Monday, 23 March 2009
|
| Re: rephial.org in trouble? Timo Pietil 00:33:02 |
| | Zachary Kline wrote:
Hi all, I have been out of the loop for a while, so don't know if this is a recent or well-known issue. rephial.org seems to be having some problems. I get 404 errors when I browse over there, though what looks like the navigation bar still shows and I can download successfully. A couple days ago I wasn't getting any site at all, just a bunch of Lua errors. Is this something which should be fixed soon? Am I bringing up an old issue?
I'm getting same 404 error in "play" and "development" -links.
"releases" and "wiki" works just fine.
Apparently something is broken in rephial.org.
Timo Pietil
|
| | Add comment |
Sunday, 22 March 2009
|
| Adom is a violent computer game that should be BANNED!!! Harald Gentexeater 01:45:46 |
| | Adom is all about violence and murder and
violent computer games such as Adom do
turn harmless people into mass murderers,
serial killers and cannibals!!!
Adom should be BANNED!!!
|
| |  1 answer | Add comment |
Friday, 20 March 2009
|
| Re: ADOM Sage - can anyone provide it? Eoghan Griffin 01:47:09 |
| | macgrabowy@gmail.com wrote:
Since its main page is down I've been searching high and low for the 0.9.1 version for Windows. I failed miserably. Can anyone please host it on one of those temporary hosting sites and link it to me? That, or declare my google-fu skillz weak and get me a link not leading to the main site? Pretty please? I long for the pretty message colors!
The main page works fine.... I just downloaded both the .msi and the .zip =P
|
| | 2 answer | Add comment |
Tuesday, 17 March 2009
|
| JADE question Guest 18:15:19 |
| | Will JADE ever see the light of day or is it Vaporware at this point
in time? Also, will there be a new version of the Windows Beta any
time soon?
|
| | 9 answers | Add comment |
|
| New discovery / abilty to Benjamin Temple 14:35:52 |
| | Hey, anyone ever wonder what the experience requirements to get
hawkslayer is. The companion which requires you to know a password
(Iceberg). I am a level 13 high elf archer, and I went into the CoC
just recently. I found 2 stair cases on the first level. Going down
one into D2, I found a black pool. I'm doing really well, but, as dumb
as I was, I risked drinking once. I got the message from God saying
the word Iceberg, or whatever it is (I wish now I had had a
screenshot. Maybe when I get hawk I'll screen that). Anyway so now,
after each level I'll return to hawk and see if he'll join my party.
I'll post back with the result (what the lowest level it is possible
to get him).  I can't wait. This game is so complicated and has so
many special things.
|
| | 3 answer | Add comment |
Sunday, 15 March 2009
|
| Re: Curses (cross post from oook) Timo Pietil 23:03:13 |
| | Nick McConnell wrote:
For a while I've been saying I'm going to revamp curses in FAangband, with the level of detail I give being roughly "remove sticky curses and add more interesting curses". So now I think I have a slightly more fleshed out scheme, and would like to hear opinions.
e. Occasional summoning
This is in vanilla potential instadeath. At least in past summoned
monsters could act in same turn they are summoned, which can kill you
instantly (for example player without FA and carrion crawler).
I don't know if this is the case in FA or in latest versions of the vanilla.
Timo Pietil
|
| | 11 answers | Add comment |
Thursday, 12 March 2009
|
| Killing small children in town Plg 16:01:23 |
| | I don't know why you have to kill small children in town. Whose idea
was that? It's not enjoyable. I changed the first and last lines of
the code from this:-
N:41:& Small child~
G:t:W
I:+0:1:1d2:6:1:0
W:0:1:0
B:WORSHIP
F:MALE | FRIENDS | RAND_25 | RUN_AWAY
D:A rather cute child with large trusting eyes.
to this:-
N:41:& Naughty goblin~
G:t:W
I:+0:1:1d2:6:1:0
W:0:1:0
B:WORSHIP
F:MALE | FRIENDS | RAND_25 | RUN_AWAY
D:A rather ugly goblin trying to trick you.
