The Tavern

For some reason I was thinking about this, possibly due to making equation sheets for exams these past couple of weeks, so I'm curious:

If you could give one standard looseleaf page, front and back, filled with DCSS advice to a new player, what would you say? (Assume reasonable writing size, not cramming three pages' into the margins in tiny script.)

You can think of this also as, "If you could send one page worth of advice back in time to when you first started playing Crawl, what advice would you give?"

You don't have to write out exactly what you'd say, just summarize.

What I'd try to fit in as succinctly as possible:

Tactics
1.) A statement and brief explanation of crate's law ("You should almost never move toward enemies upon first seeing them")
2.) An explanation of why moving is often the most tactically strong play you can make, because it affects how enemies will position themselves in highly predictable ways and lets you maintain control over LOS, and thus over the encounter in general
3.) Use consumables to survive, hoarding is false economy
4.) An explanation of how noise works and how things are happening around you in the dungeon even outside of LOS, and how you should try to anticipate/take these into account (Playing as Ash and getting to high piety is a great way to see this in action; alternatively wizmode)
5.) For each new encounter, decide to flee or fight and pursue that tactic conscientiously. Although if the fight starts going bad, switch to fleeing. You should usually consider the fight to be going badly if other, tougher enemies begin showing up, or if you are at less than 75% of your health.
6.) Although it is difficult to measure precisely, control is the most important thing to have/maintain in any situation in Crawl. Bad positioning means bad control and can mean that death is likely, even if you are at 90% or full health.
7.) The game waits indefinitely for you to take your turn. In tough situations the most important thing is to calm down and carefully consider/weigh your options. Check inventory to make sure you haven't forgotten some asset that you have.

Strategy
1.) Your goal is to get three runes then get the orb then ascend. Not to kill everything, not to get any particular spell castable, not to get the "perfect" set of equipment or anything else. Get good at getting three runes and winning before you impose more challenges or restrictions or variant play conditions on yourself.
2.) Very brief primer on min-delay with formula. damage > accuracy, base damage > +dam, best brands are vampiric, antimagic if you don't need MP yourself. Pain/electrocution is amazing on a fast weapon but still very good on any weapon (assuming necro training with pain). Otherwise, freezing/flaming, then vorpal/draining. Distortion can be extremely good but really varies depending on character/situation. Skill investment, base damage, and availability are most important considerations for what weapon you want to be using long term.
3.) Cost of skill levels increase exponentially, but don't branch out too early. Backgrounds are starting points, but in most cases you don't want to deviate too much from the path they provide for you until Lair, at the earliest. However eventually branching out is good.
4.) First train your immediate offensive options, then once those are good you need to get defenses.
5.) "Pure" anything is usually not optimal in the long term. "Casters" are only squishy if you train them to be that way, "fighters" are only unflexible/one-dimensional if you train them to be that way.
6.) Bonus AC from training armor = (Base AC) * (armor skill/22)
7.) Provided you have decent AC, then each point of AC is roughly equivalent to one point of EV for survival purposes. However everyone wants to get decent AC as soon as they can to minimize chance of large damage spikes. All characters should strive for decent melee capability eventually, and HP is always good, so everyone should pick up a weapon, eventually get some skill in it, and also get some skill in fighting.
8.) Resistances are nice, but the most important thing your armor gives you is AC, the most important thing your weapon gives you is more damage output. MR+ is most important resistance, however rF+ - rC+ - rN+ (somewhat in that order) becomes valuable by the Vaults though it is far from required. More than one pip of any resistance is not needed until Zot, at which point rF++ is very nice for orbs of fire. rPois is very nice for Snake but not strictly necessary, and outside of Snake not even a high priority most of the time. Walking around with a vulnerability however is basically asking the RNG to one- or two-shot you. Remember that you can swap jewelry in 1 turn. Remember that potions of resistance exist and provide insulation (rElec).
9.) If you are short on gold and have some great stuff available to buy in shops, acquire gold. Otherwise just acquire wands. If you have wand of heal wounds and hasting and plenty of ways to teleport, then acquire whatever. Don't hoard scrolls of acquirement.
10.) ID priority: !curing, ?teleportation, ?blinking, !heal wounds, !speed (top tier). ?fear, ?fog (mid-tier). Relevant boost potions are good but not highest priority to ID. Rest of the stuff is helpful but much more situational so don't sweat it if you haven't IDed them even by mid-game. Don't quaff !mutation. Limit how much you use-ID because it is a waste, however if you are beyond D4 and haven't IDed your top priority stuff and have 3 or more of a single unknown scroll/potion, then use-IDing those stacks becomes a reasonable play.

