So, is Showdown fun?

[Eternaly_Lost]

Assuming no other Showdown modifiers, I'd say it has to be in line with the Exogalactic Transciever event at the end of FS. If it's not, then doing that one just makes more sense considering you can more easily bottleneck it (due to not having to defend Showdown devices in out of the way places).

I think it's fine if doing it the FS way is somewhat easier, since that has way more prereqs than the showdown devices, but sure.  But how much heavier is the transceiver event than a no-superweapons showdown?

I mean, I can make the exo numbers line up to some extent, but there's a lot of things in a showdown that don't compare as readily numerically.

I personally would rather see the Showdown to be fairly close to the EXO event. To leave it as basically a third option. You can ignore the Exo event, you can do it, or you can do the show down. Or if you feeling insane, do both at the same time.

However, right now to do the Exo event, you basically need to take at least one of the homeworlds, I guess you really could just leave it on alert the whole time, but very rarely if I have enough power there to get the Shard home, do I not have enough power to go, clear up that world and stick a warpjammer on it, then bring the shard home.

Remember, Lore wise, the AI is already going to throw everything and the kitchen sink at you when you do the EXO event, because it knows that either it knocks you out, or a force will show up that will knock it out. (Outside of crashes due to running of out memory)

The Showdown device is the exact same thing in a different way. Rather then an unstopable force showing up to take it out, it is completely cut off from reinforcements, meaning that eventually,  even the weakest force that can hurt it will win out.

And if it can't take at least one of the showdown devices before then, it knows for certain that such a force already exists there.

I could see it being a little harder or a little easier on the showdown, but I feel it should be roughly in the same danger.

[Chthon]

Um... The exo event doesn't require you to even leave the AI homeworld on alert.  Nothing suggests that you even need to take the world you survey, just protect the survey ship and the shard all the way back.  Tough I know, but its possible and probably easier.

Heck, hack the AI sensors for better effect.

[Toranth]

Assuming no other Showdown modifiers, I'd say it has to be in line with the Exogalactic Transciever event at the end of FS. If it's not, then doing that one just makes more sense considering you can more easily bottleneck it (due to not having to defend Showdown devices in out of the way places).

I think it's fine if doing it the FS way is somewhat easier, since that has way more prereqs than the showdown devices, but sure.  But how much heavier is the transceiver event than a no-superweapons showdown?

Since you asked.  I played a game up to Hour 15 or so, then did the Fallen Spire campaign, then branching from that late-game save right before starting the FS, I did the Showdown instead.  The FS campaign went quickly, but the SD required significantly more preparation.  Games ended at 24:45 and 30:45 respectively.
Settings were 7/7 Random Easier/Random Easier for both AIs (Neinzul Viral Enthusiast/Turtle for AI 1, and Vanilla/Grav Driller for AI 2), FS 4, Dyson Sphere 4.  No other plots or modifiers were in effect.

Here are the totals I encountered for the Fallen Spire Transceiver event vs the Showdown.

The Showdown events

Type	Unit Count	Exo Cost	Strength
Exos	2,029	59,872	23,551.4
Waves	10,882	N/A	88,927.96
CPA 1	7,500
CPA 2	1,500
CPA 3	900

My estimate of the CPA strength is about the same as the Waves - roughly 10,000 ships of Mark II+, so another 90,000 to 100,000 strength.

The first CPA actually broke one of my chokepoints (Mil-III + 5 of 7 Core Turret sets, plus 2 Fort Is, 2 Fort IIs, and a Fort III + 25% of my fleet).  Then, of course, the GCSs each broke another chokepoint deeper into my territory, but defense in depth (yay Core Turrets and beachheading) delayed and won the day handily.


Fallen Spire Exos

	Unit Count	Exo Cost	Strength
Total @ Trans Complete	8,451	1,921,824	480,963.82
Total @ Win	12,955	2,443,168	613,249.09

I'm pretty sure a normal Timed Exowave went off during the Transceiver event, but I couldn't identify it clearly to remove it, sorry.  So that number is a little inflated.  There were also a normal wave or two, but they were insignificant.


