RogueForge

Redux

Waiting for better models

The restart moved at the speed of the models. Twice it stalled for weeks, and twice a new Claude model got it moving again.

The restart did not move at a steady pace. It went in bursts, and the quiet stretches between them were not holidays. Twice the work reached a point the model of the day could not get past, and twice it started again when a new Claude model arrived. The prompt history shows it plainly.

Rogue restoration activity by week, 2026 Prompts sent each week in the Rogue projects. Busy in March and early April, silent from mid-April to early June, busy again from June to August with a two-week pause in late June, silent from September to early November, then busy again in November. 0 400 800 1200 seven quiet weeks Fable away two quiet months Opus 4.6 Opus 4.8 and Fable 5 Opus 5.5 Week of 16 March: 565 prompts Week of 23 March: 1115 prompts Week of 30 March: 739 prompts Week of 6 April: 118 prompts Week of 13 April: 38 prompts Week of 20 April: 0 prompts Week of 27 April: 0 prompts Week of 4 May: 0 prompts Week of 11 May: 6 prompts Week of 18 May: 0 prompts Week of 25 May: 0 prompts Week of 1 June: 0 prompts Week of 8 June: 335 prompts Week of 15 June: 210 prompts Week of 22 June: 0 prompts Week of 29 June: 9 prompts Week of 6 July: 10 prompts Week of 13 July: 553 prompts Week of 20 July: 184 prompts Week of 27 July: 11 prompts Week of 3 August: 128 prompts Week of 10 August: 0 prompts Week of 17 August: 0 prompts Week of 24 August: 0 prompts Week of 31 August: 0 prompts Week of 7 September: 0 prompts Week of 14 September: 0 prompts Week of 21 September: 0 prompts Week of 28 September: 0 prompts Week of 5 October: 0 prompts Week of 12 October: 0 prompts Week of 19 October: 0 prompts Week of 26 October: 0 prompts Week of 2 November: 0 prompts Week of 9 November: 258 prompts Week of 16 November: 77 prompts 5.4 comes close 3.4 matches Linux ports Mar Apr May Jun Jul Aug Sep Oct Nov
Prompts sent each week in the Rogue projects, 2026. Shaded weeks are the pauses; the line above names the Claude models in use, as recorded in commits; diamonds mark matches.
Show the numbers
Week ofPrompts
16 Mar565
23 Mar1115
30 Mar739
6 Apr118
13 Apr38
11 May6
8 Jun335
15 Jun210
29 Jun9
6 Jul10
13 Jul553
20 Jul184
27 Jul11
3 Aug128
9 Nov258
16 Nov77

Spring: close, and stuck

The first push, in March, ran on Claude Opus 4.6 with the effort setting at its maximum. It went fast at first. In a fortnight every version was disassembled and decompiled, Rogue 3.4 was read through function by function, and a rebuild of Rogue 5.4's object file was under way. On 2 April that rebuild came within two small patches of matching.

Then it stopped converging. The last differences in 5.4 resisted everything, and the model kept concluding that what was left could not be fixed, or was close enough. The person running the work kept insisting it could be, and that "close enough" was not the standard. By the end of March it was stuck all the same. In mid-April the effort setting was turned down, turned back up, and then the work stopped. For seven weeks there was almost nothing.

June: a new model, and then the right one

Work started again on 9 June with a different approach: instead of reading Rogue 3.4 harder, build the 2.8BSD compiler and prove each function by compiling it. Claude Opus 4.8 appears in the commits the next day, and within two days the C library, curses and termcap libraries all matched the binary exactly.

On 12 June Claude Fable 5 came out, and the problems that had been set aside were handed to it the same day. The next week was the breakthrough of the whole project: the remaining differences in Rogue 3.4 were worked through one after another, with the effort setting raised again when a problem needed it.

About a week later Fable was taken away for a while, and the work waited for it. For two weeks at the end of June there are almost no prompts at all.

July: everything falls

When it came back, things fell quickly. Rogue 3.4 matched byte for byte on 3 July. In the next three weeks so did Rogue 3.6 on both machines, 4.22, 5.3 and 5.2, and the toolchains and simulators grew into the general-purpose tools they are now.

November: the ports

Then another quiet stretch, from September to November. The Linux ports had started, and the Rogue 5.2 port had been stuck for a while. Claude Fable 5.2 was first used on 11 November and Claude Opus 5.5 on 14 November, and in the week that followed all six Berkeley versions were ported, with more than five hundred separate bug fixes between them, and PC Rogue after them.

What still waits

Two spots in Rogue 5.4 are still patched by hand after compiling: the exit from do_move() and six bytes of register choice in command(). The likely cause is a compiler build slightly different from the one reconstructed. If the pattern holds, they are waiting for the next model.