What is "balance" anyway

[Timo Pietilä]

I certainly knew people who complained, and the squelch patch was written by someone who couldn't stand them.

Later, yes, but not in old days. Quite a lot was different back then, and those were no problem to anyone (in fact non-monster skeletons and monster skeletons used same letter which made things a bit interesting and also phial used same letter as shards of pottery. I don't recall anymore if broken sticks were same as torches. So you needed to at least look what those items were. There also were no "preserve mode". If you missed artifact into level you lost it, there was no getting it back.

Also IIRC "detect items" and "detect monsters" etc. didn't allow you to look items/monsters that are not in your LoS like it allows now. In fact you didn't get any other info than their symbol appearing in the map. I actually liked that. Now it is "clairvoiance" - "known items" - "alter reality/use stairs" if there is nothing interesting into level.

[Marble Dice]

Later, yes, but not in old days. Quite a lot was different back then, and those were no problem to anyone (in fact non-monster skeletons and monster skeletons used same letter which made things a bit interesting and also phial used same letter as shards of pottery. I don't recall anymore if broken sticks were same as torches. So you needed to at least look what those items were. There also were no "preserve mode". If you missed artifact into level you lost it, there was no getting it back.

Sounds a lot worse for sure. I didn't know about skeleton enemies acting as mimics though - that's interesting to think about when combined with all the other mimic-like features that could get someone in trouble if they didn't know to always look in all the mimic-possible situations. I'm glad Angband has been been moving away from that sort of thing, but it used to be a major feature of the game design.

Also IIRC "detect items" and "detect monsters" etc. didn't allow you to look items/monsters that are not in your LoS like it allows now. In fact you didn't get any other info than their symbol appearing in the map. I actually liked that. Now it is "clairvoiance" - "known items" - "alter reality/use stairs" if there is nothing interesting into level.

Being able to look at detected items/monster that aren't in your LoS and clairvoyance/alter reality being over powered are two separate issues. I like !Enlightenment, but the spell feels a little broken.

[Derakon]

Timo's definitely right about not being able to look at the monsters you'd just detected (and, in general, not being able to look at anything that wasn't in LOS). Made things much more interesting when you knew there was a black W in that vault, but had no idea what it could be beyond that (or couldn't differentiate between an ancient red dragon and a great hell wyrm).

2.4 frog-knows was definitely flawed, and I'd say that the current Angband is a better game, but not universally so. I'm going to call out two things as examples of what the old game did differently that I think was good, or at least made the game interesting.

1) Uncapped OOD limitations. IIRC it was theoretically possible to find any item or monster at any depth. Certainly you found odd items much more commonly; I had a few characters find PDSM back in the day, for example, while I don't think any character of mine from the post-2.7.x days has ever found a set. On the flipside, I also had characters running like hell from a young dragon at 600', and that was also fun! Unfair, yes, but fun. It wasn't impossible that I could have beaten the thing if I'd had the right items (!speed, !rfire, etc.), which would have gotten me a couple of levels right there. In general, this kind of placement algorithm means that you can't be certain at any time that you're ready for what the dungeon has to throw against you. This uncertainty adds danger, which in turn adds fun.

And don't even get me started on the fun times had polymorphing townsfolk into completely random monsters. Ahh, duelling AMHDs in the town...

2) Ambiguity in detection. This is basically what we're describing above (not being able to 'l'ook at monsters you've detected). You know roughly what you're going to be facing -- a fire-based dragon, or a druj, etc. But not knowing the specifics can get you in over your head, which, again, makes the game more interesting. "Oh shit", you think, "that monster has about five times more health than I thought it would and could kill me if it breathed twice in a row. How am I going to deal with this?" Similarly, you couldn't see that that ring was a Carnelian Ring of Acid; you could just see that it was yellow. Maybe it was an Amber Ring of Speed, or even a Plain Gold Ring! Seems like it should be worth going into the vault to check it out, don't you think?

What I'm seeing in current Angband is basically that a well-played character has no reason to ever put himself in harm's way. The risk-reward tradeoffs for every encounter can be determined before the encounter even starts, and the player simply avoids encounters that aren't worthwhile -- either because the items aren't good enough, or because the danger is too great. Players find the depth where they know they can handle the monsters that will be generated, and play there until they feel like they're ready for a few levels deeper in. Even the aggressive divers still regulate their levels relative to those of the monsters they encounter so that they have reasonable odds of being able to run away! That feeling of danger seems to have been lost, and IMO it's the danger that makes the fun.

[d_m]

What I'm seeing in current Angband is basically that a well-played character has no reason to ever put himself in harm's way. The risk-reward tradeoffs for every encounter can be determined before the encounter even starts, and the player simply avoids encounters that aren't worthwhile -- either because the items aren't good enough, or because the danger is too great. Players find the depth where they know they can handle the monsters that will be generated, and play there until they feel like they're ready for a few levels deeper in. Even the aggressive divers still regulate their levels relative to those of the monsters they encounter so that they have reasonable odds of being able to run away! That feeling of danger seems to have been lost, and IMO it's the danger that makes the fun.

