Diefolio
English

AMD · GPU

AMD Radeon PRO W7900

AMD Radeon PRO W7900 — Specifications and benchmarks. .

These specifications are supported by the linked sources. Core counts, clocks and memory capacities do not predict workload performance. Check benchmark versions, environments and sample counts.

Evidence review in progress · Search indexing deferred

Compare this product →

Power

ParameterValue and conditions
Board power
295 W
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Total Board Power (TBP)",
  "raw_value": "295W",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Total Board Power (TBP)",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
295 W
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Total Board Power (TBP)",
  "raw_value": "295W",
  "qualifiers": {
    "source_field": "Total Board Power (TBP)",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
Recommended system PSU
750 W
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Minimum PSU Recommendation",
  "raw_value": "750W",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Minimum PSU Recommendation",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
750 W
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Minimum PSU Recommendation",
  "raw_value": "750W",
  "qualifiers": {
    "source_field": "Minimum PSU Recommendation",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}

Compute

ParameterValue and conditions
Published process node
TSMC 5nm GCD 6nm MCD
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Lithography",
  "raw_value": "TSMC 5nm GCD 6nm MCD",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Lithography",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
TSMC 5nm GCD 6nm MCD
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Lithography",
  "raw_value": "TSMC 5nm GCD 6nm MCD",
  "qualifiers": {
    "source_field": "Lithography",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
AMD AI accelerators
192
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "AI Accelerators",
  "raw_value": "192",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "AI Accelerators",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
192
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "AI Accelerators",
  "raw_value": "192",
  "qualifiers": {
    "source_field": "AI Accelerators",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
ArchitectureSources disagree
RDNA3
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Architecture",
  "raw_value": "RDNA3",
  "qualifiers": {
    "scope": "ROCm hardware reference; listing does not alone prove runtime support"
  },
  "document_id": "amd-rocm-gpu-b7f874b2259ab04a"
}
AMD RDNA 3
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "GPU Architecture",
  "raw_value": "AMD RDNA 3",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "GPU Architecture",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
AMD RDNA 3
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "GPU Architecture",
  "raw_value": "AMD RDNA 3",
  "qualifiers": {
    "source_field": "GPU Architecture",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
Compute units
96
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Compute Units",
  "raw_value": "96",
  "qualifiers": {
    "scope": "ROCm hardware reference; listing does not alone prove runtime support"
  },
  "document_id": "amd-rocm-gpu-b7f874b2259ab04a"
}
96
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Compute Units",
  "raw_value": "96",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Compute Units",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
96
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Compute Units",
  "raw_value": "96",
  "qualifiers": {
    "source_field": "Compute Units",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
LDS
128 KiB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "LDS (KiB)",
  "raw_value": "128",
  "qualifiers": {
    "scope": "ROCm hardware reference; listing does not alone prove runtime support"
  },
  "document_id": "amd-rocm-gpu-b7f874b2259ab04a"
}
AMD ray accelerators
96
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Ray Accelerators",
  "raw_value": "96",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Ray Accelerators",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
96
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Ray Accelerators",
  "raw_value": "96",
  "qualifiers": {
    "source_field": "Ray Accelerators",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
Stream processors
6,144
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Stream Processors",
  "raw_value": "6144",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Stream Processors",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
6,144
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Stream Processors",
  "raw_value": "6144",
  "qualifiers": {
    "source_field": "Stream Processors",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}

Memory

ParameterValue and conditions
Infinity Cache
96 MiB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Infinity Cache (MiB)",
  "raw_value": "96",
  "qualifiers": {
    "scope": "ROCm hardware reference; listing does not alone prove runtime support"
  },
  "document_id": "amd-rocm-gpu-b7f874b2259ab04a"
}
Infinity Cache (documented MB)
96 MB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "AMD Infinity Cache Technology",
  "raw_value": "96 MB",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "AMD Infinity Cache Technology",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
96 MB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "AMD Infinity Cache Technology",
  "raw_value": "96 MB",
  "qualifiers": {
    "source_field": "AMD Infinity Cache Technology",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
GPU L2 cache
6 MiB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "L2 Cache (MiB)",
  "raw_value": "6",
  "qualifiers": {
    "scope": "ROCm hardware reference; listing does not alone prove runtime support"
  },
  "document_id": "amd-rocm-gpu-b7f874b2259ab04a"
}
Memory bandwidth
864 GB/s
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak Memory Bandwidth",
  "raw_value": "864",
  "qualifiers": {
    "baseline_generation": "95d4e3b61a136b693f43",
    "evidence_scope": "exact listed manufacturer SKU; not a measured benchmark or a partner-board setting",
    "evidence_type": "manufacturer-specification",
    "field_evidence": {
      "Peak Memory Bandwidth": {
        "corroborating_sources": [],
        "location": {
          "reviewed_product_index_zero_based": 0
        },
        "quantity_scope": "manufacturer peak/maximum",
        "source_label": "Peak Memory Bandwidth",
        "source_value": "864 GB/s",
        "transformation": "explicit peak GB/s value; no bus-width or clock calculation"
      }
    },
    "normalization_reviewed_at": "2026-10-11",
    "source_field": "Peak Memory Bandwidth",
    "source_row": 1,
    "source_sha256": "fee1c4108b833cf88010febc67589e17b90221932be44cd667e2f0d049ad4ac3"
  },
  "document_id": "gap-fill-2026-10-11-ff86cb4aa40a604e-fee1c4108b833cf8"
}
Memory bus
384 bit
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Memory Interface",
  "raw_value": "384-bit",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Memory Interface",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
384 bit
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Memory Interface",
  "raw_value": "384-bit",
  "qualifiers": {
    "source_field": "Memory Interface",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
Graphics memory type
GDDR6
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Dedicated Memory Type",
  "raw_value": "GDDR6",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Dedicated Memory Type",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
GDDR6
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Dedicated Memory Type",
  "raw_value": "GDDR6",
  "qualifiers": {
    "source_field": "Dedicated Memory Type",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
Dedicated VRAM
48 GB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Dedicated Memory Size",
  "raw_value": "48 GB",
  "qualifiers": {
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Dedicated Memory Size",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
48 GB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Dedicated Memory Size",
  "raw_value": "48 GB",
  "qualifiers": {
    "source_field": "Dedicated Memory Size",
    "source_row": 1
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
Dedicated VRAM (GiB)
48 GiB
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "VRAM (GiB)",
  "raw_value": "48",
  "qualifiers": {
    "scope": "ROCm hardware reference; listing does not alone prove runtime support"
  },
  "document_id": "amd-rocm-gpu-b7f874b2259ab04a"
}

