Dynamic Difficulty and Evolution

Phoenix Point Game Feedback
21

There’s no evidence that the auto scaling is based on reload, the only point is it seems based on stuff like soldiers death, so at some points not perform well in a series and not bother reload to fix it, seems low down difficulty a bit and a time.

I don’t think it works the first month, probably a bug in the formula when there isn’t enough combats done. But later I had the feeling multiple time, bad series lower difficulty, good series increase it. It’s not the only factor but it seems play a role.

It’s weird after a soldier death I did two citadels, one the Scylla was almost fully killed already, and the second was a timid Scylla, lol, never had seen that before. Ok the damaged Scylla was certainly a weird bug.

22

I don’t think anyone is currently playing PP in ironman mode. Some of us are playing honest man. In any event, no one wants to punish anybody here. I hope.

As to adaptive difficulty, I think that that and panda evolution should be separate topics.

When it comes to adaptive difficulty, personally I’m ambivalent - I can even accept that the devs might want to punish savescumming because they don’t want their game to be played like that.

However, panda evolution is not so much about difficulty as the AI reacting to the player, not just by getting a bigger budget but by spending it differently.

So, has anyone come across any evidence that this is currently happening? Can the devs say if they understand that it is currently happening? Has this idea been abandoned?

23

no one is playing in ironman mode because ironman mode was pulled from the release build because it was breaking saves

24

Ah, i’d seen someone say they were playing ironman somewhere around here, so it in truth was more of the honest man mode.

I’d hope that’s not the case. If they don’t want players to savescum why add the option to begin with? They could be honest upfront and simply say they are putting out a game with nothing but a single autosave slot. If they’re going to punish people for using a system they put into the game why would they bother with the system to begin with? Just build it how they want, then advertise it as it truly is. To me that’d be the same as somehow increasing difficulty if you dual class a soldier, carry more equipment on them, or even just spend their skillpoints. If they put it int he game don’t punish players for using it.

1 Like
25

I think you are misunderstanding what people here are saying, Dorrin. No-one thinks PP is deliberately punishing save scumming and no-one wants that. As Spite and Barleyman put it:

I agree with them. The DDA is currently malfunctioning because it’s NOT counting Restarts, it’s only counting wins and the final score of that win. So it thinks someone who takes 5 reloads to ‘Ace’ (or even just win) a scenario is a super-player who managed that in just 1 playthrough and ramps up the difficulty accordingly.

Actually, what it should be doing is counting that as 4 losses and 1 win, and adjusting difficulty to that level instead. Simple to say, but only simple to program if the system has a way of recording Restarts. I’m not a programmer, so I have no idea how straightforward that is - but to make the DDA truly reflect a player’s performance, that’s what they need to do.

26

Well, not all saving and loading is savescumming. For example, I will load a game if something happens because I don’t know a game mechanic or I want to experiment with something; I will not load a game just because something didn’t go the way I wanted it to, or some soldier got killed. Savescumming is about achieving optimal results through saving and loading. I can understand that a developer may say I don’t want my game to be played like that. Many do.

I’m not saying that I want adaptive difficulty to punish save scumming, my point is I want Panda evolution, and adaptive difficulty has nothing to do with this.

27

That’s where I partially disagree, because I think Panda Evolution is adapting the difficulty - or should be.

If/when (I hope) this works right, the Pandas will adapt to take away the crutches you rely on to win, forcing you to find other ways to deal with them. That will make the game more difficult and would probably make it impossible if the DDA still works as it does now.

However, if this ultimately replaces ‘bigger guns and more hits’, and the DDA takes into account the amount of times you ‘lost and restarted’ a mission before you won it, this could make the game very, very challenging and interesting.

28

@MichaelIgnotus Panda evolution would definitely make the game more interesting and more challenging, but it is a separate issue from adaptive difficulty.

What I’m trying to say is that right now the Pandas become stronger over time simply by getting more armor, more health and bigger guns. Call it “fake Panda evolution”.

And then there is this thing called adaptive difficulty, which we don’t know how it works exactly (for instance, I recall reading that it doesn’t affect Panda’s deployment, only :panda_face: reinforcements), but which ultimately results in more or less, and stronger or weaker enemies appearing depending on player’s performance.

We can also speculate as to why there is adaptive difficulty at all. Could well be to punish savescumming - other game developers have taken measures like that, or different ones because they are particular about how their game is going to be played.

My point is it can’t be there to substitute for “true” Panda evolution.

29

At the moment, that’s exactly what I think it’s doing.

What we were sold as ‘the Pandas will evolve in response to your tactics/activity’ is the DDA, which makes the Pandas stronger/weaker & more/less numerous depending on how well it thinks we are doing.

That’s also why I think we are given no Option to adjust the Difficulty manually, because the devs don’t want us screwing up the DDA - the fact that it’s screwing itself up at the moment is another issue altogether.

In my opinion, the two should be linked. The Pandas should buff up/down in response to our weapons and how effectively we use them - just not as much as they do at the moment; but they should also evolve their loadouts at the same time in reaction to the way we use those weapons. That way, they don’t just get stronger, they get smarter - and we have to adapt to survive.

30

If it is, it’s very disappointing. It would be nice if the devs could clarify this.

1 Like
31

I certainly hope they’re not actively punishing save scumming. Sends a very mixed message in a game that has a restart button as an ordinary tactical option and automatic saves before every mission.

