New store inventory management

[Magnate]

Following on from the "buyout button" poll, and if Takkaria is willing, I am keen to fix the store-scumming problem now, rather than see a buyout button implemented and have the problem continue.

I think, from reading the thread, that everybody agrees that if a store is ever going to stock something, it should be available. The buyout campaigners want to do this by buyouts, and the anti-buyout crowd would prefer to see some other mechanism for setting the price of the desired item.

So there are two issues here: which items should be available (ever), and how should we determine their prices.

I'm happy to hear views on the former, but to be honest I have no real problems with the items which are currently stocked. The BM is obviously excluded from this entire discussion - it's up to Takkaria if he wants to remove it, but I'm not planning on touching it.

So we're left with discussing the prices. Each item has a base price, which (for consumables) is fixed in object.txt - I'm not planning to change these. The discussion needs to focus on how prices for consumables escalate when they're restocked.

My plan is to create a "market prices" array, so that the "current" store price of every consumable can be tracked separately from the base price. This means messing with savefiles, if Takkaria will allow that. (Another way to do it is to add a market price element to the object kind structure, but it amounts to the same thing.)

My assumption is that the price increase of each item will be roughly proportional to its base price. But some things are much more commonly required than others (I'm thinking ?phase and !CSW, not to mention ?WoR and ammo) - perhaps these should not escalate at all, like at the moment? Other things (like ?enchant) should definitely escalate - perhaps as much as 50% of the base value each time the stock runs out.

I don't want to track how many of each item have been purchased (means yet more savefile bloat), but we might need to tone down the randomness of amounts - currently you can find 1 !CSW or you can find 79. We might need to agree small variances around fixed figures, so that prices escalate after roughly constant purchase numbers. For things like ?phase this would be quite a large number (30? 50?) and for things like ?+dam it would be a lot smaller (5? 10?).

Sorry if that's a bit rambling, but keen to know if people would be interested in this, or if we'd rather carry on arguing about a buyout button.

EDIT: sorry, finished that in a bit of a hurry. I meant to say that the price increase should be random *up to* something like half base price, and also that there should be a possible random reduction to represent other sellers, new products etc. So the actual market price adjustment on restock should be something like:

market price = old price + randint1(base price * 2 / 3) - randint0(base price * 1 / 3)

.... that would give an average increment of 1/6 base, with the possibility of up to two thirds, or up to 1/3 reduction. That should prevent prices of desirable items spiralling out of control, but would need testing to ensure that it's not too generous for ?+dam etc.

Then of course one could add a small (one_in_(whatever)) chance that the price drops dramatically - maybe all the way back to base - some sort of 'price reset' mechanism. This may appease those who cling to the fact that buying out stores doesn't currently increase the actual item prices ...

[Donald Jonker]

Everything seems pretty sensible. Have a few questions to throw out there, though.

1. Are all items currently capable of being generated always guaranteed?
2. If so, will there be room for everything + reasonable space for selling to stores?
3. Is it feasible to have no town source of endgame consumables w/ current dungeon generation?
4. Should BM-only, but non-endgame consumables be put on sale at a high premium at regular stores? (?tele, ?telLevel, &c)
5. Should the Armory and Weaponsmith simply be replaced by good starting equip?

[Pete Mack]

I don't think these changes belong in V for the first attempt. Angband is fun because it's not predictable (or fair). I don't see why making it more predictable is likely to make it more fun.

[Donald Jonker]

I don't think these changes belong in V for the first attempt. Angband is fun because it's not predictable (or fair). I don't see why making it more predictable is likely to make it more fun.

But the whole point of having a town is that it's more predictable than the dungeon. Just how predictable you want it to be is a matter of taste, I suppose. Even store restocking boils down to a measure designed to add a greater measure of predictability.

[buzzkill]

Sounds a little like simulated supply and demand, something I've suggested before. Maybe supply isn't involved, but the more you buy, the more it costs, which is the important bit.

My assumption is that the price increase of each item will be roughly proportional to its base price. But some things are much more commonly required than others (I'm thinking ?phase and !CSW, not to mention ?WoR and ammo) - perhaps these should not escalate at all, like at the moment? Other things (like ?enchant) should definitely escalate - perhaps as much as 50% of the base value each time the stock runs out.

I'd like to see price escalation for all items, if for no other reason, to try to lessen dependence on the stores. Perhaps the price increase for commonly needed items could be less than the average.

I don't want to track how many of each item have been purchased, but we might need to tone down the randomness of amounts - currently you can find 1 !CSW or you can find 79. We might need to agree small variances around fixed figures, so that prices escalate after roughly constant purchase numbers. For things like ?phase this would be quite a large number (30? 50?) and for things like ?+dam it would be a lot smaller (5? 10?).

Agreed. The randomness, now that unlimited amounts are available, should be tuned down. Also, I'd try tune these numbers to levels commonly needed in the early/mid game, because after gold becomes overly abundant this mechanism won't make a bit of difference.

