Just how strong is the speed brand on heavy weapons?

[Blomdor]

This is something I've always wondered a bit about, and after finding an executioner's axe of speed in my current game I figured it wouldn't hurt to ask the question. I'm currently using a +9 executioner's axe of chopping with +6 strength on top of it, and the speed branded executioner's is only +2 with +3 strength. That's a difference of +7 enchantment, the chopping brand, and +3 strength.

I have long been under the impression that the speed brand is very, very strong on hard-hitting weapons that it can't normally be found on, but I do not know exactly how strong it is or if it's been changed over the years. Obviously, one doesn't get the opportunity to test it out all too often. Is this +2 axe of speed worth thinking twice about?

[ajon]

Very strong. So weapons take only 66% as long to swing as normal. So if your big weapon is .7, its between .4 and .5 units. You essentially get 1.5X the number of swings as you normally would. So consider it a brand that adds 50% extra damage which is not affected by any resistances, but only by AC.

[duvessa]

Vorpal is +16.67% damage on average, regardless of AC.
Freezing or flaming is +25% damage on average (sans resistances), regardless of AC.
Speed is +50% damage on average, regardless of AC.

There's not much point in making it any more complicated than that.

[VeryAngryFelid]

Obviously, one doesn't get the opportunity to test it out all too often.

That's exactly type of questions fsim is supposed to answer. Spend 10 minutes to learn how to use it and then spend 5 more minutes to actually compare those weapons

Spoiler: show

I have no idea which of the weapons is better but I wouldn't be surprised if one of them is vastly superior

[Shtopit]

Obviously, one doesn't get the opportunity to test it out all too often.

That's exactly type of questions fsim is supposed to answer. Spend 10 minutes to learn how to use it and then spend 5 more minutes to actually compare those weapons

Spoiler: show

I have no idea which of the weapons is better but I wouldn't be surprised if one of them is vastly superior

From a game-design perspective, doesn't the fact that fsim is needed to have an answer represent an inherent problem, though?

[Siegurt]

Obviously, one doesn't get the opportunity to test it out all too often.

That's exactly type of questions fsim is supposed to answer. Spend 10 minutes to learn how to use it and then spend 5 more minutes to actually compare those weapons

Spoiler: show

I have no idea which of the weapons is better but I wouldn't be surprised if one of them is vastly superior

From a game-design perspective, doesn't the fact that fsim is needed to have an answer represent an inherent problem, though?

Only if you consider the question of "which is the absolute most optimal weapon to use in the close cases" to be important information for winning, which fortunately it's not, using Fsim to learn how the game works is perfectly reasonable though.

Choosing between two not-close weapons can be done easily without fsim, and choosing between two close weapons doesn't (significantly) matter for your chances of winning (it only matters for satisfying your desire to eke out every possible drop of power)

If you have two weapons and can't readily identify which one is better, given the in-game information provided and knowledge of what brands do, then it doesn't matter, for winning purposes, which one you select.

The fact is, combat is sufficiently random in crawl that tight margins aren't ones that you should ever come close to exploiting, every combat should either be a clear win by large margins, or a retreat and avoidance, if you're fighting combats with sufficiently tight margins that a few percentage points of offense matters, then you are playing riskily, and will die when rolls don't go your way, regardless of which weapon you selected.

[Airwolf]

Imagine drinking a potion of haste before every fight... (Well, OK, not quite.)

[Blomdor]

Thanks, guys. I'll try to learn fsim more and use it in these cases in the future. Duvessa's numbers are concise and helpful, so I shall keep that in mind. For the record, I lost the game in question being an idiot and either axe would have resulted in the same outcome, haha. I guess it didn't really matter which one I used. The question I really had was whether there was a clear advantage to be had here, and that seems to have been answered "no."

[ajon]

eh, I would say that 34% is not a negligible or marginal increase considering you were comparing speed to vorpal. The answer might be yes, not no. Of course executioners axe is super slow and takes tons of skill to make it usable. Generally other weapons will outdamage it until you get a pretty high skill.

[watertreatmentRL]

Huh, so not being straightforward with the player about how much damage various kinds of weapons do isn't a problem from a game design perspective because those details have no impact on winning the game... This makes me think maybe the existence of those details is a problematic from a game design perspective...

