the portion of the second subset of the workloads in the second graphics pipeline based on the determination.] It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to incorporate the teachings of Øygard to configure the processor to determine if either a portion of the first or second subset of the set of workloads has been rendered and to refrain from rendering the portion of either the first or second subsets of the set of workloads based on that determination. Checking whether or not a workload has already been rendered communicates to the system what work has already been done in order to prevent redundant rendering which takes up unnecessary computing resources.
Claim(s) 17-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Leather, Wong, and Foo (US 20180158168 A1) hereinafter Foo.
Regarding claim 17, Leather in view of Wong teaches the apparatus of claim 1. Leather in view of Wong does not teach wherein the at least one processor, individually or in any combination, is further configured to: output an indication of the rendered first subset of the set of workloads and the rendered second subset of the set of workloads. However, Foo teaches wherein the at least one processor, individually or in any combination, is further configured to: output an indication of the rendered first subset of the set of workloads and the rendered second subset of the set of workloads (“propagating an indication of the flag through stages of the graphics pipeline as the scene is rendered so that work done as part of the at least one rendering task is associated with the set flag” – Abstract [NOTE: Foo discloses a flag that gives an indication of completion of each stage in their graphics pipeline. One of those flags could be a flag to indicate the rendered workload]). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wong to incorporate the teachings of Foo to output an indication of the first and second rendered subsets of the set of workloads. Having some form of indication to convey that the first and second subset of the workload has been rendered will improve system communication and aid in the preparation for the next set of workloads.
Regarding claim 18, Leather in view of Wong teaches the apparatus of claim 17. Leather in view of Wong does not teach wherein to output the indication of the rendered first subset of the set of workloads and the rendered second subset of the set of workloads, the at least one processor, individually or in any combination, is configured to: transmit the indication of the rendered first subset of the set of workloads and the rendered second subset of the set of workloads. However, Foo further teaches wherein to output the indication of the rendered first subset of the set of workloads and the rendered second subset of the set of workloads, the at least one processor, individually or in any combination, is configured to: transmit the indication of the rendered first subset of the set of workloads and the rendered second subset of the set of workloads; or store the indication of the rendered first subset of the set of workloads and the rendered second subset of the set of workloads. (“the graphics unit may be configured to store an indication of the set flag in the parameter buffer as part of the intermediate data generated from the geometry processing performed as part of the at least one rendering task.” – Par 37, Lines 4-8). It would have been obvious to one of ordinary skill in the art before the effective filing date to modify Leather in view of Wang to incorporate the teachings of Foo to store the indication of the rendered first and second subsets of the set of workloads. By storing the indication of the rendered first and second subsets of the set of workloads, the graphics system will have a predicted result of reliably communicating what portion of the workload has been reassigned, what has been rendered, and what not to render.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DAVID V. NGUYEN whose telephone number is (571)272-6111. The examiner can normally be reached M-F 7:30-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, King Y Poon can be reached at 571-270-0728. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/DAVID VAN NGUYEN/
Examiner, Art Unit 2617 /KING Y POON/Supervisory Patent Examiner, Art Unit 2617