|
| | 5 answers | Add comment |
Wednesday, 11 March 2009
|
| [un] bug pressing * causes crash Guest 11:35:03 |
| | Steps to produce:
not sure if this is the right place for this:
when enemy is in view:
press f
select ammo
press *
unangband.exe has encountered a problem and needs to close....
occurs using shift-8 for * and numpad start
version:0.6.2 gold 2
os:windows xp
gets quite annoying
I have a save which can reliably reproduce this crash if anyone wants
it
|
| | 1 answer | Add comment |
Tuesday, 10 March 2009
|
| Finding food? Guest 18:08:47 |
| | i have been playing nethack for a while now and started playing
angband today. i LOVE it already but i found myslf starving to death
looking for food on lvl 3 of the dungeon. how do i find food in this
game? i can't see me having to go back to the town all the time
because i will be in the same situation again when i get back to where
i was. can anyone help me?
thanks
|
| | 6 answers | Add comment |
Monday, 9 March 2009
|
| [3.1.0 Beta] advice for high-elf ranger John Smith 23:06:22 |
| | Hi, I'm currently playing vanilla 3.1.0 beta (windows) and chose to create a
high-elf ranger. So far things are going well and I've managed to reach
1000' without too much pain. However I don't feel as though I'm playing this
character right as all I'm doing is going around with a blade and not
casting any spells from my magic books. I found the dagger 'Dethanc' very
early on at 600' and haven't had any problems killing off beasties with it
all the way down to 1000'. I have recently changed to using a 'Long Sword of
Extra Attacks'. I take it that this character is better suited towards
shooting than slashing. My character's intelligence is at 16, whereas all
other abilites are 18/10+.I guess thats preventing me from fully embracing
spell casting as my SP(max) is at 22. Things are getting a bit tricky at
1000' and I would appreciate any advice as to what I should be doing (or
what I am doing wrong) so that I can progress more strategically with the
game and win for once.
My character dump is located at:
http://angband.oook.cz/ladder-show.php?id=8706
Thanks for any insight.
Cheers, JS
|
| | 14 answers | Add comment |
Sunday, 8 March 2009
|
| Vanilla 3.1 beta. Bug or Murphy? Timo Pietil 23:18:57 |
| | Hello again.
I'm having strange phenomenon in my current game. I have seen all mage
dungeon spellbooks except Keleks, and most of them multiple times, but
not one dungeon priest book in dungeon. I bought Ethereal Openings from BM.
I'm playing priest.
Current depth is 2600 (going slowly and building monster memory with rod
of probing and letting things to hit me).
Is this a bug, or is it just Murphys law in effect?
Timo Pietil
|
| | 31 answer | Add comment |
|
| potion of toughness John Smith 23:07:58 |
| | Hi, I was wondering about 'potion of toughness' which i found in 3.1.0 beta
in the black market for 3000 AU. I purchased and consumed and my
constitution went up from 18/10 to 18/21. I found no reference to this on
the net. Is this normal? Any help most appreciated.
|
| | 2 answer | Add comment |
Monday, 2 March 2009
|
| Evaluation of item activations (long, spoily) Magnate 21:33:02 |
| | Hi folks,
The next set of improvements to randarts will be random activations - at the
moment a randart simply inherits its parent artifact's activation (if any).
More significantly, activations are not calculated as part of artifacts'
overall power ratings, which means some significant activations come for
free.
I'd like to gather views on which activations are worth how much. The scale
we're working on is "equivalent damage per blow". Each non-damage-related
attribute is rated according to how much damage it's worth trading for. Way
back in 2001 when the randart code was rewritten, a long debate on rgra
established the following values for resists and other abilities:
22 fire immunity
17 acid immunity
17 cold immunity
15 telepathy
14 elec immunity
14 resist poison
12 resist confusion
10 resist nether
10 resist chaos
10 resist disenchantment
8 resist dark
8 resist blindness
7 free action
7 resist sound
6 hold life
5 see invisible
5 sustain strength
5 resist nexus
4 sustain dex
4 resist shards
4 regeneration
3 sustain con
3 resist fear
3 resist light
3 resist fire
3 resist elec
3 resist cold
2 resist acid
2 sustain int
2 sustain wis
1 feather fall
1 permanent light
1 blessed
0 sustain cha
Now, of course they're not perfect (hold life worth more than regen??), but
it's what was agreed at the time, and this thread is not intended to be
about reopening that debate (though I am up for doing that separately). I
list them solely to put the following artifact power ratings into
perspective. (The evaluation of pval-related attributes is complex, so I'll
skip over them to save space.)