Connectivity and features

ParameterValue and conditions
GPU FP16 Matrix dense
123 TFLOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak Half Precision (FP16 Matrix) Performance",
  "raw_value": "123 TFLOPs",
  "qualifiers": {
    "engine": "matrix",
    "measurement_type": "manufacturer_peak",
    "precision": "FP16",
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Peak Half Precision (FP16 Matrix) Performance",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6,
    "sparsity": "dense"
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
123 TFLOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak Half Precision (FP16 Matrix) Performance",
  "raw_value": "123 TFLOPs",
  "qualifiers": {
    "engine": "matrix",
    "measurement_type": "manufacturer_peak",
    "precision": "FP16",
    "source_field": "Peak Half Precision (FP16 Matrix) Performance",
    "source_row": 1,
    "sparsity": "dense"
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
GPU FP16 Vector dense
61.3 TFLOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak Half Precision (FP16 Vector) Performance",
  "raw_value": "61.3 TFLOPs",
  "qualifiers": {
    "engine": "vector",
    "measurement_type": "manufacturer_peak",
    "precision": "FP16",
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Peak Half Precision (FP16 Vector) Performance",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6,
    "sparsity": "dense"
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
61.3 TFLOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak Half Precision (FP16 Vector) Performance",
  "raw_value": "61.3 TFLOPs",
  "qualifiers": {
    "engine": "vector",
    "measurement_type": "manufacturer_peak",
    "precision": "FP16",
    "source_field": "Peak Half Precision (FP16 Vector) Performance",
    "source_row": 1,
    "sparsity": "dense"
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
GPU FP32 Vector dense
61.3 TFLOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak Single Precision (FP32 Vector) Performance",
  "raw_value": "61.3 TFLOPs",
  "qualifiers": {
    "engine": "vector",
    "measurement_type": "manufacturer_peak",
    "precision": "FP32",
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Peak Single Precision (FP32 Vector) Performance",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6,
    "sparsity": "dense"
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
61.3 TFLOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak Single Precision (FP32 Vector) Performance",
  "raw_value": "61.3 TFLOPs",
  "qualifiers": {
    "engine": "vector",
    "measurement_type": "manufacturer_peak",
    "precision": "FP32",
    "source_field": "Peak Single Precision (FP32 Vector) Performance",
    "source_row": 1,
    "sparsity": "dense"
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
GPU INT4 Matrix dense
245 TOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak 4-bit Precision (INT4 Matrix) Performance",
  "raw_value": "245 TOPs",
  "qualifiers": {
    "engine": "matrix",
    "measurement_type": "manufacturer_peak",
    "precision": "INT4",
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Peak 4-bit Precision (INT4 Matrix) Performance",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6,
    "sparsity": "dense"
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
245 TOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak 4-bit Precision (INT4 Matrix) Performance",
  "raw_value": "245 TOPs",
  "qualifiers": {
    "engine": "matrix",
    "measurement_type": "manufacturer_peak",
    "precision": "INT4",
    "source_field": "Peak 4-bit Precision (INT4 Matrix) Performance",
    "source_row": 1,
    "sparsity": "dense"
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
GPU INT8 Matrix dense
123 TOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak 8-bit Precision (INT8 Matrix) Performance",
  "raw_value": "123 TOPs",
  "qualifiers": {
    "engine": "matrix",
    "measurement_type": "manufacturer_peak",
    "precision": "INT8",
    "release_evidence": "manufacturer-listed-date-unknown",
    "source_field": "Peak 8-bit Precision (INT8 Matrix) Performance",
    "source_product_url": "https://www.amd.com/en/products/graphics/workstations/radeon-pro/w7900.html",
    "source_row": 6,
    "sparsity": "dense"
  },
  "document_id": "amd-pro-graphics-table-1787b597aed7af7f"
}
123 TOPS
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "Peak 8-bit Precision (INT8 Matrix) Performance",
  "raw_value": "123 TOPs",
  "qualifiers": {
    "engine": "matrix",
    "measurement_type": "manufacturer_peak",
    "precision": "INT8",
    "source_field": "Peak 8-bit Precision (INT8 Matrix) Performance",
    "source_row": 1,
    "sparsity": "dense"
  },
  "document_id": "detail-2dd5a8d342d80005-fee1c4108b833cf8"
}
LLVM target
gfx1100
Source and conditions

