Randarts (was: SquareLOS experimental branch on CBRO)


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

Tartarus Sorceror

Posts: 1774

Joined: Tuesday, 23rd December 2014, 23:39

Post Saturday, 2nd May 2015, 04:23

Randarts (was: SquareLOS experimental branch on CBRO)

advil wrote:it sort of had the feel of a meleebug character (like, are y'all sure that something about randart generation or item quality in general isn't messed up in 0.17??).

I also wonder this.
streaks: 5 fifteen rune octopodes. 15 diverse chars. 13 random chars. 24 NaWn^gozag.
251 total wins Berder hyperborean + misc
83/108 recent wins (76%)
guides: safe tactics value of ac/ev/sh forum toxicity

For this message the author Berder has received thanks:
advil

Ziggurat Zagger

Posts: 5382

Joined: Friday, 25th November 2011, 07:36

Post Thursday, 7th May 2015, 23:28

Re: SquareLOS experimental branch on CBRO

Berder wrote:
advil wrote:it sort of had the feel of a meleebug character (like, are y'all sure that something about randart generation or item quality in general isn't messed up in 0.17??).

I also wonder this.

This is basically offtopic, but briefly: randart generation changes imho make randarts ever so slightly better. It isn't huge, but in games with high rng luck (so like 5-10% of games) you can notice a decent boost. This is subject to confirmation bias and all that, I may be wrong, etc, but that's how it feels to me. This isn't a "bug" per say, but they could tweak randarts later if need be.

For this message the author tasonir has received thanks: 2
advil, Rast

Tartarus Sorceror

Posts: 1739

Joined: Tuesday, 13th March 2012, 02:48

Post Friday, 8th May 2015, 01:05

Re: SquareLOS experimental branch on CBRO

tasonir wrote:This is basically offtopic, but briefly: randart generation changes imho make randarts ever so slightly better. It isn't huge, but in games with high rng luck (so like 5-10% of games) you can notice a decent boost. This is subject to confirmation bias and all that, I may be wrong, etc, but that's how it feels to me. This isn't a "bug" per say, but they could tweak randarts later if need be.


I tried to convince the dev responsible. He said that the average randart quality is the same... I pointed out that the average randart sits on the dungeon floor, and it's only the top 10% or so which get used, so increasing the variation in unique quality increases the qulity of the usable randarts.

For this message the author Rast has received thanks:
advil

Ziggurat Zagger

Posts: 8786

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

Post Friday, 8th May 2015, 01:21

Re: SquareLOS experimental branch on CBRO

unless there's a +SquareLOS randart property i suggest you take this discussion to another topic

Tartarus Sorceror

Posts: 1739

Joined: Tuesday, 13th March 2012, 02:48

Post Friday, 8th May 2015, 01:45

Re: SquareLOS experimental branch on CBRO

You know, LOS increase on a fixedart would be kinda neat.

Sar

User avatar

Ziggurat Zagger

Posts: 6418

Joined: Friday, 6th July 2012, 12:48

Post Friday, 8th May 2015, 01:48

Re: SquareLOS experimental branch on CBRO

Like Robe of the Night, only bad instead of amazing!

For this message the author Sar has received thanks: 2
archaeo, Arrhythmia

Tartarus Sorceror

Posts: 1694

Joined: Tuesday, 31st March 2015, 20:34

Post Friday, 8th May 2015, 13:27

Re: SquareLOS experimental branch on CBRO

Increasing LOS would be really good in most of the 3 rune game for a ranged weapon user, or a vehumet blaster with the right spellset.
Once you've got a lot of dangerous ranged enemies with hellfire and torment that changes, but I'd love to be able to increase LOS sometimes.

Sar

User avatar

Ziggurat Zagger

Posts: 6418

Joined: Friday, 6th July 2012, 12:48

Post Friday, 8th May 2015, 13:58

Re: Randarts (was: SquareLOS experimental branch on CBRO)

I wanted to mention how that would be unplayable in console but LoS radius was changed with squarelos merging so it's possible to make a randart that would expand your LoS by one as a malus of sorts. It better have something good on it to make me consider wearing it though, and only +1 LoS is hardly exciting overall.