Here's how the existing artifacts come out according to the current
evaluator. I've listed them by final power, with their max power in
brackets (before deductions for curses, aggravation etc.). I've also listed
their activation, if any, and for weapons their power in terms of pure
damage, before any non-damage attributes are counted.
Morgoth 22074 (silly pval breaks calculator)
The One 315(373)BIZARRE
Bladeturner 180 RAGE_BLESS_RESIST
Vilya 174 HEAL2
Deathwreaker 153(183) (100)
Grond 150(180) (154)
Ringil 129 COLD_BALL100 (55)
Nenya 129 RESTORE_LIFE
Durin 129 (47)
Aule 125 (71)
Eonwe 125 LOSKILL (48)
Ulmo 114 TELE_OTHER (44)
Anduril 107 FIRE_BOLT72 (35)
Erebor 107 (35)
Soulkeeper 104 HEAL2
Narya 100 PROTEVIL
Orome 98 STONE_TO_MUD (42)
Gondor 98 HEAL1
Fundin Bluecloak 98 DISPEL_EVIL (35)
Dor-Lomin 96
Eowyn 96 (38)
Thunderfist 96 (48)
Aeglos 94 COLD_BALL100 (51)
Feanor 93 HASTE1
Thorin 93
Balli Stonehand 93 (27)
Celeborn 88 BANISHMENT
Nain 84 STONE_TO_MUD (31)
Olorin 84 PROBING (37)
Wrath 84 (50)
Sting 81 (29)
Isildur 81
Zarcuthra 80(110) (59)
Numenor 80
Gurthang 79 (40)
Angrist 74 (27)
Rohirrim 74
Doomcaller 73(128) (74)
Caspanion 72 DESTROY_TDOORS
Hurin 72 BERSERKER (40)
Avavir 70 RECALL (29)
Eorlingas 70 (43)
Razorback 68(98) STAR_BALL
Belthronding 67 (52)
Dal-i-thalion 65 REM_FEAR_POIS
Mediator 65(95) STAR_BALL
Pain 64 (61)
Aglarang 64 (29)
Gil-galad 63 STARLIGHT2
Cubragol 62 (20)
Fingolfin 60 ARROW
Amras 59 (41)
Maedhros 58 (21)
Tuor 58
Elessar 58 HEAL1
Amrod 58 (28)
Haradekket 57 (40)
Belegennon 57 TELE_PHASE
Bard 57 (40)
Bloodspike 57 (33)
Mundwine 56 (41)
Umbar 55(85) ARROW (46)
Theoden 54 DRAIN_LIFE2 (18)
Crisdurian 54 (43)
Arunruth 53 COLD_BOLT2 (23)
Azaghal 53 (22)
Thalkettoth 52
Barukkheled 52 (31)
Tulkas 51 HASTE2
Gondricam 51 (15)
Turmil 50 DRAIN_LIFE1 (19)
Taratol 50 HASTE1 (29)
Totila 50 CONFUSE2 (31)
Cambeleg 49
Celebrimbor 49
Osondir 48 (27)
Aeglin 48 (29)
Himring 48 PROTEVIL
Hammerhand 47(77)
Til-i-arc 46 (25)
Orcrist 46 (28)
Dwarves 44
Thror 44
Anarion 43
Arvedui 43
Eriril 43 IDENTIFY (17)
Holhenneth 42 DETECT_ALL
Glamdring 41 (28)
Rilia 41 STINKING_CLOUD (16)
Nar-i-vagil 41 (29)
Nimloth 40 (19)
Anguiriel 40 (24)
Barahir 38
Belangil 38 COLD_BALL (17)
Ingwe 38 DISPEL_EVIL
Celegorm 38
Dagmor 37 (31)
Luthien 37 RESTORE_LIFE
Thranduil 37
Thingol 36 RECHARGE
Forasgil 35 (24)
Hithlomir 35
Colluin 32 RESIST_ALL
Elvagil 31 (15)
Thengel 29
Thrain 27 DETECT_ALL
Calris 26(69) (45)
Haradrim 26(56) BERSERKER
Firestar 26 FIRE_BOLT72 (21)
Lotharang 26 CURE_SERIOUS (19)
Evenstar 24 RESTORE_LIFE
Dethanc 23 ELEC_BOLT (19)
Careth Asdriag 22 (13)
Narthanc 22 FIRE_BOLT (18)
Thorongil 22
Nimthanc 21 COLD_BOLT (17)
Colannon 21 TELE_LONG
Cammithrim 19 MISSILE
Holcolleth 17 SLEEPII
Gothmog 17(81) FIRE_BALL (43)
Paurhach 16 FIRE_BOLT
Pauraegen 13 ELEC_BOLT
Paurnimmen 13 COLD_BOLT
Paurnen 12 ACID_BOLT
Carlammas 11 PROTEVIL
Elendil 10 MAPPING
Westernesse 9(54) CLAIRVOYANCE
