Home / Adreno (TM) 840 / Compare

891.11 vs Turnip_Gen8_V38

On Adreno (TM) 840 (HONOR Magic8 pro, iQOO 15, LENOVO TB324ZC, nubia NX809J, OnePlus 15, POCO F8 Ultra, POCO F9 Ultra, realme RMX5210, Samsung Galaxy S26, Samsung Galaxy S26, Samsung Galaxy S26 Ultra, Samsung Galaxy S26 Ultra, Samsung Galaxy S26 Ultra, Samsung Galaxy Z Fold8, Samsung Galaxy Z Fold8 Ultra, Xiaomi 17, Xiaomi 2512BPNDAG). The first driver named is compared against the second, which is the baseline.

StockTurnip_Gen8_V38Mesa 27.0.0-devel (git-f28352d117)

Custom891.11Driver Build: 17fab2f4fc, Iafcdf367ac, 1791270827

Graphics system version (Vulkan)1.4.363 to 1.4.351

Optional features (extensions)185 to 213+48 -20

Features that pass their check (validated)65 to 612 pass only on 891.11 6 pass only on Turnip_Gen8_V38

Speed test scenes (benchmark)2 faster 0 slower24% 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_2160p34% faster
oit_linked_lists14% 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: 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.

Turnip_Gen8_V38 details 891.11 details

Feature checks that differ (25)

Feature	Turnip_Gen8_V38	891.11	Change
extendedDynamicState3ColorBlendEnable VK_EXT_extended_dynamic_state3	pass	unsupported	lost
extendedDynamicState3ColorBlendEquation VK_EXT_extended_dynamic_state3	pass	unsupported	lost
extendedDynamicState3ColorWriteMask VK_EXT_extended_dynamic_state3	pass	unsupported	lost
extendedDynamicState3DepthClampEnable VK_EXT_extended_dynamic_state3	pass	unsupported	lost
extendedDynamicState3LogicOpEnable VK_EXT_extended_dynamic_state3	pass	unsupported	lost
extendedDynamicState3PolygonMode VK_EXT_extended_dynamic_state3	pass	unsupported	lost
compat.nested_map core	crash	pass	gained
core.compute_udiv core	fail	pass	gained
extendedDynamicState3AlphaToCoverageEnable VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3AlphaToOneEnable VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3ConservativeRasterizationMode VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3DepthClipEnable VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3DepthClipNegativeOneToOne VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3ExtraPrimitiveOverestimationSize VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3LineRasterizationMode VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3ProvokingVertexMode VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3RasterizationSamples VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3RasterizationStream VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3SampleLocationsEnable VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3SampleMask VK_EXT_extended_dynamic_state3	untested	unsupported	changed
extendedDynamicState3TessellationDomainOrigin VK_EXT_extended_dynamic_state3	untested	unsupported	changed
shaderSharedInt64Atomics VK_KHR_shader_atomic_int64	unsupported	untested	changed
shaderResourceMinLod VkPhysicalDeviceFeatures	untested	unsupported	changed
sparseResidency16Samples VkPhysicalDeviceFeatures	unsupported	untested	changed
sparseResidencyImage3D VkPhysicalDeviceFeatures	unsupported	untested	changed

Speed test scenes (benchmark)

Scene	Turnip_Gen8_V38 ms	891.11 ms	Difference	Starts	GPU clock, MHz	Picture vs stock
compute_nbody	n/a	5.67	only one side	3 of 3 vs 3 of 3	? vs ? clock unknown	no large difference 0.0 no small difference worst 0.0 64x36
compute_nbody_2x	10.64	n/a	only one side	3 of 3 vs 3 of 3	? vs ? clock unknown	no large difference 0.0 no small difference worst 0.0 64x36
hdr_2160p	8.91 typical 9.15	6.63 typical 6.62	34% faster	3 of 3 vs 3 of 3	? vs ? clock unknown	no large difference 0.0 no small difference worst 0.0 64x36
oit_linked_lists	12.89 typical 12.61	11.28 typical 11.34	14% faster	3 of 3 vs 3 of 3	? vs ? clock unknown	no large difference 0.0 no small difference worst 0.0 64x36
pipeline_barriers	1.46	n/a	only one side	3 of 3 vs 3 of 3	? vs ? clock unknown	no large difference 7.1 differs in 1337 small areas worst 48.0 64x36
pipeline_barriers_2160p	6.27	n/a	only one side	3 of 3 vs 3 of 3	? vs ? clock unknown	no large difference 6.4 differs in 1038 small areas worst 51.7 64x36
subpasses	n/a	1.85	only one side	3 of 3 vs 3 of 3	? vs ? clock unknown	not checked
subpasses_2160p	n/a	n/a	only one side	3 of 3 vs 3 of 3	? vs ? clock unknown	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 891.11 against Turnip_Gen8_V38. "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 891.11 (48)

