Over-constrained team split 40×8×1

One session: eight teams of five with department and gender targets that uneven department sizes make impossible to meet exactly, plus hard and soft pair rules; the proven optimum cost is 255.

Open in GroupMixer Results

40 people8 groups1 sessions60s limit8 CPUs

Scenario ID: internal.hard-1s-unbalanced-teams-40p-8g.v1
Problem hash: sha256:399ba4897a76e652994490d9809b5449f37ad7646dd6990b91aa4fbd5697c3bf

Comparator results

publication r20

Barasu-kun

Not run

valid —best —median —

Configuration

barasu-kun-0586a3f-sequential-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

GolfSoftware Play All

Not run

valid —best —median —

Configuration

golfsoftware-play-all-live-55-second-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

Good-Enough Golfers

Not run

valid —best —median —

Configuration

good-enough-golfers-4db20742-fixed-ga-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

GroupMixer

Feasible

valid 5/5best 255median 255

Configuration

browser-auto-8-defaults-v1

evidence r20 · eligible

Measured 11 October 2026 on an Intel Core Ultra 9 386H; eight pinned CPUs, 8 GiB physical memory, swap disabled. Current production browser Auto, eight requested lanes, automatic Extreme handoff and default CP-SAT admission.

Gurobi

Unavailable

valid —best —median —

Configuration

unavailable-license

evidence r20 · unavailable

No runnable licensed environment and reproducible formulation are available.

Halftime Team Pairings

Not run

valid —best —median —

Configuration

halftime-team-pairings-live-sequential-last-six-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

Hexaly

Unavailable

valid —best —median —

Configuration

unavailable-license

evidence r20 · unavailable

No runnable licensed environment and reproducible formulation are available.

IIENSTITU Random Group Generator

Not run

valid —best —median —

Configuration

iienstitu-live-avoid-history-seeded-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

Optimal Group Generator

Not run

valid —best —median —

Configuration

optimal-group-generator-live-fixed-ga-sequential-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

Google OR-Tools CP-SAT

Optimal

valid 5/5best 255median 255

Configuration

ortools-cpsat-measured-20261011:generic-staged-v4

evidence r20 · eligible

Measured 11 October 2026 on an Intel Core Ultra 9 386H; eight pinned CPUs, 8 GiB physical memory, swap disabled. Measured OR-Tools 9.15.6755 campaign formulation and parameters; this additional case was outside the original frozen matrix.

Social Golfer Online

Not run

valid —best —median —

Configuration

social-golfer-online-recompute-all-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

Teach Arcade Group Maker

Not run

valid —best —median —

Configuration

teach-arcade-live-roundrobin-sequential-history-v1

evidence r20 · verify

Not run on this case; capability has not been assessed.

TourneyCard Golf Randomizer

Not run

valid —best —median —

Configuration

tourneycard-v2.0.4-case-specific-iterations-v2

evidence r20 · verify

Not run on this case; capability has not been assessed.

