Role difficulty and number of ascensions

On 2008-02-06, ThomasHarrisonNelson.DeleteThis@gmail.com <ThomasHarrisonNelson.DeleteThis@gmail.com> wrote:
> I was bored, so I crunched some numbers. Here's the number of
> ...
> So here's my main question: how closely do these numbers correspond to
> your notion of difficulty? Are the roles with the most ascensions the
> easiest?

This thread came up about a year ago, maybe a little bit more. I
crunched some numbers as well; the output is at:

http://www.akrasiac.org/rachel/role_distinctions.html

The data is old, of course, but the proportions are similar. I don't
have time to sit and interpret the data in detail again at the moment,
but a good indicator of which roles are "easiest" was the rightmost
column in the top section -- players whose first ascension was that
particular role.

"hardest" is very difficult to determine; the low play counts for
Cavemen and Knights really skews their numbers in a non-useful manner.

Google archives probably has the whole thread with a lot of discussion,
if you wade through it enough.

--
Derek

Game info and change log: http://sporkhack.com
Beta Server: telnet://sporkhack.com
IRC: irc.freenode.net, #sporkhack
 >> Stay informed about: Role difficulty and number of ascensions 

On Feb 6, 2:42 pm, ThomasHarrisonNel... RemoveThis @gmail.com wrote:
> I was bored, so I crunched some numbers.  Here's the number of
> ascensions per role in the last 365 days on nethack.alt.org, scraped
> fromhttp://alt.org/nethack/mostascyear.html:
>
> Cav - 107
there isn't much to differentiate them from other combat classes and
they're not as tough as barbarians. On the other hand they can dip
for Excalibur, but then why not just play a knight or valkyrie?
> Kni - 124
I'm a little surprised these aren't more common since they can dip for
Excalibur with a starting weapon.
I have a feeling that in Sporkhack knights will start to be more
common since they are the only class that can acquire Excalibur from
fountain-dipping ... just checked the high score list and found 10
ascended knights of the hundred-plus ascensions there, so maybe I'm
just sharing my own prefence, since in fact that is my next Sporkhack
project.
> Hea - 131
> Rog - 136
> Arc - 137

My first ascension, great with spells and silver sabers, if just a
little bit fragile in the beginning.

> Tou - 138
> Mon - 139

Challenging but fun

> Ran - 140

I've never played one (probably because I don't know how to set my NAO
preferences to autopickup quivered items)

> Pri - 156
> Sam - 177
> Bar - 185
> Val - 346

Start as lawful, dip for Excalibur, ascend and start a new game all
within 25,000 turns. Quick fun, IMO.

> Wiz - 412

being able to write unknown spellbooks and scrolls makes the whole
game go so much faster.
>
> It's worth noting that Valkeryies and Wizards are ascended twice as
> often any other class.  In fact, the number of Valkeryies and Wizards

Maybe that's because they only take half the time?

> ascended this year is about the same as the number of Cavepeople,
> Knights, Healers, Rogues, Arceologists, and Tourists combined.  I
> think the reason why is pretty clear: Valkeryies are almost everyone's
> first ascension, and Wizards are the most common way to do a lot of
> the conducts (I think).  One thing that surprised me is how few

Doing an atheist-foodless conduct is difficult without a starting ring
of slow digestion so wizards are the easiest role for that type of
conduct game.

> cavepeople and knights are ascended, since they're supposed to be
> easier than Tourists, Arcs, and Rangers.
>
> So here's my main question: how closely do these numbers correspond to
> your notion of difficulty?  Are the roles with the most ascensions the
> easiest?

My personal notion of difficulty is based on the length of the
game ... if I can get in and out in 25,000 turns I can ascend, if I
need 40,000 turns then I'm going to make a mistake and die.
 >> Stay informed about: Role difficulty and number of ascensions 

ThomasHarrisonNelson.TakeThisOut@gmail.com wrote in news:6311a805-1216-4e4c-a215-
59d9886359f6.TakeThisOut@q21g2000hsa.googlegroups.com:

> I was bored, so I crunched some numbers. Here's the number of
> ascensions per role in the last 365 days on nethack.alt.org, scraped
> from http://alt.org/nethack/mostascyear.html:
>
> Cav - 107
> Kni - 124

These classes are kind of boring and don't add anything to the game other
than SMASHY SMASH

> Hea - 131

A fairly difficult class.

