The Tavern

Sure, I understand his motivations. I just didn't find it funny or interesting.

Did you just came up with a excuse to not have to win the game, yet feel better than people who does it regularly? Amazing.

I believe most people who are relatively new to the game and/or are trying to learn it consequently end up ciclying through combos and playing the early game for the most part. But sure, you are the special one.

In all seriousness, the actual elite playerbase of this game doesnt play the game around "true combos", but rather around understanding their current position in each game and deciding what to do from there.

People who try to approach this game from a theorical perspective are tavern posters.

Enjoy your stay.

Learning a lot in this thread.

Did you just create a secondary account so you could issue a rebuke without future accountability? Amazing.

> I believe most people who are relatively new to the game and/or are trying to learn it consequently end up ciclying through combos and playing the early game for the most part.

Sure, but that's because they're dying, not quitting strong characters. I assume all elites learn the early game to the extent that they rarely die. Therefore I could not discover anything innovative at that skill level.

> In all seriousness, the actual elite playerbase of this game doesnt play the game around "true combos", but rather around understanding their current position in each game and deciding what to do from there.

I did it mostly out of my own perfectionism, because I could sense a pattern but couldn't complete it. I'm pleased to have found a novel concept.

Leo_Littlebook wrote:Nobody has said anything positive about the idea of exclusively studying DCSS' early game in order to figure out true combos. Instead there's been only demands for proof of wins. Which leads me to believe that nobody has done what I did:

Play all species simultaneously, switching to the lowest XL one after doing a few floors. Quit sometime between Lair and Depths, or sooner if the combo+playstyle feels wrong. Whenever quitting a species, start a new game for that species. Infinitely loop until confident that have found true combos and playstyles.

In my 37-streak I basically came to the conclusion that Ar is the best background regardless of species. I know you disagreed with me but I firmly believe I could streak something like MuAr indefinitely as long as I didn't play while sleepy/distracted/whatever. Perfect play allows you to have a 100% winrate as long as your background can deal with nonsense RNG bs in the early game - Ar is the only background that gives you this power. Your self-made nomenclature of "true combo" might mean something different to you, I guess. For me, it's "what allows you to win the maximum number of games with perfect play." Now, with regards to what feels wrong, this is bound to be a very subjective thing.

Leo_Littlebook wrote:Sure, I understand his motivations. I just didn't find it funny or interesting.

I agree entirely, the reason I commented about it is that it's just very disappointing to see that from someone who should be setting a good example to others rather than trolling newcomers. What a shame.

Hi leo,

In your list of optimal builds, should one do 15 runes with them or just 3? maybe some are more optimal for extended than other? asking for a friend.

I'm not disagreeing with your ability to streak indefinitely, Malcolm.

True combo certainly means something different than optimal combo, although both are objective. It's defined above.

Removing the last 1% of deaths in order to achieve indefinite streaking leads to unaesthetic play. DCSS frays at the fringes. The game would be better if the difficulty were adjustable so that elites die a reasonable percentage of the time while playing normal builds.

An Orb-clock that scaled Orb-run spawn rate by elapsed game time might accomplish this with minimal game alteration. But I suspect the Orb run is too far removed from the early game to make much difference for starting combo selection. The early game takes relatively little game time. Perhaps tweaking the number of enemies per floor upward by 5-10% would do the trick.

It's certainly possible that Ar background provides sufficient power floor to avoid death and become established, given elite skill. If e.g. MuAr outperforms MuNe on win-rate, then the optimal differs from the true combo. MuNe offers a better mean power, but because DCSS' mean difficulty is too low, the primary concern for perfect streaking is to avoid rare deaths by maximizing the power floor, which Ar does with evocations consumables.

I think this is a shame and prefer to play for mean power, not streaking.

> should one do 15 runes with them or just 3? maybe some are more optimal for extended than other?

It's not a list of optimal builds, and I am an expert on the early game, not extended.

If true combos are objective, there should be some objective criteria which other players could use to reproduce your findings.

How would I go about confirming a true combo?

> How would I go about confirming a true combo?

Here are the rules:
1. Use every species once and only once
2. Use every god
3. Use every background

Looking forward to your list.

Leo_Littlebook wrote:> How would I go about confirming a true combo?

Here are the rules:
1. Use every species once and only once
2. Use every god
3. Use every background

Looking forward to your list.

You haven't given me any basis on which to make a list. What objective criteria would I be looking for?

uv4> You haven't given me any basis on which to make a list. What objective criteria would I be looking for?

You are asking me to defend the objectiveness of my list against all potential challengers. No. Pick one challenger, and I will defend against that. And it has to be your challenger, so there's skin in the game.

I'm asking you to explain the basis on which a true combo is objective.

You said: "True combo certainly means something different than optimal combo, although both are objective." (Emphasis mine.)

I understand how an optimal combo could be objective. Players might disagree on the exact criteria to use (e.g. do we look at overall winrate or do we only look at winrate among experienced players?) but that objective criteria could be used independently by different people to come to the same results.

What objective criteria do we use to determine whether a combo is true? If we agree on an objective criteria, we should be able to come to the same (or at least similar) conclusions.

In this context, something can be considered objectively true if no reasonable alternative can be proposed.

I've given three selection rules. That narrows the field quite a bit, in a fully objective way.

How you interpret "true" is up to you. You could just go by .23 elite win rate.

