fsim Zero Skill Melee Study


Ask fellow adventurers how to stay alive in the deep, dark, dangerous dungeon below, or share your own accumulated wisdom.

Tartarus Sorceror

Posts: 1822

Joined: Thursday, 31st May 2012, 15:45

Post Tuesday, 8th March 2016, 04:00

fsim Zero Skill Melee Study

These are the results of a large fsim study I did a couple months or so ago to explore the question, "What is a good zero-skill weapon for book starts to pick up off the floor?" For the sake of completeness, rare weapons were included as well--only TSO-blessed ones were excluded.

The wizmode character was XL:1, STR 12 DEX 12 (in retrospect these were a bit high), with zero in all skills. +0 weapon, no brand. The "High AC" target is an orc warrior in chain mail (I made it with &m and selected it manually for each &f, so it always had the same AC and EV). The "normal" target is an ogre, and the "High EV" target is an adder. The values in the table are fsim's "AvEffDam", averaged over three runs, in thousandths of a damage point. Thus the dagger has an AvEffDam of 0.267 against the orc warrior, 1.200 against the ogre, and 0.700 against the adder.

  Code:
         High AC  Normal  High EV
Dagger     267     1200      700
QuickBl    567     2233     1300
ShortSw    467     1733      967
Rapier     533     1867     1000
Falchion   699     1900     1000
LongSw     800     2333     1067
Scimitar   900     2667      900
DemonBl   1200     3200     1200
DoubleSw  1333     3300     1267
GreatSw   1133     3100      900
TripleSw  1167     3067      667
HandAx     500     1733      900
WarAx      833     2367     1000
BroadAx    900     2567      800
BattleAx   933     2633      500
ExecAx     800     2533      100
Whip       400     1700      800
Club       300     1200      600
Mace       600     1833     1000
Flail      733     2267      967
MStar     1000     2700      900
DemonWh   1133     3300     1500
DireFlail 1067     3067      867
EStar     1333     3267     1233
GreatMc   1100     3000      667
GiantC    1200     3500      167
GSC        867     3167      133
Spear      467     1767      967
Trident    700     2200     1067
Halberd    867     2600      733
Scythe     700     2100      433
DemonTr   1200     3000     1400
Glaive    1000     2767      767
Bardiche   800     2567      100
QStaff     900     2567     1367
Lajatang  1300     3533     1000
Stone      100      500      200
Tomahawk   467     1867      967
Javelin   1167     3267     1567
LrgRock   3300     6900     3367
HSling(s)  500     1867      900
HSling(b)  800     2400     1267
GSling(s)  767     2300      867
GSling(b)  967     2767     1000
Shortbow   767     2267     1133
Longbow   1267     2900     1267
HandCr    1167     2800     1600
Arbalest  1567     3233     1600
TripleCb  1667     3233     1367

Last edited by MainiacJoe on Tuesday, 8th March 2016, 10:57, edited 1 time in total.
Won (52). Remaining (15): 5 species: Ba, Fe, Mu, Na, Op; 5 Backgrounds: AM, Wr, Su, AE, Ar; 5 gods: Jiyv, newNem, WJC, newSif, newFedh

For this message the author MainiacJoe has received thanks: 11
Airwolf, Bart, crate, duvessa, Lasty, nago, Sprucery, Utis, yesno, ZoFy and one more user

Ziggurat Zagger

Posts: 6394

Joined: Tuesday, 30th October 2012, 19:06

Post Tuesday, 8th March 2016, 06:22

Re: fsim Zero Skill Melee Study

Note that while your numbers seem accurate and certainly reasonable enough, I'd question the choice of an ogre as a representative "middle" sample, since they have almost no AC and low EV, they show more damage done than say, a gnoll, hobgoblin, goblin or 'vanilla' orc would, an expected overall average for creatures you'd encounter with 0 skill would be closer to the 'high ev' or 'high ac' numbers than it would be to your chosen "middle" sample (all of the mentioned critters are closer to an adder than they are to an ogre, defense wise)

While you can compare numbers *within* a column easily enough, comparing the columns to each other might be misleading.

