Gold and no_selling

[Magnate]

Originally Posted by Twilight

How is this in principle any different from the slight chance that you can get an incredible drop (say Ringil) at 50' if it passes OOD and rarity checks? A longstanding tradition, and one of the good parts of Angband IMO, is that there is a chance, however small, that you can get an awesome itemdrop early. The only slight change I'd make is to make this impossible in town to make it comparable to the itemdrops. Another example is an early potion of experience - sure you bypass the early game due to that, but you soon "catch up" with some diving to make the game challenging.

Those wow-feelings you can get from time to time more than compensates for bypassing some stages of the game. The game *is* self-regulating. If you are too powerful you can just dive some, even with oldschool playstyles.

Just my 50 öre

There's nothing to stop Maggot dropping Ringil, you know. So yes, possibly-huge gold drops are in keeping with this tradition.

[Magnate]

Originally Posted by PowerDiver

I wonder if the folks collecting stats are extracting chest drops. Early on, those tend to be money. Not that they matter much.

Well I recently fixed your long-standing bug about chest drops, so they now drop items at close to the chest's dlev - you should see far fewer items which shouldn't be there (below max depth). I'm not sure if fizzix's sim actually opens chests to generate the drops - if it doesn't yet, it wouldn't be hard for him to add (fizzix - just call chest_death()).

I tested a 3.0 warrior clearing DLs 1 to 10. He collected a bit over 4K, grabbing all gold drops, but missed part of DL8 to a trapdoor. That surprised me, since my 3.0 mage who did not collect all money got near 6K. Perhaps it is related to spawned monsters, with the mage needing much more rest so killing more spawned monsters.

The 3.0 warrior found too many good [no ego patch] weapons and wands and staves to keep track of in that time, as well as average stuff I paid no attention to. At a guess, 15 to 20 each good+ weapons and magic shop sales. Playing with selling, I'd have made 3 trips to sell the stuff if I was clearing levels.

But Eddie, this is the exact point Buzzkill and I have been trying to make: this whole change is about stopping people recalling 3 times in 10 levels! If you played without doing that, you'd have less money than if you did it. So no_selling means you end up with less money, just as if you hadn't made 3 trips to sell stuff. Why is that hard to understand?

The issue with gold drops is, I think, orthogonal to this issue. Takkaria's graph shows that his original intention when changing gold drops was not to change the amounts in the first half of the dungeon, but to achieve a more sensible curve for the 2nd half. Ewert has correctly diagnosed that Pete Mack's infinite-spread change has actually vastly reduced the average money drops, by the factor of about 5 that you observed in your test chars. The fact that nobody worried about this until no_selling was introduced shows just how crazily screwed up the game of shopping aspect has become. Gold drops are totally irrelevant *unless* no_shopping is on.

So I think the x4 mult for no_selling is fine, providing we can fix the variance so that the true average comes back to where it was in 3.0 and fits takk's graph. Since that would be exactly what you did when you piloted no selling in 3.0.9e, I don't see how you can disagree with that (but confident you'll have a good try).

[ewert]

I'll grant that "omg look what dropped" is happening for items ... But the code for "omg look what dropped" is not interfering with normal item dropping code so that we have a hard time balancing the game to what we might think is proper.

If we instead of this skewed variance formula had chances for "uberdrops of gold" that do not cause distortion to the base drops, I wouldn't complain. It's just the method of the formula right now interferes with the base drop balancing very badly, resulting mostly in just trial and error. Well now with Monte Carlo sim it's not so bad I guess for those who can use the sim to quickly check what their trial&error changes result in. =)

[PowerDiver]

Originally Posted by Magnate

.But Eddie, this is the exact point Buzzkill and I have been trying to make: this whole change is about stopping people recalling 3 times in 10 levels! If you played without doing that, you'd have less money than if you did it. So no_selling means you end up with less money, just as if you hadn't made 3 trips to sell stuff. Why is that hard to understand?

My experiments were each a single trip down through DL10. If I recalled more often to sell stuff, I would have produced more money. I was specifically comparing play without multiple recalls.

Keep in mind who I am. If I tell you that this multipler results in less money than with selling, it is based on a ridiculous amount of play and you really ought to believe it.

At this point, money drops are so insignificant in 3.2 they might as well be removed entirely. That way a newbie will know up front that the only way to get money is to sell stuff.

[Magnate]

Originally Posted by PowerDiver

My experiments were each a single trip down through DL10. If I recalled more often to sell stuff, I would have produced more money. I was specifically comparing play without multiple recalls.

Keep in mind who I am. If I tell you that this multipler results in less money than with selling, it is based on a ridiculous amount of play and you really ought to believe it.

There appears to be some confusion here, but not about who you are or what I ought to believe. I accepted the results of your tests: no_selling with the current x5 multiplier to gold drops results in less money than by selling stuff. What I thought we were debating was whether this is a problem.

To be clear - I am not intending that no_selling should provide the same amount of money as someone recalling to sell a pack full of stuff every 3-4 dlevs, who can buy anything they want by the time they hit dl20 or so. I am aiming more towards the O end of the scale, where every purchase of a spare (3rd) ?WoR requires some thought: would this cash be better spent on more !CCW, or ?PD, or those neat ego arrows ... etc. Yes, this will make the game harder - this is where we're going - for now it's an option.

I accept that gold drops are probably too stingy at the moment and go beyond the level of challenge we want. This can be fixed either by increasing the multiplier or by fixing the variance problem - I don't mind which. d_m has asked for some time to look into the variance issue, so I'm happy to wait and see what he comes up with. If 3.2 is released with the variance as-is, I'll increase the multiplier to somewhere between x10 and x20 (ie. between ~2x and ~4x the old pre-variance gold drops).