> Rog - 136

I would guess most people are put off by the quest level.

> Arc - 137

i dunno

> Tou - 138

I thought more people would've played and ascended tourist. Never played
them myself.

> Mon - 139

Monk are tough. In my experience monks give the illusion of being all
smashy yet they have their own particular flavor (vegetarianism,
armorless, etc).

> Ran - 140

Can't really give an opinion on them.

> Pri - 156

Priests are easy because of the whole automatic identification of BUC
status.

> Sam - 177

Samurai are easy, probably could be considered about as boring as Caveman
or Knight but Samurai was my first and only ascension.

> Bar - 185

More smashy and stuff. I play an orc barbarian now and again since I
can't imagine barbarians being any other race (ignoring REH references)

> Val - 346

So very simple for one reason: cold resistance at level one.
Next reason: Mjollnir as first guaranteed sac gift.

> Wiz - 412

Never played wizard but I would gather their starting kit and proficiency
with spells turns them into asskickers.
 >> Stay informed about: Role difficulty and number of ascensions 

On Feb 6, 2:42 pm, ThomasHarrisonNel....RemoveThis@gmail.com wrote:

> It's worth noting that Valkeryies and Wizards are ascended twice as
> often any other class. In fact, the number of Valkeryies and Wizards
> ascended this year is about the same as the number of Cavepeople,
> Knights, Healers, Rogues, Arceologists, and Tourists combined. I
> think the reason why is pretty clear: Valkeryies are almost everyone's
> first ascension,

Is that really true? I ascended quite a few roles before Valk, but
then my play history is somewhat weird compared to what I've seen a
lot of people describe here on rgrn. (I've also been playing since
version 3.1 if it makes any difference)

> and Wizards are the most common way to do a lot of
> the conducts (I think).

Wouldn't that be Monk? I think Wizards are the most popular role
because they have a great quest item, start with magic resistance, and
have the potential to start with other very useful stuff.

> One thing that surprised me is how few
> cavepeople and knights are ascended, since they're supposed to be
> easier than Tourists, Arcs, and Rangers.

Cavepeople are generally considered to be boring, and the tight
squeeze on the quest start level is a pain. Knights have an annoying
code of conduct which though can probably be safely ignored after a
certain point it gets irritating to see "You caitiff!" over and over.
I had a hard time with knights until I learned how to make use of the
lance effectively.

> So here's my main question: how closely do these numbers correspond to
> your notion of difficulty? Are the roles with the most ascensions the
> easiest?

In my personal experience Barbarians are the easiest. I think the
starting poison resistance is more useful than the Valk's starting
cold resistance. My favorite and most-ascended class is the Priest,
but I don't claim they're easier, I just really like the inherent B/U/
C ID.

- funcrunch
 >> Stay informed about: Role difficulty and number of ascensions 

Haakon Studebaker wrote:
> ThomasHarrisonNelson DeleteThis @gmail.com wrote in news:6311a805-1216-4e4c-a215-
> 59d9886359f6 DeleteThis @q21g2000hsa.googlegroups.com:
>
>>I was bored, so I crunched some numbers. Here's the number of
>>ascensions per role in the last 365 days on nethack.alt.org, scraped
>>from http://alt.org/nethack/mostascyear.html:
>>
>>Cav - 107
>>Kni - 124
>
> These classes are kind of boring and don't add anything to the game other
> than SMASHY SMASH

The "knights jump" and riding (pet horse and saddle) distinguishes
the Knight from other classes. They're comparably easy to play.

>>Arc - 137

Difficult. (At least I often fail to advance them.)

>>Tou - 138
>
> I thought more people would've played and ascended tourist. Never played
> them myself.

Difficult.

Janis
 >> Stay informed about: Role difficulty and number of ascensions 

Janis Papanagnou <Janis_Papanagnou DeleteThis @hotmail.com> wrote in
news:fodicd$1g4$1@online.de:

>> These classes are kind of boring and don't add anything to the game
>> other than SMASHY SMASH
>
> The "knights jump" and riding (pet horse and saddle) distinguishes
> the Knight from other classes. They're comparably easy to play.