Eol 5(35) MANA_BOLT
Galadriel 5 ILLUMINATION
Wormtongue 3(27) TELE_PHASE
Melkor -4 (41)
Gorlim -19
Beruthiel -31
Mormegil -66 (30)
Camlost -89
Again, I'm not looking to debate how good or bad the calculator is overall
(again, happy to do that separately). What I want to know is, how much is
each activation worth? If we take a baseline of 1 for the crappy FOO_BOLT
activations of the Paur* and *thancs, where would we end up? Is the
1000-point heal of Vilya and Soulkeeper worth 10? 15? Is the frost ball of
Aeglos and Ringil worth more? Less? Etc.
I'm not really expecting anyone to come up with a full list. I'd just like
comments about which activations you consistently use, and which ones you
don't (and why not). Then I'll come back with a list, and a revised table
of artifacts so we can see what difference it makes.
CC
|
| | 23 answer | Add comment |
Saturday, 28 February 2009
|
| Penance dead? Timo Pietil 00:19:32 |
| | Hi all.
Does anybody know if Graig and J.D. Oliver are OK? I posted a comment in
their webpage few about week ago without reply.
Timo Pietil
|
| | Add comment |
Thursday, 19 February 2009
|
| Vanilla old savefiles and 3.1 beta Timo Pietil 17:43:35 |
| | Hi there.
Is there any way to get monster memory from old savefile when you start
playing 3.1 beta? Or do I need to start from scratch?
What about old scores?
Timo Pietil
|
| | 8 answers | Add comment |
|
| Noise-to-signal ratio Andi Sidwell 05:47:03 |
| | I'd like to make a quick request before people any kind of permathread
gets going: please consider the noise-to-signal ratio of the group when
making posts. Some posts are not worth responding to.
Cheers,
Andi
|
| | 10 answers | Add comment |
Tuesday, 17 February 2009
|
| Re: device success rates Stefan O'Rear 09:52:27 |
| | On 2009-02-13, Eddie Grove <eddiegrove@hotmail.com> wrote:
The code implementing device rates is a bit complicated and produces somewhat strange results. That is pretty much unavoidable, given its premises. Those are that everyone has non-zero success and failure with every item, and that strictly better device skill means strictly better success rate. Simpler approaches require changing (some of) those premises.
int pct = 50 + (int)(45 * tanh((double)(skill - difficulty) / 15.0));
Is that a bit complicated? Does it produce strange results?
|
| | 12 answers | Add comment |
|
| Ring of Escaping probably stronger than intended {V 3.10} Joshua Rodman 06:58:43 |
| | So I got my ranger out of the beginning game after some ignoble
deaths, and as suggested when he hit level 20 things changed with
the extra shot. The lucky Longbow of Extra Might may have had
something to do with it also.
However, I picked up a Ring of Escaping (+4), which essentially is a
Ring of Speed without the ability to attack. Is this intended to be
so strong for midgame mages and rangers? You can get through the
majority of combats without attacking at all.
