Making the game harder, take two

[Magnate]

...and there's the problem. If you have items in shops that can't be guaranteed to be there then someone will town-scum to get them. Either they need to be guaranteed or so rare that town-scumming becomes moot point. Or have some other solution to that problem.

I agree. The only sensible probabilities for items available in shops are 0% and 100%. Anything else encourages town-scumming and leads to calls for a "buyout button" etc. etc.

[bio_hazard]

As a new and not-very-skilled player, I don't personally see a need for the game to get harder, but I certainly don't have a strong objection for the game to cater to the folks willing to put in the development time. Someone mentioned this before- 3.1.2 will still be here, and if I prefer that to the newer version, I can always play that. All I'd recommend would be to make it easy to figure out what the differences are if someone wants to choose a version to play.

Another thought- If the "problem" is that experienced players know what all the risks are, then maybe something to make it more interesting would be to randomize some features of the monsters, such that players would need to learn about monsters each game.

[Derakon]

I think that's the wrong kind of "I don't know the risks" to start adding, personally. I would say the big problem that experienced players are having with the game being too easy comes down to the game almost always providing all the tools you need to avoid taking any significant risk. By the time you encounter scenarios where you might need to Teleport Other, wands have been dropping for 500'. By the time you can start getting really in over your head, ?Destruction and _Destruction have been dropping for ages.

Ideally IMO the big escapes should be so limited in number that the player would almost always rather soldier on than use them. ?Destruction should be as rare as !Life, _Destruction should go away entirely, wands of TO should only have 2-3 charges in them (and be bloody hard to recharge), etc. Back in the day the only escape I used with any frequency was ?Teleport, which is risky as hell. And that's good, because it means I constantly have to decide between fighting on against dangerous enemies or teleporting away and risking getting blown up by whatever I land next to. Eventually I learned this was suboptimal play, and now I stop using Teleport at all below 2500' or so, in favor of TO, Teleport Level, and Destruction.

Short version: downplay the guaranteed escapes in favor of escapes that are not guaranteed. Right now it's too easy to get away from trouble.

Edit: adding a response to dos350.

please to not be so radical~

why does angband need change so much, i have only played 3122 and 309, but i think is no need to change~~~~~

the item restock of town is annoying but ok~

making so stores dont regen ever is LUDA imo

also auto id is not fun~

why is wanted?

You've missed all of the earlier discussions we've had on these topics, but to recap in brief:

* Angband will keep changing so long as people can keep thinking of improvements to add. Far better for the game to keep evolving than for everyone to lose interest, causing it to die like Moria has. As Magnate noted, that doesn't mean you have to play the latest version. Angband 3.1.2 will still be available if you don't like the changes that have been made.
* It's generally accepted that the shopping aspect of the game is broken and boring. Players dive into the dungeon, find equipment they don't want that shopkeepers will nevertheless pay top dollar for, and use that to fill up on all sorts of items that vastly increase their capability to handle challenges. If the town doesn't have the items they want, then they just wait in the dungeon until the town has restocked. That's just stupid. Ideally the player would find all of the items they need to be successful, and we wouldn't need a town, but we aren't there yet; the logical next step is to remove random stocking from the town. That means that a given store will either always stock an item, never stock an item, or only stock a guaranteed limited quantity of the item (or some more exotic solutions like always stocking the item but continually increasing the price it sells for).
* The identification game is almost pointless. Experienced players know that past the first 1000' or so, the risk from wielding un-identified equipment far outweighs the benefits of saving an ID use, so nothing gets used until it has been identified. Identification is also cheap and readily available. Thus the only purpose of identification past that point is to keep players from knowing what something is until they have picked it up and spent a turn on the identification process. This largely just adds tedium to the gameplay; hence the suggestion that past a certain point the player should simply automatically know what every item they pick up is. We'll see how it plays out in practice.

[Magnate]

As a new and not-very-skilled player, I don't personally see a need for the game to get harder, but I certainly don't have a strong objection for the game to cater to the folks willing to put in the development time. Someone mentioned this before- 3.1.2 will still be here, and if I prefer that to the newer version, I can always play that. All I'd recommend would be to make it easy to figure out what the differences are if someone wants to choose a version to play.

Well, every version has had quite detailed changelogs pointing out changes from the previous version. Admittedly aggregating these in your head and working backwards would be a little tedious. Fortunately takk has already created a page at rephial.org for every release (click on Releases on the front page). I'm sure he would welcome anyone sending him a text file describing the key features of any particular release, which he could then add to that release's page. That way people wanting to browse different releases can see some basic text about that version, as well as the changes from the version before.

[d_m]

I agree. The only sensible probabilities for items available in shops are 0% and 100%. Anything else encourages town-scumming and leads to calls for a "buyout button" etc. etc.

This is only true if you assume that store restocking is the status quo.

