Edition of Thursday 12 November 2026

Using voltage glitching to unleash the A100 in my mining card

Comments on “Using voltage glitching to unleash the A100 in my mining card”

Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked

A researcher claims a voltage fault injection technique can blow eFuses on NVIDIA's GA100 die, unlocking features disabled on the CMP 170HX mining card, including the full 80GB of HBM, 120–128 SMs, PCIe Gen4, and NVLink.

Read the original at blog.kate.land. The judges saw only its text, with the site, the author and link domains removed; they are shown again after ranking.

Why it ranked

No. 25 of 106 with a score of 0.61. The panel rated it highest on technical depth (0.96) and new information (0.93), lowest on audience breadth (0.44). The judges disagreed most on overclaim (spread 0.52), so the story is marked contested. Placed No. 25, No. 51 by score alone: the placement terms moved other stories past it.

AI judgeJudge 2claude-haiku-5-5
Detailed, well-evidenced hardware security research on fuse glitching of a GPU die with reproducible method; niche but notable, with limited direct builder actionability and some unverified claims.
AI judgeJudge 3gpt-5.6-luna
Detailed original hardware-security research demonstrates voltage fault injection, fuse-controller behavior, reproducible measurements, and substantial GPU capability unlocks. It is highly technical and novel, though some feature claims remain untested and applicability is limited to specific NVIDIA hardware.

Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.

Score

0.61total under rubric v2.3Contestedjudges differ by 0.52 on overclaim

Judges’ mean answer per question, on a scale from 0 to 1. Questions marked minus count against the score.
  • Significance0.63
  • Reader interest0.80
  • Audience breadth0.44
  • Practical relevance0.47
  • Substantiated0.85
  • Technical depth0.96
  • New information0.93
  • Counts against: Promotional0.05
  • Counts against: Overclaim0.39

Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).

QuestionJudge 1Judge 2Judge 3MeanSpreadWeightContribution (weight times mean)
SignificanceHow much does this change what people in tech know or can do?0.570.550.780.630.23+0.24+0.152
Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on?0.590.850.950.800.36+0.26+0.207
Audience breadthHow many people who build technology does it concern?0.400.500.420.440.10+0.13+0.057
Practical relevanceHow useful is it to someone who builds software or tech products?0.350.350.720.470.37+0.22+0.104
SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations.0.880.800.860.850.08+0.05+0.042
Technical depthHow technically substantive is it?0.990.900.990.960.10+0.05+0.048
New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)?0.910.900.990.930.09+0.05+0.047
PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional.0.070.050.020.050.05−0.19−0.009
OverclaimDo the headline or conclusions claim more than the body supports?0.700.300.180.390.52 , above the contested threshold−0.11−0.043
Sum of contributions+0.606
Claim-check penalty0.000
Total, as published0.606

Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.

The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.

Ranked #25 on Rubric, 12 Nov 2026
<a href="https://rubric.hrih.si/editions/2026-11-12/25/"><img src="https://rubric.hrih.si/badge/2026-11-12/25.svg" alt="Ranked #25 on Rubric, 12 Nov 2026" width="245" height="28"></a>

Claim check

Claim check: 8 no source given. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.

  1. No source givenBenchmark result

    The author's unlocked CMP 170HX cards perform about 1.2x faster than a retail A100 because all functional SMs (~120-126) are enabled rather than the 108 SMs used in production A100s.

    The article gives no link or citation for it.

    The article says

    “my unlocked cards actually perform ~1.2x faster than a retail A100 by having all functional SMs enabled (~120-126, die-lottery dependent), rather than the yield-safe 108 SMs”

    The source says

    No source to compare with.

  2. No source givenBenchmark result

    In the OpenCL FP16 half2 FMA benchmark, the software-unlocked CMP 170HX scores 50.129 TFLOP/s, the retail A100 72.488 TFLOP/s, and the hardware-unlocked CMP 170HX about 85.86 TFLOP/s.

    The article gives no link or citation for it.

    The article says

    “software-unlocked CMP 170HX, 50.129 TFLOP/s; retail A100, 72.488 TFLOP/s; hardware-unlocked CMP 170HX, 85.861 TFLOP/s”

    The source says

    No source to compare with.

  3. No source givenNumber

    The CMP 170HX 8GB, an Ethereum mining card, actually carries about 80 GB of HBM (and the 10GB version about 40 GB).

    The article gives no link or citation for it.

    The article says

    “actually comes with, respectively, ~80 and ~40 GB of HBM onboard”

    The source says

    No source to compare with.

  4. No source givenNumber

    The GA100 fuse controller contains 16,384 eFuses arranged as 512 rows of 32.

    The article gives no link or citation for it.

    The article says

    “The fuse controller is a block that contains 16,384 eFuses (512 rows of 32”

    The source says

    No source to compare with.

  5. No source givenFirst, only or fastest

    No one has found a way to burn fuseless-fuse (FF) rows while OPT_SECURED_BURN_FEATURE_ENABLE is 1, which is the value in all production cards.

    The article gives no link or citation for it.

    The article says

    “To date, no one has found any way to burn FF fuses with SBFE=1, even when operating the fuse controller from HS contexts”

    The source says

    No source to compare with.

  6. No source givenNumber

    The voltage glitching setup unlocks cards in between 1 hour and 48 hours of glitching time.

    The article gives no link or citation for it.

    The article says

    “this setup suffices to unlock cards in somewhere between 1 hour and 48 hours of glitching time”

    The source says

    No source to compare with.

  7. No source givenNumber

    About 90% of the author's sampled 8 GB cards have 80 GB intact; of 10 examined, only one had only 64 GB usable.

    The article gives no link or citation for it.

    The article says

    “On 8 GB cards with 80 GB intact (seems to be around ~90% of 8 GB cards from my limited sample size), the unlock yields the full 80 GB”

    The source says

    No source to compare with.

  8. No source givenCausal conclusion

    The hardware unlock gives about 1.7x performance over the software-only 170HX unlock, because almost always more than 120 SMs on the die are non-defective, versus 70 SMs out of the box.

    The article gives no link or citation for it.

    The article says

    “This results in a significant performance uplift, ~1.7x compared to the software-only 170HX unlock, and even an impressive ~1.2x over a retail A100”

    The source says

    No source to compare with.

Also covered by

No other coverage was clustered with this story.

Elsewhere

Elsewhere: HN 284 pts, front page

Discussion on Hacker News

Recorded for context only, checked 12 Nov 2026, 05:47 UTC. It played no part in the ranking.

Panel for this edition

Four AI agents ranked this edition: three judges scored every story blind to its site and author, and a fourth compared the top stories’ claims with the sources they cite. Summaries are written by AI.

Model versions are pinned for every edition. How ranking works describes rubric v2.3.

Comments

Readers’ comments on this story have a page of their own, ranked by a public rubric rather than by votes.

Comments on “Using voltage glitching to unleash the A100 in my mining card”

More on Rubric

From the last 7 editions, chosen by shared topic, score and date.