Overall, it looks like the Showdown totalled about half the strength of the Transceiver events, not including the GCS or Defensive Exo.  However, the Showdown requires a distributed defense (at least two segments because of the GCSs), while the Transceiver can concentrate.  Certainly, I had an easier time with the FS than the SD, despite spending less time preparing.

[keith.lamothe]

Very interesting numbers, thank you for the effort (another couple multiverse instances laid waste for science).  A few questions:

1) You mention that the showdown numbers don't include the GCS's themselves; did that phase feel right strength-wise (since you don't have to do distributed-defense at that point)?

2) The distributed-defense phase of the showdown threw about half a transceiver-response's force at you.  Is it the overall weight that felt too high, or the spikes (specifically that first CPA) ?

[Toranth]

Very interesting numbers, thank you for the effort (another couple multiverse instances laid waste for science).  A few questions:

1) You mention that the showdown numbers don't include the GCS's themselves; did that phase feel right strength-wise (since you don't have to do distributed-defense at that point)?

2) The distributed-defense phase of the showdown threw about half a transceiver-response's force at you.  Is it the overall weight that felt too high, or the spikes (specifically that first CPA) ?

It was interesting to compare the two - plus, I hadn't done game with absolutely no super weapons at all or at Diff 7 in a long time.  I almost killed myself with some early recklessness.  I think I may need to play some more basic games to get back into the habit of being careful.

1)  For the post-Device part of the Showdown, the balance was fine in this game.  I knew what was coming, so I played a very "forward" defense.  I researched early what path the AI would take to my Homeworld, and then fortified heavily an adjacent-to-Core world on that path.  Then I skipped a system, then captured and fortified another system on the path, etc.  My idea was hit hard early, to give time to rebuild stuff later when the first blockade goes down, while using beachheaded Core turrets in every system on the path to whittle down the GCS.  If the AI HWs had been closer, or if I had been unable to pre-determine the path, I could have been in trouble.
A Mil-III, 2 Fort Is, 2 Fort IIs, 1 Fort III, a full cap of HBC I-IVs, turrets Needler I-III, Laser I-III, and V for the other ranged turrets plus a bunch of mines was roughly enough to kill a GCS and its escorts.  Actually, the GCS killed the Mil-III rendering the remaining Fort II and Fort III out of supply, so if I'd had an adjacent system to keep supply, all the stuff I listed would have been enough by itself.


2)  On the other hand, a Mil-III, 2 Fort Is, 2 Fort 2s, and 1 Fort 3 + a full set of ranged Core Turrets plus mines and a chunk of my fleet was crushed by the first CPA.  That was a huge spike, coming only 10 minutes into the Showdown.  By comparison, the minute 20 CPA and the final CPA were laughable.  Almost all the CPA was concentrated near one AI HW, which allowed it to overcome my blocker, but this also let me build a panic-nuke, cutting it by 90%.  Thankfully, the AI stopped to destroy every last turret in each system it passed through, which gave me time.
I suppose it mostly came down to distance and access-control.  I did a bunch of gate-raiding, and used that to prevent waves from going directly to the SD systems.  I had also hacked or captured 6 of the 8 Core Turrets, and used that to build an additional buffer line of beachhead Core Turret/MiniFort sets on every AI world I had supply on, to keep AI numbers low, and both slow and attrition the CPAs.  I still lost 2 systems to the CPA and another outlying isolated system to a Counterwave I couldn't defend against in time, but the HW and the SDs were easily kept safe, just because of the depth.

The Exos weren't a problem, because only one Exowave even had a Superweapon leader - a Mk I Siege Tower.  Only in the final Exo did a single Dire Guardian finally make an appearance.  Exos and waves also are suicide attacks - very few units ever retreat successfully, so there's little buildup to be concerned with.  The only real threat in the distributed defense phase was the first CPA.

It would be another story entirely without the Core Turrets and MiniForts, however.  Those are the heart-and-soul of distributed defense.


Overall, I was suprised at how easy the Showdown was.  I had to pay more attention to what was going on than I did with the Transceiver, because I couldn't hide behind a single super chokepoint, but without the multipliers from superweapons, everything was managable.

