Artefact Bows Nerf - Announce 3940

Santar

So my choice now is to either aim, and sacrifice about 20% of my speed per shot, or just snipe without aim, and sacrifice 20% of my accuracy. Any way you look at it, Serpents got the raw end of this deal.

Santar

What I'm saying is that snipe has -never- required the use of aiming ever before, so requiring it now is a massive nerf. 

Cahin

Daeir said:

75% accuracy with marksman, l3, aim, wind and snipe is poor, even if 20 shots is a relatively small sample size.

No, 20 shots being a small sample size has everything to do with his results.  Sample size matters a lot. 

Law of Large Numbers 

Santar

I don't know. Definitely doesn't seem like snipe is offering an accuracy boost.

I scored 75% accuracy on my test while shooting 2+ rooms away, lupine bow, snipe, marksman trait. Some serious BS.

Edit: Sample size don't matter because it shouldn't ever miss, period.

Eld

Tecton said:

Ok, so I've just loaded up some changes that bring things back up a little, some of the details:

• 4% overall increase in artefact bow accuracy.
  
• 5% overall increase in accuracy from the WIND ability (Subterfuge and Chivalry)
• 10% overall increase shooting at an "aimed" target. 

To give some rough numbers (assuming a normal, unafflicted, unobstructed target):

• Artefact bow, no trait, no abilities, 2 or more rooms away - ~40% accuracy
• Artefact bow, trait, no abilities, 2 or more rooms away - ~50% accuracy
• Artefact bow, trait, no abilities, adjacent room - ~65% accuracy
• Artefact bow, trait, abilities, no aiming, 2 or more rooms away - ~80% accuracy
• Artefact bow, trait, abilities, no aiming, adjacent room, ~95% accuracy
• Artefact bow, trait, abilities, aiming, 2 or more rooms away - 100% accuracy with a bonus through ice/stonewalls.
• Artefact bow, trait, abilities, aiming, adjacent room - 100% accuracy with a larger bonus through ice/stonewalls.

Of course, things are not always this black and white, since it's entirely possible that you'll get unlucky 5 times in a row due to bad random chance in cases where you're under 100% accuracy. 

Tecton said:

Santar said:

Disagree heavily with the decision to make snipe miss for Serpents.

I just tested 20 shots, and hit 15/20 in Serpentform, with a level 3 bow, marksman trait.

Snipe still has the same accuracy boost as it's always provided to regular shooting.

Just to clarify, in your first set of numbers, "abilities" versus "no abilities" refers to just SHOOTing with, say, a darkbow and the Darkbows ability in either Subterfuge or Chivalry? How should Snipe affect those numbers?

20 shots is indeed a small sample size, but from the rough numbers given it sounds like a serpent sniping with a Lupine, aiming, and the marksman trait, unobstructed from an adjacent room, should miss rarely enough that 5/20 misses would be pretty significant.

Cahin

Daeir said:

Cahin said:

Daeir said:

75% accuracy with marksman, l3, aim, wind and snipe is poor, even if 20 shots is a relatively small sample size.

No, 20 shots being a small sample size has everything to do with his results.  Sample size matters a lot. 

Law of Large Numbers 

You don't need to lecture me on statistics, thanks. I'm aware of the limitations surrounding a small sample size, but what I am saying is that if Tecton has extrapolated a projected accuracy of 100% with a selection of factors that imports a lesser effect, then another selection of factors with a greater effect should not reflect a lower percentage to hit, even across a comparatively small sample size. That means that there's a statistical Type I error in either what Tecton is saying or the code. It's probably the former.

Basically: if scrub archers are getting 100% accuracy (supposedly), there's no way superior serpent master race archers should be getting anything but 100%, which isn't the case from both Santar's testing and my own.

The 100% is with aiming.

eta: Santar already stated that he wasn't using the aiming ability.

eta2: I think requiring aiming to guarantee a hit is very healthy progress for balancing LoS. One of it's negatives is the ability to lean on your enter key, move into LoS with a group, and instakill the opponent. Requiring aiming to guarantee that your shot will hit provides a willpower penalty to those doing this.

Tecton

"Abilities" is for a serpent with sniping, for a knight, you probably want to knock ~10% off those numbers.

I've got some major datalogging running for every arrow shot in the live game, with all possible live scenarios. I'll look at the data tomorrow and potentially tweak things some more.

Tecton

Daeir said:

Aiming infers a 10% boost and from what Tecton posted, so does Marksman. Having marksman means there should be no need to aim effectively, unless you want to stack the accuracy.