Have fun.

Spectate good players.

You can think of this also as, "If you could send one page worth of advice back in time to when you first started playing Crawl, what advice would you give?"

I wouldn't give so much advice. Crawl is largely about learning for yourself and thinking through things. I would give the following:

1. Focus on getting good at killing dudes. Once you can kill dudes, get good at not dying to dudes.
2. It's better to die with an empty inventory than a full one.
3. When there are two options and it's not clear which is best, always take more fun one (or the one you are less familiar with).
4. LoS is the one acronym you need to really understand.

WalkerBoh wrote:
You can think of this also as, "If you could send one page worth of advice back in time to when you first started playing Crawl, what advice would you give?"

I wouldn't give so much advice. Crawl is largely about learning for yourself and thinking through things. I would give the following:

1. Focus on getting good at killing dudes. Once you can kill dudes, get good at not dying to dudes.
2. It's better to die with an empty inventory than a full one.
3. When there are two options and it's not clear which is best, always take more fun one (or the one you are less familiar with).
4. LoS is the one acronym you need to really understand.

That's good too. I'd put that up at the top of the paper as a summary that covers the most important stuff.

Now I just gotta make a time machine so I can send a piece of paper back in time, working on it now. Will report back when finished.

I'd just clear a bit the long post patashu made some time ago.

I wrote up a whole long document mostly for myself that contains most of what you've outlined. My conclusion from the process was that knowing and doing are two very different things. I personally found the illustrated MiBe guide by Psieye the best approach, because it focuses on practical tactical situations and shows rather than tells the solution: viewtopic.php?f=12&t=7699 (now slightly out of date).

One thought I've had is to savescum in order to illustrate the difference between good and bad approaches to tactics:

1) Retreat on sight of enemy versus advance (demonstrating how advancing into unexposed territory and potentially bringing more enemies into LoS is, well, bad)
2) Retreat vs advance on ranged/smiting attackers (and demonstration of the rare situations when advance is actually the right decision)
3) Retreat to heal vs. heal in place after battle
4) Retreat when battle is slightly south vs far south

Curated ttyrecs would be awesome -- the full games are tedious, but illustrations of tactical goodness are awesome.

I wrote up a whole long document mostly for myself that contains most of what you've outlined. My conclusion from the process was that knowing and doing are two very different things. I personally found the illustrated MiBe guide by Psieye the best approach, because it focuses on practical tactical situations and shows rather than tells the solution: viewtopic.php?f=12&t=7699 (now slightly out of date).

This guide is really good! It pretty much covers all the major kinds of 'good tactics' with worked examples from a real game. Only thing it's missing are some minor things like shoutless (if you attack a monster that is wandering (not asleep) in LoS it never shouts and always transitions to hunting state, but if you let it notice you it will probably shout). I would recommend reading this to a lot of people.

Re ttyrecs showing good tactics: There is an online ttyrec player somewhere that you could use to make ttyrecs an easily accessible for the average viewer part of a guide. I forget what it is called, unfortunately.

Find something else to do.

"A portal to somewhere" isn't an actual portal.

I'm working on a comprehensive guide to winning crawl at the moment, and as a matter of fact I think the final version will be less than one page (the real question is whether I will have to use 10 point font in order to fit it on one page).

Amusing because my rcfile is many pages long. :lol:

The cliffnotes version is that it is a collection of simple rules of thumb which together cover all decisions made in a game, framed around discussion of the central thesis, which is "mikee avoid".

Plus maybe a line at the bottom mentioning to look up every monster in gretell to make sure it can't kill you in one turn, because "this monster can instakill you" is not considered important enough information to display in-game.

the full games are tedious

Just like playing them! It's a perfect simulation!

My guide doesn't even mention spectating or replays. The reason for this is that they display the bad moves alongside the good moves and leave it to the viewer to tell the difference; in fact the good moves are usually hidden. When you're spectating a top player and they hammer out a dozen rapid-fire moves and then sit and stare at the screen for a few seconds before making move number 13, the most important move was actually move 7 and you didn't even see it because they just did it and cruised on to the next thing, because that's what happens when you make the right move. Truly masterful moves always look boring. In every game, but Crawl more than any other I've ever played, the best moves don't even look like moves.

