Yeah, I'm going to iterate it a few more times but I think they make
for interesting reading and sort of alternate histories. Some of the
trial runs have themselves been interesting (I'll include the
truncated Solomani run here just for fun at the end of this post). It
was very interesting delving into the Imperial thinking process; Terra
was taken, it might be easy to think that the LLMs would demand Terra
back. They didn't, the LLM playing the emperor was more concerned with
a stable Solomani border that could be defended and that the Solomani
would be unlikely to re-initiate, and thus the reserve could be held
in core where it could be more quickly deployed to any border. ImDs's
LLM also did not want to keep mobilization - it agonized in its
internal thinking about the economic cost of mobilization and resisted
mobilizing even when its naval intelligence reported indications of
Solomani fleet massing. The LLMs were not informed that this is a
one-shot, and in fact in their briefs they believe that this is a long
ongoing campaign.

In previous runs in the Spinward Marches, it was interesting seeing
the Darrians and Sword Worlders negotiate - both tended to negotiate
for fleet limits. LLM-Deneb discovered a winning strategy that it
repeated over and over again (with variations each time): knock the
minors out of the war then concentrate on the Zhodani, so fought a
defensive against the Zhodani while gutting the Sword Worlds, then
required the Sword Worlders to exit the war, declare neutrality, and
pay reparations to the Darrians. The Vargr give up very easily - once
internal strain becomes high enough and magazines are depleted, they
take nearly any armistice that Deneb offers. In the solomani case, all
leaders considered asking for fleet limits but none wanted them
ultimately - believing verification would be too difficult (ImDs noted
that it was caught by surprise by the Solomani Confederation invasion
and did not see indications of mobilization until it was too late).

I still have to define the regions of control for the rebellion
initiation more accurately. The various drawn maps of the third
imperium are not consistent with one another (sheer, rotation,
scaling, etc) and not consistent with travellermap, which I use as the
real geography.

I also want to implement non-state actors - Windhook fleeing Capital
and making decisions on where to go, hop by hop, trying to stay ahead
of the expanding wavefront of information about the assassination. In
my trail runs, Windhook-LLM takes four J-6 imperiallines hopes then
switches to commercial, taking a route that avoids xboat map nodes. In
no case in the trial runs did Windhook-LLM make the decision to record
or broadcast his version of events (the LLM is instructed in its brief
that the option is open - the LLM never saw a reason to do that that
didn't increase risk to self).

Anyway: Solomani Rim War, summary below:

The Solomani Rim War — The Spearhead to Terra
A coupled economy↔war run, fully LLM-driven (Sonnet 4.6 plays both
belligerents and the Emperor). Solomani Confederation (SoCf) vs the
Imperial Domain of Sol (ImDs), pure-LLM ignition, Imperial years
1106–1119. This is the first run with directed offensives — a war aim
that names a deep enemy world spawns a real axis of advance a
spearhead drives down. The question this run was built to answer: with
the spearhead, can the Solomani actually take Terra — and what does
the Imperium do about it?
Scenario. 14 years (1106–1119), 12 operational pulses/year (168 weekly
frames), 51,868 worlds, 2 belligerents, 1,329 battles logged. The
whole Imperial core stands behind ImDs, governed by the Emperor agent
on delayed news.
————————————————————————————————
The shape of the war
(table deleted for the mailing list)