1 Like
32

I read the point of it was to avoid a difficulty decrease, perhaps it’s because of Easy, but in fact the feeling it gives is up and down. But if it’s a setup for Easy, it makes sense I suppose, Im’ more used to play Normal than Easy.

At max difficulty there’s a problem, scaling down is weird, only scaling up seems difficult.

About save and reload, not save scuming some shoot or action, I quoted in save files weird files about restart. So it can makes wonder. But I didn’t feel any link with auto scaling.

A hiatus is all players (but me) consider RB as exploit, so some don’t use them, but then is it the difficulty intended? And some use them and feel they shouldn’t and then argue anyway the game is too difficult.

For me the only problem is during first month, and eventually a bit up. Then the difficulty scaling can be too step even at Easy.

33

I am by rights a perfectionist… I want to reveal every part of the geoscape and do every scavenging mission… According to DDA this is a perfect way to cripple my game… put it this way… I ahven’t even got close to finishing the campaign and now im stopping my attempts until patches are in place…

Count yourself lucky you haven’t had the 20+ athron spam scenario…

I have had 20 Athrons against my team of level 7’s… all with heavy armor… grenade launchers and machine guns. all with return fire. my team spawned in the open. I do not use any of the OP things like dash and rapid clearance or Priest Spam. I play the game normally like most people would without considering how to break it with abilities… I can confirm it is IMPOSSIBLE to win in this scenario… they rush me with no care in the world… staying in the open. I physically cannot kill them fast enough. there is no point disabling arms because I would have to disable both arms. Within 2 turns im crippled. 3 turns dead.

Thats not evolving difficulty curves. Thats lazy programming. I was playing on normal… and up until that point the game had been challenging but not stupid. then suddenly… BOOM. It wasn’t even labelled up as threat level extreme… That was when i decided to uninstall and wait…

4 Likes
34

Ya against the armored arthrons like you had even dash+ rapid clearance wouldn’t have done it (well maybe if it was a min/max setup with both close quarters specialist and reckless talents). If you’d had the assault/heavy setup for that extra +1 will for the team on each kill it’d be doable, but i also refuse to build that way. I want a varied team able to handle situations, not a single class combo for handling all maps and then the sniper/heavy team to handle the end boss… To me it just isn’t strategy if there is only one method to get through these situations.

2 Likes
35

I suspect that DDA is being amped-up as a crutch/proxy for the fact that there isn’t much “true” evolution of the Pandorans responding to the players tactics. The quick and easy solution to this appears to be amping-up DDA to boss level fights (I’ve never been a fan of boss fights).

36

One solution to the DDA would be to put some brakes on it. So for Easy you could change the rate of evolution by a factor of 10 or 20, for Veteran by a factor of 5 or 10, for Hero maybe 0 or 2, and for Legendary, either 0.5 or 1. I think it’s the rate of evolution that we beginners are finding difficult to adapt to. I’ve seen a lot of complaints about the difficulty spike for those playing on lower settings and I’ve restarted at least two games because of it. The paradoxical workaround is simply not to run supply missions.

I think the idea of evolving different adaptations is OK, but another aspect would simply be to evolve new species. Why is it just crabs, tritons, sirens, worms, chirons and scyllas? The oceans have the vast majority of diversity on the planet. There should be other forms that have been affected by the virus. Mixing up the Pandas could help a lot.

As it stands now, once you get to level 7, it’s pretty easy to handle large numbers of Sirens and Chirons in the missions. If anything, just increasing map size would help a lot in terms of sniping. The bomber chirons should be on a lower selection level or be present on missions with decent cover.

Wrt OP builds based on Dash or Rage Burst, I could see these being on longer cooldowns or requiring more WP. The multi-dash maneuver is fun a few times, but yeah, it’s OP and gets old, sure gets you out of a bind if you’re not min maxing… On the other hand, Rage Burst is what takes the difficulty out of “evolved” Pandas, so there should be some form of it in the game, again, just increase the WP if it’s a balance issue. WP adjustments could easily be added to the difficulty level to improve the OP aspects of “balance.” Keep WP where it’s at for the rookie mode, double it for Legendary?

Btw, Scylla serrata is probably the fiercest crab I’ve ever handled in real life. Would love to see cones, octopuses, ambush predators (some worms, fishes), serious predators such as sharks, etc. Panda strategy could mimic PP strategy, too, wrt snipers, rageburst, etc.

1 Like
37

Your proposed changes are exactly what I was hoping the DDA would be: constantly shifting genetic diversity and adaptation of tactics based on those which you use against them to always keep us on our toes.

38

The problem is that a player who wins alot will have more difficulties than one that losses more. PP punishes wins and awards loses is a problem.
Say, you lost a map. Why would your campaigh be easier then someone who won. It`s perfectly fine if loss would spike your difficulty.

39

It is fine for a loss to spike your difficulty - I think the idea though with PP is that it’s about balancing relative difficulty rather than absolute difficulty. If it worked correctly, then each player should receive the right amount of challenge based on how well they’re playing. The aim is to avoid a player finding the game too easy/hard.

I think ultimately it could do with being a feature that the player could turn on or off.

(And off course it needs to work properly too)

40

Because currently Phoenix Point is a “crab”-game … ;o)

2 Likes