FauxVkDB
English
Home / Adreno (TM) 740 / Compare

turnip_mrpurple_T26-toasted vs Qualcomm Technologies Inc. Adreno Vulkan Driver

On Adreno (TM) 740 (AYN Odin2, AYN Odin2 Portal, AYN Thor, Moorechip Retroid Pocket Nova, motorola edge plus 2023, OnePlus 11 5G, POCO F6 Pro, Samsung Galaxy S23, Samsung Galaxy S23, Samsung Galaxy S23, Samsung Galaxy S23 Ultra, Samsung Galaxy S23 Ultra, Samsung Galaxy S23+, Sony Xperia 1 V, Xiaomi 13 pro). The first driver named is compared against the second, which is the baseline.

StockQualcomm Technologies Inc. Adreno Vulkan DriverDriver Build: 3df4c357f0, I515341007c, 1769513954
vs
Customturnip_mrpurple_T26-toastedPurpleVK 26.1.0-devel (git-1c4d081f8c)
Graphics system version (Vulkan)1.3.128 to 1.4.344
Optional features (extensions)115 to 166+61 -10
Features that pass their check (validated)41 to 499 pass only on turnip_mrpurple_T26-toasted 1 pass only on Qualcomm Technologies Inc. Adreno Vulkan Driver
Speed test scenes (benchmark)3 faster 0 slower13% faster overall (geometric mean)
Games each driver could run (an estimate from its feature list): green likely, yellow maybe, orange limited, red no
StockDX92002+DX102006+DX112009+DX122015+Switch
CustomDX92002+DX102006+DX112009+DX122015+Switch
  • hdr_2160p6% faster
  • pipeline_barriers_2160p25% faster
  • subpasses_2160p9% faster

Custom driver against stock, per scene. Bars to the right are faster, to the left slower.

Small areas of the picture differ from the stock driver in 2 scenes (pipeline_barriers: 1791 of 2304 cells, pipeline_barriers_2160p: 1676 of 2304 cells). The picture matches overall, so this is a local rendering fault, not a different scene.

Did not start every time: compute_nbody (stock 3 of 3, custom 0 of 3).

Not on equal terms: the GPU clock differed during 1 of these scenes (for example pipeline_barriers_2160p: stock at 90% of the device's maximum, custom at 100%). Part of the frame-time difference comes from the clock the device chose, not from the driver.

Qualcomm Technologies Inc. Adreno Vulkan Driver details turnip_mrpurple_T26-toasted details

Feature checks that differ (35)

FeatureStockCustomChange
core.compute_udiv
core
passfaillost
borderColorSwizzle
VK_EXT_border_color_swizzle
unsupportedpassgained
extendedDynamicState3ColorBlendEnable
VK_EXT_extended_dynamic_state3
unsupportedpassgained
extendedDynamicState3ColorBlendEquation
VK_EXT_extended_dynamic_state3
unsupportedpassgained
extendedDynamicState3ColorWriteMask
VK_EXT_extended_dynamic_state3
unsupportedpassgained
extendedDynamicState3DepthClampEnable
VK_EXT_extended_dynamic_state3
unsupportedpassgained
extendedDynamicState3LogicOpEnable
VK_EXT_extended_dynamic_state3
unsupportedpassgained
extendedDynamicState3PolygonMode
VK_EXT_extended_dynamic_state3
unsupportedpassgained
inlineUniformBlock
VK_EXT_inline_uniform_block
unsupportedpassgained
shaderBufferInt64Atomics
VK_KHR_shader_atomic_int64
unsupportedpassgained
borderColorSwizzleFromImage
VK_EXT_border_color_swizzle
unsupporteduntestedchanged
extendedDynamicState3AlphaToCoverageEnable
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3AlphaToOneEnable
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3ConservativeRasterizationMode
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3DepthClipEnable
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3DepthClipNegativeOneToOne
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3ExtraPrimitiveOverestimationSize
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3LineRasterizationMode
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3ProvokingVertexMode
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3RasterizationSamples
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3RasterizationStream
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3SampleLocationsEnable
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3SampleMask
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
extendedDynamicState3TessellationDomainOrigin
VK_EXT_extended_dynamic_state3
unsupporteduntestedchanged
descriptorBindingInlineUniformBlockUpdateAfterBind
VK_EXT_inline_uniform_block
unsupporteduntestedchanged
shaderInt64
VkPhysicalDeviceFeatures
unsupporteduntestedchanged
shaderResourceMinLod
VkPhysicalDeviceFeatures
unsupporteduntestedchanged
shaderResourceResidency
VkPhysicalDeviceFeatures
unsupporteduntestedchanged
shaderTessellationAndGeometryPointSize
VkPhysicalDeviceFeatures
unsupporteduntestedchanged
sparseResidency16Samples
VkPhysicalDeviceFeatures
untestedunsupportedchanged
sparseResidency2Samples
VkPhysicalDeviceFeatures
untestedunsupportedchanged
sparseResidency4Samples
VkPhysicalDeviceFeatures
untestedunsupportedchanged
sparseResidency8Samples
VkPhysicalDeviceFeatures
untestedunsupportedchanged
sparseResidencyAliased
VkPhysicalDeviceFeatures
unsupporteduntestedchanged
sparseResidencyImage2D
VkPhysicalDeviceFeatures
untestedunsupportedchanged

