FauxVkDB
简体中文
首页 / Samsung Xclipse 920 / 对比

MdiEx_2200_Vortek_v3.2 对比 mesa_radv_xclipse

在 Samsung Xclipse 920 上 (Samsung Galaxy S22, Samsung Galaxy S22 Ultra, Samsung Galaxy S23 FE, Samsung Galaxy S24 FE, Samsung Galaxy S24+). 先列出的驱动与后一个对比，后一个是基准。

原厂mesa_radv_xclipseMesa 26.3.0-devel (git-f1ab583ed7)
对比
第三方MdiEx_2200_Vortek_v3.2Revision: 0448fa4c23
图形系统版本（Vulkan）1.4.363 → 1.3.231
可选功能（扩展）224 → 149+10 -85
通过检测的功能（已验证）52 → 6018 个仅在 MdiEx_2200_Vortek_v3.2 上通过 10 个仅在 mesa_radv_xclipse 上通过
跑分测试场景（基准测试）没有共同场景
各驱动能跑的游戏（根据功能列表估算）：绿色基本能跑，黄色可能能跑，橙色问题较多，红色跑不了
原厂DX92002+DX102006+DX112009+DX122015+Switch
第三方DX92002+DX102006+DX112009+DX122015+Switch

mesa_radv_xclipse 详情 MdiEx_2200_Vortek_v3.2 详情

结果不同的功能检测（45）

功能mesa_radv_xclipseMdiEx_2200_Vortek_v3.2变化
floatRepresentation
VK_EXT_depth_bias_control
通过不支持减少
extendedDynamicState2LogicOp
VK_EXT_extended_dynamic_state2
通过不支持减少
extendedDynamicState3ColorBlendEnable
VK_EXT_extended_dynamic_state3
通过不支持减少
extendedDynamicState3ColorBlendEquation
VK_EXT_extended_dynamic_state3
通过不支持减少
extendedDynamicState3ColorWriteMask
VK_EXT_extended_dynamic_state3
通过不支持减少
extendedDynamicState3DepthClampEnable
VK_EXT_extended_dynamic_state3
通过不支持减少
extendedDynamicState3LogicOpEnable
VK_EXT_extended_dynamic_state3
通过不支持减少
extendedDynamicState3PolygonMode
VK_EXT_extended_dynamic_state3
通过不支持减少
vertexInputDynamicState
VK_EXT_vertex_input_dynamic_state
通过不支持减少
textureCompressionBC
VkPhysicalDeviceFeatures
通过不支持减少
depthClamp
VkPhysicalDeviceFeatures
未检测通过新增
largePoints
VkPhysicalDeviceFeatures
未检测通过新增
compat.helper_loop_terminator
core
未检测通过新增
compat.nested_map
core
未检测通过新增
core.blend_advertised_formats
core
未检测通过新增
core.compute_big_stack
core
未检测通过新增
core.host_coherent_readback
core
未检测通过新增
core.mrt_partial_kill
core
未检测通过新增
core.ubo_update_buffer
core
未检测通过新增
core.ubo_written_after_record
core
未检测通过新增
core.varying_subset
core
未检测通过新增
mem.image_padding
core
未检测通过新增
perf.fb_local_barrier_large_draws
core
未检测通过新增
perf.fb_local_barrier_overdraw
core
未检测通过新增
stage.geometry_max_output
core
未检测通过新增
stage.geometry_point_size
core
未检测通过新增
stress.scratch_per_batch
core
未检测通过新增
sync.first_submit_clear_then_compute
core
未检测通过新增
depthBiasControl
VK_EXT_depth_bias_control
未检测不支持有变化
depthBiasExact
VK_EXT_depth_bias_control
未检测不支持有变化
leastRepresentableValueForceUnormRepresentation
VK_EXT_depth_bias_control
未检测不支持有变化
extendedDynamicState2PatchControlPoints
VK_EXT_extended_dynamic_state2
未检测不支持有变化
extendedDynamicState3AlphaToCoverageEnable
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3AlphaToOneEnable
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3ConservativeRasterizationMode
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3DepthClipEnable
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3DepthClipNegativeOneToOne
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3LineRasterizationMode
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3LineStippleEnable
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3ProvokingVertexMode
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3RasterizationSamples
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3SampleLocationsEnable
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3SampleMask
VK_EXT_extended_dynamic_state3
未检测不支持有变化
extendedDynamicState3TessellationDomainOrigin
VK_EXT_extended_dynamic_state3
未检测不支持有变化
shaderResourceMinLod
VkPhysicalDeviceFeatures
未检测不支持有变化

