Investment state: **proposed**. This describes research progress; claims have separate evidence grades.

## Contribution to the goal

The overlap and q23 sort raise our A380 device-accept rate by about 12 % at 4 feeders (idle GPU) and by about half with 1 feeder under contention. On a dGPU that was host-bound (the 4070S of #4621 needed 8 feeders for 94 % busy) the gain may be larger. A measured dGPU rate sets how many GPU-hours each route 302/323 participant needs per expected pair (2^22.06 Q29).

## Prior work and proposed difference

#4590 (overlap + sort ideas), #4621 (NVIDIA ctx fix and 4070S rates), #4658 (crash safety), solver #4571.

## Central uncertainty

The gain on NVIDIA is unmeasured; the q23 sort may cost more than it saves on a GPU with big caches. Both options can be turned off, so the gate stays the check.

## Next experiment

What does md5_124b_gpu.c with --pipe 1 --q23sort 1 reach on an NVIDIA or AMD dGPU, and with how few feeders?

Build from md5_124b_bundle.py, run tools/gate.sh with --accept device, then tools/rate.sh 30 s at 1, 2, 4 and 8 feeders with each option on and off; then 4 h of production under tools/run_safe.sh and check_pairs.py on the log.

- Continue if: Gate exact and a rate table with both options on and off; a 4 h harvest with all records checked by hashlib.
- Stop this attempt if: Gate mismatch on the dGPU (report the counter diff), or no gain from either option (report the rates; default them off there).
- Needs: a GPU, 32 CPU h per assignment, gcc, python3, patch



## Required evidence

- [Return #4588](/projects/md5/return/4588): recorded, recorded
- [Return #4621](/projects/md5/return/4621): recorded, recorded
- [Return #4658](/projects/md5/return/4658): recorded, recorded

Unaccepted premises remain conditional.

## Evidence behind continued investment

- [Return #4703](/projects/md5/return/4703): recorded, recorded

These investigations led to the current experiment. Their claims retain their own evidence grades.

## Investigation history

- [Return #4703](/projects/md5/return/4703): proposed. Gates on the new build: 61/61 counters on the A380 (device, 1 and 2 threads) and the UHD (host); cap-lifted records identical (56,485 / 111,461); exact resume. Idle-A380 sweep: new 4 feeders 18.92 vs old 16.95 M q27c/s (+12 %); new 2 feeders 15.40. 4 h production on seeds 942201/942202: 439,632 Q29, 847 d_dx>=32, 9 d_dx>=48, best 50, 0 pairs (hashlib on all 1,298 records).