Although a "little to no defense" sample is useful too, representing something more like a 'best case scenario' average (Although I a Jelly or slime creature has even slightly lower defense than an ogre, jelly's you can't fsim properly because of corrosion, but slime creatures show a noticeably higher avgeffdmg than ogres even)
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!
User avatar

Vestibule Violator

Posts: 1591

Joined: Saturday, 3rd August 2013, 18:59

Post Tuesday, 8th March 2016, 07:17

Re: fsim Zero Skill Melee Study

Wow, large rocks are OP, god damn.
To all new players: Ignore all strategy guides posted on the wiki, ask questions in the Advice forum, players with lots of posts normally have the best advice.

crawl.akrasiac.org:8080 <- take this link to play online or spectate.

Tomb Titivator

Posts: 885

Joined: Sunday, 28th June 2015, 14:44

Post Tuesday, 8th March 2016, 08:36

Re: fsim Zero Skill Melee Study

They're also rare enough that if you throw them at everything you run out.

Abyss Ambulator

Posts: 1221

Joined: Thursday, 10th March 2011, 19:45

Post Tuesday, 8th March 2016, 08:53

Re: fsim Zero Skill Melee Study

Aside from large rocks, it's quite interesting how good hand crossbow is.

Tartarus Sorceror

Posts: 1822

Joined: Thursday, 31st May 2012, 15:45

Post Tuesday, 8th March 2016, 10:53

Re: fsim Zero Skill Melee Study

Siegurt wrote:Note that while your numbers seem accurate and certainly reasonable enough, I'd question the choice of an ogre as a representative "middle" sample, since they have almost no AC and low EV, they show more damage done than say, a gnoll, hobgoblin, goblin or 'vanilla' orc would, an expected overall average for creatures you'd encounter with 0 skill would be closer to the 'high ev' or 'high ac' numbers than it would be to your chosen "middle" sample (all of the mentioned critters are closer to an adder than they are to an ogre, defense wise)

While you can compare numbers *within* a column easily enough, comparing the columns to each other might be misleading.

Although a "little to no defense" sample is useful too, representing something more like a 'best case scenario' average (Although I a Jelly or slime creature has even slightly lower defense than an ogre, jelly's you can't fsim properly because of corrosion, but slime creatures show a noticeably higher avgeffdmg than ogres even)


The purpose of the ogre column was to analyze just the weapon output itself. I almost used "ghost", a defunct monster that is available in fsim with AC 0 and EV 0. So the no AC and low EV was actually what I was aiming for. Whether it's useful information, that's another question.
Won (52). Remaining (15): 5 species: Ba, Fe, Mu, Na, Op; 5 Backgrounds: AM, Wr, Su, AE, Ar; 5 gods: Jiyv, newNem, WJC, newSif, newFedh

Tartarus Sorceror

Posts: 1822

Joined: Thursday, 31st May 2012, 15:45

Post Tuesday, 8th March 2016, 10:56

Re: fsim Zero Skill Melee Study

Jeremiah wrote:Aside from large rocks, it's quite interesting how good hand crossbow is.


This shows how weapon accuracy is more important than usual at zero skill. Other examples: quarterstaff vs adder, giant club vs GSC, long sword vs scimitar with adders.
Won (52). Remaining (15): 5 species: Ba, Fe, Mu, Na, Op; 5 Backgrounds: AM, Wr, Su, AE, Ar; 5 gods: Jiyv, newNem, WJC, newSif, newFedh
User avatar

Dis Charger

Posts: 2056

Joined: Wednesday, 7th August 2013, 08:25

Post Tuesday, 8th March 2016, 11:06

Re: fsim Zero Skill Melee Study

I often say GSC are useless in early game and these numbers seem to confirm that. AvEffDam or 1.33 versus an adder? [It's too slow and inaccurate to use without a lot of skill investment; you will kill yourself if you try to use it without extensive training FIRST.]
I'm beginning to feel like a Cat God! Felid streaks: {FeVM^Sif Muna, FeWn^Dithmenos, FeAr^Pakellas}, {FeEE^Ashenzari, FeEn^Gozag, FeNe^Sif Muna, FeAE^Vehumet...(ongoing)}

Ziggurat Zagger

Posts: 8782

Joined: Sunday, 5th May 2013, 08:25

Post Tuesday, 8th March 2016, 11:15

Re: fsim Zero Skill Melee Study

That is a very specific artifact of having accuracy below a certain level. You can see it also happens with bardiche and exec axe. The issue disappears once you get 3 levels of weapon or fighting skill, which is does not take long for an ogre. Notice that even at 0 skill it's fine against low ev stuff.

