I think you’re wrong. The only valid reason to use WoTF is the small (like very minor) DPS advantage. Also, near the end, you’re just acting salty and using clearly sarcastic tones. So please stop and be conducive to helping everyone instead of being a stubborn idiot who refuses to accept facts. And, if you are really in need of OHKO, use the darn ST wand. It does more damage per hit. WoTF, in my opinion (unless you really need that DPS for some reason) should be a swapout. Not a main.
As for using a tiered wand, you should carry a tiered wand for Abyss and things like that, and use the other non-piercing wand for everything else. Piercing is rarely useful (you don’t often fight huge hoards of things).
As the author of that paper, I can say that my results are dubious. I wrote a computer simulation that revealed WoF becomes consistent at 110 shots (130 after adjusting for RoF) and condu becomes consistent at 240 shots (192 after RoF adjustment).
Nobody has come forth with an explanation for the inconsistency so… it’s up to y’all to decide if you believe my math or my simulation.
fallen is simply more powerful than condu, there is no getting around that fact in any situation. it doesn’t really matter which one is ever so slightly more “consistent” than the other, fallen is still stronger. the only enemies that condu even comes close to dealing as much damage as fallen are those with 0-10 def. those enemies don’t really matter since they’re not dungeon bosses and have low enough hp that your exact weapon choice doesn’t matter very much, but even if they did, fallen would still be stronger. heck, even the gods in glands have more than 10 def (I think the average is ~15).
personally the only situations where I have issues with the range of fallen are against groups of enemies attacking at once, and if there’s multiple enemies you should be using a recomp instead anyway. I would never use condu if I also had a fallen.
I will readily admit to being sarcastic. I’ve seen people pontificate about DPS for years without bothering to look at all the relevant data and it still amazes me to no end.
I don’t see how that makes me salty, other than how calling me “salty” is a very convenient way for you to avoid addressing any point I’ve made.
No. The entire point of Fallen is to sacrifice range, rate of fire and piercing to gain more base damage. The only valid reason to use it is to deal more damage to a single higher-Def enemy (provided you either don’t have Cwand or aren’t facing an enemy against which Cwand is better).
It has even worse rate of fire than Fallen though.
I also don’t think you understand. When I’m talking about “time to kill”, I don’t mean OHKO, I mean the time it takes for you to kill an enemy. There’s rarely any point in trying to OHKO an enemy with a Wand because it’s almost always faster to use a piercing one to mow down several enemies at once instead.
Where did I say Fallen should always be used as main weapon?
You claim you’ve read the entire thread, yet somehow never picked up on all the times I’ve told people to use a tiered Wand instead of either Conducting or Fallen?
Is this a joke?
RotMG is a game where you almost never have only one enemy on screen.
There’s a reason why tiered Wands were all given pierce and CWand was given Armor Pierce instead.
I never said the maths MetaphorMe linked to were wrong, I’m just saying they don’t take a lot of factors into account.
I did not claim that Condu is “superior.” What I did is procure a mathematical rebuttal to your claim that Fallen puts out more consistent damage.
Consistency is independent of enemy def; it is purely a function of base weapon damage range and net rate of fire (taking into account character dex). There is no variable representing def in the the analysis I linked to. (The exception to def-independence would be when def > .85 * minimum damage, because then enemy def affects damage range and distribution. In the case of Condu def would have to be about 103, and that’s only a handful on enemies in realm. In any case, the effect of def over the threshold is only to make each weapon more consistent.)
Well you aren’t going to sit on anything with these wands, but, for me at least, the higher range and projectile speed on Condu only helps with staying on target longer.
The times given are just one example using an extremely high confidence level and an extremely tight standard for consistency. The general point of the analysis is that rate of fire turns out to be more important for damage consistency than base weapon damage range. The 3:2 rate of fire ratio trumps the 2:1 damage range ratio. I would expect this to remain the case for lower confidence levels and looser consistency levels, which require much lower times to reach.
It is you, my friend, who claimed that consistency is so important. Inconsistency on the part of Condu is one of two reasons you gave that supposedly invalidateeverything that can be said in favor of Condu. It is the only one of those two reasons that applies to the Fallen vs. Condu choice. I am just saying that your reason is a false proposition.
My own position is that consistency really doesn’t matter much. I choose to carry Condu rather than Fallen on Sorc for reasons already discussed by others above.