Speed test scenes (benchmark)

The two drivers were not measured on the same set of device models. Frame times from different devices do not compare drivers, so read these numbers with that in mind.

SceneStock msCustom msDifferenceStartsGPU clock, MHzPicture vs stock
compute_nbody7.84 typical 7.85n/aonly one side3 of 3 vs 0 of 3561 vs ? clock unknownnot checked
compute_nbody_2x25.49 typical 25.89n/aonly one side3 of 3 vs ?550 vs ? clock unknownnot checked
hdr_2160p26.73 typical 25.7725.33 typical 25.266% faster3 of 3 vs 3 of 3550 vs 680 no large difference 0.0 no small difference worst 0.3 64x36
oit_linked_listsn/a25.30only one side3 of 3 vs 3 of 3719 vs 680 no large difference 0.0 no small difference worst 0.0 64x36
pipeline_barriersn/a2.78too light, see pipeline_barriers_2160p3 of 3 vs 3 of 3680 vs 572 not comparableno large difference 8.8 differs in 1791 small areas worst 86.0 64x36
pipeline_barriers_2160p9.55 typical 9.227.62 typical 7.6325% faster3 of 3 vs 3 of 3495 vs 680 not comparableno large difference 11.2 differs in 1676 small areas worst 76.7 64x36
subpasses5.53 typical 4.574.46 typical 4.41too light, see subpasses_2160p3 of 3 vs 3 of 3549 vs 680 not comparablenot checked
subpasses_2160p15.58 typical 15.4014.30 typical 14.309% faster3 of 3 vs 3 of 3550 vs 680 not checked

Time to draw one frame, per scene: the average (mean) frame time of the scene's best round, which is how many frames the driver delivered; lower is faster. "Typical" beside it is the median frame, which can sit well under the average when a driver's frames come in two speeds. A pair of reports made before per-round averages were recorded is compared on the median. A scene whose rounds disagree by more than 15% is left out. A scene marked "paced" was held at that mean frame time by the driver or the display queue, together with at least one other scene of the same run; its time says how the frames were spaced, not how fast the driver rendered, so it is not compared. A scene marked "too light" took under 8 ms at the standard size (1920 by 1080) on both drivers, where the screen's frame queue spaces the frames more than the chip does; its heavier run is the one compared: the row ending in _2160p is the scene drawn at 3840 by 2160, and compute_nbody_2x is that scene with twice the particles. "Faster" and "slower" describe turnip_mrpurple_T26-toasted against Qualcomm Technologies Inc. Adreno Vulkan Driver. "Picture vs stock" compares one fixed frame per scene as a coarse colour grid (0 is identical, above 40 is a different picture); it finds a scene that was not drawn. Where both reports carry the finer 64x36 grid, the second figure is how many of its 2,304 cells differ by more than 4 of 255, and "worst" is the largest single cell; this is what finds a small rendering error in a picture that otherwise matches. A row marked 16x9 comes from a report made before the finer grid existed. "Starts" is how many of a scene's process starts reached a measured round. A GPU clock marked "not comparable" differed by more than 5% of the device's maximum between the two timings.

