feat(rip-200): add Pentium M antiquity tier (Banias/Dothan/Yonah)

[Scottcjn]

Summary

Pentium M was missing from ANTIQUITY_MULTIPLIERS and silently fell through to the modern x86 bucket (0.8x). Adds 4 entries + brand-string detection so genuine vintage Pentium M silicon (2003-2006) gets its proper multiplier.

The trigger: brought a real 2003 IBM ThinkPad T40 (2373-7CU, Pentium M Banias 1.5GHz) online today as a RustChain miner. All 7 hardware fingerprint checks pass cleanly — anti-emulation 0 indicators, SSE+SSE2 only (no SSE3), no LM (i686), 1MB L2 — but the server was about to assign it 0.8x. That's wrong: this machine is 23 years old, older than most entries in the table that already get a vintage bonus.

Changes

1. node/rip_200_round_robin_1cpu1vote.py (+11/-1)

Four new entries between pentium_iii (2.0x) and pentium4 (1.5x):

Tier	Mult	Year	Notes
pentium_m	1.9	2003-2006	Generic fallback
pentium_m_banias	1.9	2003	130nm, 1MB L2, max 1.7GHz — T40 class
pentium_m_dothan	1.8	2004	90nm, 2MB L2, up to 2.26GHz
pentium_m_yonah	1.6	2006	dual-core, first Core/Core Duo

Slots into the existing chronological curve: PIII (1999) > Pentium M (2003) > P4 (2000-2008, market-saturated, already discounted) > Core 2 (2006).

2. node/rustchain_v2_integrated_v2.2.1_rip200.py (+74)

New _detect_x86_vintage() helper called from derive_verified_device() before the final fall-through (the miner hardcodes arch='modern' for ALL x86 in _get_hw_info, so server-side brand-string lookup is the only place to recover the actual tier).

• Pentium M family split by clock speed (Banias maxed at 1.7GHz, Dothan exceeded it) for 32-bit silicon, and by machine field (x86_64 indicates Yonah).
• Regex uses \b boundaries and a (?!\d) negative lookahead to avoid false-matching Apple M4 or future Apple-style names.
• Also covers PIII / PII / Pentium Pro / Pentium MMX — all of which had multiplier entries already but no detection path.

Verification

15-case unit test of the regex passes including these false-positive guards:

• Apple M4 → None (no match)
• AMD K6-2 with MMX support 350MHz → None (no match)
• Intel(R) Pentium(R) D CPU 3.00GHz → None (not Pentium M)
• Intel(R) Pentium(R) II 350MHz → pentium_ii (not matched as III)

End-to-end against the T40:

Fingerprint snapshot (Banias-class reference):
  clock_drift cv:    0.0324
  cache L1/L2/L3:    500.94 / 463.76 / 468.90 ns (no L3 — L2 hits dominate)
  simd flags count:  26 (no AVX, no AltiVec, no NEON)
  thermal drift:     0.9932 (real silicon stability)
  anti_emulation:    0 VM indicators

Both files pass python3 -m py_compile.

Test plan

• py_compile on both edited files
• 15-case regex unit test (Pentium M Banias/Dothan/Yonah + PIII/PII/Pro/MMX + false-positive guards)
• Verified fingerprint pipeline against real T40 hardware
• Deploy to Node 1 (50.28.86.131) and confirm /api/miners shows T40 with device_arch=pentium_m_banias
• Deploy to Nodes 2/3/4
• Confirm T40 receives 1.9x weighting in next epoch settlement

🤖 Generated with Claude Code

[github-actions]

✅ BCOS v2 Scan Results

Metric	Value
Trust Score	60/100
Certificate ID	BCOS-741b8a59
Tier	L1 (met)

What does this mean?

The BCOS (Beacon Certified Open Source) engine scans for:

• SPDX license header compliance
• Known CVE vulnerabilities (OSV database)
• Static analysis findings (Semgrep)
• SBOM completeness
• Dependency freshness
• Test infrastructure evidence
• Review attestation tier

Full report | What is BCOS?

---

BCOS v2 Engine - Free & Open Source (MIT) - Elyan Labs

[Scottcjn]

@createkr — heads up, this PR adds a new Pentium M antiquity tier (Banias / Dothan / Yonah). Before it merges, your Node 4 (38.76.217.189:8099) appears unreachable from here — all probed TCP ports (443, 8099, 8088, 8080, 22) timed out. Could you confirm the host is up + apply the patch when you have a window?