--
Grim. Grom. Gr mmer.
|
| | 1 answer | Add comment |
Sunday, 15 February 2009
|
| Asgaardi Luki667 01:11:45 |
| | Let me how far can you go. No save scumming pls.
http://www.sendspace.com/file/u3yu5t
|
| | Add comment |
|
| Angband: Maintain or Develop? Andi Sidwell 00:27:04 |
| | I spent a few hours over the last day or so writing a little article
about my views on the future of Angband. I thought I should probably
post it here for people to read, otherwise it would be a bit of a wasted
effort. 
http://takkaria.org/angband-develop-or-maintain.html
I expect that it will be slightly more controversial here than it will
be on the forum, for whatever reason.
Andi
|
| | 22 answer | Add comment |
Saturday, 14 February 2009
|
| Angband 3.1.0 beta released Andrew Sidwell 20:19:55 |
| | Well, it took some time, but finally there's a beta version out... get
it while it's fresh!
Source:
http://rephial.org/downloads/3.1/angband-3.1.0beta.tar.gz
Windows compile:
http://rephial.org/downloads/3.1/angband-3.1.0beta-win.zip
OS X compile:
http://rephial.org/downloads/3.1/Angband-3.1.0beta-osx.dmg
There are a lot of changes in this release. The most notable are that
the UI has been tweaked, refined and generally twiddled so it should be
the easiest-to-use release ever. Other than that, there should be less
junk and some new item kinds. Underneath all this is some fairly
wide-ranging code cleanup and rewriting.
There is no exhaustive changelog for this release since the changes are
so wide-ranging. However, a fairly complete one is attached to the end
of this post. I welcome all criticism and bug reports; this release is
not complete, and new releases will probably stay in beta through to
3.1.3 or thereabouts, so if you have issue with anything, please make it
known!
Angband 3.1.0
=============
This is an Angband development version -- it may be buggy, so be warned!
Special thanks for this release go to: Eytan Zweig, pelpel, Kenneth Boyd,
J.D. White, Eddie Grove, Rowan Beentje, Nick McConnell, Gabriel
Cunningham, Shanoah Alkire, Alexander Philips, mikon, Antony Sidwell,
Joe Buck, Erik Osheim, d_m, Antoine, and all the alpha- and beta- testers
who downloaded the game, sometimes in an extremely shaky state, and played
it anyway.
Noticeable up-front UI changes
------------------------------
* Add EyAngband-style quickstart option.
* Notes, like NPP and FA, except with brand new code by J.D. White.
(#10)
* Add a trap detection indicator. Also, colour the edges of trap-
detected areas green. (#19)
* Make status bar display all timed effects, left-to-right, which
involves moving the dungeon level to the bottom of the left-hand
display. (#191)
* Add show_lists option, which is the same as always_show_lists in
Ey. (#25)
* Auto-wield birth items (#24)
* Change the object information screen quite a bit: (#214, #26)
- Show average damage for weapons and ammo (#26)
- Make descriptions a fair bit terser (#399)
- Show breakage chance for ammo
- Track item origin, so every item has a tag telling you where it
came from (#149)
* Monster recall changes (part of #152)
- some colouring removed, some added
- Monster's depth is in red if it's beyond maximum depth, light
green otherwise.
- more information is known about monsters before they are killed
* Add how many items the player already has when buying something at
the store. (closes #151)
* Allow use of the space bar to advance a screenful in all
menus. (#244)
* Make all menus wraparound at the top/bottom. (#244)
* Monster recall display now cycles through monsters in level order,
not in the order that the monster definition file is written
in. (#302)
* Add an indication of maximum capacity in the inventory
display. (#358)
* Menus should use 1234567890 as little as possible for "tags". (#292)
* Remove the "depth_in_feet" option, instead displaying both the
dungeon level and the depth everywhere.
* Ensure that store-bought and birth-given torches have the same number
of turns for stacking purposes.
* Death screen now uses scrollable menu. (#416)
* Drop squelched items
* Append {tried} to tried items in the object browser (r606)
* Put {tried} items before untried items in the object browser
* The monster list display has a little more colour-coding
* Death menu now uses a scrolly menu.