I think I agree with the general idea behind the first point, but disagree with the second point. The problem you mention (that there's no reason to take risks) would probably go away if:

1. fewer quality items were found on the floor
2. drops were better
3. levels varied more widely in risk and reward

Making detection weaker simply changes taking a risk from "I'm not sure if this item is good and/or this monster is hard, should I go for it?" into "This item is definitely good and this monster is definitely hard, should I go for it?"

[Marble Dice]

1) Uncapped OOD limitations.

Everything in this paragraph sounds good or neutral to me. Angband has a lot of fringe possibilities (PDSM, RoS +30, etc), and they usually add to or at least do not detract from the game.

2) Ambiguity in detection.

I disagree with this paragraph. The difference between an ancient red and a great hell wyrm, or even a baby red (100 hp, 33 breath, 7 dam/t) and a mature red (500 hp, 166 breath, 44 dam/t) is large, and this does not sound interesting or fun to me, and it places an even greater strain on the player to memorize the monster spoilers. IMO the only way something like this could even work is if your interface not only allowed you to look at any monster, but provided a list of all known possible enemies for ambiguous "non-LOS" symbols.

[Hariolor]

1) Uncapped OOD limitations.

I don't know about completely uncapped - it'd have to be well-balanced, as I'd get really annoyed if the vast majority of decent characters died at CL15 to an OOD Black Reaver or some such nonsense. If the rarity of very OOD bad guys was parallel to the rarity of very OOD artifacts, I could be sold.

2) Ambiguity in detection.

I'll stick my neck out and say I actually agree very much with this point. There's a fine line between being blindsided by an unfair death, and walking into something you should have avoided. Being given the opportunity to gamble regarding which is which is, to me, exciting.

Part of the recent trend in V seems to be driven by a sentiment that the game should be statistically sterile (this goes back to the question of what is 'balance', in fact). That is to say, an adequately cautious player should expect to virtually always win by knowing ahead of time the layout, availability, and attributes of every enemy and item. While the RNG-driven level creation means you still have to actually *play* the game to win; there is a personality type that likes to "solve" a game like Angband. For those people, intentionally adding opacity to information is probably a bitter pill.

I, however, would not mind at all if ESP and detection were severely limited. Doubly so if aggravation/stealth and vision/illumination were tweaked to compensate. My favorite moments are still when you realize you're suddenly too close to something very scary and have to devise a way to survive with what you've got with you.

[Derakon]

I don't know about completely uncapped - it'd have to be well-balanced, as I'd get really annoyed if the vast majority of decent characters died at CL15 to an OOD Black Reaver or some such nonsense. If the rarity of very OOD bad guys was parallel to the rarity of very OOD artifacts, I could be sold.

Presumably in practice some kind of front-loaded bell curve would be used, so while it'd be possible to get a Black Reaver at DL15, you wouldn't see it (or its equivalent) more often than once every 50 games, if that. I'd see it working out something like this:

Monster level relative to dungeon level / odds of being generated:
-10 / 5%
-5 / 30%
0 / 55%
5 / 8%
10 / 1.5%
15 / .3%
20 / .2%

This is probably still too heavily tweaked towards deep monsters, but it gives you some idea of what I'm talking about.

[Timo Pietilä]

I disagree with this paragraph. The difference between an ancient red and a great hell wyrm, or even a baby red (100 hp, 33 breath, 7 dam/t) and a mature red (500 hp, 166 breath, 44 dam/t) is large, and this does not sound interesting or fun to me

It is, trust me. You die a bit more often, but who cares? Added danger makes game feel more intense and you play more carefully. You might actually even look at what you are fighting, and not automatically safely suppose that it isn't anything that can kill you.

, and it places an even greater strain on the player to memorize the monster spoilers.

Spoilers? You use spoilers? Use monster memory. It is an in-game feature. Sounds to me that you are relatively new to this game.

IMO the only way something like this could even work is if your interface not only allowed you to look at any monster, but provided a list of all known possible enemies for ambiguous "non-LOS" symbols.

I don't quite understand your reasoning behind this. It worked just fine in old versions, so it could work just fine in new versions. In new versions there is info-screen that gives you all known monsters in single look. Of course ESP doesn't only tell you that there is an monster, it tells you what that monster is, so this really doesn't change gameplay that much. It does however make ESP more valuable.

[andrewdoull]

It is, trust me. You die a bit more often, but who cares? Added danger makes game feel more intense and you play more carefully. You might actually even look at what you are fighting, and not automatically safely suppose that it isn't anything that can kill you.

Timo - could you talk about your experiences with variants (in particular Oangband and Sangband)? My experience with those is that the monster design ensures that there are in level monsters which are always going to be a challenge - which monsters those are depending on which class you play. That's something I've tried to emphasize with Unangband.

I'm suggesting that this could be addressed in a large part by more careful design, not just increasing the range of monster types.

Andrew

[andrewdoull]

When he's not doing awesome things like the quiver and the 256-colour patch, d_m is working on a big level generation patch too.

