Possible "cheat" with ability gain?
APLer
more when the relevent exercise is conducted shortly before the 600 turn
interval. Has/Can anyone looked into whether the arithmetic backs this up?
sreservoir
There is no 600 turn interval. It merely begins first on turn 600.
Exercise is given equal weight regardless of timing.
--
"Six by nine. Forty two."
"That's it. That's all there is."
"I always thought something was fundamentally wrong with the universe"
APLer
> APLer wrote:
>> Recently it has come to my attention that my abilities tend to increase
>> more when the relevent exercise is conducted shortly before the 600
>> turn interval. Has/Can anyone looked into whether the arithmetic backs
>> this up?
>
> There is no 600 turn interval. It merely begins first on turn 600.
>
Yes there is. Pay more attention when you play.
> Exercise is given equal weight regardless of timing.
>
Why don't I now believe you know that?
The first increase check happens at turn 600, the next *around* 1600 and
every other is every 600 turns. If you exercise a stat on a turn around
modulus 100 or less nothing much happens at the next interval. However if
you exercise around modulus 500, you're more likely to get a stat
increase it appears. I refer you to exercise.txt (which *used* to be on
Eva's site, but the site that held it no longer exists. Someone needs to
archive these before they all disappear!). The cheat would be minor at
best I suppose, but it would decrease the amount of turns spent exercising
a stat and could affect luck too I suppose.
Now as I said does anyone actually *know* if it's true?
Patric Mueller
> sreservoir <srese...@gmail.com> wrote in news:hdcpus$c7p$1...@aioe.org:
>
>> APLer wrote:
>>> Recently it has come to my attention that my abilities tend to increase
>>> more when the relevent exercise is conducted shortly before the 600
>>> turn interval. Has/Can anyone looked into whether the arithmetic backs
>>> this up?
>>
>> There is no 600 turn interval. It merely begins first on turn 600.
>>
> Yes there is. Pay more attention when you play.
>
>> Exercise is given equal weight regardless of timing.
>>
> Why don't I now believe you know that?
>
> The first increase check happens at turn 600, the next *around* 1600 and
> every other is every 600 turns. If you exercise a stat on a turn around
Why do you claim stuff that's easy to disprove by looking at the
source code linked from
http://nethack.wikia.com/wiki/Exercise
There is no 600 turns interval, there's a 200 to 1000 turns interval:
http://nethack.wikia.com/wiki/Attrib.c#line441
Bye
Patric
--
NetHack-De: NetHack auf Deutsch - http://nethack-de.sf.net/
NetHack for AROS: http://sf.net/projects/nethack-aros/
UnNetHack: http://apps.sf.net/trac/unnethack/
Ilmari Karonen
> APLer <AP...@floor.tilde> wrote:
[stuff]
>
> Why do you claim stuff that's easy to disprove by looking at the
> source code linked from
> http://nethack.wikia.com/wiki/Exercise
Because he's chuckie. He doesn't seem to need any other reason.
But he does have a point in his original post, sort of. The exercise
counter has a hard cap at 50 exercise points, so you can't "save up"
any more exercise than that -- once you hit that limit, any further
exercise is useless until something (abuse or the next exercise check)
reduces it again.
Now, the recommended way to raise your stats is to keep exercising
regularly as you play and to avoid stat abuse entirely (unless
necessary to achieve some other goal). But from what I've seen, a lot
of inexperienced players seem to engage in infrequent periods of
frantic exercise (such as finding a boulder and pushing it around the
level until hunger forces them to stop) and pay little attention to
stat exercise and abuse otherwise.
This is an inefficient strategy, since most of the exercise is likely
to be wasted due to the 50 point cap and subsequent abuse will most
likely bring the exercise counter back down again before the next
check. But the inefficiency is minimized if one happens to time one
of those massive bouts of exercise just before a stat check, since
then there will have been less time for abuse to reduce the counter.
So, yes, if one exercises only sporadically, one should try to do it
just before a stat check. But since stat checks are not entirely
predictable, except for the first one, it's best to keep exercising
your stats all the time, whenever the opportunity arises, and to avoid
abuse so that your exercise counters will always stay fairly close to
the upper limit.
--
Ilmari Karonen
To reply by e-mail, please replace ".invalid" with ".net" in address.
Richard Bos
> Now as I said does anyone actually *know* if it's true?
As usual with you, it's not true, and a browse of the source would have
confirmed that.
Richard
Link
> On 2009-11-11, Patric Mueller <bh...@bigfoot.com> wrote:
>
> > APLer <AP...@floor.tilde> wrote:
>
> This is an inefficient strategy, since most of the exercise is likely
> to be wasted due to the 50 point cap and subsequent abuse will most
> likely bring the exercise counter back down again before the next
> check.
I'm confused with your wording. I thought the random 1-50 number is
compared to abs(1/2*2/3) the exercise counter for strength/dexterity/
constitution, and abs(1/2) the exercise counter for wisdom?
In other words, it seems like the cap for most stats is actually 150.
And if the stat is 11 or higher, then you have less than a 50% chance
to increase the counter when you do something that exercises it.
Which means that you can often exercise a stat more than 300 times
without hitting the cap?
APLer
news:lziqdhi...@bhaak.dyndns.org:
> APLer <AP...@floor.tilde> wrote:
>> sreservoir <srese...@gmail.com> wrote in news:hdcpus$c7p$1...@aioe.org:
>>
>>> APLer wrote:
>>>> Recently it has come to my attention that my abilities tend to
>>>> increase more when the relevent exercise is conducted shortly before
>>>> the 600 turn interval. Has/Can anyone looked into whether the
>>>> arithmetic backs this up?
