GOTM9 - The Battle before Edo

GOTM9 Index



The Japan Campaign
- Introduction
- Planning
- Military orders
- Initial engagements
- Far off landing
- Coastal thrust
- Main advance
- Inland sea
- Consolidate the opening
- Secure the horses
- Western port
- Moving inland
- Battle before Edo
- Combat tables
- Yokohama and Nara
- Osaka
- Kyoto’s fate
- The gambit
- The battle
- Combat tables
- Mopping up


List of updates to this article


Items below this point
are maintained seperately
from the Japan Campaign files
and may not always available.

Known Bugs and Glitches

- The Corona Bug
- The Scared2Death Bug


The pivotal battles of the Japan Campaign were fought in the forests and plains southeast of Toyama and just north of the Japanese city of Edo. We could not know exactly where Edo was located due to our lack of a complete territory map or world map of the Japanese territory but this lack of a precise city location only served to reinforce that tactics that were used to draw the Japanese forces into this battle outside of the fortified defenses of their cities.

The straw that brought Tokugawa to battle
Our advance into Japanese territory included road crews of workers to extend our road network and allow our forces to advance as rapidly as possible. These road crews had now been supplemented with captured slave workers courtesy of Tokugawa’s excess settlers and workers who had been in the captured front line cities.

In retrospective, these slave workers served as additional bait that must have combined with any military necessity that Tokugawa may have felt in needing to counterattack our forces before they could reach his major city.

At the end of 250AD, the units of the main advancing force were consolidating and healing near and in the captured towns of Nagoya and Toyama. Reconnaissance probes by war chariots allowed us to gain some extra visibility of the road network and terrain that lay in front of our troops to the south of Nagoya (click here to view the recon moves near Nagoya) and between Toyama and the final objective of Kyoto. Technically, both of these routes would be about the same distance in order to reach Kyoto but the key difference would lie in the type of terrain that we encountered as well as how well the Japanese may have already connected their road network.

A more detailed version of this animation is not available
Reconnaissance and Advancing Units east of Toyama in 250 AD

I continued to push forward more war chariot units that had come up from the reserves at full strength and engaged in a few minor skirmishes that killed a Japanese warrior and a horseman without suffering any Egyptian casualties. The advancing units along with the tempting groups of workers combined with Tokugawa’s growing anger and frustration to lead him to make a fatal mistake.

Click here to download the 250AD ending save fileIf you would like to pick up and play your own moves forward from this scenario, you may click here to download a Civ3 V1.21 save file that will position you right at the end of the year 250AD. Remember: Your replay results will vary slightly from the original game because the impact of the Random Number Generator will alter the play sequence a little bit more for each movement sequence that progresses.

Tokugawa’s counterattack in 260 AD
Click on this image for an animation of the attack sequence for 260ADEmerging out of the dark fog from the direction we suspected as hiding our next major Japanese city, Tokugawa hurled forward a stack of five (5) veteran horsemen plus one (1) archer to confront our advance guard of four (4) veteran War Chariots accompanied by a veteran swordsman and spearman. The tactical mistake made by Tokugawa was to launch these units into the forest and stop immediately in front of and in easy reach of our offensive units. Tokugawa’s position was also just within reach of additional war chariots that could launch out of Toyama and cross the nearby river using our freshly completed road connections. In total, we had the capacity to bring over eight offensive units to bear directly on Tokugawa’s advancing force of six (6) units. In addition to unknowingly giving up numerical superiority, Tokugawa had sacrificed his offensive strengths and road mobility in exchange for the slight defensive bonus of the forest. Now we would be attacking his units with a 3.00 or a 2.00 attack power versus a defensive strength of 1.25 for the Japanese units.

Even if our units failed to totally destroy the Japanese counterattack, our advance guard included stronger defensive units that could fortify and defend in the forest at a strength of 2.625 (2.00 * 1.25 * 1.25) versus the best Japanese attackers which could only muster an attack strength of 2.00.

(click here to view a combat results table from this segment of the battle)

Click on this image for an animation of the attack sequence for 270ADThe “Warrior Flutter” in 270 AD
More Japanese units can be seen to run forward in an attempt to distract our forces from the primary mission of getting into position to capture Edo. In this turn, three Japanese warriors advanced in the open terrain on either side of out strong defensive position overlooking the town. The chosen moves by the Japanese forces were clearly aimed at trying to reach the road crew workers and the wounded Egyptian units that are recovering just out of the Japanese reach.

