Knight Limb Counter v2

Aerek

Right, and that seems sub-optimal when DSL > Undercut > RSL, so I need to be certain we're at RSL prep. (I'm guessing Undercut didn't exist when Tony was active.) Also when folks break on odd numbers, it's nice to know. The change you're contemplating sounds ideal. I'll take a look when I'm home, but if you want to share your fix, that'd be swell.

I'm not knocking Tony's work here, with his S&B formula knowing breakpoints at a glance it sounds flawless. As DC/DB, with a hits:health table that's still mostly blank, correcting mistakes mid fight just feels harder than it needs to be.

Aroan

Aerek said:

his S&B formula knowing breakpoints at a glance it sounds flawless.

There are some problems with the snb formula at certain weapon damages and health ranges, unfortunately (as Archaeon mentioned in October). I've been trying to work out where the error is, but don't have enough data on it yet.

I've switched from using the snb formula in the meantime to setting the thresholds manually, so that when I do find an error I can adjust for it either midfight or immediately afterwards.

Eryl

Aerek said:

Right, and that seems sub-optimal when DSL > Undercut > RSL, so I need to be certain we're at RSL prep. (I'm guessing Undercut didn't exist when Tony was active.) Also when folks break on odd numbers, it's nice to know. The change you're contemplating sounds ideal. I'll take a look when I'm home, but if you want to share your fix, that'd be swell.

I'm not knocking Tony's work here, with his S&B formula knowing breakpoints at a glance it sounds flawless. As DC/DB, with a hits:health table that's still mostly blank, correcting mistakes mid fight just feels harder than it needs to be.

Undercut was not a thing when the limb counter was released and being updated.  I don't use undercut at all; I find it kind of clunky and awkward.  I haven't so far really found a necessity for it, either.  I'm sure it exists, but I just haven't found it yet.

This highlights the key downfall of the tracker imo, though: because it only counts confirmed hits and relays that as a percentage, you can't accurately mix and match weapons.  If you had the weapon formulas, that would be a nonissue.  I do not have them, though.

I'll gladly share the fix if I get around to it soon!  It's up there on the list of improvements to make to my personal stuff.

Tony's work is amazing, and it doesn't at all come off like you're knocking it.  It works fine out of the box, but there are always quirks to things someone else coded because everyone's thought processes are different. It's important to ask questions like this!

Correcting mistakes mid fight is difficult, yes.  A hearty amount of preparation will prevent that, though, and after enough fights the mistakes just kind of are gone because you've set the breakpoints tight enough. When I find I have a mistake in my count, I try to remember the mistake is there and adapt accordingly.  If I see the left leg broke at 92.8% (arbitrary percentage that models one hit from being broken), I make a mental not that the breakpoint is 92.8% instead of 100%.  I fix TCONFIG HITS immediately when the fight is over. Obvs not the best solution, but it works for me!

As the SnS formula, it is not accurate anymore. There were some adjustments to things during/after Tony published his tracker, so it's off by just a hair.  I've found the formula feels "one artefact level" too low.  If I'm using a forged longsword, I use the damage stat from a level 1.  Right now, I have a level 2 longsword and my TCONFIG is set to the damage of a level 3.  I haven't found a single fight where this causes problems, and all my counts have been accurate.  That is anecdotal; I obviously have not fought someone of every possible health threshold, so there may be a couple places it needs tweaked.  I simply have not had that happen.

For clarity: My level two longsword has a damage stat of 125, but setting the damage stat to 131 inside TCONFIG has given me accurate breakpoints.

Afterthought edit: There is code in the tracker that accounts for the bonus limb damage added by which lag rune does it.  I'm unsure if this is has something to do with the formula being off, but I am a runewarden. I do not have recent enough experience using paladin to know if this part of the code is being problematic. I have no intention of looking into it, though, because adjusting my damage stat makes it a nonissue.

Edit#74826: Tagging @Aroan so he gets notified about the formula adjustment.

Aroan

Thanks for the info, @Eryl !

I never had any trouble with a forged longsword in Paladin using the formula, but Archaeon reported having it with his l3. I've seen it off in runie with a l1 and l2. I'll try tweaking the weapon damage up next as you suggested.

Aerek

Eryl said:

Undercut was not a thing when the limb counter was released and being updated.  I don't use undercut at all; I find it kind of clunky and awkward.  I haven't so far really found a necessity for it, either.  I'm sure it exists, but I just haven't found it yet.

It circumvents losing a leg prep to clumsiness or dex dodging because you hit with one delph and missed with the other. That was a disadvantage of unartefacted DSL vs dodgy opponents that other specs didn't suffer, (artefacts might make it a non-issue) so I was pleasantly surprised to find Undercut when I came back. It's also slightly faster than DSL.