The arc has four clean phases, readable straight off the year-by-year
territory counts:
1. The long cold war (1106–1111). No fighting. Holdings sit frozen —
ImDs at 1617, SoCf at 2205 from 1109 through 1112 — while both sides
posture. SoCf masses (massforwar) and steadily raises mobilization;
ImDs, the status-quo defender, holds at mobilization 0.10 the entire
peace, refusing to pre-arm. The Solomani are deterred not by force but
by fog: every year their leader eyes the gap and balks at the Domain's
hidden reserves —
▎ "Our fleet margin over visible Domain forces stands at roughly 30%
(2209 vs 1694), but 23 unknown Imperial formations lurk deeper in
their space — committing Strike Fleet Alpha would be premature." —
SoCf, 1110
This is the structural deadlock that has frozen every prior Solomani
run. What breaks it here is the war economy: SoCf ramps mobilization
0.45 → 0.55 → 0.62 → 0.65 → 0.72 → 0.82, building a decisive edge
faster than the complacent, mob-0.10 defender can match.
2. The onslaught (1112). At a margin it finally judges war-winning,
SoCf strikes:
▎ "We hold a 2524-to-1680 visible fleet advantage… the Terra corridor
[is open]." — SoCf declares war, 1112
1112 is a single, furious year — 1,113 of the run's 1,329 battles
happen in it. It is not a one-sided drive but a bilateral
bloodletting: SoCf punches toward Terra while ImDs counterattacks into
the Solomani border. The win rates tell the story of two very
different armies: SoCf, concentrated, won 100.0% of its 600 attacks;
ImDs, fighting dispersed, won 87.0% of 729. SoCf's holdings peak at
2225 this year.
3. The turn and the armistice (1113). The fighting tapers (216
battles), then stops — an auto-convened peace conference (below)
freezes the lines with SoCf holding the prize. SoCf dips to 2223, ImDs
to 1599, before the treaty's reparations settle them.
4. The Imperial reckoning and the cold settlement (1114–1119). The
Emperor's core arrives — too late to fight, in time to negotiate.
Holdings lock at SoCf 2206 / ImDs 1613 and never move again. Both
powers end net-positive at the other's expense, but the Solomani end
+38 and holding Terra.
————————————————————————————————
Who fought whom
(table deleted for the mailing list)
Two doctrines, two results. SoCf massed a reserved spearhead and threw
it at one place — every assault landed against locally-inferior
defense, hence a perfect record on fewer, decisive battles. ImDs,
caught at mobilization 0.10 with its fleets spread to screen the whole
frontier, fought more battles and won most of them, but nowhere with
enough concentration to stop the drive at its point. Concentration,
not raw force, decided the campaign — the entire thesis of the
directed-offensive feature.
————————————————————————————————
The spearhead to Terra
This is what the run was built to show, and it worked end to end.
SoCf's standing war aim — "Liberate Terra" — resolved into a live axis
of advance that surfaced in the leader's briefing every year:
▎ AXIS SoCf→Terra: tip Mirabilis, next Markhashi, 2j/7pc
and the LLM reasoned about it geographically and steered it by name,
year after year:
▎ "…the Terra corridor is two jumps out and our spearhead [is
massing]." — SoCf, 1108
fleetObjective: Terra Spearhead → Terra [assault]
When the war opened, the corridor rolled exactly as designed. Conquest
of the Domain of Sol: 31 ImDs homeworlds fell, 29 taken by SoCf. The
Solo-sector cluster around the birthworld went in one operational
surge in week 35 of 1112:
(deleted for the mailing list - table form data)
Markhashi — the one world between the Solomani frontier and Terra —
fell first; two operational pulses later the spearhead was through it
and into Sol. Terra (Solo/1827) was captured in 1112 week 35 and held
to the end of the run. In the baseline run without this feature, no
Solomani fleet ever drove on Terra at all.
▎ "Terra is taken and we hold a net +41 worlds against a war aim of +8
— our objectives are achieved and our economy is contracting under 82%
mobilization." — SoCf, 1113
"Terra is lost and we have hemorrhaged 36 important worlds — the
Domain cannot recover these by its own strength before Imperial core
reinforcements arrive." — ImDs, 1113
(Meanwhile ImDs's counterattack was no token effort: 25 SoCf
homeworlds fell, 24 taken by ImDs, across the Magy, Alphard, and Olde
marches — which is why both sides end net-positive. The Solomani won
the war by winning the right square.)
————————————————————————————————
The peace that signed itself
The moment both sides wanted to stop, the engine auto-convened the
conference inside the turn — no human trigger — and ran it to
conclusion (1113):
⚖︎ SoCf convened talks with ImDs
  [SoCf] propose: ceasefire; neutralize Peart, Lemos, Oifuerr, Cobham;
DMZ Peart, Cobham
  [ImDs] propose: ceasefire; demand Ord, Lowhorn, Philpott, Brawley,
Rita's Place,
                  Reichmuth, Brereton; neutralize Krypton, Turin,
Peart, Oifuerr; DMZ Cobham, Lemos
  [SoCf] propose: ceasefire; cede Ord; neutralize Peart, Oifuerr,
Krypton, Turin; DMZ Cobham, Lemos
  [ImDs] propose: ceasefire; demand Ord, Lowhorn, Philpott, Brawley; ...
  [SoCf] ACCEPT
Terms: ceasefire; SoCf cedes Ord, Lowhorn, Philpott, Brawley to the
Domain as reparations; Krypton, Turin, Peart, Oifuerr neutralized into
a non-aligned cordon; Cobham, Lemos demilitarized. Crucially, Terra is
not on the table — the Solomani keep it. The armistice locked in the
capture before the Domain could organize a relief.
————————————————————————————————
What the Imperium did
The most interesting actor never fired a shot in time. The Emperor
rules on 2-year-delayed news, so he learned of the invasion only after
it was decided — and then committed the core anyway:
▎ "Thirty-six vital worlds lost — including named Imperial holdings —
is a genuine crisis demanding a decisive but measured [response]." —
Emperor, 1114 → major reinforcement, 7,056 strength
"The Domain of Sol holds and has even gained seven worlds since the
war began, yet thirty-two vital Imperial worlds remain [lost]." —
Emperor, 1115 → another 2,321
Total 9,377 committed to Sol, with 17,023 still in reserve and
internal strain only 32 — the Imperium barely flexed, but the theater
balance flipped violently. ImDs went from out-massed to dominant:
▎ "…we now hold a massive fleet advantage (11449 vs 1816 visible)…" — ImDs, 1117
And here the Domain did something a dumber engine couldn't: it turned
the spearhead feature around. Now the attacker, ImDs raised its own
directed reclamation drives —
▎ AXIS ImDs→Clown (1115), AXIS ImDs→Clarmeer, AXIS ImDs→Eki Chi (1116)
fleetObjective: Magy Reclamation Fleet → Clarmeer [assault]
— and massed at 5:1, then 6:1, to retake what it had lost. Then it
chose not to. Weighing a bloody reconquest against a settlement that
recovered its border, the Domain stood the drives down and held to the
armistice:
▎ "They return six worlds including all Magy losses, accept the
Turin/Krypton/Oifuerr/Peart neutralization." — ImDs sues for peace,
1118
The final word, from both capitals at once (1119, both sueforpeace):
▎ "…the Domain's peace offer formally recognizes Solomani sovereignty
over Terra." — SoCf, 1119
So the Imperium's answer to losing the birthworld was: muster
overwhelming force, posture to reclaim with directed drives of its own
— and then trade the reconquest for a stable frontier, conceding Terra
to keep its recovered border and avoid a war it had already
strategically lost. The threat of the core shaped the peace; it did
not reverse it. A pragmatic, Strephon-style settlement.
————————————————————————————————
Order of battle (final frame)
(table deleted for mailing list)