Ah, I forgot about that jumping bit. Jumping IS nice.
 >> Stay informed about: Role difficulty and number of ascensions 

On 2008-02-07, Haakon Studebaker <heptapod.TakeThisOut@gmail.com> wrote:
> Janis Papanagnou <Janis_Papanagnou.TakeThisOut@hotmail.com> wrote in
> news:fodicd$1g4$1@online.de:
>
>>> These classes are kind of boring and don't add anything to the game
>>> other than SMASHY SMASH
>>
>> The "knights jump" and riding (pet horse and saddle) distinguishes
>> the Knight from other classes. They're comparably easy to play.
>
> Ah, I forgot about that jumping bit. Jumping IS nice.

Don't forget the starting lance, along with the starting saddle and
(dangerous to mount on turn 1) mount. Lance gives a distance attack
and a decent chance at a stunning attack at melee range.

Distance attack + stunning attack + jumping + extremely fast movement
(horses are speedy) is a very nice combination.
 >> Stay informed about: Role difficulty and number of ascensions 

On Feb 7, 1:31 am, Uwe Schürkamp <hoo... RemoveThis @hoover.dyndns.org> wrote:
> Sounds interesting. I've been playing nethack since the early nineties
> and I haven't managed a single ascension so far, my best ever result was
> around 100,000 points with an elven ranger (that's the character I
> usually play). Sounds like I should give the knight a try

As dust writing is the defense you will use at game start, why not try
a Rogue?

Dust write then back toss for oodles of damage. (Up to 5d30 BEFORE
base weapon damage, skill, ST, enchantment, etc. are even considered.)

Magicbane helps not just a writing tool, but also because it inflicts
scare on the target which is very very useful for Rogues and only so-
so for everybody else.

If Stormy scared (as the Moorcock stories indicate very clearly), I
might be tempted to use it.

-HJC
 >> Stay informed about: Role difficulty and number of ascensions 

On Wed, 06 Feb 2008 21:28:09 -0600, Marc Hartstein wrote:

> Distance attack + stunning attack + jumping + extremely fast movement
> (horses are speedy) is a very nice combination.

Sounds interesting. I've been playing nethack since the early nineties
and I haven't managed a single ascension so far, my best ever result was
around 100,000 points with an elven ranger (that's the character I
usually play). Sounds like I should give the knight a try

Cheers, uwe
 >> Stay informed about: Role difficulty and number of ascensions 

Derek Ray <de....TakeThisOut@moot.its.only.a.spamtrap.org> wrote:
> ThomasHarrisonNel....TakeThisOut@gmail.com <ThomasHarrisonNel....TakeThisOut@gmail.com> wrote:
>
> > So here's my main question: how closely do these numbers correspond to
> > your notion of difficulty?  Are the roles with the most ascensions the
> > easiest?
>
> This thread came up about a year ago, maybe a little bit more.  I
> crunched some numbers as well; the output is at:
>
> http://www.akrasiac.org/rachel/role_distinctions.html

Maybe the number to look for would be percentage ascending
by type not percentage of serial ascensions among ascenders.

> The data is old, of course, but the proportions are similar.  I don't
> have time to sit and interpret the data in detail again at the moment,
> but a good indicator of which roles are "easiest" was the rightmost
> column in the top section -- players whose first ascension was that
> particular role.

That's the other good qualifier for difficulty IMO.

> "hardest" is very difficult to determine; the low play counts for
> Cavemen and Knights really skews their numbers in a non-useful manner.

Maybe combine the two numbers by multiplying? Then sort highest
to lowest as a difficulty estimate.
 >> Stay informed about: Role difficulty and number of ascensions 

On 2008-02-07, Doug Freyburger <dfreybur.RemoveThis@yahoo.com> wrote:
> Derek Ray <de....RemoveThis@moot.its.only.a.spamtrap.org> wrote:
>> ThomasHarrisonNel....RemoveThis@gmail.com <ThomasHarrisonNel....RemoveThis@gmail.com> wrote:
>>
>> > So here's my main question: how closely do these numbers correspond to
>> > your notion of difficulty?  Are the roles with the most ascensions the
>> > easiest?
>>
>> This thread came up about a year ago, maybe a little bit more.  I
>> crunched some numbers as well; the output is at:
>>
>> http://www.akrasiac.org/rachel/role_distinctions.html
>
> Maybe the number to look for would be percentage ascending
> by type not percentage of serial ascensions among ascenders.