View source ↗

{
  "state": "known",
  "raw_label": "LLVM target name",
  "raw_value": "gfx1100",
  "qualifiers": {
    "scope": "ROCm hardware reference; listing does not alone prove runtime support"
  },
  "document_id": "amd-rocm-gpu-b7f874b2259ab04a"
}

Public benchmarks

Compare results with matching versions, scenes, settings and statistics. A run count of 0 identifies an aggregate or reported value, not a newly measured run.

Blender (36)

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · classroom · HIP · Windows · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
824.59 / 824.59
Measurement dates
2024-08-15 – 2024-08-15
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-08-15T20:20:32.354750+00:00",
        "score": 824.5905423485578,
        "source_record_id": "f77fb921-4d8f-4190-a18b-e3443d9726d3"
      }
    ],
    "median": null,
    "observed_max": 824.5905423485578,
    "observed_min": 824.5905423485578,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.0",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "4.2.0"
  },
  "condition_id": "713bd8df90aec1f0a252",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f77fb921-4d8f-4190-a18b-e3443d9726d3"
  ],
  "measured_from": "2024-08-15T20:20:32.354750+00:00",
  "measured_to": "2024-08-15T20:20:32.354750+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · junkshop · HIP · Windows · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
968.51 / 968.51
Measurement dates
2024-08-15 – 2024-08-15
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-08-15T20:19:56.462483+00:00",
        "score": 968.5053574110436,
        "source_record_id": "f77fb921-4d8f-4190-a18b-e3443d9726d3"
      }
    ],
    "median": null,
    "observed_max": 968.5053574110436,
    "observed_min": 968.5053574110436,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.0",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "4.2.0"
  },
  "condition_id": "cad3ab8a8edaa07ccdac",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f77fb921-4d8f-4190-a18b-e3443d9726d3"
  ],
  "measured_from": "2024-08-15T20:19:56.462483+00:00",
  "measured_to": "2024-08-15T20:19:56.462483+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · monster · HIP · Windows · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,848.9 / 1,848.9
Measurement dates
2024-08-15 – 2024-08-15
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-08-15T20:19:05.028460+00:00",
        "score": 1848.8995187669955,
        "source_record_id": "f77fb921-4d8f-4190-a18b-e3443d9726d3"
      }
    ],
    "median": null,
    "observed_max": 1848.8995187669955,
    "observed_min": 1848.8995187669955,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.0",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "4.2.0"
  },
  "condition_id": "8990caba1091f40be323",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f77fb921-4d8f-4190-a18b-e3443d9726d3"
  ],
  "measured_from": "2024-08-15T20:19:05.028460+00:00",
  "measured_to": "2024-08-15T20:19:05.028460+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.5.0 · classroom · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
895.94 / 900.68
Measurement dates
2025-11-11 – 2025-11-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-11-11T22:28:25.716103+00:00",
        "score": 900.6835450705354,
        "source_record_id": "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53"
      },
      {
        "measured_at": "2025-11-16T02:32:45.777395+00:00",
        "score": 895.937838796507,
        "source_record_id": "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
      }
    ],
    "median": null,
    "observed_max": 900.6835450705354,
    "observed_min": 895.937838796507,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "4.5.0"
  },
  "condition_id": "899d6488f25c83bee834",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53",
    "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
  ],
  "measured_from": "2025-11-11T22:28:25.716103+00:00",
  "measured_to": "2025-11-16T02:32:45.777395+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.5.0 · junkshop · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,018.93 / 1,021.66
Measurement dates
2025-11-11 – 2025-11-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-11-11T22:27:45.790833+00:00",
        "score": 1021.6611413541544,
        "source_record_id": "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53"
      },
      {
        "measured_at": "2025-11-16T02:32:05.422455+00:00",
        "score": 1018.9271511445683,
        "source_record_id": "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
      }
    ],
    "median": null,
    "observed_max": 1021.6611413541544,
    "observed_min": 1018.9271511445683,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "4.5.0"
  },
  "condition_id": "7d23b3fe2c22b0412fb0",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53",
    "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
  ],
  "measured_from": "2025-11-11T22:27:45.790833+00:00",
  "measured_to": "2025-11-16T02:32:05.422455+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.5.0 · monster · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,867.87 / 1,874.29
Measurement dates
2025-11-11 – 2025-11-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-11-11T22:26:56.763042+00:00",
        "score": 1874.2914835334047,
        "source_record_id": "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53"
      },
      {
        "measured_at": "2025-11-16T02:31:15.582660+00:00",
        "score": 1867.8683306508415,
        "source_record_id": "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
      }
    ],
    "median": null,
    "observed_max": 1874.2914835334047,
    "observed_min": 1867.8683306508415,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "4.5.0"
  },
  "condition_id": "8ef189f112be053055d1",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53",
    "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
  ],
  "measured_from": "2025-11-11T22:26:56.763042+00:00",
