Nightlies

[buzzkill]

Well, Takkaria made the change, but as I understand it the basic idea is to reduce store dependency, and also solve the "scumming for restores" issue...

I fear, having not played with it yet, that it will make early stat draining inconsequential (if you're close to your next CL, don't worry about stat draining at all), and later stat draining... well that's not really a problem at all between sustains and stat potions and infrequent leveling (and prior to the change restore potions were abundant).

What bugs me is that it seems an runs an end around actually solving the greater scumming problem. If town scumming is undesirable, in the eyes of the maintainer, in this instance, then isn't town scumming in general undesirable, and if so then why not deal with that problem directly.

While I can't define it precisely, IMO it looks like change for the sake of change. After all, it is possible to continue on with mildly drained stats. It never necessitated town scumming... and dramatically drained stats, well that should be a huge hindrance, if not a player killer.

[Derakon]

Having actually played with the new system, here's how it's going for me:

* In the early game, drainers are not a huge deal because the next level is right around the corner. Given that, for the most part, early drainers were not a huge deal anyway for people who knew the game, I feel the main result here is that newbies can get introduced to the concept of stat drain without having to suffer massive consequences from it.

* In the late game, I have no sustains for my important stats (ironically, my warrior has sustained INT/WIS/CHA), so I've been carefully avoiding monsters that can drain stats. Stat potions are nowhere near common enough to rely on for restoration, and the only mushroom of vigor I've seen was in the black market. Drainers are now monsters to be avoided, because any draining will stay with me for a long time instead of until I next go back to town.

[Taha]

I really like the stat drain changes. It makes them less painful very early with quick level ups, but more of a pain in the mid game. At level 26 or so, with no ESP or SI, getting drained has hit me pretty hard a few times, but you live with it because there isn't an alternative. My perspective:

Plus:
Less grinding - significantly fewer trips to town, less need to scum or grind for money if a strength drainer catches you early.
Makes leveling up more interesting and exciting. Some strategic value in that near level up or with high xp monsters.
Fewer semi-useless / hard to ID potions to try when doing ID by use. Which is a mechanic I like.
Dealing with drained stats in the midgame. I never used to play for long with significantly drained stats once I had any sort of money. Now I do. Reduced str/con in particular is interesting to manage while you either play on or try to level up.
There is now a truly valuable mushroom. Vigor gets a slot in inventory when I can find one. They aren't hard to find early, but mushrooms all seem to get rarer as you drop.

Minus:
Game might be slightly easier early. This could go either way, it seems like it would smooth out the learning curve a bit for newbies, as I think it is a little more challenging afterwards.

Overall, I think it is an interesting change, and net positive. I'm curious what others think. The bug where you can lose a level from life drainers and then get your stats fixed on restore life levels needs fixed. Haven't tried to abuse it, but it has happened. Really need SI...

On unrelated notes, randarts are way way more rare pre level 30 (I know this was a bug that was fixed), and egos as well. In itself that has made the game much harder since I last played a few months ago.

The lowered selling values is big. Considering a no selling game, since most things aren't worth it now anyway. From what I hear about the size of the gold drops with no selling, I think you might end up richer in that setup now, and I have already stopped carrying anything less than ego weaponry and rings of escaping back for the money. Can you still sell items to ID?

Rods of treasure detection / rods of detection also seem much rarer. I haven't seen any this game, and only one treasure detect in the past three or four tries before dlvl thirty. Spending what money I have on stacks of scrolls at the alchemist.
________
LIVE SEX

[Derakon]

In no-selling, you can in fact sell items to the stores; you just always get 0AU for them. So basically you sell to ID, recharge items at the magic shop, or I suppose you could use the stores as risky temporary storage, though I can't imagine what scenario would provoke that.

And yeah, money's a bit too common with no-selling right now.

[Taha]

Keypad and Macro issue

The following is all going from the Feb 13 nightly to the Feb 20 nightly on OS X 10.6.

I get the following error on loading my file: "Parse error in /mypath/taha.prf line 475 column 1: P: undefined directive". The game loads, but the keypad doesn't work for motion - arrows do, as do number keys on the top of the keyboard. Interact with macros shows only options a-e, all relating to keymaps. All my macros are missing. Tried suiciding character and restarting, same issues.

Starting a new fresh character doesn't get the error, but the keypad still doesn't work. No way to interact with macros there either.

The savefile broken issue was going backwards, not forwards. Opening the modified file with the slightly older nightly gave an error about a corrupted save file. The funny part was the error: "Savefile too old. Try importing it into an older Angband first."

The old character and a new one and my prf file is attached.

Has anyone had a chance to look at this yet? I see new nightlies up, but hesitant to mess with it if nothing has been fixed. Losing the keypad hurts survival odds.

Oh, and I like no selling - I definitely have had more money and less hassle without selling than with the nerfed selling in the nightlies. I'm sure that will be balanced some before the next release, but having already adjusted to the drop in cash with selling, this feels like the logical place to go.
________
Web Shows

[scud]

Has anyone had a chance to look at this yet? I see new nightlies up, but hesitant to mess with it if nothing has been fixed.

Still there in the OS X nightly of 23 Feb. Playing on Intel 10.6.6.