VK_ANDROID_external_format_resolve
VK_ARM_data_graph
VK_ARM_tensors
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_external_memory_host
VK_EXT_fragment_density_map2
VK_EXT_fragment_shader_interlock
VK_EXT_frame_boundary
VK_EXT_image_compression_control_swapchain
VK_EXT_memory_decompression
VK_EXT_mesh_shader
VK_EXT_pipeline_library_group_handles
VK_EXT_pipeline_protected_access
VK_EXT_pipeline_robustness
VK_EXT_shader_subgroup_ballot
VK_EXT_subpass_merge_feedback
VK_EXT_vertex_attribute_robustness
VK_KHR_cooperative_matrix
VK_KHR_depth_clamp_zero_one
VK_KHR_internally_synchronized_queues
VK_KHR_maintenance10
VK_KHR_maintenance8
VK_KHR_maintenance9
VK_KHR_pipeline_binary
VK_KHR_ray_tracing_pipeline
VK_KHR_ray_tracing_position_fetch
VK_KHR_robustness2
VK_KHR_shader_maximal_reconvergence
VK_KHR_shared_presentable_image
VK_KHR_unified_image_layouts
VK_NV_optical_flow
VK_QCOM_cooperative_matrix_conversion
VK_QCOM_data_graph_model
VK_QCOM_elapsed_timer_query
VK_QCOM_filter_cubic_weights
VK_QCOM_image_processing2
VK_QCOM_image_processing3
VK_QCOM_queue_perf_hint
VK_QCOM_render_pass_store_ops
VK_QCOM_render_pass_transform
VK_QCOM_tile_memory_heap
VK_QCOM_tile_properties
VK_QCOM_tile_shading
VK_QCOM_ycbcr_degamma

Extensions only in Turnip_Gen8_V38 (20)

VK_AMD_buffer_marker
VK_ARM_rasterization_order_attachment_access
VK_EXT_attachment_feedback_loop_dynamic_state
VK_EXT_custom_resolve
VK_EXT_debug_marker
VK_EXT_dynamic_rendering_unused_attachments
VK_EXT_extended_dynamic_state3
VK_EXT_image_drm_format_modifier
VK_EXT_legacy_vertex_attributes
VK_EXT_memory_budget
VK_EXT_primitive_restart_index
VK_EXT_shader_replicated_composites
VK_EXT_shader_uniform_buffer_unsized_array
VK_GOOGLE_decorate_string
VK_GOOGLE_hlsl_functionality1
VK_KHR_shader_relaxed_extended_instruction
VK_NV_compute_shader_derivatives
VK_VALVE_buffer_device_address_allocation_alignment
VK_VALVE_fragment_density_map_layered
VK_VALVE_mutable_descriptor_type

Core features that differ (6)

multiviewGeometryShader Vulkan 1.1	pass to unsupported
protectedMemory Vulkan 1.1	unsupported to pass
shaderInputAttachmentArrayDynamicIndexing Vulkan 1.2	unsupported to pass
shaderInputAttachmentArrayNonUniformIndexing Vulkan 1.2	unsupported to pass
shaderSharedInt64Atomics Vulkan 1.2	unsupported to pass
pipelineProtectedAccess Vulkan 1.4	unsupported to pass

Limits that differ (22)

Limit	Turnip_Gen8_V38	891.11
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
maxComputeWorkGroupInvocations	2048	1024
minMemoryMapAlignment	4096	64
minUniformBufferOffsetAlignment	64	256
minStorageBufferOffsetAlignment	4	16
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)

Only in 891.11:

G10X6_B10X6R10X6_2PLANE_420_UNORM_3PACK16
G10X6_B10X6R10X6_2PLANE_422_UNORM_3PACK16
G16_B16R16_2PLANE_420_UNORM

Only in Turnip_Gen8_V38:

B8G8R8A8_SSCALED
B8G8R8A8_USCALED
B8G8R8_SINT
B8G8R8_SNORM
B8G8R8_UINT