Technical differences (Vulkan)

Extensions only in turnip_mrpurple_T26-toasted (61)
  • VK_AMD_buffer_marker
  • VK_ARM_rasterization_order_attachment_access
  • VK_EXT_attachment_feedback_loop_dynamic_state
  • VK_EXT_attachment_feedback_loop_layout
  • VK_EXT_border_color_swizzle
  • VK_EXT_conditional_rendering
  • VK_EXT_custom_resolve
  • VK_EXT_depth_clamp_zero_one
  • VK_EXT_descriptor_buffer
  • VK_EXT_device_address_binding_report
  • VK_EXT_dynamic_rendering_unused_attachments
  • VK_EXT_extended_dynamic_state3
  • VK_EXT_external_memory_dma_buf
  • VK_EXT_fragment_density_map_offset
  • VK_EXT_graphics_pipeline_library
  • VK_EXT_host_image_copy
  • VK_EXT_image_2d_view_of_3d
  • VK_EXT_image_drm_format_modifier
  • VK_EXT_inline_uniform_block
  • VK_EXT_legacy_dithering
  • VK_EXT_legacy_vertex_attributes
  • VK_EXT_map_memory_placed
  • VK_EXT_memory_budget
  • VK_EXT_multi_draw
  • VK_EXT_multisampled_render_to_single_sampled
  • VK_EXT_mutable_descriptor_type
  • VK_EXT_nested_command_buffer
  • VK_EXT_non_seamless_cube_map
  • VK_EXT_primitives_generated_query
  • VK_EXT_rasterization_order_attachment_access
  • VK_EXT_shader_module_identifier
  • VK_EXT_shader_replicated_composites
  • VK_EXT_shader_uniform_buffer_unsized_array
  • VK_EXT_shader_viewport_index_layer
  • VK_EXT_zero_initialize_device_memory
  • VK_GOOGLE_decorate_string
  • VK_GOOGLE_hlsl_functionality1
  • VK_KHR_compute_shader_derivatives
  • VK_KHR_dynamic_rendering_local_read
  • VK_KHR_index_type_uint8
  • VK_KHR_line_rasterization
  • VK_KHR_load_store_op_none
  • VK_KHR_maintenance5
  • VK_KHR_maintenance6
  • VK_KHR_maintenance7
  • VK_KHR_map_memory2
  • VK_KHR_pipeline_library
  • VK_KHR_ray_tracing_maintenance1
  • VK_KHR_shader_atomic_int64
  • VK_KHR_shader_clock
  • VK_KHR_shader_expect_assume
  • VK_KHR_shader_float_controls2
  • VK_KHR_shader_relaxed_extended_instruction
  • VK_KHR_shader_subgroup_rotate
  • VK_KHR_vertex_attribute_divisor
  • VK_KHR_workgroup_memory_explicit_layout
  • VK_NV_compute_shader_derivatives
  • VK_QCOM_multiview_per_view_render_areas
  • VK_QCOM_multiview_per_view_viewports
  • VK_VALVE_fragment_density_map_layered
  • VK_VALVE_mutable_descriptor_type
Extensions only in Qualcomm Technologies Inc. Adreno Vulkan Driver (10)
  • VK_EXT_astc_decode_mode
  • VK_EXT_blend_operation_advanced
  • VK_EXT_fragment_density_map2
  • VK_EXT_swapchain_maintenance1
  • VK_KHR_shared_presentable_image
  • VK_KHR_swapchain_mutable_format
  • VK_QCOM_render_pass_store_ops
  • VK_QCOM_render_pass_transform
  • VK_QCOM_rotated_copy_commands
  • VK_QCOM_tile_properties
Core features that differ (22)
protectedMemory Vulkan 1.1pass to unsupported
