Curses feature branch

[Ingwe Ingweron]

Bug: Morgul weapons appearing in shops. (I forgot to do a savefile before recalling to the dungeon, but, if you look at the information screen for the weaponmaster's shop, you will see the morgul blade in the shop inventory.)

Savefile: Testo 4.zip

[Ingwe Ingweron]

More of a complaint than a bug: ?Identify have been painfully rare. @ home and inventory are chock full of "??" items because of the lack of identify.

Savefile: Testo 5.zip

Edit: Just checked the object.txt file and discovered the "max" for ?Identify is 70. Given that @ has dived fast, identify has been hard to come by and once past DL 70, virtually none existent. I've modified the object.txt file to alleviate the problem for the remainder of this test run, but I would ask you to consider tweaking the ?Identify allocation.

[Ingwe Ingweron]

Second example of a cursed item that says its not cursed and thus curse can't be removed. Check out the elvenkind body armor worn by @. Remove it, put it back on, get the "deathly cold" message for a cursed item, but examine and there's no curse and reading ?*Remove Curse* does nothing.

Savefile: Testo 6.zip

Edit: I think this might possibly be my own error. Wield an item and get the "deathly cold" message and so I know it's cursed. However, remove curse will not work unless I also first use a ?Identify to identify the curse. So, a two-step process is required for curse removal. Is that right?

[Ingwe Ingweron]

CRASH on walkover bug: @ has found The Plain Gold Ring!!! Wahoo!!! But, walkover My Precious and the game instantly crashes.

Savefile: Testo 7.zip

Edit: [Also note that the description of the Plain Gold Ring goes wonky on walkover.]

That's it for my bug testing until this bug is fixed.

Remaining known bugs to get a savefile for:

1. body armor showing to-hit and to-damage values
2. curses on artifacts that aren't normally cursed
3. display message errors on remove curse
4. crash on walkover of Morgoth's crown.

[jevansau]

I've just found the plain gold ring too (in a bubble vault at L89), and see the same crash on walkover .
Bit of a shame as it is the first time I've seen it for years.
savefile available if needed.

On experimentation, I found that removing the curse:sticky:100 line from the object file allowed me to pick it up without error and the description still says that it cannot be taken off.

[jevansau]

A few more thoughts on curses after a pretty full play through:
1) I never found a cursed item that I would have considered if it wasn't cursed. I think for curses to be interesting, the items they are on need to be especially good for their depth. The only exception to this was the one ring
2) I first found the sticky curse after level 50 - pretty deep I would have thought.
3) curses seem pretty rare in general.

A question - what is the impact of a vulnerability and how does it interact with resistances and immunity?

[Pete Mack]

@jevensau--the One Ring and Calris are the obvious, with Gorlim, Cloaks of Enveloping, and Bows of Nazgul closing out the potentially useful cursed items. With randarts, unfortunately, the cursed items are almost uniformly useless. Not bad like Beruthiel, but not worth trying.

[Nick]

CRASH on walkover bug: @ has found The Plain Gold Ring!!! Wahoo!!! But, walkover My Precious and the game instantly crashes.

New version up with this bug fixed - Windows and OS X versions. Also fixes the cursor.

I believe you are correct about being unable to uncurse without knowing the curse. Morgul weapons currently don't have a curse, and may have positive cash value, so they can indeed occur in shops. If you don't mind aggravation and experience drain, you may even want to wield one - I'm interested to hear opinions on what they should be like.

A few more thoughts on curses after a pretty full play through:
1) I never found a cursed item that I would have considered if it wasn't cursed. I think for curses to be interesting, the items they are on need to be especially good for their depth. The only exception to this was the one ring
2) I first found the sticky curse after level 50 - pretty deep I would have thought.
3) curses seem pretty rare in general.

A question - what is the impact of a vulnerability and how does it interact with resistances and immunity?

Vulnerability makes the player take 1/3 more damage, cancels out single resists, reduces double resists to single, and doesn't affect immunity.

Thanks for the impressions. Cursed items are meant to be a little better than you would expect to compensate - maybe I need to increase that. And yes, I think they are probably too rare. The sticky curse is one of 27 and curses are chosen at random, so when you first find it is just down to luck

Note that I am expecting considerable adjustment to have to happen to get the balance right for curses. All suggestions are happily received. With luck there will be more bugfixes soonish too.

[Fiquin]

Removing Curses

Please correct me if I'm wrong however the last post on this topic was 2007 which is surprising considering "Leather Boots of Wormtongue" (4.0.4) can be quite useful in the early/mid game and now that I want to take them off... I'm wondering in this patch, can it be done?

I've hit them with at least three enchant armor scrolls so far and they're still frozen to my feet. So, are cursed items removable, if so, how?

Help much appreciated!

[Derakon]

Please correct me if I'm wrong however the last post on this topic was 2007...

You've confused the date Nick registered on the forums with the date his post was made. This thread is still active.

(Your question was already answered in another thread though, so I'll not recapitulate that here)

[Pete Mack]

I think I am seeing a duplicate free from the object awareness code, after looking at missing objects after *destruction*. I don't pretend to understand the object awareness code, so I can't help much more than this. (I've been running exclusively in the debugger, because I've seen a number of crashes. I'm hoping to track them down better than this.)

The warning from gdb:

#5  0x74c977a5 in msvcrt!free () from C:\WINDOWS\SysWOW64\msvcrt.dll
#6  0x01300000 in ?? ()
#7  0x004d0aa2 in mem_free (p=0x55c11f4) at z-virt.c:65
#8  0x0046470a in object_delete (obj_address=0x9ef194) at obj-pile.c:344
#9  0x00466c49 in floor_pile_know (c=0x85a851c, y=17, x=41) at obj-pile.c:1388
#10 0x004020a3 in square_note_spot (c=0x85a851c, y=17, x=41) at cave-map.c:218
#11 0x004069c4 in update_one (c=0x85a851c, y=17, x=41, blind=0)
    at cave-view.c:514
#12 0x0040712b in update_view (c=0x85a851c, p=0x2e60cf4) at cave-view.c:659
#13 0x004093fe in move_player (dir=2, disarm=true) at cmd-cave.c:1038
#14 0x00409756 in do_cmd_walk (cmd=0x534894 <cmd_queue+2100>)
    at cmd-cave.c:1134

[Pete Mack]

And a seg fault, from memory corruption:
#0 0x00458d12 in object_is_ignored (obj=0x8a0d94c) at obj-ignore.c:584
#1 0x00458e81 in ignore_item_ok (obj=0x8a0d94c) at obj-ignore.c:626
#2 0x00465d81 in drop_near (c=0x897554c, dropped=0x8a0d94c, chance=0, y=50,
x=176, verbose=true) at obj-pile.c:1010
#3 0x00449fd5 in monster_death (mon=0x89d1934, stats=false)
at mon-make.c:1385
#4 0x0044a412 in mon_take_hit (mon=0x89d1934, dam=50, fear=0x9ef22a,
note=0x0) at mon-make.c:1526

I am trying to figure out clang compiler -fsanitize on windows, quite desperately now!

[Nick]

Thanks Pete, these will be really helpful when I finally get back to working on this

[Pete Mack]

Nick--
I looked at this a little more. As far as I can tell, this says almost nothing about what is happening. By the time this fires, the memory has been freed, and initialized by the debug malloc package so that everything is the value 0xfeeefeee repeated. (Other allocators may use 0x0badbeef instead.)
When I get a chance I'll use fsanitize.

[PowerWyrm]

Other allocators may use 0x0badbeef instead.

That's it... I'm becoming a vegan!