(click here to view a combat results table from this segment of the battle)

Click here to download the 270AD ending save fileIf you would like to pick up the game at from this point in time, you may click here to download a Civ3 V1.21 save file that will position you right at the end of the year 270AD. Remember: Your replay results will vary slightly from the original game because the impact of the Random Number Generator will alter the play sequence a little bit more for each movement sequence that progresses.

Tokugawa’s 2nd Major counterattack in 280 AD
After two rounds of attacking and positioning to deal with the Japanese units and prepare for Edo, a veteran archer emerged from Edo city and attacked out defensive position directly. Our strongest spearman just barely won the battle killing the attacking archer. Then we could watch with some growing anxiety as six more Japanese units emerge from the fog and reached forward toward striking our road crew and our wounded reserves. This turn is truly the climatic moment in the whole Japanese campaign because we have drawn the major strength of the Japanese military out into the open where we can potentially destroy Tokugawa’s ability to defend his empire for all time.

The critical decision process here has to do with choosing to deal with the counterattack first or choosing to seize the town of Edo and then brace for dealing with the Japanese units in territory that would now be in our possession.

I counted the full strength Egyptian units and determined that we had five full strength attacking units that could probably capture Edo and we had five wounded units in the area as well. Our strong units might have been able to capture the town but that would have left us exposed in the next turn to the seven offensive Japanese units that we could currently see plus any further offensive units that might be advance from further back in the fog.

At this point, the Egyptian units were still really hampered by continued problems with crossing the network of rivers that lie between our resupply point back in Toyama and our roads were still almost worthless. We were still three turns away from having the Engineering technology that would free our units up to move easily between these two positions. The lack of mobility combined with the large number of Japanese units in the counter attack meant that I felt the Egyptian forces had to deal with the counter attackers first and then use further reinforcements to consolidate and capture Edo in the next turn.

From our central defensive position we could see Japanese units spread out in four locations all around us. Four of the seven Japanese units would be real offensive threats if we left them undestroyed while only one of the Japanese units (the single elite warrior to the southwest) could be classified as no real threat.

In addition to the assessment of threat, we could see that the Japanese units were stacked in twos in three of the four locations immediately adjacent to our positions. These stacks of two were not as advantageous as if Tokugawa had placed all the units in a single stack, but each stack meant that we could attack and win one battle with a one movement unit like our swordsmen without having those units advance into an exposed position. This observation is very important because it meant that each of the three stacks of two Japanese units had to be attacked with a swordsman first in order top try and keep our units consolidated as much as possible. The last remaining Japanese unit in each stack could then be destroyed with War Chariots so that the victorious War Chariot could then consolidate back onto our defensive position. I we had used the units to attack in a different order, then we probably would have ended the turn with units spread out on three or four different locations instead of just the two strong defensive positions we actually ended with.

(click here to view a combat results table from this segment of the battle)

At the end of 280 AD when Tokugawa moved his forces, more Japanese units rushed forward into the open areas to counterattack our forces. The significant change in this set of moves was that the Japanese units were now mostly defensive units instead of the offensive type units we had seen in previous turns. I interpreted this intelligence to mean that the current offensive power of the Japanese military might have been destroyed by the way we had dealt with the advancing combat up to this point. Our Egyptian forces had destroyed a total of eleven (11) Japanese horsemen and seven (7) Japanese archers with only three (3) Egyptian casualties because we had engaged most of these Japanese units in open terrain and by using the strength of our offensive units. A key point to observe in this process is the fact that, I tried to never have offensive units like horseman, archers, and war chariots just fortify and stand in a defensive position. I either used them to attack and then hide under stronger defensive units, or tried to stay out of reach of or run away from any potential attackers.

Click here to download the 280AD ending save fileIf you would like to play along with this scenario, you may click here to download a Civ3 V1.21 save file that will position you right at the end of the year 280AD. Remember: Your replay results will vary slightly from the original game because the impact of the Random Number Generator will alter the play sequence a little bit more for each movement sequence that progresses.

The Capture of Edo in 290 AD
Now was the time to capture Edo and then fortify in position while me mopped up as many of the Japanese units as possible.

Our strong units on the defensive point, where we could attack directly into Edo, included five full strength war chariots (4 regular; 1 elite) plus two elite swordsmen that were slightly wounded but still viable attack units. We also had seven other units nearby that could defend the captures town and supplement the garrison. By observing the units come and go from Edo, I felt that the town was only defended by at least two spearmen (one veteran and probably one regular). Our seven strike units, mustering at total of at least 19 strokes, could easily handle two or three defenders with 7 to 10 hit points of defense.