* Preserve mode has a sensible description again.
* Try and differentiate the level feelings more.
* Add the moderately well-tested "lazy movement" patch which allows a
movement delay to be set, during which it is possible to choose a
second direction. This means that up and left in quick succession
can be translated to an actual diagonal. This may improve some
people's laptop playing experience without resorting to the
roguelike keyset.
* Add bigscreen support to the self-knowledge screen.
* Make inscriptions like "!k!k!k" prompt multiple times again. (#492)
* Make monster XP and native depth always known, even without
kills. (#596)
* Rejig social status a bit -- it rises with level from a certain
baseline.
* Display various skills numerically on the character sheet, rather
that with text. This is like Eddie's patch, but a bit more polished.
* Add a "display item list" command and term window. (Thanks to Eric
Osheim)
* Add method of death to the character dump. (#501)
* Tell the payer when timed effects increase or decrease as the result
of effects. (idea by Eddie Grove)
* Windows versions now use hashes and percent signs for walls until
there's time to sort out all existing fonts to be Vista-compatible.
(#232)
* Squelch 'K'ind feature now available in the roguelike keyset. (#339)
Gameplay changes
----------------
There are some pretty big gameplay changes.
The object list has been reworked quite a bit. It is a bit more
varied and has less duplication. Some items have been removed and
others added. The new items have had relatively little testing and
may be gone again next release.
- The number of items that monsters drop has been reduced drastically.
- Potions tend to appear in stacks now.
- The range of food items has been reduced.
- Healing heal a proportion of wounds, not just a set value-- this
makes all the potions useful.
- Add a selection of "gain one, lose one" stat potions to replace the
old stat loss potions-- they guarantee a stat to raise and reduce one
randomly.
- The Elven Rings have a variety of healings, at the suggestion of
Timo. Also, apply some other changes from the TV-patch (r428)
- Replace the sustain {stat} rings with sustain mind and sustain body,
providing three sustains each.
- (#543) Make ammo of slay evil/venom apply to Mithril ammo too.
Gold drops have also been reworked. By the end of the game, there is
about 3.6x less gold per drop than in any earlier version or Angband
or Moria, but the amount per drop at the beginning of the dungeon
is quite similar to pre-existing levels. You may be interested in
looking at http://rephial.org/research/avg_gold_drop.pdf for an
illustration.
Magic mapping and detection use the same-sized rectangular area around
the player regardless of screen size.
Fear now carries a to-hit penalty to missile weapons of 20, as well as
giving an AC bonus of 8. It also increases the failure chance of
spells, but spells that cure fear have been made harder to fail to
balance this out.
Birth:
* Make point-based points worth less gold.
* Minimum starting gold is 200AU.
* Add an option to start the character with 500AU but no equipment.
* Point-based character generation is now equivalent to best available
from the autoroller. (Eddie Grove)
* All characters start with WoR.
Stores:
* Phase Door is now a staple item.
* Start stocking deeper items as the character descends.
* Rewrite the rules for shopkeepers
- Remove individual-shopkeeper and race-based price changes from the
store code, to make it easier to balance pricing.
- Adjust adj_chr_gold[] values upwards to represent the average-110%
inflation usually applied by the individual-shopkeeper greed
values. This should really be applied to the object.txt file
instead, though.
ID:
* Make ID reveal all powers of an object, and remove *ID*. (#158)
* Make pseudo-id not occur whilst resting, but also make pseudo-ID
lengths shorter (by a little over half) to compensate.
* Some items, when worn long enough, get automatically ID'd
* Items pseudo-id'd as {average} are automatically ID'd, and thus will
stack properly
* Make wands/staffs stack even when un-ID'd
* Pseudo-ID obviously excellent items as {excellent} when possible
(e.g. brands/slays/obvious stat bonuses).
* Remove "good" pseudo-ID, replace with "magical" -- for non-cursed
non-average non-ego non-artifacts
* Add "strange" pseudo-ID, for when items have mixed blessings
* Stop automatically cursing all bad weaponry
* Make weak pseudo-id notice the difference between magical and
excellent
* Allow any/all rings/amulets to be cursed if the game wants to
generate a "bad" item. This is liable to be rethought in the future.
