FauxVkDB
English
Home / Mali-G57 / Compare

panvk_faux123_v1.0.1 vs EXP_panvk_g57_rg556_lab16

On Mali-G57 (Anbernic RG556, Redmi A5, TECNO KL4). The first driver named is compared against the second, which is the baseline.

StockEXP_panvk_g57_rg556_lab16Mesa 26.3.0-devel (git-60220e12cd)
vs
Custompanvk_faux123_v1.0.1PanVK-kbase RC3-lowmem (Mesa 26.3.0-devel (git-f9ed07a075))
Graphics system version (Vulkan)1.3.354 to 1.3.363
Optional features (extensions)166 to 178+12 -0
Features that pass their check (validated)49 to 6516 pass only on panvk_faux123_v1.0.1 0 pass only on EXP_panvk_g57_rg556_lab16
Speed test scenes (benchmark)no shared scenes
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

EXP_panvk_g57_rg556_lab16 details panvk_faux123_v1.0.1 details

Feature checks that differ (18)

FeatureEXP_panvk_g57_rg556_lab16panvk_faux123_v1.0.1Change
robustImageAccess2
VK_EXT_robustness2
unsupportedpassgained
depthClamp
VkPhysicalDeviceFeatures
untestedpassgained
largePoints
VkPhysicalDeviceFeatures
untestedpassgained
compat.helper_loop_terminator
core
untestedpassgained
compat.nested_map
core
untestedpassgained
core.blend_advertised_formats
core
untestedpassgained
core.compute_big_stack
core
untestedpassgained
core.host_coherent_readback
core
untestedpassgained
core.mrt_partial_kill
core
untestedpassgained
core.ubo_update_buffer
core
untestedpassgained
core.ubo_written_after_record
core
untestedpassgained
core.varying_subset
core
untestedpassgained
mem.image_padding
core
untestedpassgained
stage.geometry_max_output
core
untestedpassgained
stage.geometry_point_size
core
untestedpassgained
sync.first_submit_clear_then_compute
core
untestedpassgained
extendedDynamicState2PatchControlPoints
VK_EXT_extended_dynamic_state2
unsupporteduntestedchanged
depthBounds
VkPhysicalDeviceFeatures
unsupporteduntestedchanged

Speed test scenes (benchmark)

SceneEXP_panvk_g57_rg556_lab16 mspanvk_faux123_v1.0.1 msDifferenceStartsGPU clock, MHzPicture vs stock
compute_nbody30.51 typical 30.19n/aonly one side3 of 3 vs ?? vs ? clock unknownnot checked
hdr_2160p527.62 typical 524.25n/aonly one side3 of 3 vs ?? vs ? clock unknownnot checked
oit_linked_lists153.08 typical 151.98n/aonly one side3 of 3 vs ?? vs ? clock unknownnot checked
pipeline_barriers34.14 typical 6.84n/aonly one side3 of 3 vs ?? vs ? clock unknownnot checked
subpasses55.10 typical 95.42n/aonly one side3 of 3 vs ?? vs ? clock unknownnot 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 panvk_faux123_v1.0.1 against EXP_panvk_g57_rg556_lab16. "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 panvk_faux123_v1.0.1 (12)
  • VK_EXT_conditional_rendering
  • VK_EXT_image_compression_control
  • VK_EXT_present_timing
  • VK_EXT_primitives_generated_query
  • VK_EXT_swapchain_maintenance1
  • VK_KHR_present_id
  • VK_KHR_present_id2
  • VK_KHR_present_wait
  • VK_KHR_present_wait2
  • VK_KHR_shader_constant_data
  • VK_KHR_swapchain_maintenance1
  • VK_KHR_swapchain_mutable_format
Extensions only in EXP_panvk_g57_rg556_lab16 (0)
  • None
Core features that differ (1)
shaderOutputViewportIndex Vulkan 1.2unsupported to pass
Limits that differ (5)
LimitEXP_panvk_g57_rg556_lab16panvk_faux123_v1.0.1
sparseAddressSpaceSize0140737454800896
maxGeometryShaderInvocations3264
minStorageBufferOffsetAlignment164
timestampComputeAndGraphics01
timestampPeriod038.4615
Formats supported by only one driver (0 gained, 0 lost)

Only in panvk_faux123_v1.0.1:

  • None

Only in EXP_panvk_g57_rg556_lab16:

  • None