P.S. I finally finished work for xmas and got a chance to play my first game since comp88. Trying Dwarf Paladin with no selling. Read unIDd ?WoR at 300' on first trip, now at 550' on 2nd trip, found almost 2k gp so far. Maybe this is the difference between x3 and x5 ... this is by no means luxurious but it isn't painful either.

[takkaria]

Originally Posted by PowerDiver

If you are willing to do x4 on 3.0 gold drops, you ought to be willing to do x20 on 3.2 drops. Read my numbers. If anything, that 914 total was above average for 3.2 money drops in my experience. I remember that a previous char was under 800, but didn't have a dump.

Can you do a playtest of the nightly that will be built at 0800 UTC (288cf04)? It has a change which makes the variance much rarer and should restore the average gold to the straight line on the graph I posted a while ago (e.g. pretty close to historic levels of drops).

[ewert]

IMHO, since no_selling is not a "ironman" option, I would prefer the amount of gold for use to about similar to selling. Lets face it, if you are a compulsive recall'n'seller, the amounts of gold are nuts.

I like that no_selling removes the detraction from gaming that comes from constant townjumping. But I also think that it should give somewhat comparable amounts of gold. I also understand it will make gold in the end much more plentiful versus selling, because people just stop selling after they have enough gold, but with no_selling it keeps on piling, so looking at end-game is irrelevant.

Dlvl 11 and 2k gold is piddling amounts, that's a few wands. Or one plussed heavy weapon. Or a single piece of okay armor. Etc. Unless you dive.

I remember I used to clear dlvl 10-11 a few times for wands/staves/good weapons to sell and scum for enchant scrolls when I was ... a little more patient. It didn't take long to get tons of gold for the scrolls, mostly it was scumming town for the scrolls than clearing levels for the gold ...

[PowerDiver]

Originally Posted by takkaria

Can you do a playtest of the nightly that will be built at 0800 UTC (288cf04)? It has a change which makes the variance much rarer and should restore the average gold to the straight line on the graph I posted a while ago (e.g. pretty close to historic levels of drops).

I tried it, and it felt a lot better. The numbers were a bit disappointing. I think it might have been due to luck with monsters. Mughash only had 3 kobolds with him, whereas my earlier experiements he had around 40. The only other unique with escorts was Orfax with a typical escort. In the earlier experiments I killed 2 or 3 orc uniques with large escorts and maybe an orc pit.

Now for the disappointing numbers. 2520 AU picked up clearing DLs 1 through 10. I managed to get back all stolen money. After selling what I would typically sell, I had 42,379. That's a multiplier of 17 even after the gold drops have been fixed. While high, I think it is not extreme. Even 60K would not shock me, though I would call 60K extreme. This is something to keep in mind. People here tend to underestimate the amount of gold available from selling when you have a good trip.

I'd say for now, with the fixed gold drops, a multiplier of 5 might be suitable. This allows for gold to be a bit less than was 3.0 and still achieve the minimal find/sell ratio I considered playable way back when. If the gold is actually equal to 3.0, even a small chance of the x17 I encountered this time would make x5 "fair", although I admit that fairness doesn't really apply. A multiplier of 4 is probably fine, but there is a chance it isn't, so why risk making the option unfriendly to newbies?

[Magnate]

Originally Posted by PowerDiver

I tried it, and it felt a lot better. The numbers were a bit disappointing. I think it might have been due to luck with monsters. Mughash only had 3 kobolds with him, whereas my earlier experiements he had around 40. The only other unique with escorts was Orfax with a typical escort. In the earlier experiments I killed 2 or 3 orc uniques with large escorts and maybe an orc pit.

Now for the disappointing numbers. 2520 AU picked up clearing DLs 1 through 10. I managed to get back all stolen money. After selling what I would typically sell, I had 42,379. That's a multiplier of 17 even after the gold drops have been fixed. While high, I think it is not extreme. Even 60K would not shock me, though I would call 60K extreme. This is something to keep in mind. People here tend to underestimate the amount of gold available from selling when you have a good trip.

I'd say for now, with the fixed gold drops, a multiplier of 5 might be suitable. This allows for gold to be a bit less than was 3.0 and still achieve the minimal find/sell ratio I considered playable way back when. If the gold is actually equal to 3.0, even a small chance of the x17 I encountered this time would make x5 "fair", although I admit that fairness doesn't really apply. A multiplier of 4 is probably fine, but there is a chance it isn't, so why risk making the option unfriendly to newbies?

Newbies tend not to start toggling options away from their defaults until they're no longer newbies.

Anyway, your post is interesting because the multiplier *is* currently x5, and still you think it's a factor of 17 out. My current game is progressing smoothly even without the fix to the variance - I found about 2.5k gp in the first ten levels *without clearing* any levels.

There is a lot of sound and fury from people who like collecting a lot of money by selling stuff. It's an option folks - if you like to sell stuff for money, then don't turn selling off FFS. It is nonsensical to suggest that no_selling has to provide the same amount of money as compulsive recalling-when-backpack-is-full-to-sell-stuff.

[buzzkill]

If the point is to make dives longer and keep players in the dungeon longer, add a % multiplier to GP drops that increases with each down staircase and resets upon recall or taking an up staircase (anything that but going deeper). Maybe 2% per level, so a ten level trek would result in 20% higher drops gold at the end. A 20 level, single trip, descent would yield 40% more gold on that last level.

I realize this isn't probably the "right" approach, even I'm not in love with it. But I like it better than raising the standard multiplier any higher than it has to be. It gives players a little extra incentive to stay in the dungeon. It popped into my head, so I'm just throwing it out there.