A small selection of other changes:
* Remove autoscum. (#365) -- really?
* Add minimum depths for each kind of trap. As such, only teleport and
confusion traps occur at dlev1. This should help some people.
* Make the rogue slightly better at combat than the ranger, based on
Timo's sugges
tion. (Message-ID: <651vknF2cvjl2U1@mid.individual.net>)
* Lower the distance missile weapons can fire into the 10-14 grids range.
* Make monster arrow attacks sometimes miss. (#537)
* Some improvements to monster AI.
Port
----
* X11: Add -x commandline option for choosing the location of the
settings file (#333, patch by morth)
* GTK2: Continue work; add term windows, graphics, etc. and also some
special dedicated display windows, like "messages" and "monster
list". (shanoah)
* OS X: the OS X version is now a universal binary, and runs faster,
better, quicker, and with more sound than before. (Rowan Beentje)
Code-level
----------
Monster HP is now specified as an average value around which to
randomise, instead of using a "12d5"-type calculation. Uniques always
have the number of HPs specified in the monster.txt file, like before.
This is copied from Ey.
A lot of code-level cleanup has been applied to this release, much of it
still in flux. Many new files have been split out of existing files,
which should now have much more self-explanitory names. The player/,
monster/ and object/ subdirectories have been created, to aid seperation
of code.
There is also a new generic effects handler so that all item activations
and use abilities share the same code; this replaces use-obj.c and a lot
of descriptions from lib/edit/object.txt. (#234)
Memory handling has been reviewed. Now, all code should assume that
allocation will succeed, just like it did already. There is also a
realloc()-style function along with its own hook. string_make() has
been modified to no longer return a "const char *", and so all the code
now only uses "const" when it means "const". Some macros have been
removed.
"Elly" rewrote the message and quark packages, previously in util.c,
and in the process split them out into their own z- files. The z-msg
package is now much more understandable but marginally less "efficient",
and certainly less "optimised".
Option definitions have moved to option.h, and the names and
descriptions for options have moved to option.c. Option data is now
stored in a much more visually compact way, much more suitable for
editing than the old system was. (#64)
Additionally, a new file API has been introduced which combines the
previous mishmash of fd_*() and my_f*() calls into a simpler, better
documented interface. (#137)
Store stocking is specified in lib/edit/store.txt now, in a reasonably
flexible format.
A fair bit of work has gone into allowing item names to be used in the
edit files rather than the tval/sval numbers. For example, see the
new store.txt file; but also the lib/pref/ graphics files, and the
starting items in the p_class.txt file, amongst others. This is a
fairly easy thing to do and is a massive massive usability enhancement.
It means that invalid item references are caught much sooner, since if
an item is removed, all references to its name become invalid.
Set up an event system for UI display updates, where the game
communicates with the UI by sending messages saying that something's
changed that the UI might be interested in, rather than deciding that
specific things should be redraw. The main screen updates all use this
system now. Amongst other things, this system should help things like
the borg and graphical frontends, because they can hook right into
changes in state. (#348)
object1.c and object2.c have been split out into a series of files,
obj-{desc,make,ui,util}.c. This makes it much easier to remember what
is where. Also, object_desc() has seen a rewrite, so hopefully it's a
lot more hackable now.
The beginnings of an organised code documentation scheme have started,
using doxygen. Makefile.std provides an easy mechanism to generate docs
(try "make docs").
Other assorted changes:
* Remove a whole load of useless or unused configuration options from
config.h.
* "make -f Makefile.inc depgen" will output "Makefile.new", which then
contains up-to-date dependencies for the project.
* Add "flags" back to object_type, removing the need for the hacky
"xtra" bytes. This paves the way for e.g. fireproofing scrolls
later. (#120)
* Makefile for MSVC/NMAKE command-line build. Currently works with
MSVC 6+; flags could use tweaking for MSVC 8+.