Percentage ascended by type gives incorrect numbers - it lists Caveman
and Knight as easiest because they're rarely played by anyone except
those filling out their "list of 13". You can see a few columns related
to that in there.

"Percentage of serial ascensions among ascenders" doesn't make sense as
written; run that by again, rephrased?

>> The data is old, of course, but the proportions are similar.  I don't
>> have time to sit and interpret the data in detail again at the moment,
>> but a good indicator of which roles are "easiest" was the rightmost
>> column in the top section -- players whose first ascension was that
>> particular role.
>
> That's the other good qualifier for difficulty IMO.

That one's a very revealing number. Valks and Wizards are so far ahead
of the pack it's not even funny, and this is not surprising at all given
the benefits each receive in vanilla (+3 shield, cold resistance, and
free Excalibur if they're lawful for Valks; Magicbane, all the spells
_and_ all the armor a Barb can wear, and starting MR for Wizards).

Barbs come in a distant third (also not surprising).

>> "hardest" is very difficult to determine; the low play counts for
>> Cavemen and Knights really skews their numbers in a non-useful manner.
>
> Maybe combine the two numbers by multiplying? Then sort highest
> to lowest as a difficulty estimate.

Which two numbers are you thinking of?

--
Derek

Game info and change log: http://sporkhack.com
Beta Server: telnet://sporkhack.com
IRC: irc.freenode.net, #sporkhack
 >> Stay informed about: Role difficulty and number of ascensions 

Derek Ray <de....TakeThisOut@moot.its.only.a.spamtrap.org> wrote:
> Doug Freyburger <dfrey....TakeThisOut@yahoo.com> wrote:
> > Derek Ray <de....TakeThisOut@moot.its.only.a.spamtrap.org> wrote:
>
> >> This thread came up about a year ago, maybe a little bit more.  I
> >> crunched some numbers as well; the output is at:
>
> >>http://www.akrasiac.org/rachel/role_distinctions.html
>
> > Maybe the number to look for would be percentage ascending
> > by type not percentage of serial ascensions among ascenders.
>
> Percentage ascended by type gives incorrect numbers - it lists Caveman
> and Knight as easiest because they're rarely played by anyone except
> those filling out their "list of 13".  You can see a few columns related
> to that in there.
>
> "Percentage of serial ascensions among ascenders" doesn't make sense as
> written; run that by again, rephrased?

If I read the table correctly there are columns of how many ascensions
are a part of a run of sequential ascensions. It takes very few
regular
serial ascenders to bias such numbers greatly.

> >> The data is old, of course, but the proportions are similar.  I don't
> >> have time to sit and interpret the data in detail again at the moment,
> >> but a good indicator of which roles are "easiest" was the rightmost
> >> column in the top section -- players whose first ascension was that
> >> particular role.
>
> > That's the other good qualifier for difficulty IMO.
>
> > Maybe combine the two numbers by multiplying?  Then sort highest
> > to lowest as a difficulty estimate.
>
> Which two numbers are you thinking of?

(First ascensions by type) / (Number of games by type)

It counteracts the effects of popularity and drops the effect of
serial
ascenders.

Given how popular valk and wiz are I would not be surprised if their
first asension ratio would be very low. And that really does mean
they are "harder" in some nebulous sense,

Nethack in general is so hard that measuring difficulty by class
is like using mathematics to compare different ways to generate
infinity. And then noting that a method like tangent generates
serial inifinities and thus there exist serial ascenders in Nethack.
 >> Stay informed about: Role difficulty and number of ascensions 

On Feb 7, 5:41 pm, Doug Freyburger <dfrey... RemoveThis @yahoo.com> wrote:
> Derek Ray <de... RemoveThis @moot.its.only.a.spamtrap.org> wrote:
>
> > "Percentage of serial ascensions among ascenders" doesn't make sense as
> > written; run that by again, rephrased?

I wonder, has anyone looked at 78291's NAO ascension rates by class?
He has over 200 ascensions within the past year and, by himself, might
provide a good read on the real degree of difficulty for each class.

-cdi
 >> Stay informed about: Role difficulty and number of ascensions