Dungeon Master

Posts: 3160

Joined: Sunday, 5th August 2012, 14:52

Post Friday, 8th May 2015, 15:00

Re: SquareLOS experimental branch on CBRO

Rast wrote:
tasonir wrote:This is basically offtopic, but briefly: randart generation changes imho make randarts ever so slightly better. It isn't huge, but in games with high rng luck (so like 5-10% of games) you can notice a decent boost. This is subject to confirmation bias and all that, I may be wrong, etc, but that's how it feels to me. This isn't a "bug" per say, but they could tweak randarts later if need be.


I tried to convince the dev responsible. He said that the average randart quality is the same... I pointed out that the average randart sits on the dungeon floor, and it's only the top 10% or so which get used, so increasing the variation in unique quality increases the qulity of the usable randarts.


I have doubts that I increased the variation in randart quality. I did some things that increased variation and others that decreased it. The maximum amazingness of an artefact is pretty sharply defined now, whereas it was fairly open-ended before.

For this message the author Lasty has received thanks:
advil
User avatar

Barkeep

Posts: 4435

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

Post Friday, 8th May 2015, 15:15

Re: Randarts (was: SquareLOS experimental branch on CBRO)

Sar wrote:It better have something good on it to make me consider wearing it though, and only +1 LoS is hardly exciting overall.

+1 LoS, +1 spell range?
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

For this message the author njvack has received thanks:
Rast
User avatar

Barkeep

Posts: 1788

Joined: Saturday, 29th June 2013, 16:52

Post Friday, 8th May 2015, 15:35

Re: Randarts (was: SquareLOS experimental branch on CBRO)

What if you just reflavored Thief Cloak so that it increased your LOS by 1 instead of giving you a slaying malus?

Sar

User avatar

Ziggurat Zagger

Posts: 6418

Joined: Friday, 6th July 2012, 12:48

Post Friday, 8th May 2015, 15:49

Re: Randarts (was: SquareLOS experimental branch on CBRO)

njvack wrote:+1 LoS, +1 spell range?

Sounds cool. Honestly I was thinking of something like a good heavy armour that would also kill stealth and give you a halo but my idea seems really redundant with existing randarts (and gods!).

Dungeon Master

Posts: 388

Joined: Monday, 18th August 2014, 20:04

Post Friday, 8th May 2015, 20:39

Re: SquareLOS experimental branch on CBRO

Lasty wrote:
Rast wrote:
tasonir wrote:This is basically offtopic, but briefly: randart generation changes imho make randarts ever so slightly better. It isn't huge, but in games with high rng luck (so like 5-10% of games) you can notice a decent boost. This is subject to confirmation bias and all that, I may be wrong, etc, but that's how it feels to me. This isn't a "bug" per say, but they could tweak randarts later if need be.


I tried to convince the dev responsible. He said that the average randart quality is the same... I pointed out that the average randart sits on the dungeon floor, and it's only the top 10% or so which get used, so increasing the variation in unique quality increases the qulity of the usable randarts.


I have doubts that I increased the variation in randart quality. I did some things that increased variation and others that decreased it. The maximum amazingness of an artefact is pretty sharply defined now, whereas it was fairly open-ended before.


Thanks for the response. Looking back at my morgue from the game that started this thread I actually do see several randarts generated that aren't so great, and so on the randart side there may have been some confirmation bias combined with good luck; hard to say. I also felt like I was getting fewer of the jarringly unusable/pointless randarts that I seem to see everywhere in 0.16, and if that's right this is probably a good thing. But part of what led to my (originally throwaway) comment is that it wasn't just the randarts but also general item quality luck. I had possibly the most and earliest wands of HW I've ever seen in a single game (to the point where I didn't even both going back to buy some that were in early stores), and the most and earliest rods I've ever seen in a single game, as well as some of the better acquirement draws I've had. A certain amount of this was in really good stores all over the dungeon. I didn't see anything that seemed completely impossible given my knowledge of what can happen in crawl, it just all happened in a single game (which is my only experience with 0.17) so I noticed it.

Ziggurat Zagger

Posts: 8786

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

Post Friday, 8th May 2015, 20:46

Re: Randarts (was: SquareLOS experimental branch on CBRO)

I wonder what percentage of non-Chei/Fo games generate a randart that is better to find than the first wand of hasting. I suspect it isn't very high.