Magnate and/or d_m - could you elabourate on this more (perhaps by starting another thread discussing it)? I've not seen any discussion to date, and I would be intrigued to see what you had in mind.

Andrew

[Marble Dice]

It is, trust me. You die a bit more often, but who cares? Added danger makes game feel more intense and you play more carefully. You might actually even look at what you are fighting, and not automatically safely suppose that it isn't anything that can kill you.

No, it is not fun. Maybe to you it is, to me it is not. You obviously don't know me or what is fun to me, but that's fine, I wouldn't expect you to. That's why I tried to make it clear that it was my opinion when I stated it.

Spoilers? You use spoilers? Use monster memory. It is an in-game feature. Sounds to me that you are relatively new to this game.

Spoilers == Monster memory. Monster memory is a useful in-game feature, although average hp is somewhat awkward to learn in-game, but unfortunately save files do tend to get left behind, and with them, their memories. My personal memory is suitable for many enemies, but I admit I don't have the maximum hp of every dl40+ breather memorized, and occasionally I need to be reminded of what the damage cap or divisor is for one high breath or another. According to you, anyone younger than dinosaurs of the Triassic are new to Angband, so yes, I am quite new.

I don't quite understand your reasoning behind this. It worked just fine in old versions, so it could work just fine in new versions. In new versions there is info-screen that gives you all known monsters in single look. Of course ESP doesn't only tell you that there is an monster, it tells you what that monster is, so this really doesn't change gameplay that much. It does however make ESP more valuable.

Again, if you're referring to obscure detection magic, no, I don't think it worked. Many other people don't think it worked which is why it isn't that way now. Just because you or some people you know didn't complain about it doesn't mean most people think it's a good idea - and maybe what you played was a better game for you, or even for most people, but I am simply saying it is definitely not a better game for me.

My point was this: If I cast a detection spell and I see just a red "d", that's bad if I think have to ask myself, or the memory, or the spoilers, what a red "d" could be. If I cast a detection spell and for example on the visible monsters list or when I look at that symbol, I see that it could be a baby red dragon, a young red dragon, OR a mature red dragon, that's good. The interface has prevented me from doing the mindless and unnecessary busy work of looking up or simply knowing all possible red "d" symbols.

I agree with UnAndrew and d_m that "sterile statistics-based gameplay" could be readily addressed by better design of dungeon levels and the risk and reward of the enemies and objects placed therein.

[Magnate]

Magnate and/or d_m - could you elabourate on this more (perhaps by starting another thread discussing it)? I've not seen any discussion to date, and I would be intrigued to see what you had in mind.

Andrew

All I know is that d_m was looking a while ago at http://trac.rephial.org/ticket/156 - Leon had sent Takkaria a complete replacement for generate.c, which hasn't yet made it into the V codebase. The more development elapsed, the more work was required to incorporate Leon's changes. I don't know how far it's got, but there were some excited noises from d_m about caves and other room types, so I'm looking forward to seeing it.

[andrewdoull]

All I know is that d_m was looking a while ago at http://trac.rephial.org/ticket/156 - Leon had sent Takkaria a complete replacement for generate.c, which hasn't yet made it into the V codebase. The more development elapsed, the more work was required to incorporate Leon's changes. I don't know how far it's got, but there were some excited noises from d_m about caves and other room types, so I'm looking forward to seeing it.

Ah that. I like some of the stuff that Leon does, and some of it just didn't work for me (the monster pit contents generators in particular confused me for the longest time).

I was wondering as I have been spending considerable amounts of time fixing the correctness of the Unangband generation code, and adding additional room generation types.

Andrew

[Nick]

Ah that. I like some of the stuff that Leon does, and some of it just didn't work for me (the monster pit contents generators in particular confused me for the longest time).

I find it simplest to just steal a whole variant.

[d_m]

Magnate and/or d_m - could you elabourate on this more (perhaps by starting another thread discussing it)? I've not seen any discussion to date, and I would be intrigued to see what you had in mind.

Andrew

So, here is the 5 second overview of what I was hoping to do (I'm going to lay out the things I want with the understanding that some of them can be stolen from Leon's patch):

1. Circular or other interestingly shaped rooms
2. Better "high-level" control of room placement
3. Distinct "level feels" e.g. cave levels, maze-levels, more-or-less open-feeling levels, etc.
4. More explicit determination of how "good" a level is, in terms of explicitly deciding whether to place vaults, how many, etc, etc "up front"
5. Try to vary dungeon size in interesting ways

Basically, after creating a patch to vary dungeon-size I got (at least somewhat rightly) smacked by PowerDiver for drastically changing the likelihood of finding greater vaults, based on the reduced chance that the generation algorithm would have room to place them (as well as some other more minor issues).

So my idea was to try to determine "interesting" things, like vaults, weird rooms, level difficulty (mabe generation of unique or OOD monsters) first, and then do all the boring stuff like connecting them with corridors or placing boring rooms. I think this would make useful level feelings possible, and it would also make the quality of a level less based on its size.

Whew. That was more than 5 seconds.

I can start a thread if people want to sound off on this.