The first unit to attack Edo was our 4/5 elite swordsman who literally massacred the veteran spearman defender. This combat sequence should have been a closer exchange of blows with the swordsman at 3 strength attacking a fortified defender at 2.75 strength, but the sequence of strokes was totally won by the swordsman and then his victory produced the Great Leader “Ramses”. Our next attack used the elite war chariot to defeat the regular spearman that rose as the second defender. A slight surprise forced us to attack and defeat an archer to finally capture the town.

Click on this image for an animation of the attack sequence for 290AD(click here to view a combat results table from this segment of the battle)

Capturing the key town Edo and gaining a great leader all in one turn, made this year 290 AD a truly memorable event in the history of our Egyptian civilization. To make things even more fantastic, Edo contained the additional prize of an intact barracks that would speed the healing rate for our forces and allow use to continue the forward advance into the Japanese capital of Kyoto at an even faster rate.

The consolidating moves included advancing every possible wounded unit into the town in order to take advantage of the healing power of the barracks. I also used advancing horsemen to clear the supply route between Toyama and Edo of the stray Japanese warrior who was menacing our rear units. I bypassed the Japanese spearmen who had become trapped in our territory because they could easily be destroyed in the next turn with no additional risks, if they decided not to commit suicide by attacking our defensive positions. Either option would guarantee that they would die in the next few turns.

In retrospect, I was in a bit too much of a hurry to use the great leader so that I could turn “Ramses” into an army and quickly get a victorious army all in one turn. I wanted to get the victorious army to enable our ability to build the Heroic Epic back home in one of our major cities. Completing the Heroic Epic would increase our chances of getting a second great leader before the war with Japan might come to an end. Strategically it was the right move to build an army immediately because that choice eliminated the hard coded programming block that prevents a second great leader from being produced whenever a first great leader already exists. My mistake was in loading a war chariot into the army to give it the victory chance. Since the war chariot had already partially moved it was not possible to advance that new army out of Edo and gain an immediate victory all in the same turn. Since I had to wait until the next turn to look for a target for the army, I could have waited one more turn and loaded the army with a horseman instead of the war chariot. The reason the war chariot in the army was a mistake, would not be immediately obvious, but later in the game when the army would be fully loaded with two knights in addition to the war chariot, I would come to find that the army was severely handicapped by the restrictions that prevent war chariots from moving onto mountains or through jungles without roads.

Notes on Great Leader production rate
”Ramses” the great leader, was produced by only our fifth (5th) elite combat victory of this campaign. Because the Egyptians are a Religious/Industrious civilization, I was playing a strategy that built few barracks and produced very few veteran units straight out of the box. Most of our early veteran or elite units were promoted from regular units that had survived many combat events with barbarians. Our victorious swordsman that generated the leader had just been promoted to elite three turns earlier in 260 AD.

I do not know if the events are related, but the massive barbarian uprisings that occur worldwide had just been generated in the year 230 AD as a result of the second civilization entering the middle ages. For the Egyptians, this uprising included two swarms of eight horsemen each that were all defeated by a combination of defensive victories by three different units that earned promotions to elite status and then continued to win elite victory after elite victory just prior to the attack on Edo. Probably just a coincidence, but interesting to note.

The next major sequence of moves

Battles in the Lull following the Egyptian Victory at Edo

Other related topics:
Cracker's unofficial CIV3 skills toolbox - some favorite tips and tricks
Discuss this campaign and/or ask questions in the Forums
- hosted by CivFanatics
Support information - copyrights and usage, link to this site, how images were prepared

GOTM9 Index



The Japan Campaign
- Introduction
- Planning
- Military orders
- Initial engagements
- Far off landing
- Coastal thrust
- Main advance
- Inland sea
- Consolidate the opening
- Secure the horses
- Western port
- Moving inland
- Battle before Edo
- Combat tables
- Yokohama and Nara
- Osaka
- Kyoto’s fate
- The gambit
- The battle
- Combat tables
- Mopping up


List of updates to this article


Items below this point
are maintained seperately
from the Japan Campaign files
and may not always available.

Known Bugs and Glitches

- The Corona Bug
- The Scared2Death Bug


© 2002 by cracker@CivFanatics.com
All rights reserved. Please read the Terms of Use.
Send mail to our webmaster with technical questions about or potential problems with this web site.

This page was last updated on: August 10, 2002