A lot of bad design choices outweight good things the game has

[wasteland_ghost]

After reading forums on the subject of hit chances, probabilities, mechanic design choices, etc.

25+ hours in JA3 at the moment and 500+ hours in XCOM, modding included.

XCOM hit chances are accurate. Proven multiple times by studying the code. And that game even cheats in player's favor at easier difficulties increasing your hit chances and decreasing alien hit chances silently.

JA2 from the memory appears much more fair on those - max aim level, high ground, in range of the weapon, marksman character and you hit. JA3 - nope, never can be sure.

Devs don't want you to even know your hit chances, they give you +/- breakdown instead. OK, it can work. But when you've got four (!) pluses on a melee character with martial artist perk using melee up close and missing - it's bs. The system doesn't work anymore, because those four pluses mean nothing at this point. It's a coin flip and your best chance is not having specialized mercs but just several mercs on one target to combat the randomness.

AI in XCOM is not cheating - aliens don't know where your soldiers are if you can't see them. AI in JA3 is cheating and it's clear - enemies know where all the mercs are and they know each time the exact borders of overwatch zones.

On top of that enemies are very spongy, it's overtuned and it's a shame because there are good things in this game, but absurd numbers are making it very unpleasant to play.

In terms of game design JA3 is stuck around JA2 times not willing to go further. You can't move dead bodies, though the game has stealth. Stealth is mostly done in real time, but it's build around turn-based principles (UI, actions, takedown mechanic - it's overly complicated, painful and not enjoyable at all). Maps have pathing errors, badly placed map borders misleading players with unexpected invisible walls. Inventory management with autosorting is a pain in co-op. Misplaced characters after loading a save (because they were standing on a square with a dead body). Lots of missing information on UI (fatigue level, experience). Desyncs and crashes in co-op. And so on and so forth. Bad sides of the game outweight the good ones. The game needed more development time, it's Early Access and TBH Steam should have instruments for that. Greedy publishers needs to be punished.

Thank you, THQ, for another unfinished game.

[sandman25dcsss]

Different pluses/minuses have different values, they can be extremely confusing without numbers:
https://steamcommunity.com/sharedfiles/filedetails/?id=3005282816

[wasteland_ghost]

Your link isn't working.

And no matter the exact values, four pluses with zero minuses resulting in a missed knife attack up close by a martial artist is a joke (yes, statistics, but it's not just one example in those 25+ hours). Making the game difficult and mocking your players are two different things.

[Rastlaff]

Originally posted by sandman25dcsss:

Different pluses/minuses have different values, they can be extremely confusing without numbers:
https://steamcommunity.com/sharedfiles/filedetails/?id=3005282816

We've got the idea: +/- is the big feature of the game it's proud of (broken one, BTW) but let's concentrate on the text below from OP, shall we?

In terms of game design JA3 is stuck around JA2 times not willing to go further. You can't move dead bodies, though the game has stealth. Stealth is mostly done in real time, but it's build around turn-based principles (UI, actions, takedown mechanic - it's overly complicated, painful and not enjoyable at all). Maps have pathing errors, badly placed map borders misleading players with unexpected invisible walls. Inventory management with autosorting is a pain in co-op. Misplaced characters after loading a save (because they were standing on a square with a dead body). Lots of missing information on UI (fatigue level, experience). Desyncs and crashes in co-op. And so on and so forth. Bad sides of the game outweight the good ones.

So using the language of the devs here: the game to me is actually - - - - against + +.

[Capo Rip]

Originally posted by wasteland_ghost:

Your link isn't working.

And no matter the exact values, four pluses with zero minuses resulting in a missed knife attack up close by a martial artist is a joke (yes, statistics, but it's not just one example in those 25+ hours). Making the game difficult and mocking your players are two different things.

This is the same bad argument they level against xcom. you think a 90% chance to hit should always hit and not just like 9 out of 10 times. So yeah, "statistics"

[Dick__Dastardly]

I suspected AI having a little omniscience and ran a test that reinforced that. Had an enemy inside a house who had lost site of my mercs for about 5 turns. There are exits on East and West sides of house. My mercs approach in stealth mode from South, enemy having no line of site. I stacked them on the corner covering the East side of house. Enemy comes out west side and shoots them in back. I reload and stacked on West side of house, enemy comes out East side and shoots them in back.

I kept testing, switched sides a few more times, then did 10 in a row of each. Every single time the enemy came out the unguarded exit and shot them in the back. Of course I could and should cover both exits, but this was to test the AI.

[sandman25dcsss]

Originally posted by Rastlaff:

So using the language of the devs here: the game to me is actually - - - - against + +.

I am not JA fan so for me it is the opposite: --++++. I would not buy the game if it was balanced around assumption player should move bodies. Shadow Tactics is much better game for that IMHO.

[sandman25dcsss]

Originally posted by Dick__Dastardly:

I suspected AI having a little omniscience and ran a test that seemed to confirm it. Had an enemy inside a house who had lost site of my mercs for about 5 turns. There are exits on East and West sides of house. My mercs approach in stealth mode from South, enemy having no line of site. I stacked them on the corner covering the East side of house. Enemy comes out west side and shoots them in back. I reload and stacked on West side of house, enemy comes out East side and shoots them in back.

I kept testing, switched sides a few more times, then did 10 in a row of each. Every single time the enemy came out the unguarded exit and shot them in the back. Of course I could and should cover both exits, but this was to test, and seemed to confirm, the AI knew what it should not.

For tests use mechanic girl, she can show positions of all enemies on the map.