[keith.lamothe]

The details are helpful, thanks.  It does sound like perhaps the first CPA should be limited to roughly 1/3rd the AI's total available forces, and the second limited to 1/2 (of what's left).  Possibly even 1/4 and then 1/2 to make the second theoretically larger.

Those limits would only kick in when the CPA's "natural" size was high enough to start outstripping what's actually available.  I'm not sure how often that's not the case for the second CPA in a showdown.  The third one always just sends everything, and that's the one I'd expect to be just gg-large in a very hard scenario where the AI's had tons of reinforcements, etc.  Not sure if that actually happens, though, as the natural size of the CPA goes up with diff too.

The more pressing question, however, is: given the two scenarios you did ("play to point A, do FS to can-do-FS-endgame, do FS endgame", and "play to point A, do showdown endgame"), how much stronger _should_ the showdown forces be (compared to your second test) if you do "play to point A, do FS to can-do-FS-endgame, do showdown endgame"?

Sounds like right now it's like 10x harder or whatever, when really it should be maybe 2x or 3x.  Not sure how much the FS fleet and cities help, given that if the capital fleet wipes you probably won't have the econ to rebuild it, etc.  My guess is that the fleet helps a lot vs GSC by virtue of "how does it feel to get three thousand photon lances in the back right after a wormhole jump?", but it sounds like the GSC isn't the primary problem.

[chemical_art]

While not surprising, I'm not sure what to think that core turrets, by themselves, make or break showdowns, period. To the point games revolve around acquiring them. I can't think of no single thing as important aside from maybe gaming gates.

[Chthon]

While not surprising, I'm not sure what to think that core turrets, by themselves, make or break showdowns, period. To the point games revolve around acquiring them. I can't think of no single thing as important aside from maybe gaming gates.

Don't forget you can build permanent beachheads in systems without reducing your system's defenses so long as you have the power to spare.

[Toranth]

The details are helpful, thanks.  It does sound like perhaps the first CPA should be limited to roughly 1/3rd the AI's total available forces, and the second limited to 1/2 (of what's left).  Possibly even 1/4 and then 1/2 to make the second theoretically larger.

Those limits would only kick in when the CPA's "natural" size was high enough to start outstripping what's actually available.  I'm not sure how often that's not the case for the second CPA in a showdown.  The third one always just sends everything, and that's the one I'd expect to be just gg-large in a very hard scenario where the AI's had tons of reinforcements, etc.  Not sure if that actually happens, though, as the natural size of the CPA goes up with diff too.

Actually, the message "All AI ships launched" appeared for each of the CPAs, rather than the normal breakdown.  That may be the problem, if it is accidentally triggering the final response each time.  Let me check... yeah, I enabled Full Visibility and sure enough, all the guardposts throughout the universe are gone as soon as the first CPA triggers.

Getting a more balanced set of CPAs will certainly help avoid the huge-spike-o-Doom that can overwhelm even chokepoints, much less distributed defenses.  Unfortunately, when you reworked the Threatfleet to be an actual threat, CPAs became a lot nastier.  Under the old rules, you'd have a bunch of ships outside each human system, but no great danger from any one of them.  Now, after a few minutes, they'll regroup and come down like a hammer on something - likely a SD device world.  Unless you turn all AI ships into Zombies at the same time they release - which would be amusing - they'll gather again for the same net result.  Also, the reserve and SF chunks already spawn all together.
Distributed defense would be helped if there was a mechanic that forced the AI to actually distribute its attacks more evenly.

The more pressing question, however, is: given the two scenarios you did ("play to point A, do FS to can-do-FS-endgame, do FS endgame", and "play to point A, do showdown endgame"), how much stronger _should_ the showdown forces be (compared to your second test) if you do "play to point A, do FS to can-do-FS-endgame, do showdown endgame"?

Sounds like right now it's like 10x harder or whatever, when really it should be maybe 2x or 3x.  Not sure how much the FS fleet and cities help, given that if the capital fleet wipes you probably won't have the econ to rebuild it, etc.  My guess is that the fleet helps a lot vs GSC by virtue of "how does it feel to get three thousand photon lances in the back right after a wormhole jump?", but it sounds like the GSC isn't the primary problem.