Tartarus Sorceror

Posts: 1739

Joined: Tuesday, 13th March 2012, 02:48

Post Friday, 8th May 2015, 21:25

Re: Randarts (was: SquareLOS experimental branch on CBRO)

That says more about wands of hasting than it does about randarts.

Slime Squisher

Posts: 415

Joined: Monday, 8th December 2014, 10:31

Location: Sweedledome

Post Friday, 8th May 2015, 21:29

Re: Randarts (was: SquareLOS experimental branch on CBRO)

With randart stats able to go to +12 you can have some pretty epic stuff appear. Sure its usually tempered with a malus, but for the right character some of them can be massively powerful. (+12dex on a ring was a favorite of mine, i'll put up with a lot of other random crap for 12 dex on 1 slot. +2 boots {+11Str} on a blaster caster was fun too.)

Haste is Haste. Always be a strong effect. But theres only so many zaps in a wand and an artifact can last forever.
Online Wins: MiFi, GrFi, TrMo, HOGl, SpEn, {DEWz, DDNe}, {OgBe, CeHu}, FoFi, VSFi, MfGl, {HaGl, VpEn, HESk}, KoAs, DsFi, TeMo

Ziggurat Zagger

Posts: 5382

Joined: Friday, 25th November 2011, 07:36

Post Saturday, 9th May 2015, 01:01

Re: Randarts (was: SquareLOS experimental branch on CBRO)

I'm fairly sure the randart changes went into .16, so this isn't just limited to .17 games, by the way. Here's a normal, general case .16+ randart

-1 cloak of Weightlessness {rElec rF+ rC+ Str+3 Dex-4}
ring "Emaczeh" {rN+ MP+9 Str+4 Int-2}

This is fairly typical of the new randart style: strong positives, with some downsides (-1 armor enchantment, and the negative dex). Still overall very usable. The ring has some decent properties and one negative one.

General case for .15 and before:

+3 pair of boots of the Flute {Str+3 Int+3}
+1 buckler of Defence {+Rage Dex+2 Slay+5}

All positives, not too many of them, but solid items. Then there's the things that just weren't possible before:

+2 robe "Dacesh" {rElec rC+++ Str-3}

Having three pips of a resist is something I don't think old randarts could do. This isn't actually that strong (it'd be better to have rC++ and rF+), but it has a psychological impact on players (or at least me). Similarly, regen and + mana on gloves/boots/etc:

+1 pair of gloves "Jerphunai" {rF- rC+ Regen+ Int+3 Slay-3}

With the -slay and rF-, they aren't that great of a pair of gloves, but getting regeneration on gloves is something new you couldn't do before, so it feels strong.

It seems to be reasonably common to get things that have their one obiligatory negative, and then a whole ton of great mods as well (these are all from one game):

amulet "Haacet" {rMut rPois rC++ Dex-4 SInv} (curse)
ring of Xollio Fohu {rElec rF++ rN+ MR+ Str+4 Dex-2}
ring "Stey" {rElec rC+ rN+ Int+5 Dex-4 Slay+3}

Apparently dex was my dump stat. Never noticed how those all lined up ;)

Conclusions: I'm not sure that randarts are any stronger than they were before on average, but there are new properties that couldn't generate before, like regen or +11 dex, which certainly feel strong. I do think if you get a particularly lucky randart it is better post .16 than pre .16, but not hugely so. Still, getting a ring with -2dex isn't really going make it significantly worse when it has 6 positive properties as well (ring of Xollio Fohu above, counting rF++ as two properties). I don't think this has to be nerfed, but if it was tweaked slightly down, that would probably be fine too.

For this message the author tasonir has received thanks:
Eyesburn

Ziggurat Zagger

Posts: 8786

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

Post Saturday, 9th May 2015, 01:26

Re: Randarts (was: SquareLOS experimental branch on CBRO)

For comparison, here are some randarts I found looking at a few 0.7 wins.
R - the ring of Greed (right hand) {rPois AC+6 EV+5}
Q - the +3 pair of boots "Keoz" (worn) {EV+5 Int+2}
j - the amulet "Phafas" (around neck) {cFly rF+ EV+6 Str+2 Dam+3}
M - the ring of Charity {Hunger +Rage rF+ AC+6 EV+4 SInv}

Also, +6 gloves/helmets/boots/cloaks

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 123 guests

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