  "measured_to": "2025-11-16T02:31:15.582660+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.4.0 · classroom · HIP · Windows · 1P · 64 cores · 128 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
887.45 / 887.45
Measurement dates
2025-05-10 – 2025-05-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-05-10T11:26:27.803419+00:00",
        "score": 887.4482235204747,
        "source_record_id": "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
      }
    ],
    "median": null,
    "observed_max": 887.4482235204747,
    "observed_min": 887.4482235204747,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 64,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 128,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "4.4.0"
  },
  "condition_id": "f08af7e696cae534a435",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
  ],
  "measured_from": "2025-05-10T11:26:27.803419+00:00",
  "measured_to": "2025-05-10T11:26:27.803419+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.4.0 · junkshop · HIP · Windows · 1P · 64 cores · 128 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,021.75 / 1,021.75
Measurement dates
2025-05-10 – 2025-05-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-05-10T11:25:48.173988+00:00",
        "score": 1021.7481653014589,
        "source_record_id": "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
      }
    ],
    "median": null,
    "observed_max": 1021.7481653014589,
    "observed_min": 1021.7481653014589,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 64,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 128,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "4.4.0"
  },
  "condition_id": "6a88da860c3f3198814d",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
  ],
  "measured_from": "2025-05-10T11:25:48.173988+00:00",
  "measured_to": "2025-05-10T11:25:48.173988+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.4.0 · monster · HIP · Windows · 1P · 64 cores · 128 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,909 / 1,909
Measurement dates
2025-05-10 – 2025-05-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-05-10T11:24:54.398392+00:00",
        "score": 1909.000895959085,
        "source_record_id": "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
      }
    ],
    "median": null,
    "observed_max": 1909.000895959085,
    "observed_min": 1909.000895959085,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 64,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 128,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "4.4.0"
  },
  "condition_id": "9c611dba131d566e86c6",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
  ],
  "measured_from": "2025-05-10T11:24:54.398392+00:00",
  "measured_to": "2025-05-10T11:24:54.398392+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.2.0 · classroom · HIP · Windows · 1P · 96 cores · 192 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
805.59 / 823.23
Measurement dates
2024-11-18 – 2024-11-18
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-18T13:55:31.065354+00:00",
        "score": 805.5944499965938,
        "source_record_id": "d3753458-4117-4916-be77-c535429c356b"
      },
      {
        "measured_at": "2024-11-18T14:23:16.591348+00:00",
        "score": 823.2281798742332,
        "source_record_id": "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
      }
    ],
    "median": null,
    "observed_max": 823.2281798742332,
    "observed_min": 805.5944499965938,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 96,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 192,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "4.2.0"
  },
  "condition_id": "b69a38993f45f6b8d226",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d3753458-4117-4916-be77-c535429c356b",
    "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
  ],
  "measured_from": "2024-11-18T13:55:31.065354+00:00",
  "measured_to": "2024-11-18T14:23:16.591348+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.2.0 · junkshop · HIP · Windows · 1P · 96 cores · 192 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
929.8 / 955.64
Measurement dates
2024-11-18 – 2024-11-18
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-18T13:54:52.087068+00:00",
        "score": 929.8022837974069,
        "source_record_id": "d3753458-4117-4916-be77-c535429c356b"
      },
      {
        "measured_at": "2024-11-18T14:22:37.547588+00:00",
        "score": 955.6417371778447,
        "source_record_id": "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
      }
    ],
    "median": null,
    "observed_max": 955.6417371778447,
    "observed_min": 929.8022837974069,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 96,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 192,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "4.2.0"
  },
  "condition_id": "f5806e9dad28bb0640ef",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d3753458-4117-4916-be77-c535429c356b",
    "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
  ],
  "measured_from": "2024-11-18T13:54:52.087068+00:00",
  "measured_to": "2024-11-18T14:22:37.547588+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.2.0 · monster · HIP · Windows · 1P · 96 cores · 192 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,797.84 / 1,826.75
Measurement dates
2024-11-18 – 2024-11-18
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-18T13:53:56.148435+00:00",
        "score": 1797.8421902945788,
        "source_record_id": "d3753458-4117-4916-be77-c535429c356b"
      },
      {
        "measured_at": "2024-11-18T14:21:42.098252+00:00",
        "score": 1826.7494819717795,
        "source_record_id": "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
      }
    ],
    "median": null,
    "observed_max": 1826.7494819717795,
    "observed_min": 1797.8421902945788,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 96,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 192,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "4.2.0"
  },