>>>
>>> There is no 600 turn interval. It merely begins first on turn 600.
>>>
>> Yes there is. Pay more attention when you play.
>>
>>> Exercise is given equal weight regardless of timing.
>>>
>> Why don't I now believe you know that?
>>
>> The first increase check happens at turn 600, the next *around* 1600
>> and every other is every 600 turns. If you exercise a stat on a turn
>> around modulus 100 or less nothing much happens at the next interval.
>> [...]
>
> Why do you claim stuff that's easy to disprove by looking at the
> source code linked from
> http://nethack.wikia.com/wiki/Exercise
>
> There is no 600 turns interval, there's a 200 to 1000 turns interval:
> http://nethack.wikia.com/wiki/Attrib.c#line441
>
So there's a check at 600 turns *as* *I* said, and then random from 200 to
800 turns later from then on. Oh *I* see *big* difference.
APLer
news:4afaac47...@news.xs4all.nl:
Your ability to be wrong is simply staggering.
APLer
news:slrnhflbdc....@melkki.cs.helsinki.fi:
> On 2009-11-11, Patric Mueller <bh...@bigfoot.com> wrote:
>> APLer <AP...@floor.tilde> wrote:
> [stuff]
>>
>> Why do you claim stuff that's easy to disprove by looking at the
>> source code linked from
>> http://nethack.wikia.com/wiki/Exercise
>
> Because he's chuckie. He doesn't seem to need any other reason.
>
No, other than being completely right within any accuracy you care to
choose.
> But he does have a point in his original post, sort of. The exercise
> counter has a hard cap at 50 exercise points, so you can't "save up"
> any more exercise than that -- once you hit that limit, any further
> exercise is useless until something (abuse or the next exercise check)
> reduces it again.
>
> Now, the recommended way to raise your stats is to keep exercising
> regularly as you play and to avoid stat abuse entirely (unless
> necessary to achieve some other goal). But from what I've seen, a lot
> of inexperienced players seem to engage in infrequent periods of
> frantic exercise (such as finding a boulder and pushing it around the
> level until hunger forces them to stop) and pay little attention to
> stat exercise and abuse otherwise.
>
> This is an inefficient strategy, since most of the exercise is likely
> to be wasted due to the 50 point cap and subsequent abuse will most
> likely bring the exercise counter back down again before the next
> check. But the inefficiency is minimized if one happens to time one
> of those massive bouts of exercise just before a stat check, since
> then there will have been less time for abuse to reduce the counter.
>
> So, yes, if one exercises only sporadically, one should try to do it
> just before a stat check. But since stat checks are not entirely
> predictable, except for the first one, it's best to keep exercising
> your stats all the time, whenever the opportunity arises, and to avoid
> abuse so that your exercise counters will always stay fairly close to
> the upper limit.
>
Yes, as a matter of fact they *are* which is my point. every 200-800 turns
later there will be another. So if you target the 500th turn each time,
you gain.
The truly amazing thins is how absolutely disfunctional this group has become.
When I first posted here there were people who shit on the trolls. Not anymore.
No wonder half of the then posters have left this mess.
7aboir
rn1(200,800)=rn2(200)+800 as in
http://nethack.wikia.com/wiki/Hack.h#line291
So between 800 and 999.
See the difference?
Ilmari Karonen
> On Nov 11, 7:27ï¿œam, Ilmari Karonen <usen...@vyznev.invalid> wrote:
>> On 2009-11-11, Patric Mueller <bh...@bigfoot.com> wrote:
>>
>> > APLer <AP...@floor.tilde> wrote:
>>
>> This is an inefficient strategy, since most of the exercise is likely
>> to be wasted due to the 50 point cap and subsequent abuse will most
>> likely bring the exercise counter back down again before the next
>> check.
>
> I'm confused with your wording. I thought the random 1-50 number is
> compared to abs(1/2*2/3) the exercise counter for strength/dexterity/
> constitution, and abs(1/2) the exercise counter for wisdom?
I was referring to the 50-point cap imposed in exercise(), on line 275
of attrib.c, where abs(AEXE(i)) is compared with AVAL, which is
#defined a few lines earlier as 50.
But the more I look at that line, the more it looks like it's either
buggy or at least behaving very differently than how I assumed it
should work. The problem is that this comparison is done before the
exercise counter is adjusted, so, once the counter hits +50 or -50,
it's effectively frozen at that value until the next stat check.
Which, if true, suggests that a frantic burst of exercise just after
the previous stat check might not be such a bad idea after all, as
long as you do enough of it to make the counter reach +50 with a
reasonably high probability. (Assuming a starting value of +25, this
should take about 475/(18-X) exercise steps to achieve with 50%
probability, where X is the current value of the stat being
exercised.)
Is this really correct, and if so, is it intentional?
Richard Bos
> On 2009-11-11, Link <chill...@yahoo.com> wrote:
> > I'm confused with your wording. I thought the random 1-50 number is
> > compared to abs(1/2*2/3) the exercise counter for strength/dexterity/
> > constitution, and abs(1/2) the exercise counter for wisdom?
>
> I was referring to the 50-point cap imposed in exercise(), on line 275
> of attrib.c, where abs(AEXE(i)) is compared with AVAL, which is
> #defined a few lines earlier as 50.
Yes, you're correct, and no, I can't imagine that this is intentional.
Best file a bug report, if it's not already in the bug list (which I
haven't checked, so it might be).
Richard