Yeah, that's a fair point RE: swapping weapons. With Svo's counter you can just edit triggers to add in the difference. Not that I switch weapons around often, but Blunt benefits from being able to swap star/flails sometimes.

Atalkez

Arties does not help much against dodging. I refunded my level 3 scimitars because I saw no difference in DPS for hunting or PK, and no difference in miss rate. I still think knight weapons are completely broken on scaling and need an overhaul.

Phaestus

Atalkez said:

Arties does not help much against dodging. I refunded my level 3 scimitars because I saw no difference in DPS for hunting or PK, and no difference in miss rate. I still think knight weapons are completely broken on scaling and need an overhaul.

At a glance, knight weapons seem fine. At present, going from no arties to l3 scimitars should be about a 13-14% dps increase on a dsl against mobs (minor variance depending on str), a 23-24% dps increase against players, and should definitely increase chance to hit. If you think scaling on a particular weapon or attack is broken, you're welcome to file a bug with the details.

Cooper

I guess knight weapons seeming fine or not depends on whether you agree that those benefits are worth a $500 purchase.

Atalkez

Phaestus said:

Atalkez said:

Arties does not help much against dodging. I refunded my level 3 scimitars because I saw no difference in DPS for hunting or PK, and no difference in miss rate. I still think knight weapons are completely broken on scaling and need an overhaul.

At a glance, knight weapons seem fine. At present, going from no arties to l3 scimitars should be about a 13-14% dps increase on a dsl against mobs (minor variance depending on str), a 23-24% dps increase against players, and should definitely increase chance to hit. If you think scaling on a particular weapon or attack is broken, you're welcome to file a bug with the details.

Going from level 2 to level 3 scimitars showed no difference that I could see, that Mezghar could see. Going from level 2 longsword to level 3 longsword didn’t show anything that Mizik could see. I didn’t mean 0 to 3, but I see the confusion. I refunded my level 3s and went back to level 2s, the 1600 credits didn’t seem to do anything for me in the upgrade.

Phaestus

Going from 2 to 3 is still about a 5% against both players and denizens, which I think is pretty close to standard across artefact weapons (excluding lash, because I don't know exact lash numbers off the top of my head but I'm pretty sure they're not that).

This is all said with the disclaimer that weapons are nightmarish abominations and they have minds of their own, and there may be some edge cases where they're just not seeing the performance they should. If you have specifics, we can look into it, but for the most part it seems okay.

Eryl

Phaestus said:

Going from 2 to 3 is still about a 5% against both players and denizens, which I think is pretty close to standard across artefact weapons (excluding lash, because I don't know exact lash numbers off the top of my head but I'm pretty sure they're not that).

This is all said with the disclaimer that weapons are nightmarish abominations and they have minds of their own, and there may be some edge cases where they're just not seeing the performance they should. If you have specifics, we can look into it, but for the most part it seems okay.

I can't really get on board with considering a weapon's dps against players, because no one gets to achieve maximum dps against another player unless they are standing still and doing nothing.  The problem with dropping 1600 credits to upgrade from level 2 scims to level 3 scims is that the potential 5% dps increase against players will never be worth it.

Spending 1600 credits to upgrade scimitars from level 2 to level 3 should feel impactful, not just be a minor damage increase on paper given perfect circumstances. It needs to lower limb breakpoints, make me do more damage per hit, and increase the speed of my dsls.

I've been told by multiple respected ACC members that this is not even close to the case.  Upgrading my weapons - the primary thing that makes me a knight - is literally the last thing on my last of things I want to purchase because of the lack of impact per credit spent.

1600cr for a 5% "pvp dps" increase is not at all a good metric to go by.

EDIT (for a little clarification): This is not a balance argument. This is a personal opinion. I am also tackling this from the arena of 1v1 combat, not groups (where 5% pk dps can potentially be incredibly impactful if you are not being targeted. Even then, I would rather just change to a class that has better dedicated damage.

Also, I think the reason looking at "pvp dps" on dual cutting doesn't sit well with me is because I consider dual cutting an affliction class, not a damage class. DSB doesn't scale with weapon damage, and dual cutting's affliction output is one its most alluring features.

Phaestus

Eryl said:

Phaestus said:

Going from 2 to 3 is still about a 5% against both players and denizens, which I think is pretty close to standard across artefact weapons (excluding lash, because I don't know exact lash numbers off the top of my head but I'm pretty sure they're not that).

This is all said with the disclaimer that weapons are nightmarish abominations and they have minds of their own, and there may be some edge cases where they're just not seeing the performance they should. If you have specifics, we can look into it, but for the most part it seems okay.

It needs to lower limb breakpoints, make me do more damage per hit, and increase the speed of my dsls.

