VR2R-0727_r1

VR2R-0727

① SA Source

• Source: 開啟完整 SA 文章
• Section: VR NVL72 TCO: BoM and Power Budget Analysis
• Line hint: 635

Context Before

As always, one caveat is that this comparison is done based on a marketed dense PFLOP basis. Effective dense PFLOP (i.e. the real world chip throughput) can differ based on Model Flops Utilization % (MFU), and

Evidence

in general we have seen NVIDIA chip operate at a higher MFU % vs AMD chips, suggesting that performance per TCO based on effective dense PFLOPs could be better for NVIDIA systems vs AMD—however, MFU is dependent on actual workloads with no one-size-fits-all MFU % that is consistently applicable to either systems.

Context After

Indeed, real world use of FP4 Sparsity will probably not reach 50 PFLOPS but it will probably deliver better effective FLOPs than FP4 Dense, but we have yet to evaluate what that real-world performance could be. Running VR NVL72 on 1800W would probably mean lower FP4 Sparse FLOPs than on 2300W.

② Atomic Claim

整體而言，SemiAnalysis 觀察到 NVIDIA chips 的 MFU 通常高於 AMD chips，因此若以 effective dense PFLOPs 衡量，NVIDIA systems 的 performance per TCO 可能更好；但 MFU 會隨實際 workload 改變，並不存在能固定套用到任一系統的單一 MFU%。

• Epistemic Mode: HYPOTHETICAL
• Mapping Status: COMPLETE

③ Semantic Frame

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "02_companies/NVDA",
        "label": "NVIDIA"
      },
      "role": "vendor"
    },
    {
      "node": {
        "id": "02_companies/AMD",
        "label": "AMD"
      },
      "role": "comparison_vendor"
    },
    {
      "node": {
        "id": "04_knowledge_base/Model FLOPS Utilization",
        "label": "MFU"
      },
      "role": "utilization_metric"
    }
  ],
  "qualifiers": {
    "condition_text": "MFU is workload-dependent; effective-dense performance-per-TCO implication is hypothetical",
    "numeric_mentions": [],
    "temporal_mentions": []
  },
  "relation_type": "NVIDIA_AMD_MFU_AND_EFFECTIVE_PERFORMANCE_PER_TCO_COMPARISON"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
vendor	NVIDIA	NVDA
comparison_vendor	AMD	AMD
utilization_metric	MFU	04_knowledge_base/Model FLOPS Utilization

⑤ Human Review

請在 Properties 逐項確認：

• 原文 → Atomic Claim 是否忠實
• Atomic Claim → Semantic Frame 是否忠實
• Canonical Entity mapping 是否正確
• Epistemic mode 是否保留原文語氣
• 最後選擇 review_action

Review state

Markdown 內文不是正式 approval。只有 Apply bridge 寫入的 Decision Ledger event 才是正式決策。