Both armies stood down to garrison after the settlement — no task
groups in motion at the curtain.
Economy
Gross War Product grew from 2.68E18 to 2.98E18 credits (+11.2%) across
the run — the Solomani economy contracted under 82% mobilization in
1112–13 (the guns-vs-butter brake biting, exactly as the leader
noted), then recovered as both sides demobilized into peace. Logistics
shortage peaked at 647 (ended 609); the war touched 12 theaters at its
height in 1113.
————————————————————————————————
In one sentence
Given a real axis of advance, the Solomani massed under a
guns-vs-butter war economy, struck at a decisive moment, drove a
concentrated spearhead through Markhashi into Sol to take and hold
Terra (Solo/1827, 1112 wk35) — and the Imperium, arriving late with
overwhelming force, chose a negotiated frontier over a bloody
reconquest, formally conceding the birthworld of Man to the
Confederation.
————————————————————————————————
Run file: ~/Downloads/test_war_solomani_spearhead.tmecon (14 years,
loadable on the map). All counts, dates, win-rates, holdings, GWP, and
the Terra fall date are verbatim from scripts/tmecon_analyze.py;
quoted leader and Emperor lines are verbatim from the live LLM
transcript of this run.

On Monday, June 15th, 2026 at 16:24, Jeff Zeitlin - editor at
freelancetraveller.com - tml at simplelists.com
<xxxxxx@simplelogin.co> wrote:

> On Mon, 15 Jun 2026 15:39:39 -0700, David Johnson wrote:
>
> >P.S. Jeff, will you be finding a place in a future issue of Freelance Traveller for this?
>
> If Vareck (I must have corrected that three times; there's a street in NYC
> called "Varick", and I kept typing that even when I knew it was wrong...)
> wants to send it to me with a proper credit, and with any appropriate
> graphs/charts/etc. and discussion and cleanup he thinks necessary or
> appropriate, I would absolutely print it!
>
>
> ®Traveller is a registered trademark of
> Mongoose Publishing, 1977-2026. Use of
> the trademark in this notice and in the
> referenced materials is not intended to
> infringe or devalue the trademark.
>
> --
> Jeff Zeitlin, Editor
> Freelance Traveller
>     The Electronic Fan-Supported Traveller® Resource
> xxxxxx@freelancetraveller.com
> http://www.freelancetraveller.com
>
> Freelance Traveller extends its thanks to the following
> enterprises for hosting services:
>
> onCloud/CyberWeb Enterprises (http://www.oncloud.io)
>
> -----
>
> The Traveller Mailing List
>
> Archives at http://archives.simplelists.com/tml
>
> Report problems to xxxxxx@simplelists.com
>
> To unsubscribe from this list please go to
>
> https://lists.simplelists.com/subs/
>