  "condition_id": "9d67efe4f4ed4117b653",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d3753458-4117-4916-be77-c535429c356b",
    "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
  ],
  "measured_from": "2024-11-18T13:53:56.148435+00:00",
  "measured_to": "2024-11-18T14:21:42.098252+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · classroom · HIP · Linux · Ubuntu · 22.04 · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
823.47 / 823.47
Measurement dates
2024-11-10 – 2024-11-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-10T16:15:49.579433+00:00",
        "score": 823.4672966908023,
        "source_record_id": "edc905f7-0399-4b61-ae69-32850d89881d"
      }
    ],
    "median": null,
    "observed_max": 823.4672966908023,
    "observed_min": 823.4672966908023,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "Ubuntu",
      "dist_version": "22.04",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "4.2.0"
  },
  "condition_id": "ab98d3666d0440c32137",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "edc905f7-0399-4b61-ae69-32850d89881d"
  ],
  "measured_from": "2024-11-10T16:15:49.579433+00:00",
  "measured_to": "2024-11-10T16:15:49.579433+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · junkshop · HIP · Linux · Ubuntu · 22.04 · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
974.71 / 974.71
Measurement dates
2024-11-10 – 2024-11-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-10T16:15:17.015977+00:00",
        "score": 974.7142405750548,
        "source_record_id": "edc905f7-0399-4b61-ae69-32850d89881d"
      }
    ],
    "median": null,
    "observed_max": 974.7142405750548,
    "observed_min": 974.7142405750548,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "Ubuntu",
      "dist_version": "22.04",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "4.2.0"
  },
  "condition_id": "880c80d2ac261392b568",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "edc905f7-0399-4b61-ae69-32850d89881d"
  ],
  "measured_from": "2024-11-10T16:15:17.015977+00:00",
  "measured_to": "2024-11-10T16:15:17.015977+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · monster · HIP · Linux · Ubuntu · 22.04 · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,879.27 / 1,879.27
Measurement dates
2024-11-10 – 2024-11-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-10T16:14:34.471263+00:00",
        "score": 1879.2736286573645,
        "source_record_id": "edc905f7-0399-4b61-ae69-32850d89881d"
      }
    ],
    "median": null,
    "observed_max": 1879.2736286573645,
    "observed_min": 1879.2736286573645,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "Ubuntu",
      "dist_version": "22.04",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "4.2.0"
  },
  "condition_id": "bb64229fa4bc8c7ed678",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "edc905f7-0399-4b61-ae69-32850d89881d"
  ],
  "measured_from": "2024-11-10T16:14:34.471263+00:00",
  "measured_to": "2024-11-10T16:14:34.471263+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · classroom · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
899.46 / 899.46
Measurement dates
2026-06-25 – 2026-06-25
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-25T11:09:54.113452+00:00",
        "score": 899.4619534314825,
        "source_record_id": "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
      }
    ],
    "median": null,
    "observed_max": 899.4619534314825,
    "observed_min": 899.4619534314825,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "5.0.0"
  },
  "condition_id": "167f21b73d3b8d3a7d9f",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
  ],
  "measured_from": "2026-06-25T11:09:54.113452+00:00",
  "measured_to": "2026-06-25T11:09:54.113452+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.1.1 · classroom · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,189.8 / 1,189.8
Measurement dates
2026-06-26 – 2026-06-26
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-26T10:51:30.439222+00:00",
        "score": 1189.804918594043,
        "source_record_id": "0d8b7a84-25be-4d63-8181-98e771a206c8"
      }
    ],
    "median": null,
    "observed_max": 1189.804918594043,
    "observed_min": 1189.804918594043,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "5.1.1"
  },
  "condition_id": "bd66526c9cc55c8692fc",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "0d8b7a84-25be-4d63-8181-98e771a206c8"
  ],
  "measured_from": "2026-06-26T10:51:30.439222+00:00",
  "measured_to": "2026-06-26T10:51:30.439222+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · junkshop · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,060.42 / 1,060.42
Measurement dates
2026-06-25 – 2026-06-25
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-25T11:09:14.153693+00:00",
        "score": 1060.4154071050025,
        "source_record_id": "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
      }
    ],
    "median": null,
    "observed_max": 1060.4154071050025,
    "observed_min": 1060.4154071050025,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "5.0.0"
  },
  "condition_id": "8872c4d64653c276bc21",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
  ],
  "measured_from": "2026-06-25T11:09:14.153693+00:00",
  "measured_to": "2026-06-25T11:09:14.153693+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.1.1 · junkshop · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,447.89 / 1,447.89
Measurement dates
2026-06-26 – 2026-06-26
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-26T10:50:52.402516+00:00",
        "score": 1447.8919411947506,
        "source_record_id": "0d8b7a84-25be-4d63-8181-98e771a206c8"
      }
    ],
    "median": null,
    "observed_max": 1447.8919411947506,
    "observed_min": 1447.8919411947506,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "5.1.1"
  },
  "condition_id": "bbed4b27112df337d76e",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "0d8b7a84-25be-4d63-8181-98e771a206c8"
  ],
  "measured_from": "2026-06-26T10:50:52.402516+00:00",