Slime Squisher

Posts: 387

Joined: Monday, 15th August 2011, 16:31

Location: Frankfurt

Post Tuesday, 8th March 2016, 11:21

Re: fsim Zero Skill Melee Study

MainiacJoe wrote:[...]averaged over three runs [...]


You can increase fsim's sample rate in your config with, e.g.

  Code:
fsim_rounds = 10000
"... while we / Unburden'd crawl toward death." -- King Lear I,1
User avatar

Dis Charger

Posts: 2056

Joined: Wednesday, 7th August 2013, 08:25

Post Tuesday, 8th March 2016, 11:26

Re: fsim Zero Skill Melee Study

duvessa wrote:That is a very specific artifact of having accuracy below a certain level. You can see it also happens with bardiche and exec axe. The issue disappears once you get 3 levels of weapon or fighting skill, which is does not take long for an ogre. Notice that even at 0 skill it's fine against low ev stuff.
And it's delay is so high you virtually HAVE to killhole with it...
I'm beginning to feel like a Cat God! Felid streaks: {FeVM^Sif Muna, FeWn^Dithmenos, FeAr^Pakellas}, {FeEE^Ashenzari, FeEn^Gozag, FeNe^Sif Muna, FeAE^Vehumet...(ongoing)}

Tartarus Sorceror

Posts: 1822

Joined: Thursday, 31st May 2012, 15:45

Post Tuesday, 8th March 2016, 12:28

Re: fsim Zero Skill Melee Study

The issue, I think, is not whether GSC does more AvEffDam than say a flail. That's obvious: it does. The issue is, is it actually possible to wring this full potential out of the weapon in the early dungeon? The effect of the long delay is that it gives enemies opportunities to swing more than once for each of your swings, and unlike in fsim, with a couple early bad rolls you can run out of HP before the law of averages catches up with your damage output.

If I understand duvessa correctly, he thinks it *is* possible to use a GSC to full effect at very low skill. I'm sympathetic to bcraden's argument (daggaz made it too in the CYC "thought experiment" thread) because it makes sense that you might die before you get a chance to swing again. But for all its obviousness, it may well be wrong.
Won (52). Remaining (15): 5 species: Ba, Fe, Mu, Na, Op; 5 Backgrounds: AM, Wr, Su, AE, Ar; 5 gods: Jiyv, newNem, WJC, newSif, newFedh
User avatar

Barkeep

Posts: 4434

Joined: Tuesday, 11th January 2011, 12:28

Post Tuesday, 8th March 2016, 17:24

Re: fsim Zero Skill Melee Study

The point I kind of get from this is "if you're going to hit popcorn with a 0-skill weapon, bigger weapons are better but don't sweat it too much because it's popcorn anyhow."

Fair synopsis?
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

Tomb Titivator

Posts: 885

Joined: Sunday, 28th June 2015, 14:44

Post Tuesday, 8th March 2016, 17:49

Re: fsim Zero Skill Melee Study

njvack wrote:The point I kind of get from this is "if you're going to hit popcorn with a 0-skill weapon, bigger weapons are better but don't sweat it too much because it's popcorn anyhow."