[insert food here]

Originally posted by Dick__Dastardly:

I suspected AI having a little omniscience and ran a test that seemed to confirm it.

Yes I have seen the same many times. Done lots of stealthy sneaking up on unaware enemies to get setup in good flanking positions only for one misstep to get me detected and all enemies instantly not only know where my one spotted merc is, but all the unspotted ones too. Total nonsense when the AI is not even held to the same rules I have to follow as the player.

This combined with the total fantasy weapon stats and poor weapon balancing is really sapping the fun out of a game I so badly wanted to be able to enjoy. I don't expect any game to be perfect but I just cannot get past the many glaring flaws and poor design choices here.

[Rastlaff]

Originally posted by Capo Rip:

Originally posted by wasteland_ghost:

Your link isn't working.

And no matter the exact values, four pluses with zero minuses resulting in a missed knife attack up close by a martial artist is a joke (yes, statistics, but it's not just one example in those 25+ hours). Making the game difficult and mocking your players are two different things.

This is the same bad argument they level against xcom. you think a 90% chance to hit should always hit and not just like 9 out of 10 times. So yeah, "statistics"

How hypocritical it is to argue with such a comment removing all the complexity of well though out combat mechanics and well written algorithm. "Huh, so statistics."

We are talking about game mechanics here, we're not in "telenovelas".

[wasteland_ghost]

Originally posted by Capo Rip:

Originally posted by wasteland_ghost:

Your link isn't working.

And no matter the exact values, four pluses with zero minuses resulting in a missed knife attack up close by a martial artist is a joke (yes, statistics, but it's not just one example in those 25+ hours). Making the game difficult and mocking your players are two different things.

This is the same bad argument they level against xcom. you think a 90% chance to hit should always hit and not just like 9 out of 10 times. So yeah, "statistics"

Tried to find back the link, but apparently that topic at Firaxis forums no longer exists.

So. Yes, years and years ago we were discussing the probabilities in XCOM. Those are fair. And yes, 90% chance to hit means 10% chance to miss. And yes, there is a 0.01% chance to miss four times in a row and yes, since it's non-zero it can happen. And yes, I myself was writing that random number generator has no memory. It's all true.

But we're not talking probabilities here. JA3 is having different game design. And without mod you don't even have those probabilities, you rely on your mercs' voice lines ("It seems as a good shot to me") and the +/- descriptions. This is how the game was supposed to work from the original design point of view and this is why you have all those perks. You see the negatives and you work to remove them: you get into weapon's range, you get your sniper to high ground, you take the one with night vision to compensate for the dark, etc, etc. And you still miss. OK, it's a shot, can happen. But it happens way to often! (We were not just resting on sat view all those 25+ hours.) And when your melee character misses his knife attack when you were specifically building it to gain all those pluses and positioning so he has no minuses and he still misses - it's bad game design. Because it means that system is not working. It's not teaching the player and even worse - it's teaching wrong things, things that are not true. And when the players realize that, they stop playing JA game with all those very individual mercs and start playing XCOM with units and hit chances. And that is the part that is wrong.

PS For clarity: we're playing the game as it was designed to be played - no mod for chances. Because we've read the explanations from the devs and decided to believe them and go with their vision.

[Vibe High!]

Faulty design from the start.

Please patch the game! Maybe some JA honour can still be salvaged!

[Rastlaff]

Originally posted by sandman25dcsss:

Originally posted by Rastlaff:

So using the language of the devs here: the game to me is actually - - - - against + +.

I am not JA fan so for me it is the opposite: --++++. I would not buy the game if it was balanced around assumption player should move bodies. Shadow Tactics is much better game for that IMHO.

Another post with no constructive arguments just a point of view ... another "telenovelas".

Don't get it wrong, sometimes I can watch telenolevas while lying brain dead on a sofa and not being a fan of JA as well.

[Space of AIDS]

Originally posted by wasteland_ghost:

But when you've got four (!) pluses on a melee character with martial artist perk using melee up close and missing - it's bs.

Why game devs keep doing this?

The reason why knife fights IRL are so deadly is that even a toddler can figure out how to shank someone to death and no amount of training can reliably protect a person from a fatal stab in the guts. Many martial artist believed in themselves, got into knife fights and were killed by some random thugs with a $5 knifes on the streets.

In games we are controlling martial arts masters or seasoned mercenaries with lightning-fast reflexes that, apparently, can't figure out how the thing with the pointy end works.

This "approach" to melee is not exclusive to JA3, btw, just thinking out loud.

[wasteland_ghost]

Originally posted by 1+1=10:

Originally posted by wasteland_ghost:

XCOM hit chances are accurate. Proven multiple times by studying the code. And that game even cheats in player's favor at easier difficulties increasing your hit chances and decreasing alien hit chances silently.

I am not sure about that. People started to believe that the RNG decides if there was a hit or not. But that is wrong. The RNG does nothing else but generating numbers.
These numbers are then used by a different system to decide about the hit.
Yes, that system might be fair and even in favor of the player by increasing the hit-chance, but that only works properly if the underlying RNG is providing evenly distributed numbers. If that is not the case, then the "cheating" in favor of the player might be there, but not working at all.

Don't worry - it was proven multiple times. Look at this article, for example[sinepost.wordpress.com] (btw, read the other ones on probabilities in games by the same guy - they're quite useful). We decompiled and studied XCOM to its very core while working on tools and mods years ago, we know how random generator and AI work in that one - and it IS fair.

But yet again: here, in JA, the problem has nothing to do with hit chances.

[Lightning]

Would like a change to be able to sell equipment if you wanted rather than scrapping only