  "measured_to": "2026-06-26T10:50:52.402516+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · monster · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,565.25 / 1,565.25
Measurement dates
2026-06-25 – 2026-06-25
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-25T11:08:27.354029+00:00",
        "score": 1565.2539933667128,
        "source_record_id": "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
      }
    ],
    "median": null,
    "observed_max": 1565.2539933667128,
    "observed_min": 1565.2539933667128,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "5.0.0"
  },
  "condition_id": "a644956df2459c70f11b",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
  ],
  "measured_from": "2026-06-25T11:08:27.354029+00:00",
  "measured_to": "2026-06-25T11:08:27.354029+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.1.1 · monster · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,848.7 / 1,848.7
Measurement dates
2026-06-26 – 2026-06-26
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-26T10:50:09.208863+00:00",
        "score": 1848.6993476072291,
        "source_record_id": "0d8b7a84-25be-4d63-8181-98e771a206c8"
      }
    ],
    "median": null,
    "observed_max": 1848.6993476072291,
    "observed_min": 1848.6993476072291,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "5.1.1"
  },
  "condition_id": "d2f0121ffa77ca5dfd9b",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "0d8b7a84-25be-4d63-8181-98e771a206c8"
  ],
  "measured_from": "2026-06-26T10:50:09.208863+00:00",
  "measured_to": "2026-06-26T10:50:09.208863+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · classroom · HIP · Windows · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
935.61 / 935.61
Measurement dates
2026-03-16 – 2026-03-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-16T16:58:00.457590+00:00",
        "score": 935.6086354549288,
        "source_record_id": "97d37e7c-73f4-415d-ba06-a416262044c7"
      }
    ],
    "median": null,
    "observed_max": 935.6086354549288,
    "observed_min": 935.6086354549288,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "5.0.0"
  },
  "condition_id": "e2634ba3ccc206b85a96",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "97d37e7c-73f4-415d-ba06-a416262044c7"
  ],
  "measured_from": "2026-03-16T16:58:00.457590+00:00",
  "measured_to": "2026-03-16T16:58:00.457590+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · junkshop · HIP · Windows · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,140.33 / 1,140.33
Measurement dates
2026-03-16 – 2026-03-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-16T16:57:20.976077+00:00",
        "score": 1140.3281121215923,
        "source_record_id": "97d37e7c-73f4-415d-ba06-a416262044c7"
      }
    ],
    "median": null,
    "observed_max": 1140.3281121215923,
    "observed_min": 1140.3281121215923,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "5.0.0"
  },
  "condition_id": "a37a144fd19b71a9af6a",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "97d37e7c-73f4-415d-ba06-a416262044c7"
  ],
  "measured_from": "2026-03-16T16:57:20.976077+00:00",
  "measured_to": "2026-03-16T16:57:20.976077+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · monster · HIP · Windows · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,550.87 / 1,550.87
Measurement dates
2026-03-16 – 2026-03-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-16T16:56:30.923565+00:00",
        "score": 1550.8660615583383,
        "source_record_id": "97d37e7c-73f4-415d-ba06-a416262044c7"
      }
    ],
    "median": null,
    "observed_max": 1550.8660615583383,
    "observed_min": 1550.8660615583383,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "5.0.0"
  },
  "condition_id": "14eba940491ac4ce084b",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "97d37e7c-73f4-415d-ba06-a416262044c7"
  ],
  "measured_from": "2026-03-16T16:56:30.923565+00:00",
  "measured_to": "2026-03-16T16:56:30.923565+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · classroom · HIP · Linux · Arch Linux · rolling · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
802.98 / 802.98
Measurement dates
2026-03-17 – 2026-03-17
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-17T12:00:19.894693+00:00",
        "score": 802.9758603394421,
        "source_record_id": "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
      }
    ],
    "median": null,
    "observed_max": 802.9758603394421,
    "observed_min": 802.9758603394421,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon Pro W7900",
      "dist_name": "Arch Linux",
      "dist_version": "rolling",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "classroom",
    "scene_checksum": "936f0a528f03eba291c67ea87743a0ebde58938fb87471b262721b89b46b875d",
    "version": "5.0.0"
  },
  "condition_id": "2657270af061eec8f59e",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
  ],
  "measured_from": "2026-03-17T12:00:19.894693+00:00",
  "measured_to": "2026-03-17T12:00:19.894693+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · junkshop · HIP · Linux · Arch Linux · rolling · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
978.12 / 978.12
Measurement dates
2026-03-17 – 2026-03-17
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-17T11:59:47.939558+00:00",
        "score": 978.1164090086237,
        "source_record_id": "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
      }
    ],
    "median": null,
    "observed_max": 978.1164090086237,
    "observed_min": 978.1164090086237,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon Pro W7900",
      "dist_name": "Arch Linux",
      "dist_version": "rolling",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "junkshop",
    "scene_checksum": "f5515a21a1337b908212d8d76733530df83014464a24c585ab7b96000da0cdce",
    "version": "5.0.0"
  },
  "condition_id": "d32a141c1cabf53dae5b",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
  ],
  "measured_from": "2026-03-17T11:59:47.939558+00:00",
  "measured_to": "2026-03-17T11:59:47.939558+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender scene rendering

