Page 7 of 8

Re: Scrollreader's House Rules

Posted: Wed Aug 31, 2016 2:50 pm
by atgxtg
Corvo wrote: To be fair, Atgxtg, Deadmanwalking gave Fell and Grievous to all the weapons in the test.
All three examples got a 3-rewards weapon.
But that's not "fair" that's biased towards the elf. The elf needs the extra 2 points of damage from grievous to be able do 12 points of damage on a great success and drop an addercop Otherwise he'd need to either get the extraordinary success , or score a wound.

The extra 2 points from grievous doesn't do the Beorning much good at all. It's overkill for the addercop, and isn't a factor against the great or or black uruk. Now the Beorning would be much better off taking something else, for instance something like skin-coat or great strength, but neither of those factor into the simulator.

Re: Scrollreader's House Rules

Posted: Wed Aug 31, 2016 3:30 pm
by Scrollreader
This is something that should be taken into account, IMO. Especially with new, and (theoretically) more valuable rewards for an archer. The Elf needs Grievous (or at least Fell Handed) to hit the magic '12' threshold for many enemies. If the Shepherd's Bow or Fierce Shot user can add 'made in Anvil Way' instead, since we know called shots are the key once you hit weapon skill 4, how much does that increase the effectiveness of the Greatbow vs the Woodland Bow?

Re: Scrollreader's House Rules

Posted: Wed Aug 31, 2016 3:40 pm
by atgxtg
Glorelendil wrote:If you're suspicious of the simulation (which you should be...voting machines, too) then run it with a iterations set to "one" and study the log. If you find anything anomalous please let me know and I'll fix it.
(BTW folks - nothing personal in this. Just disagreeing with the simulations and the conclusions. :) )

Thanks. I't not so much that I'm suspicious of the results, more that I'm suspicious of the limited situations being presented and the conclusions drawn from them. For instance, if the character were in a situation where he faced the opposition alone, and had to deal with more than a single addercop (just ask Bilbo), then the other factors that are included in the simulator would come into play. As the simulator stands, it's kinda like claiming that shark repellent is better than a magic sword, and then throwing everybody into a shark tank. Basically, what I believe the results are really showing is how great ranged combat and opening volleys are. Especially once skill is high enough so that the characters can reliably hit (which is to the detriment of stances).




Let me try a different way to illustrate my point. Now the claim is that the elf is a powerful build. Now, just about any character can have a 5 Body (sorry Hobbits), and any character can have a 4 skill with Bow. And anybody can get a fell, grievous or keen bow. So the only thing in the simulator (and in the examples posted) that backs up the claim is the woodland bow. Right? So the whole thing rests on how powerful an advantage that is.

I ran some runs on the simulator for characters with Body 5, Bow 4, and no additional perks except for the woodland bow. I ran the tests both with and without an opening volley, and I ran the tests a few times each so I could be fairly certain that the results given were approximate averages. This is what I got:

No Opening Volley
vs. Attercop
Average Rounds: 2.34
Average Hope: 0
vs. Black Uruk
Average Rounds: 4.11
Average Hope: 0
vs. Great Orc
Average Rounds: 40.74
Average Hope: 0

No Opening Volley, Woodland Bow
vs. Attercop
Average Rounds: 1.35
Average Hope: 0
vs. Black Uruk
Average Rounds: 3.11
Average Hope: 0
vs. Great Orc
Average Rounds: 40.21
Average Hope: 0


With Opening Volley
vs. Attercop
Average Rounds: 1.34 (-43%)
Average Hope: 0
vs. Black Uruk
Average Rounds: 3.04 (-25%)
Average Hope: 0
vs. Great Orc
Average Rounds: 40.28 (-1%)
Average Hope: 0

With Opening Volley, Woodland Bow
vs. Attercop
Average Rounds: 0.55 (-77%)
Average Hope: 0
vs. Black Uruk
Average Rounds: 2.15 (-47%)
Average Hope: 0
vs. Great Orc
Average Rounds: 39.97 (-2%)
Average Hope: 0


So what I see from this is that a woodland bow lets the elf drop the first addercop or Black Uruk with about one fewer shot than any other archer with 5 Body, and a 4 skill, and gives a good probability of killing two addercops in an opening volley. Against a Great Orc the advantage is not all that much (from around .3 to .5 of a "shot" difference).