A lord once asked his physician, a member of healers,
which of them was the most skilled in the art.
The physician, whose reputation was such
that his name became synonymous with medical science replied,
“My eldest brother sees the spirit of sickness and removes it before it takes shape,
so his name does not get out of the house.”
“My elder brother cures sickness when it is still extremely minute,
so his name does not get out of the neighbourhood.”
“As for me, I puncture veins, prescribe potions, and massage skin,
so from time to time my name gets out and is heard among the lords.”

That's the one nagging doubt about writing a Crawl guide: some dude in China already wrote the best Crawl guide a long long time ago, and if I posted it in here it would be totally ignored. But then given what I just wrote, maybe that should be my goal.

TheDefiniteArticle wrote:That's the one nagging doubt about writing a Crawl guide: some dude in China already wrote the best Crawl guide a long long time ago

did he play dgwn?

or is that japan only (top chinese players favor ghck)

zardo wrote:
TheDefiniteArticle wrote:That's the one nagging doubt about writing a Crawl guide: some dude in China already wrote the best Crawl guide a long long time ago

did he play dgwn?

or is that japan only (top chinese players favor ghck)

Sun Tzu? Probably played MiBe.

"Crawl is all about trial and error." (bart)

Patashu wrote:Re ttyrecs showing good tactics: There is an online ttyrec player somewhere that you could use to make ttyrecs an easily accessible for the average viewer part of a guide. I forget what it is called, unfortunately.

jettyplay works well:
http://patch-tag.com/r/ais523/jettyplay ... ent/pretty

I like this thread. I think that until you embrace the community, encounter their many experienced practices and the clouds of ideas we float between each other, Crawl will remain an unsolvable enigma for you.

Over 6 years ago I played offline and had only seen Lair once. When I rediscovered crawl by playing Webtiles in 2012, some people were kind enough to help me unravel the dungeon's mysteries and I ascended.

So I think to keep it simple, the advice I would give is to seek advice. The fun of DCSS just can't be spoiled, unless the person advising you to "wear gong!", "eat those contam chunks!", and "switch to long blades" is a dude named notcluie.

Hubris kills. Even though you're feeling powerful and splatting monsters left and right at the moment, things can change very rapidly and you should be on your toes all the time.

Whenever you see a new monster (especially unique), use xv and possibly other sources of information to decide a) whether to fight or run b) how to fight it, if selected a).

don't open it

WalkerBoh wrote:Crawl is largely about learning for yourself and thinking through things.

I can't agree more. My only advice to my past self would be the advice I still desperately need to remember as my present self:

1: If you're bored or frustrated or in danger, stop and think things through and perhaps leave it alone for a while.
2: Don't start thinking you're invincible, no matter what situation you find yourself in. Even if you're right, all you gain by thinking it is feeling bored by tabbing through easy chaff.

Everything else is just details that are fun to discover.

I'd say 5 simple things:
1. Play Nagas, all day, eerey day
2. Manually train skills
3. All those Gods you think you're going to hate(Ashenzari, Nemelex, Chei, etc), try them, they will end up being your favorites
4. 15-rune win is not as glamorous as it sounds
5. Don't get your hopes up for Djinni, they're never coming

I have been planning on writing such a guide for a few days now, and had asked patashu about reusing his tactics guide as part of it. I've posted a first draft here: viewtopic.php?f=5&t=12240

I'll take input in both this thread and it's thread, it needs a lot of work still, I know

Take your time and enjoy the ride.

TheDefiniteArticle wrote:That's the one nagging doubt about writing a Crawl guide: some dude in China already wrote the best Crawl guide a long long time ago, and if I posted it in here it would be totally ignored. But then given what I just wrote, maybe that should be my goal.

I'd be curious to see this guide.

http://suntzusaid.com/

Obviously not tailored specifically to Crawl (which is why my guide is able to be much more concise about all of it). In fact it could accurately be called the best guide for an endless list of activities. But every single bit of it will help you win Crawl. Even/especially the classic line "All warfare is based on deception."

Should have known it'd be Art of War. It's very applicable to many other fields, not just Crawl (or warfare), but it's more of a broad-strokes thing. Not many people have the overall experience (or ability) to convert broad-strokes advice into precise tactical movements.

Well, as I try to express when talking about Crawl, it's a game of broad strokes. Precision really doesn't matter, and when it does matter, it's only as a smaller-scale version of the same broad rules. So while there are some leaps to be made to read it through the lens of playing Crawl, the advice ("in difficult country, do not encamp" etc) can be applied straightforwardly, to great success. Most of my effort is spent trying to remove those leaps, particularly concerning the purely mental aspects.

