VR2R-0730_r1

VR2R-0730

① SA Source

• Source: 開啟完整 SA 文章
• Section: VR NVL72 TCO: BoM and Power Budget Analysis
• Line hint: 637

Context Before

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.

Evidence

Running VR NVL72 on 1800W would probably mean lower FP4 Sparse FLOPs than on 2300W.

Context After

Inference throughput too can diverge materially from marketed peak FLOPS, meaning spec-sheet compute does not directly translate into real-world token generation performance. While the B300 is rated at 4,500 Dense FP8 TFLOPs versus 5,000 Dense FP8 TFLOPs for MI355, implying a 10% theoretical compute disadvantage and with same marketed memory bandwidth of 8TB/s measured inference token throughput from our InferenceX benchmarks shows B300 delivering roughly 6.3x the performance at 100 interactivity for Deepseek R1, using 8k input tokens and 1k output tokens.

② Atomic Claim

VR NVL72 若以 1800W 運作，其 FP4 Sparse FLOPs 很可能低於 2300W configuration。

• Epistemic Mode: ASSERTED
• Mapping Status: PARTIAL

③ Semantic Frame

{
  "frame_type": "NARY_RELATION",
  "participants": [
    {
      "node": {
        "id": "04_knowledge_base/VR NVL72",
        "label": "VR NVL72"
      },
      "role": "platform"
    },
    {
      "node": {
        "id": "04_knowledge_base/FP4",
        "label": "FP4"
      },
      "role": "datatype"
    },
    {
      "node": {
        "id": "04_knowledge_base/Sparsity",
        "label": "Sparsity"
      },
      "role": "feature"
    }
  ],
  "qualifiers": {
    "condition_text": "probabilistic operating-mode comparison",
    "numeric_mentions": [
      "1800W",
      "2300W"
    ],
    "temporal_mentions": []
  },
  "relation_type": "VR_NVL72_FP4_SPARSE_FLOPS_LOWER_AT_1800W_THAN_2300W"
}

④ Canonical Entity Mapping

Role	Surface Label	Canonical Target
platform	VR NVL72	04_knowledge_base/VR NVL72
datatype	FP4	FP4
feature	Sparsity	Sparsity

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。