They get into lines and groups only if you or another player drag them. Now if you prefer to rush through abyss while shooting behind you, or if you prefer to clear each room one at a time by dragging back into the previous room, then yes they will clump and you should use tiered. If my pet was much stronger I might do the former, and when it was weaker I did the latter, but presently my preference is to scepter ahead about every 1.5-2 rooms and then clean up whatever is left as I enter. With that strategy they are often not clumped and I find Condu most useful for cleanup. For Tomb Scarabs of course I used tiered.
Thanks for the clarification. You didn’t share those numbers in the other thread and I took you to be saying that most of the time Condu achieved consistency at about 50 shots. Is this a different simulation now? You said in the other thread that
Half the gap between minimum and maximum damage, three times the minimum damage.
Fallen is more consistent.
Yes, that’s exactly what I meant when I said your analysis only worked at 0 Def.
Against an enemy with higher Def, having a bigger gap between minimum and maximum damage as well as a much lower minimum damage means you’re going to need more luck or more time in order to do the same damage.
I admit that wasn’t an argument for Fallen, rather a reminder that your analysis doesn’t include Projectile Speed and Range as well as the impact of the shooting pattern on the weapon’s effectiveness (see also: Bulwark).
… which is, again, a fancier way to say “Rate of Fire is more important if you hit the enemy with 100% accuracy for an extended period of time” and “You need more luck if you want it to work on lower time intervals”.
You’re not breaking any new ground here. You’re not even disproving anything I’ve said.
Does this mean you never encounter crossroads and never have to react to other players dragging on you?
Speaking of tiered Wands:
I don’t think you understood point 1. If Conducting is just a worse Recomp while Fallen at least has its own niche, then why would you use Conducting and Recomp instead of Fallen and Recomp?
To be honest, all this “damage consistency” shit is stupid.
Yeah, your damage might end up lower than it should have been for one instance. But it could just as easily have been higher than its theoretical average damage.
For n > 50, the only thing that matters is the mean. Everything rapidly balances out (unless you’re using something like sprite wand).
Edit: This doesn’t even matter anymore but these are the average number of shots (data collected over 10 thousand trials) it takes to get within 1 of weapon theoretical average. But like I said, 24 or 57… does it really matter? You shoot that amount in like 10 seconds.
You’re still focused on DPS when it’s not what you should be looking at.
What you should be asking yourself is, for each enemy:
-how long does it take to kill it?
-how long does it take to reach the SB threshold for it?
The conclusions reached might not necessarily differ. Warrior has such a crazy damage output that he’s going to look great regardless of what approach you use, for example.
But the difference would be that you’d be looking at how damage actually works in gameplay rather than toying with a theoretical number that doesn’t mean anything by itself.
import java.lang.Math;
import java.util.Scanner;
public class weaponVar {
public static void main(String[] args) {
System.out.println(" ");
int wMin = input("Min weapon damage: ");
int wMax = input("Max weapon damage: ");
int shots = input("Number of shots: ");
int trials = input("Number of trials: ");
double var[] = new double[trials];
System.out.println(" ");
for(int trial = 0; trial < trials; trial++) {
int varArray[] = new int[shots];
int total = 0;
for(int i = 0; i < shots; i++) {
varArray[i] = (int)(Math.random() * (wMax - wMin)) + wMin;
total += varArray[i];
}
double mean = total / shots;
double totalVar = 0;
for(int i = 0; i < shots; i++) {
double tmp = varArray[i] - mean;
totalVar += Math.pow(tmp, 2);
}
var[trial] = totalVar / shots;
}
for(int trial = 0; trial < trials; trial++) {
System.out.println("Variance for trial " + (trial + 1) + ": " + var[trial]);
System.out.println("Standard deviation for trial " + (trial + 1) + ": " + Math.pow(var[trial], 0.5));
}
System.out.println(" ");
}
public static int input(String str) {
Scanner s = new Scanner(System.in);
System.out.print(str);
int tmp = s.nextInt();
return tmp;
}
}
alright, so i made a little program of my own to calculate variance and standard deviation for weapon damages over a certain number of trials. @awdreg i’m not sure specifically what your program was designed for but i’m not too familiar with python so i couldn’t help you either way.
anyways, on the topic of the variance and deviation of the shots, here’s what i got:
condu:
fallen:
so fallen has more consistent damage than condu. 10 trials each, 12500 shots for condu, 8500 shots for fallen. it should all be correct; you could look over the code yourself if you wanted. i just used the standard methods of calculating variance and standard deviation.
A nonsensical argument. Really, that is irrelevant to the point I was making. Yes, fallen has a lower fire rate, but its higher damage per shot more than compensates. It simply does more damage than Condu against all enemies.