It does all of these things, to varying degrees, the sum total of which comes out to about 5% more dps overall. You're right that dps isn't wholly relevant to pvp, but it's a good, central bar to compare weapon efficacy without having to break every stat down into precise numbers.

Eryl

Phaestus said:

Eryl said:

Phaestus said:

Going from 2 to 3 is still about a 5% against both players and denizens, which I think is pretty close to standard across artefact weapons (excluding lash, because I don't know exact lash numbers off the top of my head but I'm pretty sure they're not that).

This is all said with the disclaimer that weapons are nightmarish abominations and they have minds of their own, and there may be some edge cases where they're just not seeing the performance they should. If you have specifics, we can look into it, but for the most part it seems okay.

It needs to lower limb breakpoints, make me do more damage per hit, and increase the speed of my dsls.

It does all of these things, to varying degrees, the sum total of which comes out to about 5% more dps overall. You're right that dps isn't wholly relevant to pvp, but it's a good, central bar to compare weapon efficacy without having to break every stat down into precise numbers.

Aye, I agree with all of that.  The issue in this particular circumstance is that the marginal increases across the board don't really do anything to break new thresholds. @Atalkez recounted to me his experience of the speed increase being so marginal that it was effectively nonexistent, reporting the same balance speed on dsls for level 2 and level 3 scims.

It just comes down to it being an on paper increase worth the credits compared to the increase you receive in practice which does not feel worth the credits.  It doesn't need changed and nothing is broken.  In a damage group meta, that 5% dps increase spread out across three knights is a pretty massive deal.  It's just that the expectations need to be tailored to that and not 1v1.

Eryl

@Aerek and I worked through this!

The following change continues to highlight the limb on your prompt in orange if it is prepped, but now will also highlight that limb in yellow if it is one hit away from being prepped:

antonius.targetting.promptlimb = function(limb)
	local target = antonius.target
  return (antonius.targetting.isbroken(target, limb) and "<red>" or antonius.targetting.isprepped(target, limb) and "<orange>" or antonius.targetting.ispreppednext(target, limb) and "<yellow>" or "<white>") .. target.hits[limb]
end

This next part only applies to dual cutting.

This additional fix changes the purpose of the "isprepped" function: this function now returns TRUE if the limb would break with ONE RAZESLASH.

Is also changes the purpose of the "ispreppednext" function: this function now returns TRUE if the limb would break with ONE DOUBLESLASH.

Search for the "ispreppednext" function, then locate this piece:

elseif spec == "Dual Cutting" then
			local perhit = antonius.targetting.calculatedslperc(thits.required)
			return thits[limb] < 100.0 and thits[limb] + (perhit * 4) >= 100.0<br>

Change * 4 to * 2.

Next, search for the "isprepped" function, then locate this piece:

elseif spec == "Dual Cutting" then
			local perhit = antonius.targetting.calculatedslperc(thits.required)
			return thits[limb] < 100.0 and thits[limb] + (perhit * 2) >= 100.0<br>

Change * 2 to *1.

When you are attacking, the limb color on your prompt will now turn YELLOW if a DSL will break the limb, ORANGE if a RSL will break the limb, and RED when the limb is broken.

Aerek

I like that you give me half credit, when it was more like

Aerek: Will it work if I just change this one value?
Eryl: Yes, as long as you also insert [all the actual code work].

Trendor

Can anyone help me out with setting this up for monk?

Eryl

It will work out of the box for monk, with the exception of a change you need to make because of the recent additional line in limb attacks.

Find the Tekura folder. Look at each individual trigger. These are multi line triggers that do not yet account for the additional line. You have an easy option and a hard option:

Easy mode: Increase the line delta of each multi line trigger by 1.

Hard mode: Add the new limb damage feedback line into each multi line trigger in the correct location and also increase the line delta of each multi line trigger by one.

I just got home from work, but will provide picture examples shortly!

Eryl

@Trendor :

Verdell

When I switched from Runewarden to monk, the script started giving me trouble if I set a target.

By disabling "antonius.targetting.target(matches[2])" in my targetting alias, the error went away.

SVOF said: (svof): The problem was: stuff after the actual work failed ([string "Script: Tekura"]:45: attempt 

to perform arithmetic on local 'modifier' (a nil value))

Any thoughts?

Taryius

Just took a brief look at that line in the tekura script, and it looks like you haven't set a modifier for what level knuckles you have or that you're unknuckled. They're part of the formula used, so if it has no value the script will fire an error when it tries to perform the math, can't multiply by nothing.

I couldn't find any way to actual set that through an alias (maybe it shows up under tconfig for monks?) but if you need you can just define it in your targetting alias and it'll probably be fine.

antonius.targetting.conf.knuckles = 0

0 for unartied, 1 for Level one knuckles, 2 for two, 3 for three. Update the number in your alias/script if you ever upgrade your knuckles.