Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
Same-sign midpoint separately halves subnormal endpoints · case 01
Same-sign midpoint separately halves subnormal endpoints.
Same-sign midpoint separately halves subnormal endpoints · case 02
Same-sign midpoint separately halves subnormal endpoints.
Same-sign midpoint separately halves subnormal endpoints · case 03
Same-sign midpoint separately halves subnormal endpoints.
Same-sign midpoint separately halves subnormal endpoints · case 04
Same-sign midpoint separately halves subnormal endpoints.
Same-sign midpoint separately halves subnormal endpoints · case 05
Same-sign midpoint separately halves subnormal endpoints.
Midpoint computes an exact negative-zero endpoint through addition · case 01
Midpoint computes an exact negative-zero endpoint through addition.
Midpoint computes an exact negative-zero endpoint through addition · case 02
Midpoint computes an exact negative-zero endpoint through addition.
Midpoint computes an exact negative-zero endpoint through addition · case 03
Midpoint computes an exact negative-zero endpoint through addition.
Midpoint computes an exact negative-zero endpoint through addition · case 04
Midpoint computes an exact negative-zero endpoint through addition.
Midpoint computes an exact negative-zero endpoint through addition · case 05
Midpoint computes an exact negative-zero endpoint through addition.
Midpoint silently sorts reversed endpoints · case 01
Midpoint silently sorts reversed endpoints.
Midpoint silently sorts reversed endpoints · case 02
Midpoint silently sorts reversed endpoints.
Midpoint silently sorts reversed endpoints · case 03
Midpoint silently sorts reversed endpoints.
Midpoint silently sorts reversed endpoints · case 04
Midpoint silently sorts reversed endpoints.
Midpoint silently sorts reversed endpoints · case 05
Midpoint silently sorts reversed endpoints.
Decimal notation lays out unrounded coefficient digits · case 01
Decimal notation lays out unrounded coefficient digits.
Decimal notation lays out unrounded coefficient digits · case 02
Decimal notation lays out unrounded coefficient digits.
Decimal notation lays out unrounded coefficient digits · case 03
Decimal notation lays out unrounded coefficient digits.
Decimal notation lays out unrounded coefficient digits · case 04
Decimal notation lays out unrounded coefficient digits.
Decimal notation lays out unrounded coefficient digits · case 05
Decimal notation lays out unrounded coefficient digits.
Decimal notation uses coefficient exponent as scientific exponent · case 01
Decimal notation uses coefficient exponent as scientific exponent.
Decimal notation uses coefficient exponent as scientific exponent · case 02
Decimal notation uses coefficient exponent as scientific exponent.
Decimal notation uses coefficient exponent as scientific exponent · case 03
Decimal notation uses coefficient exponent as scientific exponent.
Decimal notation uses coefficient exponent as scientific exponent · case 04
Decimal notation uses coefficient exponent as scientific exponent.
Decimal notation uses coefficient exponent as scientific exponent · case 05
Decimal notation uses coefficient exponent as scientific exponent.
Engineering notation truncates negative exponent division · case 01
Engineering notation truncates negative exponent division.
Engineering notation truncates negative exponent division · case 02
Engineering notation truncates negative exponent division.
Engineering notation truncates negative exponent division · case 03
Engineering notation truncates negative exponent division.
Engineering notation truncates negative exponent division · case 04
Engineering notation truncates negative exponent division.
Engineering notation truncates negative exponent division · case 05
Engineering notation truncates negative exponent division.
Engineering notation uses exponent residue as a zero-based digit count · case 01
Engineering notation uses exponent residue as a zero-based digit count.
Engineering notation uses exponent residue as a zero-based digit count · case 02
Engineering notation uses exponent residue as a zero-based digit count.
Engineering notation uses exponent residue as a zero-based digit count · case 03
Engineering notation uses exponent residue as a zero-based digit count.
Engineering notation uses exponent residue as a zero-based digit count · case 04
Engineering notation uses exponent residue as a zero-based digit count.
Engineering notation uses exponent residue as a zero-based digit count · case 05
Engineering notation uses exponent residue as a zero-based digit count.
Engineering notation fails to pad an integer mantissa · case 01
Engineering notation fails to pad an integer mantissa.
Engineering notation fails to pad an integer mantissa · case 02
Engineering notation fails to pad an integer mantissa.
Engineering notation fails to pad an integer mantissa · case 03
Engineering notation fails to pad an integer mantissa.
Engineering notation fails to pad an integer mantissa · case 04
Engineering notation fails to pad an integer mantissa.
Engineering notation fails to pad an integer mantissa · case 05
Engineering notation fails to pad an integer mantissa.
Engineering notation always places its decimal after one digit · case 01
Engineering notation always places its decimal after one digit.
Engineering notation always places its decimal after one digit · case 02
Engineering notation always places its decimal after one digit.
Engineering notation always places its decimal after one digit · case 03
Engineering notation always places its decimal after one digit.
Engineering notation always places its decimal after one digit · case 04
Engineering notation always places its decimal after one digit.
Engineering notation always places its decimal after one digit · case 05
Engineering notation always places its decimal after one digit.
Decimal notation strips significant trailing coefficient zeros · case 01
Decimal notation strips significant trailing coefficient zeros.
Decimal notation strips significant trailing coefficient zeros · case 02
Decimal notation strips significant trailing coefficient zeros.
Decimal notation strips significant trailing coefficient zeros · case 03
Decimal notation strips significant trailing coefficient zeros.