Yes, if you miss with one Fallen shot it’s more of a waste. But if you are firing more shots with e.g. Condu you are going to miss more often. If your accuracy is 90%, say, then you are going to do 90% of maximum damage with each. So it’s irrelevant.
Except you are going to miss more often with Condu, because of its variability. You are going to line up to shoot something with a few hundred HP and not know whether it takes 1 or 3 shots to kill. So you fire 3 shots to be sure, half the time wasting the 1 or 2 that were not needed. This is an issue with all weapons to some extent (except Ray Katana) but it is far worse for weapons with high variability.
I can’t offer insight on the main question whether/where Cond/Fallen is better over the other, I don’t rate either highly enough to carry over Recomp-Cwand combo.
I don’t see the point in limiting your range so badly even for the higher DPS; if you want this, swap your class not your weapon, and go not-sorc.
OP you gave your own answer, for me, in the question - if you find it annoying to swap, and notice not much more damage, then don’t swap. Don’t even carry the other item.
I dislike continually swapping because of wasted time during each swap. Theoretical discussions often overlook this and it can be significant time lost (same as DPS discussions overlook bad aiming). Decide what is best before you start, and only swap mid-fight to gain a LARGE benefit from the other item.
All the maths/numbers is fine to bring in, but it can only be a signpost to help your decision since there are other factors going on that the pure DPS numbers don’t take account of (what enemies you’re fighting, how you’re fighting them (playstyle), is Cond’s +WIS critical for sceptre dmg, are there enemies for which Cond’s 8 range is fine but dropping to 6.6 is uncomfortable). Trying out the items for yourself can’t be beat. What works for one player might be a terrible combo for another.
This is not the same thing that @Awdreg 's program analyzes. It is no surprise that the standard deviation of damage-per-shot for thousands of shots very closely approximates the standard deviation of the weapon’s base damage range. All you are showing is that the standard deviation of a sample many times the size of the population matches the standard deviation of the population itself.
What Awdreg analyzed is how quickly the mean of continuous, cumulative sampling approaches the population mean. This what you need to know in order to assess how likely you are to actually get a weapon’s avg DPS over a period of time. That is what I have always taken “consistent damage” to mean. It is second-order function of rate-of-fire and base damage range. It is more relevant to
than is
or
So it seems we have been meaning two different things by “more consistent damage”:
damage range and avg damage of the weapon (or a function of the two such as minimum DPS)
chance-of-getting-avg-dmg-over-time (which is a function of damage range and rate of fire, and is independent of either avg damage or DPS)
Beyond that confusion, I think we actually agree on quite a bit. I think we agree that enemy Def is relevant to the former and not to the latter. I think we agree that Fallen is better on the former and Condu might be better on the latter. We agree that other factors like range and projectile speed are relevant to a weapon’s effectiveness. And I do agree with you on what the most gameplay-relevant questions are.
I think what we disagree on is which mathematical value or function is most relevant to how quickly you can kill/get SB on an enemy.
We also disagree on the importance of niches, but it really comes down to personal preference. Sure Fallen has the most DPS, but to me that isn’t reason enough to use it over the wand that is 2nd-highest in DPS and also has a wis stat, more range, and is easier to aim. You seem to agree that some niches are not so important, insofar as you prefer tiered over Abe’s for range. One could just as easily say tiered (or everything) is just a worse Abe’s. When you call Condu a worse tiered you’re just making a subjective judgment about which attributes to consider and/or how to value them relative to each other. That’s a legitimate opinion to have and I respect your preferences, but they certainly don’t justify saying
Personally, I fire at an enemy until I see that it dies, and I don’t shift my attention to the next target until I see that the first is eliminated. I don’t waste time due to not being able to predict # of shots because I am not paying attention to # of shots to begin with. I only pay attention to individual shots if I am expecting to one-shot. Otherwise I find it more comfortable to think in terms of a cluster or beam of shots. Since that is how I approach shooting and aiming, the variability in damage doesn’t make much difference. You just shoot each thing until it is ded then move on.
I do that sometimes; especially when I am not sure of the HP of enemies, or it’s high enough I know it will take half a dozen or more shots. When you can line them up and drag them such as in dungeons you can just keep firing into the crowd until they all die.
But other times I try and use just the number of shots needed. Godlands is one place where it’s useful to use just the right number of shots as the quicker you can direct your attention to another god the better, for optimal fame/pot farming, as well as for survivability. Gods have fixed and easily rememberable amounts of HP, making it easy to predict how many shots it should take.