Strange problem. Started a new character with 23 Feb issue. I'm *convinced* that the numpad worked until I 5ed to sit out confusion and discovered nothing was happening; since 5ing the odd (ie diagonal movement) numbers don't work, whereas the even (vert/horiz) numbers work intermittently.

Backtracked to 20 Feb issue with same character file, and the even numbers worked but the odd numbers didn't.

13 Feb issue is fine.

[PowerDiver]

Still there in the OS X nightly of 23 Feb. Playing on Intel 10.6.6.

Strange problem. Started a new character with 23 Feb issue. I'm *convinced* that the numpad worked until I 5ed to sit out confusion and discovered nothing was happening; since 5ing the odd (ie diagonal movement) numbers don't work, whereas the even (vert/horiz) numbers work intermittently.

Backtracked to 20 Feb issue with same character file, and the even numbers worked but the odd numbers didn't.

13 Feb issue is fine.

I tried the latest code a day or two ago, and the numpad never worked for stepping, but did work for running with the shift key. This is using X11 under Linux. I opened a ticket. Perhaps you should add your comments to that ticket.

[scud]

Perhaps you should add your comments to that ticket.

Done.









*adds words to hit the required character count*

[myshkin]

The input layer refactor of February 15 introduced the keypad problem and removed macros. I don't yet understand the code, and so have not come up with a code-level fix yet for the keypad problem. However, if you add the lines below to your .prf file, your keypad should work again. (Change the 0 in the C: lines to a 1 if you use the roguelike keyset.)

A:;1
C:0:{K}1

A:;2
C:0:{K}2

A:;3
C:0:{K}3

A:;4
C:0:{K}4

A:;5
C:0:{K}5

A:;6
C:0:{K}6

A:;7
C:0:{K}7

A:;8
C:0:{K}8

A:;9
C:0:{K}9

In addition, to convert your macros to keymaps, edit your .prf file. You will need to convert every P: line to a C: line. The format of the C: line is C:mode:keypress, where mode is 0 for the normal keyset and 1 for the roguelike keyset. The keypress syntax is somewhat different from the macros. Experimenting with the query mode may be the easiest way to understand the syntax for now, unless you want to read keypress_from_text() in ui-event.c. Here are a few examples:

Macro P: line                           Keymap C: line
P:\[F15]                                C:0:[F15]
P:\[shift-F15]                          C:0:{S}[F15]
P:\[KP_8]                               C:0:{K}8
P:\[option-KP_0]                        C:0:{AK}0
P:\[control-KP_0]                       C:0:{^K}0
P:\[Right]                              C:0:[Right]

[myshkin]

Actually, it turns out that the correct fix for the keypad bug is simply to add those entries into pref.prf. This should go into the nightlies soon. If you would like to test before then, you can get the raw version from github and drop it into .../lib/pref.

[PowerDiver]

Do those keypad lines need to go in any particular place? I pasted them into lib/pref.prf near the top, just before the # Movement comment, and my keypad problem remains.

[myshkin]

Do those keypad lines need to go in any particular place? I pasted them into lib/pref.prf near the top, just before the # Movement comment, and my keypad problem remains.

They should work fine anywhere, as long as you don't have another overriding keymap elsewhere. It is probably cleaner if you download a new pref.prf from the link above, though.

If you query the keymap for keypad 1 (Interact with macros => Query a keymap => KP 1), what key and action do you find, if any?

[PowerDiver]

They should work fine anywhere, as long as you don't have another overriding keymap elsewhere. It is probably cleaner if you download a new pref.prf from the link above, though.

If you query the keymap for keypad 1 (Interact with macros => Query a keymap => KP 1), what key and action do you find, if any?

When I press the 1 on the numpad, the Key line changes to

Key: {K}[End]

and below

No keymap with that trigger. Press any key to continue.

I got out, changed the line in the pref from {K}1 to {K}[End] , restarted and the 1 worked.

[myshkin]

When I press the 1 on the numpad, the Key line changes to

Key: {K}[End]

and below

No keymap with that trigger. Press any key to continue.

I got out, changed the line in the pref from {K}1 to {K}[End] , restarted and the 1 worked.

Thanks. I am a bit surprised that running worked for you, as I don't see any default mappings for {SK}[End]. Does anything change if you toggle numlock? I can't think of any harm to adding the Home, Page Up, etc. keypad mappings to pref.prf, but I may be missing something.

[PowerDiver]

I manage to get everything except the numpad 5 key to work. When I am in query keymap mode, and press it, nothing happens. I remember troubles with 5 in some older versions or maybe Moria. I think that is likely to be much harder to fix then adjusting a pref file. This is one of the reasons I was opposed to the change to searching. We need a redundant way to wait one turn, not using the 5 key. Hmm, maybe ',' currently fills that role.

These lines work for me, except for 5:

A:;1
C:0:{K}[End]

A:;2
C:0:{K}[Down]

A:;3
C:0:{K}[PageDown]

A:;4
C:0:{K}[Left]

A:;5
C:0:{K}5

A:;6
C:0:{K}[Right]

A:;7
C:0:{K}[Home]

A:;8
C:0:{K}[Up]

A:;9
C:0:{K}[PageUp]