1) Try different things
2) Learn from your mistakes
3) Learn from others mistakes
4) Realize this is game of chance at times and don't beat yourself down over deaths, its about knowing RNG and deciding how to overcome it. Sometimes there is no right or best answer to a situation, only favourable choices.
5) Develop your own playstyle and your own reason for playing. Some people win for highscores, some for streaks, some for win rate, some to just do stupid things. Whatever it is, be the judge of your own progress.

I agree that the purpose of a game is to enjoy it, but (especially for a single player game with much less of a social component) striving to improve/master a game is often a means to getting more enjoyment from it. Certainly I feel that way about Crawl.

I have to disagree somewhat with the "just try it out and learn the lessons." That is true of many things in Crawl, but I drew lots of incorrect conclusions about stuff when I started playing, most of it due to being oblivious or slow, some of it due to bad sources of advice, some of it due to certain things in Crawl being unintuitive. (And some of those things are only unintuitive because they differ from other, superficially similar games, and are thus not a problem with Crawl's design per se. But new players do not exist in a vacuum, and they are going to bring their previous expectations and experiences to DCSS for better and for worse.)

It's true that you won't necessarily just figure everything out on your own, but that's why this advice forum exists, and can be used to learn more when you decide that you need more info than you've been able to extract from the game itself.

I think the most important thing you need to know that they game does not necessarily convey well is noise. Almost everything else is either reasonably deduceable/explained or else isn't that important.

What/where is gretell?

With hansel in a better place now, RIP.

I've been playing for about a month and a half, and what I've learned the hard way is that

1) the notion that you can play this game based on the information that it gives you, without outside help or spoilers, is just not remotely true.

and

2) autoskilling is apparently some kind of prank by the developers at the expense of naive new players

doubtofbuddha wrote:What/where is gretell?

IRC bot on the ##crawl channel.

There's a web interface here: http://crawl.develz.org/wordpress/bots

But it's currently not up-to-date.

edit: oh wait, is the bot down?

The bot has been down for a month or so, and shows no signs of a speedy recovery. I'm not in charge of the bots in any fashion, I believe there's a different bot that the cool kids are using these days, but it isn't hooked up to that website. This make things inconvenient for people like me who don't use IRC very much, but I tend to not need to look things up that often anymore, so I just have been tabbing away and ignoring monster stats. This tends to work decently well because I play characters with absurd defenses.

Cheibriados has taken over Gretell's monster stats, but the latter has not been disabled (presumably out of laziness).

Ugh, why does every thread end up about Cheibriados.

under_bridge wrote:I've been playing for about a month and a half, and what I've learned the hard way is that

1) the notion that you can play this game based on the information that it gives you, without outside help or spoilers, is just not remotely true.

and

2) autoskilling is apparently some kind of prank by the developers at the expense of naive new players

I'm afraid I disagree with you on both points. I won my first game of crawl unspoiled a long time ago, when the game was much harder than it is now. Also, I actually think automatic skill training is good for new players, as it weights skill training towards skills that are used the most, which are most likely to be skills you need the most. So players using automatic skill training have a safety net that limits how badly they can actually screw up their skills, which manual training does not provide. And when you are learning the game, being able to focus on more important features such as movement and turn use rather than looking at the skills screen all the time is a real benefit.

mikee wrote:
under_bridge wrote:I've been playing for about a month and a half, and what I've learned the hard way is that

1) the notion that you can play this game based on the information that it gives you, without outside help or spoilers, is just not remotely true.

and

2) autoskilling is apparently some kind of prank by the developers at the expense of naive new players

I'm afraid I disagree with you on both points. I won my first game of crawl unspoiled a long time ago, when the game was much harder than it is now. Also, I actually think automatic skill training is good for new players, as it weights skill training towards skills that are used the most, which are most likely to be skills you need the most. So players using automatic skill training have a safety net that limits how badly they can actually screw up their skills, which manual training does not provide. And when you are learning the game, being able to focus on more important features such as movement and turn use rather than looking at the skills screen all the time is a real benefit.

I initially had this impression too and now that I actually understand a bit more about how skills work I don't use the auto much. But I think what tricked me into thinking this is that I didn't realize that the partial skills that the auto selector would level up were actually worth while. I still don't like the %s used by the auto selector but it does a pretty good job.

remotely related to spoilers: be as gammafunk as possible in your attitude as far as crowdsourcing wins is concerned - don't always follow what spectators say, and if they ask you to acquire armour just acquire something else to piss them off. Honestly make your own judgement on what to do, don't blindly follow what people say especially on webtiles chat. Especially that guy VengefulParrot , he doesn't know anything.

vengefulcarrot wrote:remotely related to spoilers: be as gammafunk as possible in your attitude as far as crowdsourcing wins is concerned - don't always follow what spectators say, and if they ask you to acquire armour just acquire something else to piss them off. Honestly make your own judgement on what to do, don't blindly follow what people say especially on webtiles chat. Especially that guy VengefulParrot , he doesn't know anything.