Now I don't consider that to be such a big deal, especially since it completely ignores what happens in the rest of the fight, or even things that happen outside of combat. It's certainly not so much that altering Deadly Archery would turn the elves into action movie stars superior to the other builds. In fact, I'd say that with the right build the other characters can be as good or better in combat, especially if we were to look at the other factors in combat, such as defense,allies, or results against multiple foes.

Re: Scrollreader's House Rules

Posted: Wed Aug 31, 2016 7:20 pm
by Glorelendil
Well, yeah. Different builds shine in different circumstances. But I don't think "against a group of weak-to-medium enemies, in a fellowship of 3 or larger" is so that unusual that it should be considered contrived.

Sure, an elf archer by himself will not be as effective in close combat as a Dwarf with a sword a shield. (Then again, that extra volley makes him potentially more effective by himself than any other archer except a Woodman with a dog.)

But how often does that happen? About as often as the Dwarf finds himself alone and unable to get at the goblin archers shooting at him?

The point of setting the Body score of all of them to 5 is to normalize as much as possible so that you can test select variables. Yeah, a lot of characters will have Body greater than 5, doing extra damage with Tengwars. But they probably have lower Wits as a result, so they're going to get hit more often. How do you even compare these things?

Re: Scrollreader's House Rules

Posted: Wed Aug 31, 2016 9:09 pm
by Corvo
Deadmanwalking wrote:
You want math? Okay. The following uses this calculator that Glorelendil whipped up and linked in the Deadly Archery thread.

So, here's the calculation for 'first enemy killed' for a Beorning with a Greataxe, an Elf with a Woodland Bow, and a Hobbit with a Bow of the North Downs and Fair Shot:

Elf (26 XP)
Body: 5(7), Heart: 4(4)
Virtues & Rewards: Woodland_bow
Hacks:
Weapon: Fell, Grievous Bow: 4
Stance: Defensive/Rearward
vs. Attercop
Average Rounds: 0.15
Average Hope: 0.08
vs. Black Uruk
Average Rounds: 1.22
Average Hope: 0.41
vs. Great Orc
Average Rounds: 31.95
Average Hope: 4.42

(...)
By the way, I tried applying the "technique: called shot" to the above setup.
It's counterproductive against the Attercops, slightly negative vs the Black Orc, but damn useful againt the Great Orc:

Average Rounds: 18.29
Average Hope: 2.54


If we take out "Grievous", it takes
vs. Great Orc
Average Rounds: 19.29 (-1%)
Average Hope: 2.69


Note: I'm not comparing this to other "builds", just tinkering a bit with Glorelendil's tool. Apparently, against the Great Orc, it's the wounding hit the decisive weapon.
On the other hand, Grievous is more useful against smaller opponents.

Re: Scrollreader's House Rules

Posted: Wed Aug 31, 2016 11:59 pm
by Deadmanwalking
atgxtg wrote:Yes I do. And I still want math. Not just posted results that cannot be verified.
I linked the calculator. Feel free to verify.
atgxtg wrote:And I'd say those stats are pretty dubious. You pretty much rigged the contest. First off, you don't just run the Elf with a woodland bow, but you give the bow fell and grievous, rewards that are designed to back up that theory.
I gave everyone the same number of Rewards. The intent is to show how well the item (and characters) work from mid-xp on. You want a low xp version, I can do that, too...but it's actually a lot less useful for determining how effective a character is gonna be in practice.
atgxtg wrote:Secondly, you "nerfed" the Beorning. Severely.
1) You give him a 5 Body, even though only one of the six backgrounds presented starts off that low.
2) You arm him with a great axe, rather than the far deadly great spear, or even a splitting axe.
I gave everyone a 5/5/4 attribute spread. Otherwise you're measuring the attributes more than the Rewards or anything else.

