| Precision sensing | Quality, energy and reading cost | 3.25–23.04% less modeled sensing energy Eight selected cases versus equal reading counts; readings use 50% of energy and planning costs 3%. Exact conventional planners match the reading counts. A ninth case saves 1.17%; three equal-noise cases use 3% more. Sensing-energy model. | Eight selected reading-plan cases save 3.25–23.04% modeled sensing energy versus equal counts after the added planning cost. Exact conventional planners match the reading counts. Related model results: D120: 33.25% longer completion; 1.4747% more energy at its worst tested condition. No winner saved energy across the required robustness cases. Related model jobs: D120. |
|---|
| Rugged field computers | Battery, weight, heat and deadlines | Product budget open. | In the main modeled case, the defined seismic data-processing job uses 0.08617% less energy. Related model results: D118: 4.07% longer completion; 0.8152% more energy at its worst tested condition. No winner saved energy across the required robustness cases. Related model jobs: D118. |
|---|
| Solar control | Housekeeping power, charge quality and protection | Product budget open. | At the main modeled operating point, the solar comparisons use the same energy as the strongest conventional controls tested. Related model jobs: D001, D002, D003, D004, D005. |
|---|
| Orbital flight computing | Power, heat path, fault recovery and deadlines | Product budget open. | The defined payload-data reporting job uses 0.05941% less energy in its main modeled case. The other spacecraft comparisons use the same energy as their conventional controls. Related model results: D043: 3.95% longer completion; 0.9803% more energy at its worst tested condition. No winner saved energy across the required robustness cases. Related model jobs: D043, D042, D045, D046. |
|---|
| Portable communications and recorders | Power, weight, required coverage and complete records | 19.97% less combined recorder/input-history storage 1,453,784,064→1,163,480,256 bytes for the reviewed design boundary. Writes fall 10.71%, reads 17.65%, timing-record checksum steps 60% on separate axes. Conditional design arithmetic; original full-service qualification remains open. | At the main modeled operating point, the Wi-Fi task uses 0.2563% less energy and the public-safety evidence-recorder task uses 0.03085% less than their strongest conventional controls. Related model results: D027: 51.49% longer completion; 0.4356% more energy at its worst tested condition; D099: 3.95% longer completion; 0.5095% more energy at its worst tested condition. No winner saved energy across the required robustness cases. Related model jobs: D027, D099. |
|---|
| Phones and mobile devices | Size, weight, heat and compute; radio and battery service | 0.65–4.61% less whole-phone energy in an illustration Sensing uses 20% of phone energy, other 80% stays fixed and no further costs are added. Same-battery life rises 0.65–4.83%. Exact conventional allocation planners match the reading counts. Worked phone-energy budget. | The sensing illustration assumes a 20% phone-energy share and no further costs. Exact conventional planners match the reading counts. Separate wearable and Wi-Fi reference jobs save 0.12418% and 0.25626% nominal modeled energy. Related model results: D027: 51.49% longer completion; 0.4356% more energy at its worst tested condition; D037: 28.70% shorter completion; 1.1896% more energy at its worst tested condition. No winner saved energy across the required robustness cases. Related model jobs: D027, D037. |
|---|
| Telephone towers and cellsite radios | Power drawn, heat and required coverage | Product budget open. | D021 has no candidate that meets its constraints. D022 has a selected capability but no conventional baseline meeting the comparison constraints. Both cellsite comparisons remain unmatched. Related model jobs: D021, D022, D023, D024, D025. |
|---|
| Industrial computing and control | Energy, control quality, faults and deadlines | Product budget open. | In the main modeled cases, the magnetic-bearing control task uses 0.01029% less energy and brushless-motor control uses 0.00354% less. The other listed industrial comparisons use the same energy as their conventional controls. Related model results: D019: 43.01% longer completion; 0.2022% more energy at its worst tested condition; D015: 37.37% longer completion; 0.0679% more energy at its worst tested condition. No winner saved energy across the required robustness cases. Related model jobs: D015, D019, D061, D062, D063, D064, D065. |
|---|
| Engineering software | Exact output, time, memory and complete logs | Product budget open. | Engineering reference models tie the strongest tested ordinary controls at nominal. In the separate API pilot, gmpy2 led all 26 preserved timing groups. The report-cache design removes 737,280 expression evaluations. Related model jobs: D066, D081, D082, D083, D084. |
|---|