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · monster · HIP · Linux · Arch Linux · rolling · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
1,447.78 / 1,447.78
Measurement dates
2026-03-17 – 2026-03-17
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-17T11:59:05.212328+00:00",
        "score": 1447.7789520803992,
        "source_record_id": "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
      }
    ],
    "median": null,
    "observed_max": 1447.7789520803992,
    "observed_min": 1447.7789520803992,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon Pro W7900",
      "dist_name": "Arch Linux",
      "dist_version": "rolling",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "monster",
    "scene_checksum": "8ff812a02f720433de1403dab447be9541eecca71cc919224a7790d017c1b2ce",
    "version": "5.0.0"
  },
  "condition_id": "0a97ceb4e6de61d5efc5",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
  ],
  "measured_from": "2026-03-17T11:59:05.212328+00:00",
  "measured_to": "2026-03-17T11:59:05.212328+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · monster + junkshop + classroom · HIP · Windows · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,642 / 3,642
Measurement dates
2024-08-15 – 2024-08-15
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-08-15T20:20:32.354750+00:00",
        "score": 3641.995418526597,
        "source_record_id": "f77fb921-4d8f-4190-a18b-e3443d9726d3"
      }
    ],
    "median": null,
    "observed_max": 3641.995418526597,
    "observed_min": 3641.995418526597,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.0",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "4.2.0"
  },
  "condition_id": "a758e839baa42abc6875",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f77fb921-4d8f-4190-a18b-e3443d9726d3"
  ],
  "measured_from": "2024-08-15T20:20:32.354750+00:00",
  "measured_to": "2024-08-15T20:20:32.354750+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.5.0 · monster + junkshop + classroom · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,782.73 / 3,796.64
Measurement dates
2025-11-11 – 2025-11-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-11-11T22:28:25.716103+00:00",
        "score": 3796.636169958095,
        "source_record_id": "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53"
      },
      {
        "measured_at": "2025-11-16T02:32:45.777395+00:00",
        "score": 3782.7333205919167,
        "source_record_id": "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
      }
    ],
    "median": null,
    "observed_max": 3796.636169958095,
    "observed_min": 3782.7333205919167,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "4.5.0"
  },
  "condition_id": "5ccb0ca639bb6dcba048",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "c0f61e5a-7c6a-4ae4-a6bd-9ac9a54dfd53",
    "3c6dbc5a-1c90-469b-a97d-3991be6fa42f"
  ],
  "measured_from": "2025-11-11T22:28:25.716103+00:00",
  "measured_to": "2025-11-16T02:32:45.777395+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.4.0 · monster + junkshop + classroom · HIP · Windows · 1P · 64 cores · 128 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,818.2 / 3,818.2
Measurement dates
2025-05-10 – 2025-05-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2025-05-10T11:26:27.803419+00:00",
        "score": 3818.1972847810184,
        "source_record_id": "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
      }
    ],
    "median": null,
    "observed_max": 3818.1972847810184,
    "observed_min": 3818.1972847810184,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 64,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 128,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "4.4.0"
  },
  "condition_id": "867a6e4090a2905e642c",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "f5c84fc4-9913-4a01-a57a-fa6e330e410c"
  ],
  "measured_from": "2025-05-10T11:26:27.803419+00:00",
  "measured_to": "2025-05-10T11:26:27.803419+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 2