A little off topic: Is there any chance of decreasing sale price of items, making it dependent on CL, and in exchange, increase the size of gold deposits/drops/finds in the dungeon. I'm looking for less gold from selling and more gold from actually finding it and killing stuff.

[Magnate]

A little off topic: Is there any chance of decreasing sale price of items, making it dependent on CL, and in exchange, increase the size of gold deposits/drops/finds in the dungeon. I'm looking for less gold from selling and more gold from actually finding it and killing stuff.

I'd just go with Eddie's no selling, with 3x or 4x gold drops. Anything less still leads to optimisation of selling.

[Magnate]

Everything seems pretty sensible. Have a few questions to throw out there, though.

1. Are all items currently capable of being generated always guaranteed?

That's my thinking, yes. AFAICT there isn't anything terribly unbalancing available from the non-BM stores, if we have the right price curve for ?enchant etc. I'm only talking about consumables btw - I'm not talking about the occasional ego items which show up in the weapon and armour stores (I'm quite happy to leave them be), nor am I talking about rings/amulets from the magic shop. In fact, I guess we're really only talking about stores 1/4/5, plus ammo from 3 and town books from 6. I was originally thinking wands/staves/rods too, but as there are so many and only one non-BM place you can buy or sell them, perhaps we leave them be also. Quick poll: has anyone ever bought out the magic shop, and if so what were you after?

2. If so, will there be room for everything + reasonable space for selling to stores?

Well, as Pete said, you should never sell anything to the alchemist (5) or the temple (4) anyway. I guess the only problem will arise with the magic shop (6), and this solution might make it more difficult to sell wands/staves/rods/rings/amulets. So perhaps we don't mess with 6 apart from town spellbooks.

3. Is it feasible to have no town source of endgame consumables w/ current dungeon generation?

?? Other than the BM, which stores offer endgame consumables at the moment??

4. Should BM-only, but non-endgame consumables be put on sale at a high premium at regular stores? (?tele, ?telLevel, &c)

Not unless you want an infinite supply of them. Arguably you could allow ?tlev or ?tself with a high enough price ramp (so only a few are affordable until the late game).

5. Should the Armory and Weaponsmith simply be replaced by good starting equip?

Sorry, I wasn't clear - I don't see those as involved, since they're not about consumables (with the exception of ammo from the weaponsmith). My assumption is that the only stores ever bought out are 1/4/5/BM, but I'm interested to hear otherwise.

[Donald Jonker]

I was originally thinking wands/staves/rods too, but as there are so many and only one non-BM place you can buy or sell them, perhaps we leave them be also. Quick poll: has anyone ever bought out the magic shop, and if so what were you after?

Yes. Spellbooks, _telSelf, and (less often) _dEvil.

Well, as Pete said, you should never sell anything to the alchemist (5) or the temple (4) anyway.

Despite making good sense, this is sloppy. And you really only should avoid selling to 4 or 5 if you plan to buy out the store.

I guess the only problem will arise with the magic shop (6), and this solution might make it more difficult to sell wands/staves/rods/rings/amulets. So perhaps we don't mess with 6 apart from town spellbooks.

This seems even sloppier. Many items at 6 are just as essential as those at 4 or 5.

?? Other than the BM, which stores offer endgame consumables at the moment??

I missed the fact that you want to keep the BM. In that case I still want the buyout button. But I'd like to hear from folks with more experience against Morgoth - maybe buying out the BM isn't as necessary as I think.

Not unless you want an infinite supply of them. Arguably you could allow ?tlev or ?tself with a high enough price ramp (so only a few are affordable until the late game).

I'm just thinking that they're probably too rare already (especially ?tself). ?tself has been made readily available in other variants w/o being unbalancing.

[Pete Mack]

Teleport Level was too rare for a while, after the change in black market depth. The addition of Deep Descent has mostly fixed it. You really shouldn't need more than a dozen or two escapes in a game.

[Magnate]

Yes. Spellbooks, _telSelf, and (less often) _dEvil.

Well, you teleport much more than I do, but fair enough for spellbooks and _dEvil.

Despite making good sense, this is sloppy. And you really only should avoid selling to 4 or 5 if you plan to buy out the store.

That's a fair cop - I should not rely on aspects of the problem I'm trying to solve. So the answer to your original question is that each store has a fixed number of slots (24 I think), so as long as there aren't 24 items people want guaranteed to buy, then yes there will be room to sell. I guess the alchemist and magic shop will be fullest, but even then there shouldn't be 24 guaranteed consumables in either. Of course, since we're changing stores, we could just make sold items not appear in shop inventories ...

This seems even sloppier. Many items at 6 are just as essential as those at 4 or 5.

