Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Money game competition against XG
49 views
Skip to first unread message
peps...@gmail.com
May 21, 2022, 7:44:16 AM5/21/22
to
I have thought up a set of rules for how we can compete against each
other in money play against XG. Please let me know what you think of
the concept, or would like to suggest alternative rules.
1) Choose any common sense setting that you like. Obviously, let XG
perform at near-optimal world-class level.
2) Before starting the session, write (to yourself) some type of note
saying something like "I will start the session".
3) Then when you've had enough, write a similar note saying "I've finished
the session."
4) The reason that these notes are mandated is to prevent people
cheating (perhaps inadvertently) by having a disastrous game and then
saying "Oh, I wasn't taking the last game seriously. To get my real result,
let's look at the P and L before the last game, when I was ahead by 10."
5) A win is defined as "undeserved" if the following three conditions all hold:
A) Your PR for that game was worse than 5.00
B) Either there was a time when you doubled where XG thought you were not good enough to
double or there was a time when you took where XG thought you should have passed, or there was a time where XG thought you beavered incorrectly.
C) You were not able to counter XG's verdict under B) by rollout.
If you gain points from an undeserved win, then your score from that game is replaced by zero.
6) The reason 5) is necessary is to prevent massively escalating the
stakes until you get lucky.
7) Simply report your points total for the session, using adjustments (if any)
from clause 5).
Paul
other in money play against XG. Please let me know what you think of
the concept, or would like to suggest alternative rules.
1) Choose any common sense setting that you like. Obviously, let XG
perform at near-optimal world-class level.
2) Before starting the session, write (to yourself) some type of note
saying something like "I will start the session".
3) Then when you've had enough, write a similar note saying "I've finished
the session."
4) The reason that these notes are mandated is to prevent people
cheating (perhaps inadvertently) by having a disastrous game and then
saying "Oh, I wasn't taking the last game seriously. To get my real result,
let's look at the P and L before the last game, when I was ahead by 10."
5) A win is defined as "undeserved" if the following three conditions all hold:
A) Your PR for that game was worse than 5.00
B) Either there was a time when you doubled where XG thought you were not good enough to
double or there was a time when you took where XG thought you should have passed, or there was a time where XG thought you beavered incorrectly.
C) You were not able to counter XG's verdict under B) by rollout.
If you gain points from an undeserved win, then your score from that game is replaced by zero.
6) The reason 5) is necessary is to prevent massively escalating the
stakes until you get lucky.
7) Simply report your points total for the session, using adjustments (if any)
from clause 5).
Paul
MK
May 22, 2022, 1:22:49 AM5/22/22
to
On May 21, 2022 at 5:44:16 AM UTC-6, peps...@gmail.com wrote:
> I have thought up a set of rules for how we can compete
> against each other in money play against XG. Please let
> me know what you think of the concept, or would like to
> suggest alternative rules.
> 1) Choose any common sense setting that you like.
> Obviously, let XG perform at near-optimal world-class level.
If you really mean "compete against one another" (the
> I have thought up a set of rules for how we can compete
> against each other in money play against XG. Please let
> me know what you think of the concept, or would like to
> suggest alternative rules.
> 1) Choose any common sense setting that you like.
> Obviously, let XG perform at near-optimal world-class level.
correct English for more than two people competing;)
then everyone needs to use the same settings and the
same sequence of dice rolls.
> 5) A win is defined as "undeserved" if the following three
> conditions all hold:
gamblegammon: "undeserved win"... Ha ha! Funny. :)
> A) Your PR for that game was worse than 5.00
> B) Either there was a time when you doubled where.....
Cube skill is meaningless bullshit. Ask Tim if you don't
believe me. ;)
> C) You were not able to counter XG's verdict under B)
> by rollout.
equity estimate calculations. They are also bullshit.
In your other thread about "casino gamblegammon" you
said: "The point of my game is that every human would
lose in the long run" and now you put all kinds of stupid
restrictions on human players so that they can't beat the
bot...?? Tragicomedy! :) :(
> 6) The reason 5) is necessary is to prevent massively
> escalating the stakes until you get lucky.
There is no guarantee either that after escalating the cube,
the bot won't end up winning it. Saying that only the human
will get lucky is against logic and insulting to the players.
> 7) Simply report your points total for the session, using
> adjustments (if any) from clause 5).
in such a "competition"...
MK
MK
May 26, 2022, 10:34:36 PM5/26/22
to
On May 21, 2022 at 11:22:49 PM UTC-6, MK wrote:
> On May 21, 2022 at 5:44:16 AM UTC-6, peps...@gmail.com wrote:
>> I have thought up a set of rules for how we can
>> compete against each other in money play against
>> XG. Please let me know what you think of the
>> concept, or would like to suggest alternative rules.
>> 1) Choose any common sense setting that you like.
> If you really mean "compete against one another".....
> On May 21, 2022 at 5:44:16 AM UTC-6, peps...@gmail.com wrote:
>> I have thought up a set of rules for how we can
>> compete against each other in money play against
>> XG. Please let me know what you think of the
>> concept, or would like to suggest alternative rules.
>> 1) Choose any common sense setting that you like.
> then everyone needs to use the same settings and
> the same sequence of dice rolls.
Also, everyone needs to use the same version of the
> the same sequence of dice rolls.
bot. My version says 2.19.211.2671.pre-release. I'm
not sure if it's the latest one. Does anyone else have
a later version? And why so many people here are still
using version 2.10, etc.??
>> 5) A win is defined as "undeserved" if the following
>> three conditions all hold:
> Ha ha! :) This is amusing. Now we have a new concept
> in gamblegammon: "undeserved win"... Ha ha! Funny. :)
wanted to encourage your idea, along with your casino
play idea, and I hope that something will come out of it.
>> 7) Simply report your points total for the session....
> Will be interesting to see who will be the idiots to
> participate in such a "competition"...
have made some constructive suggestions. How about
combining the two and doing it in online virtual meetings?
If done without ridiculous rules and limitations, I would
participate and I would even pay money to participate in
non-gambling competitions, experiments, etc.
Many people are trying to make a buck by doing similar
online operations. Could this be profitable enough for
someone to organize?
Let's say 10 people join a virtual meeting and share their
screens, playing against a bot with everything being the
same for all of them. How much would be a fair price to
make it worthwhile for the host/operator? I'm sure a good
number of games can be played in an hour, for example,
using common clock settings.
We can get into more details if there is any interest at all.
Some bot and/or online game site developers/operators
could get into this as the next step in online playing..??
MK
MK
Jun 18, 2022, 9:58:58 PM6/18/22
to
On May 26, 2022 at 8:34:36 PM UTC-6, MK wrote:
> We can get into more details if there is any interest at all.
> Some bot and/or online game site developers/operators
> could get into this as the next step in online playing..??
Too bad nobody other me commented on Paul's idea.
> We can get into more details if there is any interest at all.
> Some bot and/or online game site developers/operators
> could get into this as the next step in online playing..??
I guess I only wasted my time by doing so because I
never believed that any competitions, experiments or
similar events open to public's sucrutiny would never
happen, as they would expose/reveal too much about
the bots and their developers. ;)
I'm hereby adding this to my various past predictions
about Jellyfish, Snowie, Gnubg, XG, etc. which all held
true thus far...
MK
0 new messages