[VeryAngryFelid]

Huh, so not being straightforward with the player about how much damage various kinds of weapons do isn't a problem from a game design perspective because those details have no impact on winning the game... This makes me think maybe the existence of those details is a problematic from a game design perspective...

By the way I don't think devs treat small difference as irrelevant.
Otherwise we wouldn't have a 4% buff.

  Code:

Melee weapons with the vorpal brand do between +0% and +33% additional damage each hit, averaging +16.6%.
Prior to 0.13 this brand increased your damage between 0% and 25% each hit, averaging 12.5%.


http://crawl.chaosforge.org/Vorpal

[watertreatmentRL]

Right. It's odd grounds on which to defend the status quo: "actually, it doesn't make any difference."

[rigrig]

Vorpal is +16.67% damage on average, regardless of AC.
Freezing or flaming is +25% damage on average (sans resistances), regardless of AC.
Speed is +50% damage on average, regardless of AC.

There's not much point in making it any more complicated than that.

This seems pretty straightforward, but what about the enchantment?
Ignoring the str+, does simply adding 2 or 9 to the base damage give a somewhat decent picture, or does the higher to-hit have a significant influence?

[Siegurt]

I think a reasonable solution would be to make a mode that is scored, works offline, and has fsim, but none of the other wizmode options.

Another helpful thing would be to display the effect of brands in the in game help somewhere. I suspect if someone took the time to do that, and write up the code to make them findable from say ?/ that would likely be accepted.

[Siegurt]

Vorpal is +16.67% damage on average, regardless of AC.
Freezing or flaming is +25% damage on average (sans resistances), regardless of AC.
Speed is +50% damage on average, regardless of AC.

There's not much point in making it any more complicated than that.

This seems pretty straightforward, but what about the enchantment?
Ignoring the str+, does simply adding 2 or 9 to the base damage give a somewhat decent picture, or does the higher to-hit have a significant influence?

To hit doesn't make much difference, however the base damage is multiplied by your stats and skills, the enchantment is not.

If you want a rough approximation, each of fighting and weapon skill will roughly double your base damage at 27 (weapon skills do more than fighting for damage, and increase the speed of attacks as well)

Remember that the two multiply, so at 13.5 in each skill, each increases the damage by very approximately 50% so together it would be 1.5×1.5 = 2.25 (125% more base damage)

Generally speaking i usually just mentally double or triple the base damage when i dont want to sit down and do the math, because its easy, and for a fair sized chunk of the game it is close enough given that my skills are in that general zone.

The actual fractions are fighting/30 and weapon skill/25 if you dont want approximations.

Strength also plays a role, so you can probably bump up your base damage mentally by another 10-20% or so if you have higher strength.

[duvessa]

Enchantment doesn't change that vorpal is +16.67% damage on average, freezing/flaming is +25%, and speed is +50%. Those always hold, regardless of your enchantment, strength, accuracy, slaying, base damage, etc.

And fsim doesn't preserve the state of the game. It is very likely to change your HP, statuses, etc. not to mention the HP, statuses, etc. of monsters or even which monsters exist and where they are. If you allow it in non-wizmode games you may as well allow all the other wizmode features in non-wizmode games. It's also a terrible solution to the problem; you should be able to see how much damage your weapon does without running any trials at all, let thousands of trials.

[mattlistener]

I enjoy the variety of weapons and randarts as it currently exists, and would hate to see that variety curtailed because it is capable of producing, say, a +10 morningstar of crushing and a +0 eveningstar {+5 Str, +5 Dex}, which were not significantly different in their contribution toward winning (tweaking numbers as needed for the example).

Other games with random weapon generation and lots of possible properties calculate DPS and display it on the weapon, as an aid to comparison between weapons with many differences. It's understood that this number does not take into account playstyle differences (eg multitarget, stabbing, riposte) -- that would be up to the player.

Depending on getting that actual damage output in a specific combat is of course at the player's peril.

[Siegurt]

Enchantment doesn't change that vorpal is +16.67% damage on average, freezing/flaming is +25%, and speed is +50%. Those always hold, regardless of your enchantment, strength, accuracy, slaying, base damage, etc.