shaderBufferInt64Atomics Vulkan 1.2unsupported to pass
shaderInputAttachmentArrayDynamicIndexing Vulkan 1.2pass to unsupported
shaderInputAttachmentArrayNonUniformIndexing Vulkan 1.2pass to unsupported
shaderOutputLayer Vulkan 1.2unsupported to pass
shaderOutputViewportIndex Vulkan 1.2unsupported to pass
bresenhamLines Vulkan 1.4unsupported to pass
dynamicRenderingLocalRead Vulkan 1.4unsupported to pass
globalPriorityQuery Vulkan 1.4unsupported to pass
hostImageCopy Vulkan 1.4unsupported to pass
indexTypeUint8 Vulkan 1.4unsupported to pass
maintenance5 Vulkan 1.4unsupported to pass
maintenance6 Vulkan 1.4unsupported to pass
pipelineRobustness Vulkan 1.4unsupported to pass
pushDescriptor Vulkan 1.4unsupported to pass
rectangularLines Vulkan 1.4unsupported to pass
shaderExpectAssume Vulkan 1.4unsupported to pass
shaderFloatControls2 Vulkan 1.4unsupported to pass
shaderSubgroupRotate Vulkan 1.4unsupported to pass
shaderSubgroupRotateClustered Vulkan 1.4unsupported to pass
vertexAttributeInstanceRateDivisor Vulkan 1.4unsupported to pass
vertexAttributeInstanceRateZeroDivisor Vulkan 1.4unsupported to pass
Limits that differ (19)
LimitStockCustom
bufferImageGranularity164
sparseAddressSpaceSize562949953421311274877906944
maxPerStageResources5033164816777216
maxDescriptorSetUniformBuffersDynamic3216
maxDescriptorSetStorageBuffersDynamic168
maxVertexInputAttributeOffset40964095
maxGeometryInputComponents12864
maxGeometryOutputVertices1024256
maxGeometryTotalOutputComponents1310721024
maxFragmentInputComponents112124
maxComputeWorkGroupInvocations10242048
minMemoryMapAlignment644096
minStorageBufferOffsetAlignment644
maxFramebufferLayers20481024
timestampPeriod52.083333.1126
discreteQueuePriorities32
optimalBufferCopyOffsetAlignment64128
optimalBufferCopyRowPitchAlignment64128
nonCoherentAtomSize164
Formats supported by only one driver (29 gained, 1 lost)

Only in turnip_mrpurple_T26-toasted:

  • A1B5G5R5_UNORM_PACK16
  • A2B10G10R10_SSCALED_PACK32
  • A2B10G10R10_USCALED_PACK32
  • A2R10G10B10_SSCALED_PACK32
  • A2R10G10B10_USCALED_PACK32
  • A8B8G8R8_SSCALED_PACK32
  • A8B8G8R8_USCALED_PACK32
  • A8_UNORM
  • B8G8R8A8_SSCALED
  • B8G8R8A8_USCALED
  • B8G8R8_SINT
  • B8G8R8_SNORM
  • B8G8R8_UINT
  • R16G16B16A16_SSCALED
  • R16G16B16A16_USCALED
  • R16G16B16_SSCALED
  • R16G16B16_USCALED
  • R16G16_SSCALED
  • R16G16_USCALED
  • R16_SSCALED
  • R16_USCALED
  • R8G8B8A8_SSCALED
  • R8G8B8A8_USCALED
  • R8G8B8_SSCALED
  • R8G8B8_USCALED
  • R8G8_SSCALED
  • R8G8_USCALED
  • R8_SSCALED
  • R8_USCALED

Only in Qualcomm Technologies Inc. Adreno Vulkan Driver:

  • G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16