And I armed him with an axe because I didn't want to get into the fact that Great Spear and Great Bow users (and such people only) are miles better vs. a Great Orc using Called Shots than anyone else (a fact I listed in my first post on this topic). It's a distraction from the main point, and an advantage of the weapon category in question.

As for Splitting Axe...Fell is better, as are Keen and Grievous. Which is why I gave him the good stuff. No, really, I've checked. Feel free to do so yourself.
atgxtg wrote:You also set the example up "in a vacuum" and completely ignore the fact that the characters are going to half to defend against enemy attacks, and that the elf needs two other characters to shield him form the opposition in order to keep using that bow. Or that the woodland bow's advantage is only good for the opening volley, and don't factor in for longer fights, where the elf's numbers aren't nearly so good. So basically, under ideal conditions for the elf with the bow he does very well.
Uh...I actually did a version of the Elf without the Woodland Bow specifically to show how they did in longer fights. That said, 7-8 rounds is a pretty long fight, and I measured them killing three black uruks over right about that time frame.

As for the other complaints, those are true of all archers and thus basically irrelevant to the point under discussion (which is whether Elven Archers are worse than other archers or worse offensively than melee characters). Archers in TOR need two melee people...but that's all archers, not just the melee ones, and all also receive the excellent defensive advantage of not being able to be engaged in melee.
atgxtg wrote:Oh, and you example does more to show how important "Grevious" is to the elf, rather than Woodland Bow. It's what's letting him kill addercops with great successes.
Sure. Grievous is great. Of course, you can easily have a Grievous Woodland Bow as a 6 xp character (10 xp if you raise Wisdom to 2 in there), so using it is by far the best way to get a read on how an Elven archer will do over most of their actual career as a character.
atgxtg wrote:But that's not "fair" that's biased towards the elf. The elf needs the extra 2 points of damage from grievous to be able do 12 points of damage on a great success and drop an addercop Otherwise he'd need to either get the extraordinary success , or score a wound.
Absolutely. But ignoring this is silly. As noted above, you can have this combination at very low XP, the fact that it's such a good one means that most Elven archers will do so. Therefore, if doing a comparison of Elven Archers against others, this combo needs to be in use or you're comparing meaningless abstractions, not how characters actually function.
atgxtg wrote:The extra 2 points from grievous doesn't do the Beorning much good at all. It's overkill for the addercop, and isn't a factor against the great or or black uruk.
Actually, it means he kills the Uruk in one blow on an Extraordinary Success, so it's a little relevant. But yeah, it's less so than it is for the Elf. And if going at a lower XP level, I wouldn't have grabbed it (with one Reward, I'd go Keen)...but that doesn't really matter at the hypothetical xp level we're talking (which is at 40 xp or so).
atgxtg wrote:Now the Beorning would be much better off taking something else, for instance something like skin-coat or great strength, but neither of those factor into the simulator.
Feel free to assume he has them, while the Elf has 'Elvish Dreams' and 'The Speakers'. That's what I've been doing.
atgxtg wrote:Thanks. I't not so much that I'm suspicious of the results, more that I'm suspicious of the limited situations being presented and the conclusions drawn from them. For instance, if the character were in a situation where he faced the opposition alone, and had to deal with more than a single addercop (just ask Bilbo), then the other factors that are included in the simulator would come into play. As the simulator stands, it's kinda like claiming that shark repellent is better than a magic sword, and then throwing everybody into a shark tank. Basically, what I believe the results are really showing is how great ranged combat and opening volleys are. Especially once skill is high enough so that the characters can reliably hit (which is to the detriment of stances).
This is certainly true to some extent...except that, by the math, the elf is doing on par with other archers as well. And the archers are doing on par with melee fighters (at least those using two-handed weapons). In short, outside one specific circumstance (foes with Great Size) everyone does pretty close to equally well with equal investment. Breaking that pattern is thus bad, and makes one build clearly superior to all others.

