[Announce] PosChengband 4.0.0 Released

[elliptic]

The new changes look good - I'm especially happy about removal of clone monster and of 555 (I was getting tired of watching other people abuse 555, and I use no_wilderness myself...).

As one of the people involved with making the "mpa changes", I thought I'd give my perspective on the three points of contention listed by chris. Other people involved might see things differently of course.

[1] Monster lore: I'd simply prefer to be able to look at monster memory in-game to check each time exactly which eight breaths the Yamata-no-Orochi has. The alternatives are to play with r_info.txt open or to occasionally get breathed on for 600 damage because I didn't remember the precise set of attacks some monster had. Monster memory that persists between games and grows as you encounter new monsters is a cool idea, but there are too many ways to lose access to your monster memory and have to start over with it.

[2] Item lore: I added the L25 automatic stone of lore effect because I almost always found Stone of Lore by around then (usually between L20 and L30) anyway. So if you are going to have access to unlimited ID anyway, why not simplify things and not make players carry it around forever and worry about accidentally dropping it somewhere? That way players can focus more on parts of the game that actually matter.

For ID = *ID*, there are a couple things I dislike about typical *band *ID*. First, with fixed artifacts it encourages looking at artifact spoilers to save *ID* for other items. This leads to wandering around with resistances inaccurately listed in your character sheet (but you know what they are anyway). Second, it is usually pretty confusing and complicated to figure out which items need to be *ID*ed because they might have extra properties - or to figure out which properties you can detect by convoluted testing.

I believe in Poschengband there is a town *ID* service, so these issues aren't as bad as in some other *bands - they just mean that you have to go back to town frequently and spend a little money. This still isn't very interesting.

[3] no_selling: Bringing back items to town to sell them is tedious and uninteresting. There are already more interesting tradeoffs for how to spend your pack space - items that actually do something or that you might want to use some day.

That said, digging for gold is just as bad. Personally I basically never dig for gold, and I think most people who use no_selling don't either. I would be very happy myself if digging for gold just wasn't a thing, and players had to explore and fight in areas of the game that actually contain monsters to get gold.

If there are concerns about monsters dropping too much gold in no_selling, that can just be reduced. I would happily play with no_selling even if it didn't get any bonuses to gold drops compared with selling.

[debo]

If every player got something like alchemy or a lot more gold came from drops, I'd be very happy to play with selling on. I just hate going back to town every 10 mins because my inventory is full.

These changes look great. Looking forward to the eventual release.

[Derakon]

Keep in mind that the logic for Vanilla's no-selling option was that lugging items back to town to sell for cash wasn't fun and therefore shouldn't be rewarded. It's not that the players weren't getting enough cash from the dungeon, it's that they were getting any remotely significant amount of cash from selling stuff. Even if the distribution was 99% dungeon, 1% selling, I strongly suspect you'd still see players bringing back "worthless but valuable" items to town to sell, to get that extra little bit of cash.

The idea was that if you chose no-selling, then you'd have roughly the same amount of cash as a "selling" player would at any given point at the game, without being required to carry around useless crap and make repeated return trips to town just to lighten your pack. I won't try to tell you how PosChengband should play, but I do think that just increasing the amount of cash that drops in the dungeon won't accomplish what the no-selling option was intended to add to the game.

[clouded]

Yay chris! These changes look good and I'm REALLY glad you got rid of 555 and clone monster. MPA no selling isn't as agreed upon as you might think, I definitely agree that it encourages too much digging and later on you get absurd amounts of gold (spend 10 minutes digging in the Mine for 1.5M) so I'm interested to try your approach.

[chris]

A new beta release (4.0.2) is ready in the usual location. I really wanted to hold off releasing until this version is stable and balanced (that would be when I would call it 4.1) but after another flurry of development, I need to take a bit of a break. Perhaps by the end of the year?

So, what changed?

First of all, I see many people are playing with "mpa changes", so I decided to look and see what those might be. Aside from bug fixes, the 3 main issues I see are:
[1] Complete removal of monster lore
[2] Almost complete removal of item lore (Id = *Id* and everybody gets automatic *Id* of every object at CL25)
[3] No Selling

Well, I can't say I agree with these, but I went ahead and added an easy_id birth option. It is not so munckinesque as mpa, but it makes Id the same as *Id* and also grants instantaneous pseudo-id at CL1. The automatic Loremaster change is too much for me to stomach, though, so you'll need to suffer a bit!