跑分测试场景（基准测试）

两个驱动都还没有有效的跑分测试（基准测试报告）。

技术差异（Vulkan）

仅 MdiEx_2200_Vortek_v3.2 有的扩展（10）
  • VK_AMD_shader_fragment_mask
  • VK_AMD_shader_info
  • VK_EXT_external_memory_host
  • VK_EXT_pageable_device_local_memory
  • VK_EXT_progress_marker
  • VK_KHR_acceleration_structure
  • VK_KHR_ray_query
  • VK_KHR_ray_tracing_maintenance1
  • VK_KHR_ray_tracing_pipeline
  • VK_KHR_shared_presentable_image
仅 mesa_radv_xclipse 有的扩展（85）
  • VK_EXT_attachment_feedback_loop_dynamic_state
  • VK_EXT_cooperative_matrix_maintenance1
  • VK_EXT_custom_resolve
  • VK_EXT_depth_bias_control
  • VK_EXT_depth_clamp_control
  • VK_EXT_descriptor_buffer
  • VK_EXT_descriptor_heap
  • VK_EXT_discard_rectangles
  • VK_EXT_dynamic_rendering_unused_attachments
  • VK_EXT_extended_dynamic_state3
  • VK_EXT_external_memory_acquire_unmodified
  • VK_EXT_host_image_copy
  • VK_EXT_image_compression_control
  • VK_EXT_image_compression_control_swapchain
  • VK_EXT_image_sliced_view_of_3d
  • VK_EXT_image_view_min_lod
  • VK_EXT_legacy_vertex_attributes
  • VK_EXT_map_memory_placed
  • VK_EXT_multi_draw
  • VK_EXT_multisampled_render_to_single_sampled
  • VK_EXT_mutable_descriptor_type
  • VK_EXT_nested_command_buffer
  • VK_EXT_physical_device_drm
  • VK_EXT_pipeline_protected_access
  • VK_EXT_pipeline_robustness
  • VK_EXT_present_timing
  • VK_EXT_primitive_restart_index
  • VK_EXT_primitives_generated_query
  • VK_EXT_sampler_filter_minmax
  • VK_EXT_shader_atomic_float
  • VK_EXT_shader_atomic_float2
  • VK_EXT_shader_module_identifier
  • VK_EXT_shader_object
  • VK_EXT_shader_replicated_composites
  • VK_EXT_shader_uniform_buffer_unsized_array
  • VK_EXT_vertex_input_dynamic_state
  • VK_EXT_ycbcr_2plane_444_formats
  • VK_EXT_zero_initialize_device_memory
  • VK_GOOGLE_user_type
  • VK_INTEL_shader_integer_functions2
  • VK_KHR_calibrated_timestamps
  • VK_KHR_compute_shader_derivatives
  • VK_KHR_cooperative_matrix
  • VK_KHR_copy_memory_indirect
  • VK_KHR_depth_clamp_zero_one
  • VK_KHR_device_address_commands
  • VK_KHR_device_fault
  • VK_KHR_dynamic_rendering_local_read
  • VK_KHR_extended_flags
  • VK_KHR_index_type_uint8
  • VK_KHR_internally_synchronized_queues
  • VK_KHR_line_rasterization
  • VK_KHR_load_store_op_none
  • VK_KHR_maintenance10
  • VK_KHR_maintenance11
  • VK_KHR_maintenance5
  • VK_KHR_maintenance6
  • VK_KHR_maintenance7
  • VK_KHR_maintenance8
  • VK_KHR_maintenance9
  • VK_KHR_map_memory2
  • VK_KHR_performance_query
  • VK_KHR_pipeline_binary
  • VK_KHR_present_id
  • VK_KHR_present_id2
  • VK_KHR_present_wait
  • VK_KHR_present_wait2
  • VK_KHR_robustness2
  • VK_KHR_shader_abort
  • VK_KHR_shader_constant_data
  • VK_KHR_shader_expect_assume
  • VK_KHR_shader_float_controls2
  • VK_KHR_shader_fma
  • VK_KHR_shader_maximal_reconvergence
  • VK_KHR_shader_quad_control
  • VK_KHR_shader_relaxed_extended_instruction
  • VK_KHR_shader_subgroup_rotate
  • VK_KHR_shader_untyped_pointers
  • VK_KHR_swapchain_maintenance1
  • VK_KHR_unified_image_layouts
  • VK_KHR_vertex_attribute_divisor
  • VK_MESA_image_alignment_control
  • VK_NV_compute_shader_derivatives
  • VK_VALVE_mutable_descriptor_type
  • VK_VALVE_shader_mixed_float_dot_product
不同的核心功能（23）
multiviewGeometryShader Vulkan 1.1通过 → 不支持
storagePushConstant16 Vulkan 1.1通过 → 不支持
samplerFilterMinmax Vulkan 1.2通过 → 不支持
storagePushConstant8 Vulkan 1.2通过 → 不支持
bresenhamLines Vulkan 1.4通过 → 不支持
dynamicRenderingLocalRead Vulkan 1.4通过 → 不支持
globalPriorityQuery Vulkan 1.4通过 → 不支持
hostImageCopy Vulkan 1.4通过 → 不支持
indexTypeUint8 Vulkan 1.4通过 → 不支持
maintenance5 Vulkan 1.4通过 → 不支持
maintenance6 Vulkan 1.4通过 → 不支持
pipelineProtectedAccess Vulkan 1.4通过 → 不支持
pipelineRobustness Vulkan 1.4通过 → 不支持
pushDescriptor Vulkan 1.4通过 → 不支持
rectangularLines Vulkan 1.4通过 → 不支持
shaderExpectAssume Vulkan 1.4通过 → 不支持
shaderFloatControls2 Vulkan 1.4通过 → 不支持
shaderSubgroupRotate Vulkan 1.4通过 → 不支持
shaderSubgroupRotateClustered Vulkan 1.4通过 → 不支持
smoothLines Vulkan 1.4通过 → 不支持
stippledBresenhamLines Vulkan 1.4通过 → 不支持
vertexAttributeInstanceRateDivisor Vulkan 1.4通过 → 不支持
vertexAttributeInstanceRateZeroDivisor Vulkan 1.4通过 → 不支持
不同的限制（36）
限制mesa_radv_xclipseMdiEx_2200_Vortek_v3.2
maxTexelBufferElements5368709124294967295
maxPushConstantsSize256128
maxSamplerAllocationCount655361048576
sparseAddressSpaceSize140737488355328140707423584256
maxPerStageDescriptorSamplers83886064294967295
maxPerStageDescriptorUniformBuffers83886064294967295
maxPerStageDescriptorStorageBuffers83886064294967295
maxPerStageDescriptorSampledImages83886064294967295
maxPerStageDescriptorStorageImages83886064294967295
maxPerStageDescriptorInputAttachments83886064294967295
maxPerStageResources83886064294967295
maxDescriptorSetSamplers83886064294967295
maxDescriptorSetUniformBuffers83886064294967295
maxDescriptorSetUniformBuffersDynamic1624
maxDescriptorSetStorageBuffers83886064294967295
maxDescriptorSetSampledImages83886064294967295
maxDescriptorSetStorageImages83886064294967295
maxDescriptorSetInputAttachments83886064294967295
maxVertexInputAttributes3264
maxVertexInputBindingStride204816383
maxGeometryShaderInvocations32126
maxGeometryInputComponents64128
maxGeometryOutputVertices2561023
maxGeometryTotalOutputComponents10244095
maxFragmentCombinedOutputResources83886064294967295
maxComputeSharedMemorySize6553632768
maxComputeWorkGroupCount4294967295, 65535, 6553565535, 65535, 65535
maxSamplerLodBias1615.9961
minMemoryMapAlignment409664
minUniformBufferOffsetAlignment416
minTexelOffset-32-64
maxTexelOffset3163
maxInterpolationOffset21
maxFramebufferLayers10248192
lineWidthRange0, 80, 8191.88
nonCoherentAtomSize64128
只有一个驱动支持的格式（多了 4 种，少了 33 种）