How much harder should it be?  My guess would be "Enough to counter a 5-city Spire Capital Fleet".  That's 12 shipyards, or 52 frigates, 12 destroyers, 6 cruisers, 3 battleships, and 1 dreadnaught.  En-masse, that'd take a Mothership to counter, and we certainly don't want every 2:30 mini-Exo having one of those (Well, you may enjoy that...).  But, in the worst case, the player may need to be defending 5 locations (4 SD + HW).  Distributed, that's about 10 frigates, 2 destroyers, 1 cruiser, less than 1 battleship, and the dreadnaught at only one spot (probably on the HW).

What does 10 frigates, 2 destroyers, and 1 cruiser counter?  Well, a golem or two, but that's about it.  I guess it is also possible that a Spire City could be built on each SD device world - that'd add some significant defenses, too.  Very map dependent, though.  Still, if the Exos became big enough to get Golem leaders (that's what, 12500 points minimum per battle group?) then come close. 
Of course, the FS timed Exowaves will continue during the Showdown, and they can be much larger than event waves late in the game.


I think I'll try out what you've suggested - I'll take my FS save from right before the Transceiver, and play it out to a Showdown finish and see what happens.  More data should help.

[Toranth]

Well, I went ahead and did a Showdown with 5 Fallen Spire cities and the Capital, with the 6th Shard sitting on my Homeworld.  Results were odd.

The Showdown with 5 Fallen Spire Cities

Type	Unit Count	Exo Cost	Strength
Exos	3,712	372,432	85,501.9
Waves	12,401	N/A	91,773.85

Exo unit count about doubled, while Strength almost quadrupled.
Waves, however, only went up a trivial amount.  Same number of waves (38) as last time, so I'm not sure what happened.

If my notes are correct, Fallen Spire should have increased the multiplier by 5 * (0.25 + 6 * 0.125) + 1 = 6.
Waves should have had a 12x multiplier rather than a 2x, and Exos a 6x rather than 1x.

The budget for Exowaves did go up 6x, so that seems right.  But the waves...

In the Wave Logs I see
SD only:
WaveSize = MultiplierFromWaveInterval * MultiplierFromHumanHomePlanetAndChampionCount: 2 * 1 = 2
FS + SD:
WaveSize = MultiplierFromWaveInterval * MultiplierFromHumanHomePlanetAndChampionCount: 8 * 1 = 8

So there's an additional +6 there, if that's the right spot.  Everything else seems the same, and the end result seems the same.  Randomness seemed to play a much larger part in determining wave sizes during the Showdown.



The Exos were much tougher.  Golems or Dire Guardians (which don't count as Massive Ships yet) in almost every wave, and they wrecked some serious havoc on my defenses.  However, the Spire Fleet was a match for 1 GCS, even though there were very few survivors.  Just the Dreadnaught, 2 BBs, 2 CAs, and 4 DDs.
That let my fleetships all concentrate on the other GCS, and it went down fast.
Still lost 3 planets to Counterwaves, though.

[Toranth]

To reply to myself yet again:

Did you know that if you build 5 Spire cities, build a massive Spire Capital Fleet, then scrap all your hubs/shipyards/etc, you can cut the Showdown exowave sizes by 2/3s, while keeping your Spire fleet?  Sure, you can't rebuild anything, but it's not like you were going to have the time or money to do much of that, anyway.

[keith.lamothe]

To reply to myself yet again:

Did you know that if you build 5 Spire cities, build a massive Spire Capital Fleet, then scrap all your hubs/shipyards/etc, you can cut the Showdown exowave sizes by 2/3s, while keeping your Spire fleet?  Sure, you can't rebuild anything, but it's not like you were going to have the time or money to do much of that, anyway.

The cheese is strong with this one   I think the trade is fair in that case, as the defensive benefit of the cities is substantial.  Or, at least, the trade is fair enough that I'm not reaching for the nerfhammer

Very interesting results in the other posts.

All CPAs reporting as clean-outs?  Eep.  Yea, that would hurt.  Will have to investigate that.