Make a differing list and then we can get more specific. Maybe there is no disagreement in our lists, in which case further debate is pointless.

Leo_Littlebook wrote:I am an expert on the early game, not extended.

OK. What percent of your characters make it to level 12?

Please provide links to your morgues. Do you play at all on public servers?

that is NOT a true combo, just a string. samus could have up-Bed out after the second back air, its a nice string but its not a combo.

Leo_Littlebook wrote:In this context, something can be considered objectively true if no reasonable alternative can be proposed.

You can say anything you'd like if you change the meanings of words to mean things they don't.

In any case, I'm not sure how to disagree with any of your true combos because you haven't told us how to determine true combos in the first place.

Maybe it would be easier if I simply ask you to define a true combo. What is the definition?

Leo_Littlebook wrote:I've given three selection rules. That narrows the field quite a bit, in a fully objective way.

None of the rules you gave assist in making selections. They boil down to "make a list of combos which uses every god and background and doesn't duplicate species". Many lists would satisfy these constraints but that doesn't help to determine which combos are true or which list would be most valid.

Suppose I gave you my list. How would you compare them? What objective criteria would you use to say one list is more true than another?

Leo_Littlebook wrote:How you interpret "true" is up to you. You could just go by .23 elite win rate.

If the interpretation of "true" is left to the person making the assessment, we're talking about a subjective measure and not an objective one. An objective measure is independent of the particular person doing the measuring.

Here's a zip of all the morgue files on my Macbook:
https://ufile.io/4eas8lbv

For most of that, I was playing all species in parallel, as described previously, during downtime, to discover true combos.

Looks like there's a complete selection of games still current, for every species. I just got tired of playing that way, and switched to iMac, which has two games, one current. Those two machines contain my entire DCSS record.

Uv4, my definition of objective truth as regards DCSS is appropriate. If you want a complete defense of my list, you can wait until I organize my notes during the demo. If you want to debate it now, you can provide a competing list to focus the discussion. The adversarial process is good enough to execute convicts, so it's good enough for DCSS.

Leo_Littlebook wrote:Uv4, my definition of objective truth as regards DCSS is appropriate. If you want a complete defense of my list, you can wait until I organize my notes during the demo. If you want to debate it now, you can provide a competing list to focus the discussion. The adversarial process is good enough to execute convicts, so it's good enough for DCSS.

I declare that your true combos are purely subjective.

Should you offer an objective criteria by which true combos can be verified, I'll reassess.

> I declare that your true combos are purely subjective.

Who cares? One could equally argue that a list of optimal combos is purely subjective, because the word optimal can be interpreted in different ways: optimal for beginners, intermediate or current experts, for theoretically perfect player, for win-rate, for real-time, etc.

I do not trust you to be an intelligent dialogue partner communicating in good faith, so I require you to abide by the adversarial process, which creates an equality of exposure for the two sides. And I am obviously correct in my distrust, since you have just called me either too stupid or too dishonest to communicate my own position correctly.

OK, here's my true combo list, the rules being "each species only once, each background and god":

Code:: Barachi Abyssal Knight Lugonu Centaur Air Elementalist Ashenzari Deep Dwarf Arcane Marksman Cheibriados Deep Elf Artificer Dithmenos Demigod Assassin Demonspawn Berserker Trog Draconian Chaos Knight Xom Felid Conjurer Elyvilon Formicid Earth Elementalist Fedhas Gargoyle Enchanter Gozag Ghoul Fighter Hepliaklqana Gnoll Fire Elementalist Jiyva Halfling Gladiator Kikubaaqudgha Hill Orc Hunter Beogh Human Ice Elementalist Makhleb Kobold Monk Nemelex Merfolk Necromancer Okawaru Minotaur Skald Qazlal Mummy Summoner Ru Naga Transmuter Sif Octopode Venom Mage Uskayaw Ogre Wanderer Vehumet Spriggan Warper The Wu Jian Council Tengu Wizard Yredelemnul Troll Abyssal Knight Lugonu Vampire Air Elementalist Ashenzari Vine Stalker Arcane Marksman Zin

Edit: fixed a bug.

Spoiler: show

I do not get the algorithm. Why OpWn instead of OpVM? I also do not really get why the algorithm chose Ashenzari for VpAE instead of Zin.

sanka wrote:I do not get the algorithm. Why OpWn instead of OpVM? I also do not really get why the algorithm chose Ashenzari for VpAE instead of Zin.

There was a bug, thanks for noticing! Vp can't worship Zin, that's why Ash.

Oh, yes, I totally forgot the good god restriction. Good that the algorithm has the restrictions built in!

I actually like your true combos, maybe I'll try to play them.

> The list was made with a simple algorithm with no thought of what combos are good.

I don't require you to play by adversarial rules. But if I did, this wouldn't count as having skin in the game. I also don't think it's interesting to compare my list to a random one; I may as well just write about my list.

Leo_Littlebook wrote:I do not trust you to be an intelligent dialogue partner communicating in good faith, so I require you to abide by the adversarial process, which creates an equality of exposure for the two sides.

Leo_Littlebook wrote:I don’t require you to play by adversarial rules.

???

Leo_Littlebook wrote:I also don't think it's interesting to compare my list to a random one

Sorry, but my list is by no means a random one, it is completely deterministic.

This topic has run its course and isn't going anywhere productive, closing the thread.

The Tavern

Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread

Re: Leo Littlebook's thread