Decimal notation strips significant trailing coefficient zeros · case 04
Decimal notation strips significant trailing coefficient zeros.
Decimal notation strips significant trailing coefficient zeros · case 05
Decimal notation strips significant trailing coefficient zeros.
Decimal notation drops the sign of a zero coefficient · case 01
Decimal notation drops the sign of a zero coefficient.
Decimal notation drops the sign of a zero coefficient · case 02
Decimal notation drops the sign of a zero coefficient.
Decimal notation drops the sign of a zero coefficient · case 03
Decimal notation drops the sign of a zero coefficient.
Decimal notation drops the sign of a zero coefficient · case 04
Decimal notation drops the sign of a zero coefficient.
Decimal notation drops the sign of a zero coefficient · case 05
Decimal notation drops the sign of a zero coefficient.
Decimal notation derives sign from comparison with zero · case 01
Decimal notation derives sign from comparison with zero.
Decimal notation derives sign from comparison with zero · case 02
Decimal notation derives sign from comparison with zero.
Decimal notation derives sign from comparison with zero · case 03
Decimal notation derives sign from comparison with zero.
Decimal notation derives sign from comparison with zero · case 04
Decimal notation derives sign from comparison with zero.
Decimal notation derives sign from comparison with zero · case 05
Decimal notation derives sign from comparison with zero.
Split base-two exponential uses an unconverted natural exponential · case 01
Split base-two exponential uses an unconverted natural exponential.
Split base-two exponential uses an unconverted natural exponential · case 02
Split base-two exponential uses an unconverted natural exponential.
Split base-two exponential uses an unconverted natural exponential · case 03
Split base-two exponential uses an unconverted natural exponential.
Split base-two exponential uses an unconverted natural exponential · case 04
Split base-two exponential uses an unconverted natural exponential.
Split base-two exponential uses an unconverted natural exponential · case 05
Split base-two exponential uses an unconverted natural exponential.
Split base-two exponential underflows its scale before multiplication · case 01
Split base-two exponential underflows its scale before multiplication.
Split base-two exponential underflows its scale before multiplication · case 02
Split base-two exponential underflows its scale before multiplication.
Split base-two exponential underflows its scale before multiplication · case 03
Split base-two exponential underflows its scale before multiplication.
Split base-two exponential underflows its scale before multiplication · case 04
Split base-two exponential underflows its scale before multiplication.
Split base-two exponential underflows its scale before multiplication · case 05
Split base-two exponential underflows its scale before multiplication.
Base-two exponential flushes subnormal outputs · case 01
Base-two exponential flushes subnormal outputs.
Base-two exponential flushes subnormal outputs · case 02
Base-two exponential flushes subnormal outputs.
Base-two exponential flushes subnormal outputs · case 03
Base-two exponential flushes subnormal outputs.
Base-two exponential flushes subnormal outputs · case 04
Base-two exponential flushes subnormal outputs.
Base-two exponential flushes subnormal outputs · case 05
Base-two exponential flushes subnormal outputs.
Base-two exponential overflows the highest finite power · case 01
Base-two exponential overflows the highest finite power.
Base-two exponential overflows the highest finite power · case 02
Base-two exponential overflows the highest finite power.
Base-two exponential overflows the highest finite power · case 03
Base-two exponential overflows the highest finite power.
Base-two exponential overflows the highest finite power · case 04
Base-two exponential overflows the highest finite power.
Base-two exponential overflows the highest finite power · case 05
Base-two exponential overflows the highest finite power.
Split base-two exponential reverses its fractional residual · case 01
Split base-two exponential reverses its fractional residual.
Split base-two exponential reverses its fractional residual · case 02
Split base-two exponential reverses its fractional residual.
Split base-two exponential reverses its fractional residual · case 03
Split base-two exponential reverses its fractional residual.
Split base-two exponential reverses its fractional residual · case 04
Split base-two exponential reverses its fractional residual.
Split base-two exponential reverses its fractional residual · case 05
Split base-two exponential reverses its fractional residual.
Cube root feeds a negative value to a fractional power · case 01
Cube root feeds a negative value to a fractional power.
Cube root feeds a negative value to a fractional power · case 02
Cube root feeds a negative value to a fractional power.
Cube root feeds a negative value to a fractional power · case 03
Cube root feeds a negative value to a fractional power.
Cube root feeds a negative value to a fractional power · case 04
Cube root feeds a negative value to a fractional power.
Cube root feeds a negative value to a fractional power · case 05
Cube root feeds a negative value to a fractional power.
Cube root discards the residual exponent modulo three · case 01
Cube root discards the residual exponent modulo three.
Cube root discards the residual exponent modulo three · case 02
Cube root discards the residual exponent modulo three.
Cube root discards the residual exponent modulo three · case 03
Cube root discards the residual exponent modulo three.
Cube root discards the residual exponent modulo three · case 04
Cube root discards the residual exponent modulo three.
Cube root discards the residual exponent modulo three · case 05
Cube root discards the residual exponent modulo three.
Cube root rescales by the negated exponent quotient · case 01
Cube root rescales by the negated exponent quotient.
Cube root rescales by the negated exponent quotient · case 02
Cube root rescales by the negated exponent quotient.
Cube root rescales by the negated exponent quotient · case 03
Cube root rescales by the negated exponent quotient.
Cube root rescales by the negated exponent quotient · case 04
Cube root rescales by the negated exponent quotient.
Cube root rescales by the negated exponent quotient · case 05
Cube root rescales by the negated exponent quotient.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