Too few submissions · Individual results onlyn = 2

4.2.0 · monster + junkshop + classroom · HIP · Windows · 1P · 96 cores · 192 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,533.24 / 3,605.62
Measurement dates
2024-11-18 – 2024-11-18
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-18T13:55:31.065354+00:00",
        "score": 3533.2389240885796,
        "source_record_id": "d3753458-4117-4916-be77-c535429c356b"
      },
      {
        "measured_at": "2024-11-18T14:23:16.591348+00:00",
        "score": 3605.619399023857,
        "source_record_id": "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
      }
    ],
    "median": null,
    "observed_max": 3605.619399023857,
    "observed_min": 3533.2389240885796,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 2,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 96,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 192,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "4.2.0"
  },
  "condition_id": "877c0865161f719f154e",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d3753458-4117-4916-be77-c535429c356b",
    "a1ba32c3-bb88-4f61-b9f2-c0bea78f54ac"
  ],
  "measured_from": "2024-11-18T13:55:31.065354+00:00",
  "measured_to": "2024-11-18T14:23:16.591348+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 2,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

4.2.0 · monster + junkshop + classroom · HIP · Linux · Ubuntu · 22.04 · 1P · 12 cores · 24 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,677.46 / 3,677.46
Measurement dates
2024-11-10 – 2024-11-10
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2024-11-10T16:15:49.579433+00:00",
        "score": 3677.4551659232216,
        "source_record_id": "edc905f7-0399-4b61-ae69-32850d89881d"
      }
    ],
    "median": null,
    "observed_max": 3677.4551659232216,
    "observed_min": 3677.4551659232216,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.1",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "Ubuntu",
      "dist_version": "22.04",
      "driver": null,
      "num_cpu_cores": 12,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 24,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "4.2.0"
  },
  "condition_id": "a00133841c428949222b",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "edc905f7-0399-4b61-ae69-32850d89881d"
  ],
  "measured_from": "2024-11-10T16:15:49.579433+00:00",
  "measured_to": "2024-11-10T16:15:49.579433+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · monster + junkshop + classroom · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,525.13 / 3,525.13
Measurement dates
2026-06-25 – 2026-06-25
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-25T11:09:54.113452+00:00",
        "score": 3525.131353903198,
        "source_record_id": "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
      }
    ],
    "median": null,
    "observed_max": 3525.131353903198,
    "observed_min": 3525.131353903198,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "5.0.0"
  },
  "condition_id": "76085092bdb7ec75c86a",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "acb5c119-6e01-4455-9e5c-2e0ac36f7438"
  ],
  "measured_from": "2026-06-25T11:09:54.113452+00:00",
  "measured_to": "2026-06-25T11:09:54.113452+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.1.1 · monster + junkshop + classroom · HIP · Windows · 1P · 16 cores · 32 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
4,486.4 / 4,486.4
Measurement dates
2026-06-26 – 2026-06-26
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-06-26T10:51:30.439222+00:00",
        "score": 4486.396207396023,
        "source_record_id": "0d8b7a84-25be-4d63-8181-98e771a206c8"
      }
    ],
    "median": null,
    "observed_max": 4486.396207396023,
    "observed_min": 4486.396207396023,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 16,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 32,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "5.1.1"
  },
  "condition_id": "131665ab7919836682b4",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "0d8b7a84-25be-4d63-8181-98e771a206c8"
  ],
  "measured_from": "2026-06-26T10:51:30.439222+00:00",
  "measured_to": "2026-06-26T10:51:30.439222+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · monster + junkshop + classroom · HIP · Windows · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,626.8 / 3,626.8
Measurement dates
2026-03-16 – 2026-03-16
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-16T16:58:00.457590+00:00",
        "score": 3626.8028091348597,
        "source_record_id": "97d37e7c-73f4-415d-ba06-a416262044c7"
      }
    ],
    "median": null,
    "observed_max": 3626.8028091348597,
    "observed_min": 3626.8028091348597,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon PRO W7900",
      "dist_name": "",
      "dist_version": "",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Windows"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "5.0.0"
  },
  "condition_id": "f1c843990cfd8d14953b",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "97d37e7c-73f4-415d-ba06-a416262044c7"
  ],
  "measured_from": "2026-03-16T16:58:00.457590+00:00",
  "measured_to": "2026-03-16T16:58:00.457590+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Blender complete benchmark

Too few submissions · No representative score

Run count: 1

Too few submissions · Individual results onlyn = 1

5.0.0 · monster + junkshop + classroom · HIP · Linux · Arch Linux · rolling · 1P · 8 cores · 16 threads

Counts represent submissions; the number of independent systems is unknown. Stock or overclocked status, RAM and power settings are also unknown.

Measurement scope and interpretation

Sample thresholds are an initial display policy, not a guarantee of statistical confidence. P95 is not a theoretical limit. OS submission shares do not represent user market share.

Observed minimum / maximum
3,228.87 / 3,228.87
Measurement dates
2026-03-17 – 2026-03-17
Duplicate submissions removed
0
Quantile calculation
linear interpolation · (n−1) × p

View source ↗

{
  "analysis": {
    "duplicates_removed": 0,
    "independent_system_count": null,
    "individual_results": [
      {
        "measured_at": "2026-03-17T12:00:19.894693+00:00",
        "score": 3228.871221428465,
        "source_record_id": "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
      }
    ],
    "median": null,
    "observed_max": 3228.871221428465,
    "observed_min": 3228.871221428465,
    "policy": "submitted-runs-v1",
    "quantile_method": "linear interpolation at (n-1)*p",
    "sample_count": 1,
    "sample_status": "individual",
    "settings_status": "unknown"
  },
  "comparison_readiness": "observational-only",
  "condition": {
    "benchmark": "Blender",
    "environment": {
      "backend": "HIP",
      "benchmark_script": "3.1.2",
      "device_name": "AMD Radeon Pro W7900",
      "dist_name": "Arch Linux",
      "dist_version": "rolling",
      "driver": null,
      "num_cpu_cores": 8,
      "num_cpu_sockets": 1,
      "num_cpu_threads": 16,
      "power_limit": null,
      "ram_bytes": null,
      "settings_status": "unknown",
      "system": "Linux"
    },
    "scene": "monster + junkshop + classroom",
    "scene_checksum": "c3453da9769193d968fb1cc675fd7f92a3424abe9ac730218f2e00b953389356",
    "version": "5.0.0"
  },
  "condition_id": "4c9e8a30fc33e9caa5bf",
  "document_id": "blender-open-data-3b13b06e2e84b715",
  "example_record_ids": [
    "d564cc8a-f969-48b5-ab89-b9f8cb5a7b2c"
  ],
  "measured_from": "2026-03-17T12:00:19.894693+00:00",
  "measured_to": "2026-03-17T12:00:19.894693+00:00",
  "median": null,
  "parameter_id": "benchmark.blender.total_samples_per_minute",
  "population": "public submitted runs; not independent systems",
  "product_id": "radeon-pro-w7900",
  "run_count": 1,
  "sample_status": "individual"
}

Sources

Source terms and retrieval dates are listed below. Published values and conditions may change.