Topic: Problems with rng

  • Author
  • #4044
    Avatar photoewhalecancer

    Hey I just bought the game and enjoying at the moment. As a fan of Fire emblem and Xcom I’m well aware of the joys and annoyances of rng. Quickly I’ve noticed the rng is quite messed up it seems. I miss 70-90 more often than not and my first play session on normal seems utterly plagued by this. I don’t believe its just bad luck. My latest two battles I’ve had two bandits literally dodged 7 attacks in a row from my entire warband. Neither had shields at 80% for every strike. The next was an undead battle. I stopped counting and continued raging at the amount of over 70% attacks i missed so bad that i started losing units just do to sheer amount of rng bull. Properly set up shield wall and defensive position and could barely could hit any of them. Asking if anybody else is noticing this and if its a real problem right now.

    Avatar photoSky

    Someone wrote about similar issue a while back. Personally I have not noticed or experienced this kind or rng behavior except for few times earlier in EA. Maybe it is just a personal curse from the rng gods or perhaps it is something triggered by specific game setup. I’d say try equiping better tohit chance weapons early game and slowly change em to whatever you like.

    Avatar photoRap

    There is nothing wrong with the RNG. The game uses Mersenne Twister (MT19937) if you want to read up on it.

    Overhype Studios - follow us!
    Facebook Youtube

    Avatar photoguidon101

    There is this:
    Gambler’s Fallacy

    But in general, I notice the “odd” RNG result here and there, but I agree with Sky. I’m not sure that the system is rigged, or you just have bad luck streak there (happens to me too, but I chock it up to luck, not system imbalance because it also happens the other way around, where a giant badass Orc Warrior whiffs consecutively on my new level 1 recruit with no shield on even even ground– I am usually holding my breath as I watch the whiffs in slow motion (read: interface lag), so I can count on that happening to me as many times to save my arse with good RNG as also to ruin my day with bad RNG).

    Maybe because of my expectation, I also “see so many” misses at 95%, with my melee, some with over 100+ melee attack rating, but I think in the long run, if I actually logged the statistics, the actual successful connects would probably be up there in the 90%. It’s just super frustrating and very noticeable when I miss at 95%, but I don’t usually notice or remember the 90 out of 100 times that I hit. My opinion/conclusion is that the melee RNG system is probably fine, although I wouldn’t mind if someone came out with a log showing actual hits vs. misses ratio from in-game associated with different % chances to prove it.
    It is helpful they include the factors in the tooltip! I strongly encourage this to help players understand the details and give us the WYSIWYG impression.

    On the other hand… What seems really strange to me in terms of the RNG are the RANGED vs. MELEE %…
    Ranged seems to hit waaay more often at low percentages (I’m talking like 5%-30%) — whereas I know my melee pretty much never land those kinds of hits, I can actually rely on ranged, even at those low chances. But it also seems to play the other way around, where high hit-chance ranged, still feels lower, so in the end Ranged attacks always feel like a 50-50, which is reliable enough for me. This is very noticeable at low levels, where my level 1 45-ranged-attack archer has 8-10% Chance-To-Hit, connects 3 times in a row (hits 3 out of 4 shots over two turns), insta-killing 2 thugs at 4+ range. Of course, I was like, “Cheers!” (read: I have the opposite reaction when I have 92% CTH and I miss 3 times in a row). And that only happened once in probably 30+ hrs of gameplay, but of course I remember the incident very well.

    final comment/suggestion:
    1) I shouldn’t say you don’t see an RNG problem, so I wouldn’t discourage other people and the devs to take a look at the RNG; let’s prove it with statistics!
    2) I feel like the RNG problem is more in our psychology than in the system, currently
    3) If there is a problem with the system, I sure hope we find it and fix though!
    4) More transparency WYSISWYG: I like how they show a lot of info on tooltips, so I encourage this trend; but also keep in mind that some players like more abstraction or blind/noinfo for immersion purposes (e.g. instead of saying “90% CTH”, say “High Chance to Hit”), so the current system currently caters to us who like more info, and it helps understand/debug the mechanics

    Avatar photoSky

    Agree, it probably is more of a psychological feel than an error. All due to the numbers in the tohit and our expectation of how it supposed to be or should be, that confronts how it actually is based on rng.

    Avatar photoJeuparfois

    I know this post is kinda old, but I did experienced the same “issue”.

    I find myself calculating 10 to 15% below in melee to counterpart this. Ranged is not a problem, because of non-retaliation.

    Maybe rng are “true” on the long term, but battle-lenght they don’t seem about right, in my experience of course.
    I mean, if on a d6 I get a hundred times 1 in a row, on 600 rolls, it would still be right but not gameplay-wise appreciable.
    A way to adjust them on short term maybe ? For a flatter chance, opposing to sheer luck/bad luck.

    Losing a character due to some turns against all odds is more “game-breaking”, and marking, than the opposite, obviously.
    I don’t mind losing a character (despise consequent time to level up) but I dislike when it strikes as overwhelming unfair (and tends to retreat/load for a cheap reason, wich I don’t enjoy either).
    Would gladly trade plain good luck for no more plain bad luck.

    Well, that’s the only thing I became fustrated with this game, at times, so it’s not such a major flaw, but still.
    Of course, again, my opinion in what I subjectively notice.

Viewing 6 posts - 1 through 6 (of 6 total)
  • You must be logged in to reply to this topic.