And fsim doesn't preserve the state of the game. It is very likely to change your HP, statuses, etc. not to mention the HP, statuses, etc. of monsters or even which monsters exist and where they are. If you allow it in non-wizmode games you may as well allow all the other wizmode features in non-wizmode games. It's also a terrible solution to the problem; you should be able to see how much damage your weapon does without running any trials at all, let thousands of trials.

Yeah, your right about fsim, i had forgotten about the stupid holes in it.

The problem with just calculating it is that deriving the amount of damage you do cant be done automatically from the actual damage dealing code, you have to hand craft a formula, which requires someone with decent math understanding, is subject to human error and has to be maintained if/when the attack code changes, I suspect nobody qualified has felt like adding a feature that has limited actual use, that is a possible maintenance headache.

It's probable that since the attack code has been stable for a while they could probably do this now without it causing any significant problems, but i think they had been treating the attack code as unstable since the redesign a while back.

While it is possible that they might so more changes, i personally don't think that they will happen frequently enough or in large enough measure that it should still present a real obstacle if someone wants to take the time to do so.

I suspect at this point the objections are more about someone qualified wanting to take the time to do it, rather than any real structural objection, also if we are going to provide damage feedback for weapons, it should probably happen at the same time as damage feedback for spells.

[syringe]

Remember that the two multiply, so at 13.5 in each skill, each increases the damage by very approximately 50% so together it would be 1.5×1.5 = 2.25 (125% more base damage)

It takes maxed skills to get that bonus. Base damage is good but not that good, Siegurt has a habit of lying about it. Strength was probably understated, a mere +8 adds 20%, and at around 25 str and max skills you can triple your base damage, quintuple at 60 str.

[Siegurt]

Remember that the two multiply, so at 13.5 in each skill, each increases the damage by very approximately 50% so together it would be 1.5×1.5 = 2.25 (125% more base damage)

It takes maxed skills to get that bonus. Base damage is good but not that good, Siegurt has a habit of lying about it. Strength was probably understated, a mere +8 adds 20%, and at around 25 str and max skills you can triple your base damage, quintuple at 60 str.

13.5 isnt maxed i mention both a maxed and "middle" point (even pointing out that the max was maxed). I consider +8 to be significantly higher strength (if you drop all your bonus stats into strength, that is +9) 60 strength is really out there in terms of things to think about, if your pointing out maxed bonuses.

Fwiw the scale i am using when i talk about high strength is:

  Code:

10- baseline
15- above average
20- high
25- very high
30- extremely high
60- wtf zig scummer post endgame strength


For comaprison:
27 fighting and weapon skill give x3.952 base damage at strength 10 (+ 295%)

My contention was that "if you have high strength you can mentally add another +10-20% to the base damage"

There doesn't seem to be a large inconsistency between that and

a mere +8 adds 20%, and at around 25 str and max skills you can triple your base damage, quintuple at 60 str.

Other than you claiming that at 25 strength and maxed skills you have less base damage than you have at maxed skills and 10 strength (maybe you weren't including fighting in your "maxed skills" definition, i am not sure?)

I dunno, if you want to say my suggestion about strength was understated (maybe it is, that is fine if so) you might as well actually use consistent numbers.

However if i did understate the impact of strength, that just makes base damage more valuable, not less, what do you think the bonus multiplier from strength is multipling?

[duvessa]

27 fighting and weapon skill give x3.952 base damage at strength 10 (+ 295%)

That's the MAXIMUM of the randomized bonus. The probability of getting that bonus is 0.0000137%. More info

[Siegurt]

27 fighting and weapon skill give x3.952 base damage at strength 10 (+ 295%)

That's the MAXIMUM of the randomized bonus. The probability of getting that bonus is 0.0000137%. More info

You're right, it's been long enough since I looked at that code I forgot that fighting and weapon skill are randomized when applying to base damage. I retract my earlier statement I did lie, apologies.
(Of course strength, enchantment and slaying are *also* randomized at that point, so when comparing slaying and enchantment to base damage, it doesn't really matter if you use the *average* bonus or the *max* bonus)