8Elite2-842.6 vs stable-turnip-sync-V2
On Adreno (TM) 825 (Nothing Phone (3), OnePlus Nord 6, Oppo K13 Turbo Pro 5G, Poco F7, Poco F7, REDMI Turbo 4 Pro). The first driver named is compared against the second, which is the baseline.
Stockstable-turnip-sync-V2
vsCustom8Elite2-842.6
Graphics system version (Vulkan)1.4.354 to 1.4.295
Optional features (extensions)174 to 174+27 -27
Features that pass their check (validated)49 to 441 pass only on 8Elite2-842.6 6 pass only on stable-turnip-sync-V2
Speed test scenes (benchmark)5 faster 0 slower61% 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
- compute_nbody80% faster
- hdr_2160p139% faster
- oit_linked_lists36% faster
- pipeline_barriers_2160p29% faster
- subpasses_2160p42% faster
Small areas of the picture differ from the stock driver in 2 scenes (pipeline_barriers: 1337 of 2304 cells, pipeline_barriers_2160p: 1038 of 2304 cells). The picture matches overall, so this is a local rendering fault, not a different scene.
stable-turnip-sync-V2 details 8Elite2-842.6 details
Feature checks that differ (24)
| Feature | stable-turnip-sync-V2 | 8Elite2-842.6 | Change |
|---|---|---|---|
| extendedDynamicState3ColorBlendEnable | pass | unsupported | lost |
| extendedDynamicState3ColorBlendEquation | pass | unsupported | lost |
| extendedDynamicState3ColorWriteMask | pass | unsupported | lost |
| extendedDynamicState3DepthClampEnable | pass | unsupported | lost |
| extendedDynamicState3LogicOpEnable | pass | unsupported | lost |
| extendedDynamicState3PolygonMode | pass | unsupported | lost |
| core.compute_udiv | fail | pass | gained |
| extendedDynamicState3AlphaToCoverageEnable | untested | unsupported | changed |
| extendedDynamicState3AlphaToOneEnable | untested | unsupported | changed |
| extendedDynamicState3ConservativeRasterizationMode | untested | unsupported | changed |
| extendedDynamicState3DepthClipEnable | untested | unsupported | changed |
| extendedDynamicState3DepthClipNegativeOneToOne | untested | unsupported | changed |
| extendedDynamicState3ExtraPrimitiveOverestimationSize | untested | unsupported | changed |
| extendedDynamicState3LineRasterizationMode | untested | unsupported | changed |
| extendedDynamicState3ProvokingVertexMode | untested | unsupported | changed |
| extendedDynamicState3RasterizationSamples | untested | unsupported | changed |
| extendedDynamicState3RasterizationStream | untested | unsupported | changed |
| extendedDynamicState3SampleLocationsEnable | untested | unsupported | changed |
| extendedDynamicState3SampleMask | untested | unsupported | changed |
| extendedDynamicState3TessellationDomainOrigin | untested | unsupported | changed |
| shaderResourceMinLod | untested | unsupported | changed |
| shaderResourceResidency | untested | unsupported | changed |
| sparseResidency16Samples | unsupported | untested | changed |
| sparseResidencyAliased | untested | unsupported | changed |
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.
