Bringing some love to evokable items


Although the central place for design discussion is ##crawl-dev on freenode, some may find it helpful to discuss requests and suggestions here first.

Mines Malingerer

Posts: 47

Joined: Thursday, 1st March 2012, 21:31

Post Sunday, 1st April 2012, 06:00

Bringing some love to evokable items

In my games of crawl and in reading the forums/wiki I've come to believe that some of the evokable (misc) items could use some love. Here are my assertions

  • Evocation provides some magical options to non-spell casters who acquire the items (Spells are generally better than the equivalent item, use food instead of charges or chance of breaking, but require light armor)
  • Evocation effects should get better as xp is applied to the skill. Somebody with 20 levels in a skill should have much better luck with an item than 0 levels.
  • I'm excluding wands, rods [except one], and decks in this discussion, since I believe they do get better with higher skill levels

The remaining items include:

  • Box of Beasts
  • Efreet Flask
  • Disc of Storms
  • Tome of Destruction (Tome of Power in source code)
  • Elemental Summoning (Air [Fan], Fire [Lamp], Earth[Stone]
  • Rod of Striking

I'd assert that for many of these items, having 20 levels of evocation skill would have some small effect, but make most of them unreliable enough to be truly useful. Here's what I can tell from reading the source code, and my proposal for each.

Box of Beasts
  Code:
     60 + Ev / 100 chance of summoning random animal, 1/6 chance of breaking on fail.
     100 / ((EV * 100) + 500) of NOT being friendly
     Given 0 skill, you'll get 11 uses out of it before you have a 50% chance of it breaking. Each use has 80% chance of being friendly.
     Given 20 skill, you'll get 20 uses out of it before you have a 50% chance of it breaking. Each use has 96% chance of being friendly.
     Animal doesn't get better with skill

Roughly, it summons something between spammables (3 of 11) and ice beast/yak strength. While I think the break rate is too high, it at least gets better with additional skill points.
I'd propose making the chance of breaking on fail scale with evocation, so skilled users are less likely to break it.

Efreet Flask
  Code:
 300 + (Ev * 10)/ 600 chance of summoning friendly
     Given 0 skill, 50% chance of friendly
     Given 20 skill, 83% chance of friendly
     Efreet doesn't get better with skill

"Hydra weapon in a bottle" is what the wiki calls this. While a friendly Efreet is nice, it is single use, has only have a 4/5s chance of being friendly even with massive skill investment.
I'd propose a 20 skill user should have 95%+ chance of friendly, else why would you actually use it in battle? If it isn't to be used in battle, we should remove and have it drop flaming weapons instead.

Disc of Storms
  Code:
(30 - Ev) / 100 chance of failure [checked twice]
     Given 0 skill, 49% chance of working
     Given 20 skill, 81% chance of working
     Damage scales with skill.

Scales with damage and chance of working. No theoretical complaints.

Tome of Destruction (Tome of Power in source code)
  Code:
     7/50, then 2/43, then 2/41, then 3/39, then 1/36 chance of failing - 70% chance of working, else bad effect.
     3 / 39 * 1 / 5 chance of breaking 1.53% chance - 44 uses before 50% chance of it breaking
     Damage scales with skill but breaking chance does not get better

A rare "uber wand" effect, able to cast anything from crystal spear down to magic dart, randomly, scaling on evocations. Even with 20 skill, you chance is 1d23 + 1d6 > 24 of getting crystal spear. However, the 70% chance of working and possible breakage would as the wiki puts it: "The only characters who might find this useful are Trog worshipers, and even for them it is of somewhat limited use. High Evocations skill does not make this item any safer, though it improves its destructive capacity significantly."
I'd like to propose making the chance of failure and chance of breaking scale based on evocation.

Elemental Summoning
  Code:
 (EV * 100 / 3000) chance of working
     Given 0 skill, 0% chance of working
     Given 20 skill, 66% chance of working
     Friendly based on # summoned and elemental skill
     Air = ((Air_Skill - (3 * Number_of_Air_Elementals)) / 15) * 96/100 chance of friendly
     Earth = ((Earth_Skill - (5 * Number_of_Earth_Elementals)) / 5 * 96/100 chance of friendly

And we come to the three elemental summoning items. Even with 20 skill (and 66% chance of working on a given invocation), you need elemental skills to have a chance to get friendly elementals. Those with elemental skills most likely can use the spell, which is only level 4, and those without would be crazy to use it.
I have two suggestions. I'd propose making the summoned elementals friendly based on the greater of elemental magic or evocations (divided by some number, say 2). That way, if you have the elemental skills, you can use them, but if you don't it isn't useful given enough practice. I'd also like to propose allowing xp to be trained on the appropriate elemental magic school of the owner of this (less important if first proposal accepted).