Or to put it another way: The builds are all close enough to each other that a serious power-up to any one of them is a problem. I'd be complaining just as much if you were suggesting powering up Hobbits or Beornings or Rohirrim.
atgxtg wrote:Let me try a different way to illustrate my point. Now the claim is that the elf is a powerful build. Now, just about any character can have a 5 Body (sorry Hobbits), and any character can have a 4 skill with Bow. And anybody can get a fell, grievous or keen bow. So the only thing in the simulator (and in the examples posted) that backs up the claim is the woodland bow. Right? So the whole thing rests on how powerful an advantage that is.
Not exactly. It's based on the fact that Woodland Bow, in combination with the stuff everyone gets, is as powerful as the stuff other people get to supplement the general stuff with. It puts them right on par with the other cultures...and that's all it needs to do for other additions to be a bad idea.
atgxtg wrote:I ran some runs on the simulator for characters with Body 5, Bow 4, and no additional perks except for the woodland bow. I ran the tests both with and without an opening volley, and I ran the tests a few times each so I could be fairly certain that the results given were approximate averages. This is what I got:


So what I see from this is that a woodland bow lets the elf drop the first addercop or Black Uruk with about one fewer shot than any other archer with 5 Body, and a 4 skill, and gives a good probability of killing two addercops in an opening volley. Against a Great Orc the advantage is not all that much (from around .3 to .5 of a "shot" difference).
Oh yeah, it's useless against a Great Orc. And mostly only useful against the first foe...but almost all combats in TOR there are three or fewer foes per PC. And most it's 2 or less, making the 'first foe every combat' something like half the foes you ever fight. That's a pretty nice number of people to have bonuses against.
atgxtg wrote:Now I don't consider that to be such a big deal, especially since it completely ignores what happens in the rest of the fight, or even things that happen outside of combat. It's certainly not so much that altering Deadly Archery would turn the elves into action movie stars superior to the other builds. In fact, I'd say that with the right build the other characters can be as good or better in combat, especially if we were to look at the other factors in combat, such as defense,allies, or results against multiple foes.
But the thing is that, if you build them with about mid-range xp (like I did) all the builds pretty much even out at around the same numbers, give or take a little (again, except vs. stuff with Great Size). And thus powering up only one build breaks a really well balanced combat engine.

Re: Scrollreader's House Rules

Posted: Thu Sep 01, 2016 2:42 pm
by atgxtg
Glorelendil wrote:Well, yeah. Different builds shine in different circumstances.... How do you even compare these things?
By simulating more than just the offensive potential of the characters under a limited set of circumstances. You have to factor in other things such as defense to get a truer picture of things.

Re: Scrollreader's House Rules

Posted: Thu Sep 01, 2016 5:29 pm
by Glorelendil
atgxtg wrote:
Glorelendil wrote:Well, yeah. Different builds shine in different circumstances.... How do you even compare these things?
By simulating more than just the offensive potential of the characters under a limited set of circumstances. You have to factor in other things such as defense to get a truer picture of things.
Well, if you can come up with an analytic way to compare those things I'll be eager to see it.

EDIT: FWIW, my first combat simulator (see my .sig) let the monsters hit back and computes not time-to-die but win:loss ratios. But that's still for only a single hero, which is kind of unrealistic.

Re: Scrollreader's House Rules

Posted: Thu Sep 01, 2016 7:24 pm
by atgxtg
Glorelendil wrote:Well, if you can come up with an analytic way to compare those things I'll be eager to see it.
Good point. The best way I can think of is to generate a party of characters and run them through mock combats. It's about a close to the real thing as I can think of - other than the fact that both sides tend to act a bit more organized and coodinated that they should. But it's close.

I used to do that back when I wanted to see how a new game handles things, or if the PCs could deal with an encounter.

Glorelendil wrote: EDIT: FWIW, my first combat simulator (see my .sig) let the monsters hit back and computes not time-to-die but win:loss ratios. But that's still for only a single hero, which is kind of unrealistic.
Is that still out there somewhere? I'd like to see how the elf stacks up compared to a melee fighter.
I wonder if it could be expanded to handle more than one hero?

Re: Scrollreader's House Rules

Posted: Thu Sep 01, 2016 7:30 pm
by zedturtle
Just click on "Combat Simulator" under Glor's post in his signature area.