| Scene | stable-turnip-sync-V2 ms | 8Elite2-842.6 ms | Difference | Starts | GPU clock, MHz | Picture vs stock |
|---|---|---|---|---|---|---|
| compute_nbody | 18.10 | 10.07 | 80% faster | 3 of 3 vs 3 of 3 | ? vs ? | no large difference no small difference |
| hdr_2160p | 26.14 | 10.92 | 139% faster | 3 of 3 vs 3 of 3 | ? vs ? | no large difference no small difference |
| oit_linked_lists | 23.38 | 17.14 | 36% faster | 3 of 3 vs 3 of 3 | ? vs ? | no large difference no small difference |
| pipeline_barriers | 10.96 | 3.07 | too light, see pipeline_barriers_2160p | 3 of 3 vs 3 of 3 | ? vs ? | no large difference differs in 1337 small areas |
| pipeline_barriers_2160p | 10.01 | 7.75 | 29% faster | 3 of 3 vs 3 of 3 | ? vs ? | no large difference differs in 1038 small areas |
| subpasses | 5.17 | 3.90 | too light, see subpasses_2160p | 3 of 3 vs 3 of 3 | ? vs ? | not checked |
| subpasses_2160p | 20.46 | 14.45 | 42% faster | 3 of 3 vs 3 of 3 | ? vs ? | not checked |
Technical differences (Vulkan)
Extensions only in 8Elite2-842.6 (27)
- VK_ANDROID_external_format_resolve
- VK_EXT_astc_decode_mode
- VK_EXT_blend_operation_advanced
- VK_EXT_depth_range_unrestricted
- VK_EXT_device_fault
- VK_EXT_external_memory_acquire_unmodified
- VK_EXT_fragment_density_map2
- VK_EXT_image_compression_control
- VK_EXT_image_compression_control_swapchain
- VK_EXT_pipeline_protected_access
- VK_EXT_pipeline_robustness
- VK_EXT_shader_subgroup_ballot
- VK_EXT_shader_subgroup_vote
- VK_EXT_subpass_merge_feedback
- VK_KHR_pipeline_binary
- VK_KHR_ray_tracing_position_fetch
- VK_KHR_shader_maximal_reconvergence
- VK_KHR_shader_quad_control
- VK_KHR_shared_presentable_image
- VK_NV_optical_flow
- VK_QCOM_filter_cubic_weights
- VK_QCOM_image_processing2
- VK_QCOM_render_pass_store_ops
- VK_QCOM_render_pass_transform
- VK_QCOM_rotated_copy_commands
- VK_QCOM_tile_properties
- VK_QCOM_ycbcr_degamma
Extensions only in stable-turnip-sync-V2 (27)
- 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_custom_resolve
- VK_EXT_debug_marker
- VK_EXT_depth_clamp_control
- 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_image_drm_format_modifier
- VK_EXT_legacy_dithering
- VK_EXT_legacy_vertex_attributes
- VK_EXT_memory_budget
- VK_EXT_nested_command_buffer
- VK_EXT_non_seamless_cube_map
- VK_EXT_primitive_restart_index
- VK_EXT_shader_replicated_composites
- VK_EXT_shader_uniform_buffer_unsized_array
- VK_EXT_zero_initialize_device_memory
- VK_GOOGLE_decorate_string
- VK_GOOGLE_hlsl_functionality1
- VK_KHR_shader_relaxed_extended_instruction
- VK_NV_compute_shader_derivatives
- VK_VALVE_fragment_density_map_layered
- VK_VALVE_mutable_descriptor_type
Core features that differ (5)
| multiviewGeometryShader | pass to unsupported |
| protectedMemory | unsupported to pass |
| shaderInputAttachmentArrayDynamicIndexing | unsupported to pass |
| shaderInputAttachmentArrayNonUniformIndexing | unsupported to pass |
| pipelineProtectedAccess | unsupported to pass |
Limits that differ (21)
| Limit | stable-turnip-sync-V2 | 8Elite2-842.6 |
|---|---|---|
| maxTexelBufferElements | 536870911 | 134217728 |
| bufferImageGranularity | 64 | 1 |
| sparseAddressSpaceSize | 274877906944 | 4294967295 |
| maxPerStageResources | 16777216 | 50331648 |
| maxDescriptorSetUniformBuffersDynamic | 16 | 32 |
| maxDescriptorSetStorageBuffersDynamic | 8 | 16 |
| maxVertexInputAttributeOffset | 4095 | 4096 |
| maxGeometryInputComponents | 64 | 128 |
| maxGeometryOutputVertices | 256 | 1024 |
| maxGeometryTotalOutputComponents | 1024 | 131072 |
| maxFragmentInputComponents | 128 | 112 |
| maxComputeSharedMemorySize | 65536 | 32768 |
| minMemoryMapAlignment | 4096 | 64 |
| minUniformBufferOffsetAlignment | 64 | 256 |
| minStorageBufferOffsetAlignment | 4 | 64 |
| maxFramebufferLayers | 1024 | 2048 |
| discreteQueuePriorities | 2 | 4 |
| pointSizeRange | 1, 4092 | 1, 4095 |
| optimalBufferCopyOffsetAlignment | 128 | 64 |
| optimalBufferCopyRowPitchAlignment | 128 | 64 |
| nonCoherentAtomSize | 64 | 1 |
Formats supported by only one driver (3 gained, 5 lost)
- G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16
- G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16
- G16_B16R16_2PLANE_420_UNORM
- B8G8R8A8_SSCALED
- B8G8R8A8_USCALED
- B8G8R8_SINT
- B8G8R8_SNORM
- B8G8R8_UINT