Rod of Striking
  Code:
 80% chance of hit (?? hard to figure out)
     Does what looks like 1d(8 + pow / 5) but I can't figure out of power is always zero. The wiki thinks so, but that's the wiki.

Not choosing this for your starting artificer package was the power gamer question in previous versions of crawl.
I'd like to propose making it suck less at the very least, and possibly re-adding it as a part of the artificer starting package (or an option).

I really like the idea behind these items, but from what I can tell many of them are never used by good players, unreliable enough to use in difficult combat. Wands, rods, and decks all get reasonably better with increased skill, I think these items should as well.

One last note, if we made these items actually useful, I'd like to fixup the acquirement code to be more likely to give these items (including the book) if you selected misc if you have high evocation.

I'm very happy to hack up a patch with my opinions coded if people on the dev team think these ideas have some merit.
Won 2 Games - MiFi (2)

Ziggurat Zagger

Posts: 4055

Joined: Tuesday, 10th January 2012, 19:49

Post Sunday, 1st April 2012, 06:13

Re: Bringing some love to evokable items

All Tome of Destruction needs is to not destroy itself (and a massive price drop in shops, since it should really cost about 600 or so gold instead of 2400). It's pretty good as-is but the fact that it blows itself up is quite disappointing.

All Disc of Storms needs to be good is to be usable with rElec.

Ziggurat Zagger

Posts: 6393

Joined: Friday, 17th December 2010, 18:17

Post Sunday, 1st April 2012, 13:03

Re: Bringing some love to evokable items

Please check out the many ideas for evocables on the dev wiki.

Mines Malingerer

Posts: 47

Joined: Thursday, 1st March 2012, 21:31

Post Sunday, 1st April 2012, 14:09

Re: Bringing some love to evokable items

Grimm, that looks awesome. I'm going see if I can't hack up a patch implementing some of those.
Won 2 Games - MiFi (2)

Dungeon Master

Posts: 1531

Joined: Saturday, 5th March 2011, 06:29

Post Sunday, 1st April 2012, 15:01

Re: Bringing some love to evokable items

Regarding the Bottled Efreet ... it's actually a permanent companion, so you don't have to use it mid-battle. At 27 Evo it has a 95% chance of being friendly which is already very good.

Anyway, would be great to see some new ideas from the wiki implemented. I should just note - none of the ideas on that page have been "officially" developer approved, it's just brainstorming. Before you spend any great time on any of them, drop in on ##crawl-dev and get some developer feedback.

Also note: Rolling boulder beetles are now in trunk, so some of the evokable ideas for boulders are immediately doable!
User avatar

Dungeon Master

Posts: 4031

Joined: Thursday, 16th December 2010, 20:37

Location: France

Post Sunday, 1st April 2012, 20:05

Re: Bringing some love to evokable items

mumra wrote:Anyway, would be great to see some new ideas from the wiki implemented. I should just note - none of the ideas on that page have been "officially" developer approved, it's just brainstorming.

The first ideas (each item has a permanent and single-use effect) are from dpeg himself, so you wouldn't call them unapproved. Even if they are 2 years old and he left, it's at least a guarantee of solid design.

mumra wrote:Before you spend any great time on any of them, drop in on ##crawl-dev and get some developer feedback.

Good advice.
<+Grunt> You dereference an invalid pointer! Ouch! That really hurt! The game dies...

Mines Malingerer

Posts: 47

Joined: Thursday, 1st March 2012, 21:31

Post Sunday, 1st April 2012, 20:44

Re: Bringing some love to evokable items

We talked on crawl-dev for awhile, and it sounds like the devs were more favorable of ripping out these items and creating new items that don't duplicate existing spells.
Won 2 Games - MiFi (2)
User avatar

Dungeon Master

Posts: 4031

Joined: Thursday, 16th December 2010, 20:37

Location: France

Post Sunday, 1st April 2012, 20:52

Re: Bringing some love to evokable items

donblas wrote:We talked on crawl-dev for awhile, and it sounds like the devs were more favorable of ripping out these items and creating new items that don't duplicate existing spells.

Discussions on ##crawl-dev often feels like that, but don't despair (also, Eronarn isn't a dev). Right now, many evoc items are more in need of new effects than new formulae. Tome of Destruction is probably the exception here, I think it could be improved by adjusting the risk of bad effects and what they do, like what has been done with the crystal ball of energy.
Maybe come back another time to try to get feedback from MarvinPA, evilmike, kilobyte and elliptic.
<+Grunt> You dereference an invalid pointer! Ouch! That really hurt! The game dies...

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 106 guests

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