Congratulations, pilot, on making it to the third Flying Solo article. In the first article we took an initial look at the new Solo Play Open Alpha rules, focusing on immediate impressions and the movement mechanics. In the second article we dove into how actions work and played two test games – one against the new Solo AI, and one against the older fan-made Heroes of the Aturi Cluster (HotAC) AI. Now, with the third article, we reach the logical final stage of test-play: how would two totally symmetrical forces, one flown by the Solo AI and one flown by the HotAC AI, fare against each other?
Keep in mind that this savagery is not for the faint of heart. Sympathizers of droid-rights advocate L3-37 are advised to turn away now, and highly encouraged to send vociferously indignant emails to contact@goonhammer.com.
Game 3: Solo vs. HotAC
To offset the significant advantages and disadvantages of “First Player” in this otherwise symmetrical match, we flipped the first player token each round. This went surprisingly smoothly, especially since we were tracking each turn and knew which AI went on odd/even turns.
We deployed the initial TIE Fighters symmetrically, but reinforcements arrived randomly. Each turn we tracked the following metrics:
- Shots Fired: If a ship had the opportunity to shoot an enemy, it was counted as one shot regardless of how many dice were rolled in attack/defense. This stat is more about ascertaining intelligent maneuvering than anything else.
- Damage Inflicted: This is purely damage caused by comparing red dice/green dice. Overkill was counted (if a ship with one hull remaining took two damage from a shot, it was counted as two damage inflicted) but damage from critical hit cards or obstacle collisions was not counted.
- Obstacle Collisions: How many times a ship overlapped an obstacle, regardless of whether or not the overlap causes any damage. If a HotAC ship overlapped an obstacle with its maneuver but then swerved to avoid it successfully, it was not counted as a collision.
- Board Edge Collisions: While the consequences of leaving the table are more dire for HotAC ships, they’re still not pleasant for Solo ships. We counted each time a ship would leave the table – as with obstacle collisions, HotAC “near misses” weren’t counted.
- Target Locks: A big concern after reading the Solo AI was that omitting the lock action from two attitudes – and burying it in the third – would make munitions very hard to get off. Each TIE Bomber had the number of times it acquired a target lock tracked.
- Munitions Fired: Acquiring locks is one thing, but successfully firing your valuable munitions is another. We counted how many times each TIE Bomber was able to launch a proton torpedo.
The following photos are intended less as a comprehensive battle report and more to give an overall feel for the battle. After the dogfight photos conclude, we’ll review the outcomes of the metrics described above.
So there you have it – the battle concluded. The entire thing lasted 22 turns to complete annihilation of one team, with Red HotAC being the victor over Blue Solo. This is even with HotAC losing that full-health fighter early on when some extremely bad luck flew it right off the board edge. Take a look below for a more detailed breakdown of the numbers.
Post-Game Numbers
As promised, here are the metrics that we said we’d be tracking.
- Shots Fired: Games one and two seemed to indicate that the HotAC AI was better at maneuvering than the Solo AI, and this game three confirmed it. The Solo AI only got 11 shots off the entire game, while the HotAC AI more than tripled that score with 36 shots. This has nothing to do with one AI being “first player,” since we alternated each turn, and pretty solidly confirms that the HotAC AI is better at both homing in on a target and, possibly, using reposition actions to dodge arcs.
- Damage Inflicted: Not only did HotAC get more shots off (indicating smarter maneuvers and defensive action selection), but it also had significantly higher damage output from those shots (indicating smarter offensive action selection and better range control). 45% of shots fired by Solo AI caused damage, while 58% of shots fired by HotAC AI caused damage. While the total shot/damage disparity could be explained away by HotAC pulling an early lead and taking ships off the table, the percentages are more likely to be the result of genuine smarter AI. The Solo AI only caused 5 damage the entire game, and all of it was from their TIE Interceptor on the enemy Bomber. Adding insult to injury? The Bomber flew into a rock and died, so the AI didn’t even get the glory of the kill! In contrast the HotAC AI did 21 damage to enemy fighters, which is the sum of each side’s total hit points… meaning that they scored exactly enough damage to kill the entire Solo AI force without help from critical hit damage, asteroids, or other sneaky means (although some of that damage was overkill, and asteroids did indeed chip in to their victory).
- Obstacle Collisions: Oddly enough, even with a robust swerving mechanic, the HotAC AI actually had more asteroid collisions than the Solo AI. Solo collided with 5 asteroids, HotAC with 7. Admittedly, three of the HotAC collisions came from the same hapless TIE Fighter who flew through two rocks in one turn, landed squarely on the second, and then had to overlap it again in its next activation. So that was essentially a triple-whammy, all from one bad maneuver! The difference here is much smaller than other stats, though, especially when you consider that treating the unlucky HotAC triple-collision as a single one makes them tied.
