Skill Points / Base Damage Data [UPDATED]

[MainiacJoe]

[EDIT] Take minmay's correction and suggestion into account.

Here is each weapon sorted not only by base damage / 10-aut turn at mindelay, but also the skill points needed to reach mindelay divided by that number. Columns are base damage per turn, skill points divided by base damage / turn, percent ranks for each column, and then sum-squared of the percent ranks. the list is sorted descending by SumSq.

Somewhat doubtful this is useful information, but still, I did it so I may as well post it.

  Code:

Weapon  BD/T  SP/(BD/T)  BD/T  SP/(BD/T) SUMSQ
SacSc    24      175      84      97      129
DWhip    22      191      79      95      123
Laj      23      263      82      90      121
EudBl    22      194      71      92      116
GClub    29      378      97      63      116
DireF    22      277      71      87      112
Trish    22      277      71      87      112
Qblad    17      108      37     100      107
GSC      31      565     100      24      103
DBlad    20      300      50      82       96
DTrid    20      300      50      82       96
ExecX    26     1050      92       5       92
Bard     26     1050      92       5       92
GrtMc    24      574      87      21       89
3xSwd    24      908      87       8       87
EStar    21      383      61      58       84
Whip     12      350      13      76       78
Spear    12      350      13      76       78
ShSwd    12      350      13      76       78
Dagg      8      350       0      76       76
QStaf    17      360      37      66       75
2xSwd    20      410      50      53       73
GrtSw    21      504      61      37       71
Scim     16      382      32      61       68
BatAx    21      651      61      16       63
Glaiv    21      651      61      16       63
Trid     15      400      29      55       62
MStar    19      442      42      45       61
Halb     19      442      42      45       61
Flail    14      420      26      50       56
Rpier    14      429      24      47       53
Scyth    20     1350      50       0       50
BrdAx    19      582      42      19       46
LngSw    13      467      21      40       45
WarAx    16      522      32      29       43
HndAx    12      514       8      34       35
Falch    12      514       8      34       35
Mace     11      525       5      26       27
Club      8      720       3      11       11


[duvessa]

Your base stats for scimitar are wrong; it only has 11 base damage, whereas demon blades and tridents have 12.

You should really multiply the numerator by the minimum delay of the weapon, since different weapons have different min delays. A weapon with 10 base damage and 1.3 base delay is rated the same as a weapon with 10 base damage and 1.4 base delay in this system, even though the first one is strictly better at all skill levels. This will stop the chart from overrating scimitars, for example.

[4Hooves2Appendages]

Looking at this table makes me want to construct a 'club challenge', but I'm not quite sure how yet.

Perhaps:
Choose club or pick up and wield the first club you find.
Never change weapon.
Don't do damage with anything other than the club (no damaging spells, wands, scrolls or god powers, no summons), or aux attacks / mutations I guess.
No cats, sorry.
Agnostic on statue form.
Hexes / utility spells are fine.
Items are OK too.

It's not particularly punishing, but probably too tedious for most people. But then, people try MuMo, so...

[Airwolf]

I think it's done like this:
There is no way anyone could win using only a club for damage!

[Leszczynek]

ITT we scientifically prove that Maces & Flails are not only the most common and the most boring weapons, but also the best. I honestly don't use any other melee weapons unless I'm looking for a challenge or specifically want to use something else for reasons unrelated to winning the game.

[Siegurt]

ITT we scientifically prove that Maces & Flails are not only the most common and the most boring weapons, but also the best. I honestly don't use any other melee weapons unless I'm looking for a challenge or specifically want to use something else for reasons unrelated to winning the game.

Note that while I don't specifically disagree with you, this chart doesn't factor in cleave, reaching, riposte, slaying, weapon enchantment, aux attacks or brands (probably other stuff too), taking it as even moderately representative of the whole picture is unlikely to have a high degree of accuracy, drawing those sorts of conclusions from this data is specious.

[4Hooves2Appendages]

In a lot of cases it would surely depend on two other important factors:
1) what you actually have available
2) aptitude (Mf, HO, etc.)

[tasonir]

I think it's done like this:
There is no way anyone could win using only a club for damage!

I'm pretty sure this has already been done with spears. Now spears have one more base damage and reaching, so they're clearly better, but I imagine someone out there could probably manage to do it with a club.

Given enough patience and time, eventually someone is going to come across a +6 slaying ring early on, and then the club starts to be decent. If they later also get a +4 slaying randart boots, now you're looking like it's actually possible.

No, I won't be doing this myself

[tasonir]

While I agree it isn't terrible accurate or useful, if you wanted to make the chart more complete you could include unarmed in the chart at specific levels for comparison? Say, include UC 10, UC 15, UC 20, UC 25, UC 27?

[duvessa]

While I agree it isn't terrible accurate or useful, if you wanted to make the chart more complete you could include unarmed in the chart at specific levels for comparison? Say, include UC 10, UC 15, UC 20, UC 25, UC 27?

The chart uses raw base damage without taking skill bonus into account. If you included unarmed in that it would be incredibly misleading and make unarmed look far better than it actually is. Including unarmed would only make sense if you included the skill bonus for weapons.

[tasonir]

Ah, true. You'd want to add another column then, and the UC entries should likely have the rest of the columns blank except for the final column? This may be more work than it's worth, but hey