Comparator status, attempts, score, and configuration
Tool	Status	Valid	Best	Median	Evidence
Barasu-kun	Not run	—	—	—	Configuration `barasu-kun-0586a3f-sequential-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
GolfSoftware Play All	Not run	—	—	—	Configuration `golfsoftware-play-all-live-55-second-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
Good-Enough Golfers	Not run	—	—	—	Configuration `good-enough-golfers-4db20742-fixed-ga-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
GroupMixer	Feasible	5/5	255	255	Configuration `browser-auto-8-defaults-v1` evidence r20 · eligible Measured 11 October 2026 on an Intel Core Ultra 9 386H; eight pinned CPUs, 8 GiB physical memory, swap disabled. Current production browser Auto, eight requested lanes, automatic Extreme handoff and default CP-SAT admission.
Gurobi	Unavailable	—	—	—	Configuration `unavailable-license` evidence r20 · unavailable No runnable licensed environment and reproducible formulation are available.
Halftime Team Pairings	Not run	—	—	—	Configuration `halftime-team-pairings-live-sequential-last-six-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
Hexaly	Unavailable	—	—	—	Configuration `unavailable-license` evidence r20 · unavailable No runnable licensed environment and reproducible formulation are available.
IIENSTITU Random Group Generator	Not run	—	—	—	Configuration `iienstitu-live-avoid-history-seeded-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
Optimal Group Generator	Not run	—	—	—	Configuration `optimal-group-generator-live-fixed-ga-sequential-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
Google OR-Tools CP-SAT	Optimal	5/5	255	255	Configuration `ortools-cpsat-measured-20261011:generic-staged-v4` evidence r20 · eligible Measured 11 October 2026 on an Intel Core Ultra 9 386H; eight pinned CPUs, 8 GiB physical memory, swap disabled. Measured OR-Tools 9.15.6755 campaign formulation and parameters; this additional case was outside the original frozen matrix.
Social Golfer Online	Not run	—	—	—	Configuration `social-golfer-online-recompute-all-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
Teach Arcade Group Maker	Not run	—	—	—	Configuration `teach-arcade-live-roundrobin-sequential-history-v1` evidence r20 · verify Not run on this case; capability has not been assessed.
TourneyCard Golf Randomizer	Not run	—	—	—	Configuration `tourneycard-v2.0.4-case-specific-iterations-v2` evidence r20 · verify Not run on this case; capability has not been assessed.

Measured attempts

GroupMixer · seed-101feasible timeout

local browsercost 25560.829sRSS —

GroupMixer · seed-202feasible timeout

local browsercost 25560.812sRSS —

GroupMixer · seed-303feasible timeout

local browsercost 25560.853sRSS —

GroupMixer · seed-404feasible timeout

local browsercost 25560.804sRSS —

GroupMixer · seed-505feasible timeout

local browsercost 25560.79sRSS —

Google OR-Tools CP-SAT · seed-101optimal proven

native processcost 2550.659sRSS —

Google OR-Tools CP-SAT · seed-202optimal proven

native processcost 2550.633sRSS —

Google OR-Tools CP-SAT · seed-303optimal proven

native processcost 2550.709sRSS —

Google OR-Tools CP-SAT · seed-404optimal proven

native processcost 2550.652sRSS —

Google OR-Tools CP-SAT · seed-505optimal proven

native processcost 2550.632sRSS —

Per-attempt benchmark status, execution class, canonical cost, runtime, and memory
Tool / attempt	Execution	Status	Cost	Wall	Peak RSS
GroupMixer · seed-101	Browser	feasible timeout	255	60.829s	—
GroupMixer · seed-202	Browser	feasible timeout	255	60.812s	—
GroupMixer · seed-303	Browser	feasible timeout	255	60.853s	—
GroupMixer · seed-404	Browser	feasible timeout	255	60.804s	—
GroupMixer · seed-505	Browser	feasible timeout	255	60.79s	—
Google OR-Tools CP-SAT · seed-101	Native	optimal proven	255	0.659s	—
Google OR-Tools CP-SAT · seed-202	Native	optimal proven	255	0.633s	—
Google OR-Tools CP-SAT · seed-303	Native	optimal proven	255	0.709s	—
Google OR-Tools CP-SAT · seed-404	Native	optimal proven	255	0.652s	—
Google OR-Tools CP-SAT · seed-505	Native	optimal proven	255	0.632s	—

Pair-frequency distribution

GroupMixer · seed-101

Number of participant pairs by meeting frequency
Meetings	Participant pairs
1	80

Unique contacts: 80
Repeat excess: 0
Repeated pairs: 0
Worst pair count: 1
Min contacts / person: 4
Mean contacts / person: 4

Evidence and reproduction

gm-scenarios materialize internal.hard-1s-unbalanced-teams-40p-8g.v1 --pretty --output case.json

Archive Checksums