仅 MdiEx_2200_Vortek_v3.2 有：

  • B8G8R8G8_422_UNORM
  • G12X4_B12X4R12X4_2PLANE_422_UNORM_3PACK16
  • G8B8G8R8_422_UNORM
  • R4G4_UNORM_PACK8

仅 mesa_radv_xclipse 有：

  • A1B5G5R5_UNORM_PACK16
  • A8_UNORM
  • B8G8R8_SINT
  • B8G8R8_SNORM
  • B8G8R8_SSCALED
  • B8G8R8_UINT
  • B8G8R8_UNORM
  • B8G8R8_USCALED
  • BC4_SNORM_BLOCK
  • BC4_UNORM_BLOCK
  • BC5_SNORM_BLOCK
  • BC5_UNORM_BLOCK
  • BC6H_SFLOAT_BLOCK
  • BC6H_UFLOAT_BLOCK
  • BC7_SRGB_BLOCK
  • BC7_UNORM_BLOCK
  • G10X6_B10X6R10X6_2PLANE_444_UNORM_3PACK16
  • G12X4_B12X4R12X4_2PLANE_444_UNORM_3PACK16
  • G16_B16R16_2PLANE_422_UNORM
  • G16_B16R16_2PLANE_444_UNORM
  • G16_B16_R16_3PLANE_420_UNORM
  • G16_B16_R16_3PLANE_422_UNORM
  • G16_B16_R16_3PLANE_444_UNORM
  • G8_B8R8_2PLANE_444_UNORM
  • G8_B8_R8_3PLANE_422_UNORM
  • G8_B8_R8_3PLANE_444_UNORM
  • R10X6G10X6_UNORM_2PACK16
  • R10X6_UNORM_PACK16
  • R12X4G12X4_UNORM_2PACK16
  • R12X4_UNORM_PACK16
  • R16G16B16_SFLOAT
  • R16G16B16_SINT
  • R16G16B16_UINT