On top of that, Snipe is supposed to offer a significant accuracy boost - which obviously isn't happening, or it isn't significant at all.

Aiming gives a lot more than 10%, but I capped things at 100% for the sake of the table.

Jarrod

Based purely on the % table Aiming is at least 20%, which is huge.

Katzchen

Seriously, 50% chance to hit someone 2 or more rooms away with an artefact bow? There is absolutely NO way I would have paid for shooting to use my bow in dragon if that was what it gave. If it's going to be that poor, offer refunds on bows/shooting upgrades.

Tecton

You still have a (practically) 100% chance to hit with meteor arrows, which is traditionally one of the main reasons people have bought artefact bows if they're a non-archery class. 

That said, as I mentioned above, those are just how things currently stand, the numbers may change, but don't expect them to be at the value they were when the bug was in place (which was ~90% chance to hit without any bowmanship abilities). 

Once all of the changes are finalised, we can investigate the option of refunds on the bow/shooting if you purchased your bow during the period where the bug was active.

Katzchen

I have always been able to hit people in dragon pretty much 100% unless walls. I bought it so I could raid in either. I can shoot meteors in lesser, so I would just lesser always to raid now, because of the huge difference in accuracy, hence the shooting is now useless.

Arador

Are the numbers before things like dex, weaving or avoidance skills are considered?

Blujixapug

Pretty sure bow accuracy for non-bow classes, via artifact bows, was always around 50% before the Marksman trait came into existence. Unimpressive, but an infinite improvement over having no ranged capabilities at all for several classes.

Jhui

Arador said:

Are the numbers before things like dex, weaving or avoidance skills are considered?

don't think any of those matter.  Possibly just the shooter's capability and arrowcatch/projectiles/wall

Sasiya

When posting those numbers in this thread, are you referring to a level 1 bow or a level 3 bow?

Hirst

Coming from someone with a level 3 bow...thank you Tecton!

Hopefully this weeds out the people who rely on LoS for combat (which is boring as hell) instead of melee.

Melodie

I have a question!

In HELP ARTEFACTS OFFENSIVE under bows, it mentions "... suffice to say that the more expensive bows are more accurate and do more damage."

Does that hold true, or is it intended to be a damage/speed increase only? If the latter, I'll likely not even bother upgrading my bow, honestly.

Simoln

9 shots out of how many? The number of misses is irrelevant without something to compare against.



(Though, I do agree, missing against prone/sleeping is lolwrong)

Tecton

Melodie said:

I have a question! In HELP ARTEFACTS OFFENSIVE under bows, it mentions "... suffice to say that the more expensive bows are more accurate and do more damage." Does that hold true, or is it intended to be a damage/speed increase only? If the latter, I'll likely not even bother upgrading my bow, honestly.

Artefact bows are more accurate than non-artefact bows, yes. Each level of artefact bow is more accurate than the last as well (although this isn't a significant jump)

Tecton

Daeir said:

I've missed 9 shots across 3 kills on people that are asleep. This is with aiming, sniping, marksman trait and a l3 bow, 1 room adjacent.

Something is seriously wrong.

Sleeping has never increased the chances of arrows hitting, paralysis does. This may change in the future, but this is the way it has always functioned.

Simoln

Beause people roll around THAT much in their sleep?

Jules

Dunn said:

Not large enough of a nerf. You still can barely (if at all) tank a Lupine. I'd like to see more firewall kills other than rainstorm so we can shift the meta away from LoS because that shit is boring.

So does anyone know if more ways to take down firewall are being considered or not?  I'd asked about this awhile back in quick questions, so maybe not, but thought I'd take the opportunity to ask again.  

Arador

Hack into it. Or sleep with the IT guy and try to tease the password out of him.

Tecton

The bonus accuracy added by the SNIPE ability has been increased slightly. 

Dunn

The only way to kill firewalls now is rainstorm in Sylvan elementalism. I wouldn't mind if standard flood killed them too. It makes sense.

@Tecton

Tesha

Allowing Magi to make an expensive (lots of ice, few thousand gold, long time of enchanting) sigil for it sounds neat. Okay with flood, too.

Aliath

Or could just let an icewall cast by a magi 'overwrite' it

Nakari

everybody's out to take my abilities...

Tecton

Firewalls don't currently interfere with archery accuracy, just stone and ice. As we look at more ways to try and bring ranged and melee combat into unison, there may be changes in that regard.