Yeah I've had a couple weird backseat quarterbacks in webtiles chat. Best bet is to ask advice on IRC or post morgue on tavern. The best use for webtiles is watching good players to get a "feel" for how they move and how they approach different situations.

mikee wrote:
under_bridge wrote:I've been playing for about a month and a half, and what I've learned the hard way is that

1) the notion that you can play this game based on the information that it gives you, without outside help or spoilers, is just not remotely true.

and

2) autoskilling is apparently some kind of prank by the developers at the expense of naive new players

I'm afraid I disagree with you on both points. I won my first game of crawl unspoiled a long time ago, when the game was much harder than it is now. Also, I actually think automatic skill training is good for new players, as it weights skill training towards skills that are used the most, which are most likely to be skills you need the most. So players using automatic skill training have a safety net that limits how badly they can actually screw up their skills, which manual training does not provide. And when you are learning the game, being able to focus on more important features such as movement and turn use rather than looking at the skills screen all the time is a real benefit.

People have won Nethack without spoilers, too. I was a good 100+ hours away from figuring out that e.g. Lair should come before Orc at the time I went wiki-diving.

The benefit of spoilers in Crawl is in no way similar to Nethack. You don't need spoilers to figure out that it is better to do lair before orc, although I do think that their depths should be changed. I don't know what else I can tell you; I know you don't want to believe me.

mikee wrote: although I do think that their depths should be changed.

this happened a while ago; orc is now (on average) deeper than lair

What about banish and paralysis in previous versions? Was it possible to be paralyzed by wizard or did it have a good description?

vengefulcarrot wrote:remotely related to spoilers: be as gammafunk as possible in your attitude as far as crowdsourcing wins is concerned - don't always follow what spectators say, and if they ask you to acquire armour just acquire something else to piss them off. Honestly make your own judgement on what to do, don't blindly follow what people say especially on webtiles chat. Especially that guy VengefulParrot , he doesn't know anything.

One reason I stopped using webtiles was the endless stream of sometimes useless, sometimes bad and very rarely good advice that spewed forth from the chat. Now I realize one could just ignore the chat box entirely and I did for a while but I am inherently curious so that didn't work too well.

under_bridge wrote:People have won Nethack without spoilers, too. I was a good 100+ hours away from figuring out that e.g. Lair should come before Orc at the time I went wiki-diving.

One verifiable person has won Nethack without using spoilers, and he did it by writing a nearly complete set of spoilers from scratch based on his observations from previous games, which he organized into multiple notebooks.

KoboldLord wrote:
under_bridge wrote:People have won Nethack without spoilers, too. I was a good 100+ hours away from figuring out that e.g. Lair should come before Orc at the time I went wiki-diving.

One verifiable person has won Nethack without using spoilers, and he did it by writing a nearly complete set of spoilers from scratch based on his observations from previous games, which he organized into multiple notebooks.

Is his story written somewhere? I would like to read that.

KoboldLord wrote:One certifiable person has won Nethack without using spoilers....

https://groups.google.com/forum/#!msg/r ... JScF7qaHUJ

I wouldn't said it's verifiable, though it looks believable.

Beating Nethack completely from scratch took roughly 1700 games,
played over a period of four years and three months. This figure
includes roughly 200 savescums (in explore mode once I figured out
how) and 170 wizard mode games

Hopeless wrote:
vengefulcarrot wrote:remotely related to spoilers: be as gammafunk as possible in your attitude as far as crowdsourcing wins is concerned - don't always follow what spectators say, and if they ask you to acquire armour just acquire something else to piss them off. Honestly make your own judgement on what to do, don't blindly follow what people say especially on webtiles chat. Especially that guy VengefulParrot , he doesn't know anything.

One reason I stopped using webtiles was the endless stream of sometimes useless, sometimes bad and very rarely good advice that spewed forth from the chat. Now I realize one could just ignore the chat box entirely and I did for a while but I am inherently curious so that didn't work too well.

Your problem is probably that you listened to notcluie. Avoid doing this.

The Tavern

One page worth of advice to give... What would it be?

One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?

Re: One page worth of advice to give... What would it be?