* Clean up racial and class skill values into arrays. (Kenneth Boyd)
* Use header file guards in all header files, and use a consistent
naming scheme for them ("Elly")
* Add a very basic test harness for "unit testing" of the game's code,
though this is barely used at present. Thanks to Fennec for cleaning
this up a bit.
* Introduce the one_in_() notation to the game, for percentile dice.
* Move to using randint0() and randint1() for random numbers.
* Add proper transparency to the PNG files. (fixes #270; thanks to
Rowan Beentje)
* Move to using flags[6] for monster flags rather than individual
flagsets.
* Score code moved to score.c and extensively refactored.
--
http://rephal.org/
|
| | 60 answers | Add comment |
|
| Ranger in 3.1.0 {v} Joshua Rodman 12:02:46 |
| | I'm trying the 3.1.0 game, and felt it would be more fun to come
back to the game with a fresh class. So I tried the ranger, which
I haven't played in the past.
This guy shoots.. well.. not far enough to damage thigns at a
significant distance. And doesn't do enough damage to avoid
meleeing any moderately strong foe.
Were rangers always this gimpy? If not, what classes are treated
well by 3.1.0?
-josh
--
Grim. Grom. Gr mmer.
|
| | 4 answer | Add comment |
Friday, 13 February 2009
|
| [V, dev] Artifact rarity Magnate 13:24:41 |
| | Hi all,
The increasingly innacurately named "new randart code" (first released in
2001 as a patch for V2.9.4) is now in trunk and will be part of V3.1.1
It's a complete rewrite of Greg Wooledge's original randart generator, with
a few retained features. One of these is the calculation of a randart's
rarity. GW posited that the "effective rarity" should remain constant, which
means:
new artifact rarity = old artifact rarity * old artifact's base item rarity
/ new artifact's base item rarity
(In case anyone is concerned about the shift from rarity to "alloc_prob",
for the purposes of this conversation I use rarity = 100 / k_ptr->alloc_prob
with a minimum of 1. Apologies if this is a ham-fisted conversion, but the
principle is more important than the numbers.)
Effective rarities in artifact.txt range from 3 (all the 'thancs have rarity
three, and a dagger has rarity 1) to 1600 (Bladeturner has rarity 16, base
PDSM has rarity 100 now, though it used to be 64). Artifact rarity is capped
at 255, so the Bladeturner-based randart can be about six times as common as
Bladeturner.
In the next set of changes to randarts (probably not before 3.2), I am
thinking of changing this. It seems silly to go to the trouble of evaluating
every single aspect of an artifact to calculate its overall power and then
completely ignore that calculation and base its rarity on whatever happens
to be in artifact.txt - I am instead thinking of making effective rarity
directly related to power (though probably not linearly).
On the plus side, this means that in any given game using randarts,
generation should be internally consistent: the better items should show up
proportionally less often. This appeals to me as a coder, but does it really
matter when it's all such a lottery in the first place?
On the minus side, it could be argued that breaking the link with
artifact.txt somehow unbalances the game, as finding randarts could be
significantly harder or easier than finding the standard artifacts (i.e. the
sum of effective rarities will be different). Does anyone care about
comparing the two?
A related issue is that of artifact depth. GW used max(old artifact depth,
old base item depth). Eight years of playtesting has revealed the hidden
truth that this tends to push randarts a bit deeper than they were before
randomisation. More significantly, it can result in two nearly-identical
randarts having significantly different depths simply because one of them
happened to be based on something deeper. My feeling is that if rarity is
properly computed and consistent, the artifact depth can be left equal to
(base item depth + 5), or something equally simple. I don't fancy trying to
incorporate the effects of OoD checks on effective rarity, but interested to
know if anyone has any strong views here.
Regards,
CC
|
| | 2 answer | Add comment |
|
| Funny randart Josh Hayes 01:27:08 |
| | I've got my usual dwarven warrior pretty deep in Zangband, and found this
little beauty:
The Sling 'Ulklechmur' (x3) (+19,+20) (+3)
It provides resistance to blindness. It affects your ability to tunnel. It
fires missiles with extra might.
Tunneling? With a sling? Really?
I don't see this as a problem, but it does strike me as pretty silly.
-JAH
|
| | 1 answer | Add comment |
|