If you'd like a hands-off rollout, the production-safe patch script that was used on Nodes 1, 2, and POWER8 is below. It's idempotent (re-running is a no-op), backs up both files before modifying, and reverts if the post-edit py_compile fails. Same +85 lines that landed in this PR's diff, applied surgically against the older code on production nodes (avoiding the larger main-branch drift):

# 1. Drop https://gist.github.com/... [or scp from a Scott-controlled box] to /tmp/banias_patch.py on your node
# 2. Dry-run first
python3 /tmp/banias_patch.py --dry-run --base /path/to/your/rustchain
# 3. Apply
python3 /tmp/banias_patch.py --base /path/to/your/rustchain
# 4. Restart your node service (whatever service unit you use)
systemctl restart <your-rustchain-service>
# 5. Verify
curl -s http://localhost:8099/health

What the patch does:

• rip_200_round_robin_1cpu1vote.py: adds 4 entries (pentium_m, pentium_m_banias, pentium_m_dothan, pentium_m_yonah) between pentium_iii and pentium4.
• rustchain_v2_integrated_v2.2.1_rip200.py: adds _detect_x86_vintage() helper called from derive_verified_device() before the final return, so vintage Pentium silicon stops falling into the 'modern' (0.8x) bucket.

Happy to send the script via DM/Telegram/whatever channel works best for you. Trigger for this work was an actual 2003 IBM ThinkPad T40 coming online as a RustChain miner today — full fingerprint signature in the PR body.

[jaxint]

Great work! Thanks for contributing to RustChain! 🦀

[Scottcjn]

Session 2026-05-27 series — this PR is one of five related fixes from a single session-arc that started with bringing a 2003 IBM ThinkPad T40 (Pentium M Banias) online as a RustChain miner. Each PR is independently reviewable + mergeable, but they're all the same root pattern: server validation was written for the v2 fingerprint format; the v3 miner uses different field names + locations:

• #6423 — feat: Pentium M Banias antiquity tier (1.9x in both ANTIQUITY_MULTIPLIERS and HARDWARE_WEIGHTS)
• #6426 — fix: accept canonical-JSON v3 signature alongside legacy 4-field MAC
• #6428 — fix: hw-binding extract_entropy_profile accepts v3 field names
• #6429 — fix: _has_powerpc_cache_profile accepts arch from simd_identity (POWER8 fix)
• #6430 — scripts: backport settle_fix_prefer_epoch_enroll.py for production nodes behind main

Production deployment status: all five fixes deployed surgically to Node 1 (50.28.86.131), Node 2 (50.28.86.153), and POWER8 (rustchain-node.service on 100.75.100.89). Nodes 3 (Ryan, offline 6d) + 4 (createkr, no access) pending. T40 earning at 1.9x top share in epoch 176 confirming end-to-end fix. POWER8 attestation now passing after #6429.

[crystal-tensor]

✅ Code Review: APPROVED

Summary

Adds Pentium M (Banias/Dothan/Yonah) to antiquity multipliers. Real 2003 IBM ThinkPad T40 was getting modern (0.8x) instead of proper pentium_m (1.9x) tier.

Changes Reviewed

1. node/rip_200_round_robin_1cpu1vote.py (+11/-1):

   • ✅ Adds 4 Pentium M entries: generic (1.9), Banias (1.9), Dothan (1.8), Yonah (1.6)
   • ✅ Correctly positioned between Pentium III (2.0) and Pentium 4 (1.5)
   • ✅ Yonah lower multiplier (dual-core, 2006, closer to Core architecture)

2. node/rustchain_v2_integrated_v2.2.1_rip200.py (+82/-1):

   • ✅ _detect_x86_vintage(): Brand string detection with regex
     • Pentium M: \bpentium(?:\(r\))?\s+m\b(?!\d) — avoids false-matching "Pentium M4"
     • Clock speed parsing: MHz/GHz from brand string
     • Banias (≤1700MHz, 32-bit) / Dothan (>1700MHz, 32-bit) / Yonah (64-bit)
   • ✅ Also adds Pentium III, II, Pro, MMX detection
   • ✅ Adds matching entries to HARDWARE_WEIGHTS in init_db()
   • ✅ Verified against real ThinkPad T40 hardware

Code Quality

• ✅ Thorough regex patterns with boundary guards
• ✅ Hardware-verified multipliers (real ThinkPad T40)
• ✅ Comprehensive docstring with verification details
• ✅ Consistent with existing antiquity multiplier structure

Result: APPROVED ✅

---

Reviewed by QClaw AI Agent
Bounty claim: 3-25 RTC per CONTRIBUTING.md