One thing I've wanted to try is a game where shops never restock... they have a random initial inventory and then that's it. So if the Black Market has boots of speed, great, you'll buy them eventually. If not, you will never see them in a store.

This removes any scumming aspect while keeping things somewhat random. It also means that you can make things available in limited quantities without people being able to always have a giant stack of them. And it means that selling stuff you might want to shopkeepers is a great strategy, because you will always be able to rebuy it later. It has the effect of making all the stores semi-pawnbrokers.

This might end up being more of a variant idea, but I will probably try it in a branch.

[Magnate]

I've been playing since 2.4.frog-knows and was playing Moria before that, and I do worry that much change is being made for the sake of change.

No, it isn't. Nobody has enough time on their hands for that. It may not always be clear to everybody why any given change has been made, but it's always got some reasoning behind it. Most (but not all) that reasoning has been captured at trac.rephial.org, so anyone who cares can look there for enlightenment. Or failing that they could ask here and be quite likely to get an answer.

Magnate mentioned earlier in the thread that the "days where an iron-fisted dev runs V single-handedly are gone". Frustrating as that might be for some people, that method did have its advantages. Change to V was very slow and incremental, and almost nothing was put in experimentally - most additions were things that had already been added to one or more variants and demonstrably worked there.

So far, my experiences with 3.0.9 and 3.1.2v2 have been great - but I am concerned that V will end up being "designed by committee".

The comment about the iron-fisted maintainer was d_m rather than me, but your point is well made. Yes, the danger of the new collaborative development model is precisely this one of ending up "designed by committee" (i.e. overly compromised). But the plus side of collaborative development is that you can achieve a lot more development with a team's time and energy than with that of a single individual. So it's a trade-off of quality for quantity: purity of vision for quicker bugfixes and responses to feature requests. For example, I sent my randart patch to the then maintainer before 2.9.6 was released: by the time he stepped down three years later (after 3.0.6) he still hadn't tested and incorporated it - he just wasn't really interested in fixing the previous randart code and had too many other calls on his time. Nowadays anyone who sends a patch can expect a response within a few weeks, even if for whatever reason it won't be used.

Takk is quite open about the fact that there isn't a grand strategy or vision for V. There is a roadmap (again, go to trac.rephial.org) which is in the process of being updated and gives a little insight into what the active devs feel like working on for the next release. But that's not the same thing, and nobody's pretending it is. Anyone who wants to publish such a vision is welcome to do so - the whole point about moving to git is that it's now easier than ever to "do a Ben" and wow the world with your cleaned-up, focused, better version of Angband. Don't be shy.

For the record, while I accept Eddie's logic that removing rather than adding things will eventually result in a harder game, I don't agree that that's the correct overall approach. People who want to go backwards can play previous versions - that's why they're kept available. IMO it is more creative and better long-term to rebalance without going backwards. That's not to say that nothing should ever be reverted: some things have been and I'm sure others will be in future (Haradrim no longer gives an extra shot in recent nightlies, magi and rogues no longer get *CLW). But in general I prefer to devote my energies to solving problems a different way. Someone described this as "changing something to fix something else" - I'm quite comfortable with that, and I think I can avoid the escalation pitfalls associated with it. Especially with the helpful feedback I usually get here. (N.B. I draw a significant distinction between toning down the impact of something and removing it from the code, and I like a lot of fizzix's suggestions for toning down the availability or impact of various things.)

There is one set of issues I will be leaving completely to other devs: anything associated with shopping. I have very detailed views on how Angband's economic elements should work, but I don't have the energy to try and persuade others of them, and I'd rather leave that for my own never-to-happen variant.

Anything else I'm happy to discuss, fix and improve. I do like the idea of getting rid of WoR and having those "portal rooms" instead. That really would be a whole world of more challenging.

[TJS]

This is only true if you assume that store restocking is the status quo.

One thing I've wanted to try is a game where shops never restock... they have a random initial inventory and then that's it. So if the Black Market has boots of speed, great, you'll buy them eventually. If not, you will never see them in a store.

This removes any scumming aspect while keeping things somewhat random. It also means that you can make things available in limited quantities without people being able to always have a giant stack of them. And it means that selling stuff you might want to shopkeepers is a great strategy, because you will always be able to rebuy it later. It has the effect of making all the stores semi-pawnbrokers.

This might end up being more of a variant idea, but I will probably try it in a branch.

Yes this is pretty much exactly what I was saying earlier. It would allow consumables to be plentiful at first but not so much later on, solve town scumming, make ammo more valuable and reduce the amount of time you spend deciding what to throw out of your home.

If you do decide to make a patch I would happily test it out thoroughly.