Other than town spellbooks, I don't instinctively agree with that. Just about the only essential wand/staff/rod is tOther, which is BM-only AFAIK. I'd be interested in what else you consider essential, other than _tSelf and _dEvil as mentioned above.

I missed the fact that you want to keep the BM.

I don't want to keep the BM, but I'm separating the issue of guaranteed consumables from that of the BM. Nothing in the BM is guaranteed, so it is not part of this discussion. If Takkaria wants a buyout button purely for the BM, that's fine with me. I'd still like to fix the other shops.

I'm just thinking that they're probably too rare already (especially ?tself). ?tself has been made readily available in other variants w/o being unbalancing.

As I said earlier, you clearly teleport a lot more than I do. I don't quite agree with Pete's two dozen escapes, but I don't consider tSelf an escape (as monsters get the first move when you land, unlike a level change).

[Donald Jonker]

Well, you teleport much more than I do, but fair enough for spellbooks and _dEvil.

I would strongly urge making _telSelf guaranteed. It's utterly necessary for survival, no matter how many you go through. For me it tends to be pretty variable - however lucky I get with the hounds & vortices around statgain. After that it gets better.

I guess the alchemist and magic shop will be fullest, but even then there shouldn't be 24 guaranteed consumables in either. Of course, since we're changing stores, we could just make sold items not appear in shop inventories ...

Lets see... there's:

1.!rStrength
2.!rInt
3.!rWis
4.!rDex
5.!rCon
6.!rChr
7.!rHeat
8.!rCold
9.?dTrap
10.?phase
11.?dTreasure
12.?dStairs
13.?mapping
14.?recharging
15.?ID
16.?to_hit
17.?to_dam
18.?enchant_armor
19.?confuse
20.?light
21.?WoR
22.?Satisfy_Hunger
23.?dInv

Phew! Just made it. I think that's all of them. I don't think disallowing sold items to appear is acceptable. What if you want to buy them back? What if you sold an early unidentified !STR and had the cash to rebuy it? You can probably cut this list down, though, but not by a lot. ?confuse and ?Satisfy can probably go. Putting the restoration potions at the Temple would be intuitive enough.

Just about the only essential wand/staff/rod is tOther, which is BM-only AFAIK. I'd be interested in what else you consider essential, other than _tSelf and _dEvil as mentioned above.

They're essential if you don't have the scrolls. If the scrolls are guaranteed but the staves aren't, then they're marginalized to the point of uselessness. You'll never use _mapping if ?mapping is always available. Ditto perception, dStairs, dTreasure (not technically essential, but insanely useful)... you get the idea.

[miyazaki]

16.?to_hit
17.?to_dam
18.?enchant_armor

I don't think these are essential items. More like luxuries. If you want a better weapon, find one in the dungeon, don't manufacture it.

I think a better solution would be to make these scrolls dungeon-only but give them a 100% success rate (even for Ringil!)

[Atarlost]

If you're going to change the shops you don't need most of the current system's baggage.

There's a suggestion to collapse the restores to two potion types. I'd also move them to the temple. There are only, what, 2 versions of blessing, 3 healing pots, and 4 books that need to be fixed for the temple. If we add the restores collapsed down to SDC and IWC that's 11 items. Alchemist is left with 17. Move dInv, Light, and WoR and they have 14 items each.

The magic store has 4 books, _teleSelf, _dEvil, perhaps another stave, and probably not more than half a dozen wands. It buys scrolls and potions. I don't see any good reason not to shift ?recharge over freeing up another slot at the alchemist for something like ?heroism. If there are fewer wand perhaps some other scroll can be shifted over as well. Perhaps the enchant scrolls can even be moved to the armorer and weaponsmith.

For that matter there is no store 9 or 0. A scribe could be added and all the scrolls removed from the alchemist to make more room if we're redoing shops anyways. You could even put all the high end consumables that you want available lategame but not early on in a new shop that you can't enter unless your recall depth is greater than 65.

[Donald Jonker]

I don't think these are essential items. More like luxuries. If you want a better weapon, find one in the dungeon, don't manufacture it.

I think a better solution would be to make these scrolls dungeon-only but give them a 100% success rate (even for Ringil!)

Possibly. But Magi and Priests rely on building a decent launcher for surviving the early game. Maybe people who are actually good at playing them don't.

Still I think in order to maximize the chances of pushing these store changes through we should assume that all else remains constant. That includes condensing restore potions, as much as I'd like to see it happen.

[Magnate]

For that matter there is no store 9 or 0. A scribe could be added and all the scrolls removed from the alchemist to make more room if we're redoing shops anyways. You could even put all the high end consumables that you want available lategame but not early on in a new shop that you can't enter unless your recall depth is greater than 65.

If Takkaria were to remove the BM, it would be trivial to replace the BM with the Scribe, leaving plenty of space at both scribe and alchemist (and temple). That just leaves the magic shop, though if it's just _tSelf and _dEvil plus four spellbooks it's no problem.