- Board Edge Collisions: Solo ships hit the board edge 3 times, while HotAC hit it 1 time. Of course, the repercussions were much more severe for that one HotAC TIE Fighter – the full-health ship was removed from the table entirely.
- Target Locks: Solo Bomber got 2 locks over the course of the game while, the HotAC Bomber narrowly beat it out with 3 lock acquisitions.
- Munitions Fired: Solo Bomber got 0 torpedo shots off during the battle, while the HotAC Bomber was able to take 2 shots (after which it ran out of ammunition). In defense of the Solo Bomber, it finally was about to get a torpedo off at the enemy HotAC Interceptor, but the friendly Solo Interceptor initiative-killed it before the Bomber could pull the trigger. After that the Bomber got dejected and sadly flew off to the middle of nowhere to pull k-turn donuts until it was put out of its misery.
In summary, Heroes of the Aturi Cluster is the clear winner.
Final Recommendations for Solo
It’s been a lovely journey crafting these three articles, and I wanted to wrap up the series thus far with my verbatim feedback on the Fantasy Flight Games Solo Play Alpha Feedback form. It’s critically important that you, dear reader, also personally leave feedback for FFG. If you played – what did you think? If you didn’t play – why not, and what would they have to do to get you interested? Every response indicates interest in these rules, and increases the odds that Solo Play will be supported with more content. If you choose not to leave feedback, you are personally stabbing a knife into the back of future Solo Play content. It’s that simple.
It’s important to reiterate that while HotAC AI is extremely good, it is also extremely limited in the ships and upgrades it works for. If nothing else, the Solo AI is valuable as a universal AI that can fly any ship, upon release, straight out of the box, without having to wait weeks or months for fan-made content to catch up. It also allows for AI ships with unique upgrade builds that the HotAC AI can’t account for. For all those reasons, the creation of the Solo AI is an incredible asset to the community that we need to encourage, even as we celebrate the brilliance of HotAC.
Take a look at my comments below and feel free to copy-paste them into your feedback to FFG if you agree. Together, we can make the Solo Play ruleset as elegant and enjoyable as possible.
FEEDBACK FORM QUESTION:
What was your least favorite aspect of the solo game?
BUFFALOCHICKEN’S ANSWER:
As mentioned earlier, the lack of “distance to tally” as a factor in determining decisions. The maneuver dice roll should determine the BEARING (straight, bank, turn, advanced), and the distance to tally should determine the SPEED (slowest, average, fastest). This would be an elegant way to have smarter AI without clunky rules.
The action trees are also frustrating in that the AI often does sub-optimal actions. Get rid of the “balanced” attitude and have blank and eyeball results be “offensive” while an evade result remains “defensive,” adding coordinate as a defensive option and a target lock option at the very bottom (so the AI will never select it as the first action when defensive, but ships like TIE Bombers can still link into it after barrel rolling). Right now enemies really struggle to get target locks, but the changes above would fix that while making the enemy more exciting/aggressive.
AI ships hit asteroids too frequently. An easy fix would be to say “if a Solo ship’s maneuver would overlap an obstacle that it is not already at range 0 of, re-roll the dice. Use this second maneuver, even if it would also overlap an obstacle.” This is a super-quick and simple way to allow ships to “swerve” away from asteroids, and the closer they are to the asteroid the harder it will be to dodge it. Fun and thematic!
Finally, the document formatting was unwieldy to use. The final document needs to feature clear, condensed tables that are formatted for easy printing and cutting out. Ideally suitable tables will share an edge so that after printing, they can be cut and folder over to make double-sided cards. Card 1: Bullseye Arc (front) and Forward Arc (back) maneuver tables. Card 2: Side Arc (front) and Rear Arc (back) maneuver tables. Card 3: Offensive Attitude action selection (front) and Defensive Attitude action selection (back). If done properly, players would be able to play the entire game with just those three double-sided cards in front of them (instead of a 15-page document with no page numbers!).
Now that we’ve given Solo a chance to learn from HotAC, let’s hope that the next revision incorporates at least a few of these suggestions.
More to Come
This is not the end of the series – I plan on doing another article when the next version of Solo Play drops, so be sure to check back then. In the meantime, stay tuned… a very special article, guest-starring the creator of Heroes Of the Aturi Cluster himself, is coming up soon!
Flying Solo Part 4: An Interview with Josh Derksen
If you have any questions or feedback, drop us a note in the comments below or email us at contact@goonhammer.com.