[EpicMan]

On the subject of shopping, I think most of the stores should have limited varieties of objects, so town scumming is pointless. Items that players have to have (food, word of recall, oil for torches) should just be infinite stock, so you can always buy how much you need, and probably the weapon/armor shops should stock some basic (i.e. neither EGO, enchanted, or rare base item) weapons and armor in infinite quantities for flavor (so the Weapons Shop will have some weapons). Anything else (enchanted items, egos) might be randomly generated at the start of the game and should be limited.

Ammo I'm not sure about, but allowing inifinite +0,+0 arrows and bolts and not having restocking scrolls of enchant shouldn't be too overpowering.

If we end up removing things like hunger, ID scrolls, etc and add alternate ways to get up and down the dungeon we could even remove the infinite stock items and just have a random assortment of non-restocking items.

[Magnate]

This is only true if you assume that store restocking is the status quo.

One thing I've wanted to try is a game where shops never restock... they have a random initial inventory and then that's it. So if the Black Market has boots of speed, great, you'll buy them eventually. If not, you will never see them in a store.

This removes any scumming aspect while keeping things somewhat random. It also means that you can make things available in limited quantities without people being able to always have a giant stack of them. And it means that selling stuff you might want to shopkeepers is a great strategy, because you will always be able to rebuy it later. It has the effect of making all the stores semi-pawnbrokers.

This might end up being more of a variant idea, but I will probably try it in a branch.

I quite like this, but it suffers from a metagaming exploit: if I don't like what the RNG provides on turn 1, I just keep restarting until the BM contains Boots of Speed (+10). So I still think that randomness is quite dangerous. I'd fully support non-restocking of fixed inventory though: the Temple a 5 restore stat potions of each flavour, and that's it ...

I also think that stores keeping stuff you sell is tantamount to expanding the home (since there is no way to make money scarce enough to matter), which I've always opposed.

Clearly, MM does V. Anyway, I need to resist the temptation to respond to shopping posts!

[EpicMan]

So make the black market only accessible after paying a large fee. You can't see what's in it when you start your character, and the time you've invested in a character accumulating the gold to open it up should make you reluctant to restart the game without a good Black Market roll.

[Magnate]

So make the black market only accessible after paying a large fee. You can't see what's in it when you start your character, and the time you've invested in a character accumulating the gold to open it up should make you reluctant to restart the game without a good Black Market roll.

That's a brilliant idea - it makes d_m's approach viable at a stroke. I'm thinking about half a million gp, which you reach somewhere in the third quarter of the game.

[bio_hazard]

Is the goal to make the RNG a bigger part of the game? I think the stores (and particularly BM) have a useful purpose to supply things that you need (FA, rPois, etc) that sometimes the dungeon just refuses to give you no matter how fast you dive.

[Dean Anderson]

This is only true if you assume that store restocking is the status quo.

One thing I've wanted to try is a game where shops never restock... they have a random initial inventory and then that's it. So if the Black Market has boots of speed, great, you'll buy them eventually. If not, you will never see them in a store.

This removes any scumming aspect while keeping things somewhat random. It also means that you can make things available in limited quantities without people being able to always have a giant stack of them. And it means that selling stuff you might want to shopkeepers is a great strategy, because you will always be able to rebuy it later. It has the effect of making all the stores semi-pawnbrokers.

This might end up being more of a variant idea, but I will probably try it in a branch.

I tried that kind of system once - not in a variant, but in a non-band roguelike I wrote from scratch (and I must get around to finishing one day)...

I used a combination of:

1) Limited inventory space in each shop

2) No new items in the shops

3) Limited funds for each shopkeeper to buy stuff with

This made an interesting shopping mini-game.

In the very early game, you're buying and selling stuff as normal.

But as the game goes on, the shopkeeper runs out of inventory space far quicker than they run out of money, so you can no longer sell them stuff. You end up buying cheap stuff from them and throwing it away just to free up the shop space so that you can sell them more valuable stuff.

And then eventually they run out of money to buy things with, and you end up using them as effectively extra houses. You've already cleared them out of junk, and each one contains a bunch of high level stuff you sold them and that you can buy back if you need it, promptly getting back the money you just spent by selling them something else to replace the item you just bought from them.

It works well, in its own self-contained way; but it might be a leap too far for Vanilla and more suited for a variant.

[Dean Anderson]

The comment about the iron-fisted maintainer was d_m rather than me, but your point is well made. Yes, the danger of the new collaborative development model is precisely this one of ending up "designed by committee" (i.e. overly compromised).

Yeah - I don't want to come on too strongly about the issue. It's not something I have a bee in my bonnet about, and I'm certainly not saying I think its happening now; but it is something that people should be aware of because it can kind of creep up on collaborative projects unnoticed...

[PowerDiver]

For the record, while I accept Eddie's logic that removing rather than adding things will eventually result in a harder game, I don't agree that that's the correct overall approach. People who want to go backwards can play previous versions

Removal is not reversion. Taking satisfy hunger away from mage-casters is a step forward, not a step backward.