I did nothing at the moment to address monster lore. I suppose the cheat_know option could be replaced by a birth option called easy_lore?

For no selling, I previously removed this option since, in my experience, it promotes/requires digging for gold. In fact, the one time I tried it, I spent almost 80% of my game time digging (in the end game), but I was using Angband's gold boosting code at the time (not mpa's) and I was also playing an old time "One and Done" devicer, where you needed to recall to town to recharge after killing each notable end game monster (cost 20k to 30k to recharge rockets, etc). So my experience was probably unfair.

But honestly, I don't understand why we need an option for no selling. Instead, I think the problem might be the gold distribution. As monsters grow more powerful, they stop dropping gold. This is considered a "bonus". Also, deeper chests don't drop gold at all. This means your only sources of gold in the late game are digging and the occasional floor drop, along with a handful of monsters like hydras.

For this release, I boosted the amount of gold drops from monsters and chests, but *not* from digging. (I even made digging less effective.) Also, DROP_GOOD and DROP_GREAT monsters now usually drop (extra) gold as well. You should be finding much more gold than before, and digging should only be considered as an early game option or just as a casual diversion to pass the time as you wander down long corridors ...

So that we can make intelligent judgments, I added statistics for this release displaying how much gold you find versus how much you get from selling, etc. My current character is finding about 50% as much gold as that gained from selling, which seems to negate the need for a no_selling option at all (but I made this change only yesterday, so my experience is very limited).

Pack usage is a gameplay choice, and if you want to use more and sell less, you should be able to do so, but there should be consequences to this. But this playstyle should be possible as well, and I don't think it was before I boosted the gold drop amounts. So play with the changes and let me know what you think ... I'd like the distribution of gold gained to be balanced, perhaps 33% gold found and 66% from selling? If selling is too overweighted, we can boost the drop amounts a bit more or decrease the resale values of objects to even things out.

I also grabbed a bunch of bug fixes from mpa. I'd prefer not to get pull requests, since they invariably rot to the point of mega merge conflicts by the time I get to them, and I end up spending more time figuring out a merge than it would take to implement myself. Also, when I access the internet, it is for a restricted amount of time (2 hours) which means I simply don't have enough time to play with github. Issues would be nice ... But it might take me months to a year or so to get to them. Please be patient

Next of all, I did some work with the wilderness, including:
[1] I fixed the 555 and 666 bugs. Talk about unanticipated consequences from a minor code change! Wow. Sorry about that ... And if you were enjoying these as a feature, I'm sure they'll show up shortly on the mpa branch
[2] The wilderness is now more difficult. Be careful, especially early on. Monsters are much more alert and will act over longer distances on the surface, so you are likely to get into trouble. This change has not been thoroughly playtested yet ... I no longer scum the surface like I used to.

Penultimately, I did alot of work on objects, especially on object distribution. This is a major gameplay rebalance effort. You might want to finish playing out on 4.0.1 before upgrading, as the early game feel will be much different. Speed and Spellbooks have been addressed, so the need for 555 and 666 exploits should be mitigated. Rockets *should* be rare, and end game devices are OP, so are intentionally hard to get. _Healing is still a bit deep, but have you tried _Holiness lately?

Finally, I recoded the Weaponsmith. This is a very enjoyable Hengband classic, and I thought it could do with a bit (actually a lot) of love, especially in the UI department. I'm still playtesting this one, so there may be future changes, but you might want to check this class out. (You won't be able to upgrade Weaponsmiths from old versions, though). The Judgment power has been removed and Smithing is now a class power rather than a hijack of the 'm' command. When you smith an object, you automatically judge its quality (pseudo-id until CL10, then id until CL30, then *id*). You can view your essences on the Character Sheet (Cee) or using the Class Extra Knowledge command (~x). Also, on the smithing top level menu, you can look at your current essences (L). Absorption is available from the Smithing command, of course, but it also is applied automatically whenever you destroy an object. So, the autodestroyer is your friend

[chris]

Because I haven't yet gone through the effort of reading and understanding every change made in it (more difficult what with changes getting bunched into a single commit). Breaking shoggoth speech was pretty bad, as was removing no-selling, so I'm extra leery of blindly pulling.
I guess I should just use mpa-poschengband, since that's technically 4.0.1 and fixes the most glaring issues, but I would still have a lot of reading to do.