Fair synopsis?

yes
User avatar

Slime Squisher

Posts: 354

Joined: Thursday, 14th April 2011, 17:28

Post Tuesday, 8th March 2016, 21:34

Re: fsim Zero Skill Melee Study

My synopsis is this:

Use Hand Crossbows and Large Rocks.
infinitevox on akrasiac & berotato
Busy dying horrible deaths from chugging too many pots of Mutation.
User avatar

Dis Charger

Posts: 2056

Joined: Wednesday, 7th August 2013, 08:25

Post Tuesday, 8th March 2016, 21:45

Re: fsim Zero Skill Melee Study

I'd previously done a skill curve with Crossbows, Hand Crossbows were better than Arbalests until skill 9, then Arbalests were better than triple crossbows until skill 21. [Assuming all of them were same enchantment level and brand.]
I'm beginning to feel like a Cat God! Felid streaks: {FeVM^Sif Muna, FeWn^Dithmenos, FeAr^Pakellas}, {FeEE^Ashenzari, FeEn^Gozag, FeNe^Sif Muna, FeAE^Vehumet...(ongoing)}
User avatar

Slime Squisher

Posts: 354

Joined: Thursday, 14th April 2011, 17:28

Post Tuesday, 8th March 2016, 23:15

Re: fsim Zero Skill Melee Study

I'm assuming your skill curve was strictly for damage and didn't take defense (shields) into account? (For the purposes of "better" )
infinitevox on akrasiac & berotato
Busy dying horrible deaths from chugging too many pots of Mutation.

Vestibule Violator

Posts: 1601

Joined: Sunday, 14th July 2013, 16:36

Post Wednesday, 9th March 2016, 00:38

Re: fsim Zero Skill Melee Study

njvack wrote:The point I kind of get from this is "if you're going to hit popcorn with a 0-skill weapon, bigger weapons are better but don't sweat it too much because it's popcorn anyhow."

Fair synopsis?

When I looked at the topic previously, my question was "if I attack with a weapon when my Wz finds itself next to a jackal/adder with no MP, which weapon should I use?"
User avatar

Barkeep

Posts: 4434

Joined: Tuesday, 11th January 2011, 12:28

Post Wednesday, 9th March 2016, 14:43

Re: fsim Zero Skill Melee Study

Hurkyl wrote:When I looked at the topic previously, my question was "if I attack with a weapon when my Wz finds itself next to a jackal/adder with no MP, which weapon should I use?"

So, triple sword or eveningstar ;)
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

Tomb Titivator

Posts: 885

Joined: Sunday, 28th June 2015, 14:44

Post Wednesday, 9th March 2016, 15:37

Re: fsim Zero Skill Melee Study

Hurkyl wrote:
njvack wrote:The point I kind of get from this is "if you're going to hit popcorn with a 0-skill weapon, bigger weapons are better but don't sweat it too much because it's popcorn anyhow."

Fair synopsis?

When I looked at the topic previously, my question was "if I attack with a weapon when my Wz finds itself next to a jackal/adder with no MP, which weapon should I use?"

The answer is probably "whatever weapon you're actually carrying"

Slime Squisher

Posts: 395

Joined: Monday, 28th April 2014, 19:50

Post Wednesday, 9th March 2016, 22:42

Re: fsim Zero Skill Melee Study

ydeve wrote:
Hurkyl wrote:
njvack wrote:The point I kind of get from this is "if you're going to hit popcorn with a 0-skill weapon, bigger weapons are better but don't sweat it too much because it's popcorn anyhow."

Fair synopsis?

When I looked at the topic previously, my question was "if I attack with a weapon when my Wz finds itself next to a jackal/adder with no MP, which weapon should I use?"

The answer is probably "whatever weapon you're actually carrying"


Although the player has a choice of what to carry. It's interesting to know that, for example, a 0-skill whip is better than a club for fighting adders, but a mace is better than either. The quarterstaff is by far the best among the easily available weapons, and several ranged weapons are surprisingly worthwhile.

Abyss Ambulator

Posts: 1217

Joined: Sunday, 14th April 2013, 04:01

Post Thursday, 10th March 2016, 20:33

Re: fsim Zero Skill Melee Study

Three wins: Gargoyle Earth Elementalist of Ash, Ogre Fighter of Ru, Deep Dwarf Fighter of Makhleb (0.16 bugbuild :( )

For this message the author TeshiAlair has received thanks:
MainiacJoe

Return to Dungeon Crawling Advice

Who is online

Users browsing this forum: No registered users and 9 guests

Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software for PTF.