Will consider on "hinting" the AI to spread out a bit more on the distributed attacks.  It's one of those things where strategically it really should be trying to concentrate (and on the point with the best combination of low-defense and high-value, at that), but also probably one of those things where it's more fun if the AI doesn't send its running plays over your weak left tackle all night long.

Odd that the waves came out to such a similar size with the 5-city+SD test compared with the 0-city+SD test.  If you have a complete wave log from each that might help.  Though it did look like it was factoring in that +600% just fine.

So it looks like on Diff 7 the 5-city+SD thing is actually quite winnable.  Though perhaps still harder than a diff 7 game ought to be?

It may be that the other tests people did where it was producing "I'm not sure this is winnable with gratuitous cheating" results was diff 10 or so.  The exo multipliers get really atrocious up thataway.

[Toranth]

To reply to myself yet again:

Did you know that if you build 5 Spire cities, build a massive Spire Capital Fleet, then scrap all your hubs/shipyards/etc, you can cut the Showdown exowave sizes by 2/3s, while keeping your Spire fleet?  Sure, you can't rebuild anything, but it's not like you were going to have the time or money to do much of that, anyway.

The cheese is strong with this one   I think the trade is fair in that case, as the defensive benefit of the cities is substantial.  Or, at least, the trade is fair enough that I'm not reaching for the nerfhammer

Very interesting results in the other posts.

All CPAs reporting as clean-outs?  Eep.  Yea, that would hurt.  Will have to investigate that.

Will consider on "hinting" the AI to spread out a bit more on the distributed attacks.  It's one of those things where strategically it really should be trying to concentrate (and on the point with the best combination of low-defense and high-value, at that), but also probably one of those things where it's more fun if the AI doesn't send its running plays over your weak left tackle all night long.

Odd that the waves came out to such a similar size with the 5-city+SD test compared with the 0-city+SD test.  If you have a complete wave log from each that might help.  Though it did look like it was factoring in that +600% just fine.

So it looks like on Diff 7 the 5-city+SD thing is actually quite winnable.  Though perhaps still harder than a diff 7 game ought to be?

It may be that the other tests people did where it was producing "I'm not sure this is winnable with gratuitous cheating" results was diff 10 or so.  The exo multipliers get really atrocious up thataway.

Here are the logs for the Waves and Exos for all three playthroughs.

Under normal circumstances, the AI concentrating its forces is appropriate.  Unfortunately, it's the nature of the Showdown that you're forced to defend more locations (like in multi-HW games) but with no additional defenses available to compensate.  The Ai still gets to concentrate, though.  It only needs to attack the right spot once, while the player needs to defend the right one every time.
I don't know if forcing the AI to spread out the attacks more is right, or would make it too easy.  But even if you didn't change the behavior, forcing the CPA to spawn spread out, rather than predominantly in one chunk, would give a chance to intercept and trim the total size.


I think AI difficulty may be a very large factor in Showdown strength, along with multiple homeworlds.  I know that when I tried a 9/9 - 2HW game on a Hex map, it was almost unwinnable.  It took several tries and lots of save-scumming (even for minor things, like a counter-wave going to a bad location) before that one was finished.  Is the HW multiplier added to or multiplied by the base Superweapon multiplier?
But at Diff 7, 1 HW, no Superweapons, the Showdown is not too difficult for an experienced player to win.  That first Showdown will end badly, though  >D

[Eternaly_Lost]

It may be that the other tests people did where it was producing "I'm not sure this is winnable with gratuitous cheating" results was diff 10 or so.  The exo multipliers get really atrocious up thataway.

Well, my game was a 9/9 game, not a 10/10 game.

If that was a 10/10 game, then I would not have even bothered to suggest that it need turning down. I asked for the pain there, and the AI was more then willing to give it.

9/9 as I understand it, should be winnable, but with a lot of cheese. I was able to prove at that point however, no amount of chease (I had a warp gate for missles sending Mk2 nukes to the planet that they were coming from over and over) could hold back that much AI at that point. Even the Fallen Spire Exo Fleet could not stand upto it.

[Histidine]

+1 to making the AI distribute its attacks evenly over all objectives. Among other things, it keeps Showdown from being "play X map to win."