Maybe once I finish my Ring.


Edit: Messed with more things, and discovered my display bug was from the command-line option to have multiple windows/terminals, my full command being—
./poschengband -f -gcu -- -right 57x26,* -bottom *x20
— where removing everything after -- 'fixes' it, at the cost of no additional terminals. Looking at main-gcu.c my syntax is correct (or at least as it says in the comment) so the issue is probably in the bunch of code that was completely re-written starting with that comment.
I looked at the commit titles of 4.0.1 at least, and there aren't any pertaining to this.

I can't repro a problem, but your command line is wrong. You should be using -mgcu not -gcu. For example:
./poschengband -f -mgcu -- -right 57x26,* -bottom *x20

Otherwise, its probably just standard Linux incompatibility issues ...

[chris]

By the way, yesterday I gave the Vapor Quest some Doom Quest II treatment. This quest always takes forever and it has annoying hidden monsters, so here's a new, smaller, more lethal version: https://bpaste.net/raw/7ecede0e7d99

The most notable monster remains (weird fume) and there's now a shimmering vortex which can shriek and wake up all the air elementals which will quickly swarm you. It's harder so I upped the danger level to 40 and also added six rings and six amulets as loot inside, it could be worth trying it early for those.

May update or make some new quests too, I did this one impulsively.

I got a 404 on this one ... I know, I'm a bit slow with no internet access these days. Perhaps you could add a new issue on the github project? I promise to eventually get around to those

[wobbly]

Anyone know if !good works properly with weak pseudo-id? It's auto-destroying un-ided (enchanted) for my priest & both good & egos show as (enchanted) until ided with weak pseudo-id.

I think I found the answer to this. "You feel the Augmented Chainmail in your pack is enchanted. You can not destroy the Augmented Chainmail (special)". So it won't destroy artifacts but it's still auto destroying egos. That's an annoying hassle till you reach level 35 or find the stone of lore.

Edit: Worked it out ?!good does what I want

[debo]

hello!
Can you guys suggest me how to solve this issue?

https://i.imgur.com/ETAs4Gn.png

How can I make -more- prompt go away in the latest version?If not,how can I atleast avoid it (through a macro or something) when resting?When I rest,my "clear mind" ability automatically sets in,which is awesome.What is not awesome though,is the spam of it. Pls help ;_;

Automore is gone, RIP.

[poschengbandplayer]

hello!
Can you guys suggest me how to solve this issue?

https://i.imgur.com/ETAs4Gn.png

How can I make -more- prompt go away in the latest version?If not,how can I atleast avoid it (through a macro or something) when resting?When I rest,my "clear mind" ability automatically sets in,which is awesome.What is not awesome though,is the spam of it. Pls help ;_;

[krazyhades]

I think the exp gain from destroying spellbooks only works with Paladins.

Definitely incorrect. It works for warriors. I can't speak to the auto-destroy vs. manual destroy issue, but I know I've manually destroyed high books as a warrior for XP. (It's really low xp too so the selling cash is better)

[HugoVirtuoso]

I think the exp gain from destroying spellbooks only works with Paladins.

[wobbly]

Does auto-destroy not give a warrior the spellbook xp? I just auto-destroyed book of the unicorn & didn't seem to get any.

[HugoVirtuoso]

Note sure if this is mentioned, but there is a major inconsistency in the turncount between PosChengband 4.0.1 chardumps in .txt format vs. .html format. Chardumps in .html format have *way* higher turncounts than the .txt versions. For instance, have a look at the difference in turncounts between Hugo$oul 8356 (winning character with its chardump in .txt format) and my preceding 4.0.1 Hugo$ouls (#8355, #8354, #8353, etc.) (chardumps in .html format). The same would apply even for both types of chardumps given the same exact game day + game time.

And of course, the oook ladder doesn't recognize winners in .html chardump format. i.e. No GREEN highlights

[wobbly]

Minor convenience issue: Resting for the night at the inn doesn't heal your mount