--- In zertz@yahoogroups.com, Dave Dyer <ddyer@r...> wrote:
<< A real draw by repetition occurred in a game on boardspace.net
>>

Do you have a transcript of the game? It'd be interesting to see how
that came about.

Hello Zèrtz folks,
the second season of the Zèrtz league will start at September, 18th.
You can still sign up at http://www.assimilate.de/Default.aspx?tabid=37.

The league will be played at boardspace.net with the same rules as in the
first season, but there are some organizational changes:

Depending on the number of sign ups, we will play the league in divisions -
probably Europe and North America - in order to minimize the timezone
difference between players.
If the Zèrtz+11 league/divisions are filled (with up to 7 players each),
additional leagues/divisions - Zèrtz and Zèrtz+24 - will be opened.
All leagues/divisions will be filled according to the Max-Ever Ranking that
you have in the related Zèrtz rankings at boardspace.net.

Michael Reitz

Hello Zertz folks,
the Zertz+11 PbeM championship is about to start.

16 players signed up and were divided into 4 groups.
The players were first ordered according to their current Zertz+11 rating at
Richard's PbeM Server (provisional ratings were reduced by 1000), and then
distributed into 4 groups with the following scheme:
Group A: 1,8,12,16
Group B: 2,7,11,15
Group C: 3,6,10,14
Group D: 4,5,9,13

Please have a look at the current standings and the detailed chamionship
rules at:
http://assimilate.int.webmatrixhosting.net/DesktopDefault.aspx?tabid=38

The PbeM championship starts as soon as Richard has setup the tournament
pools and boards on his PbeM server. So, just wait until you get your boards
by email.

Michael

Hello Zertz folks,
the Zertz+11 PbeM championship will probably start at May, 24th.
12 players have already signed up, and 4 slots are still free.
So, take the chance of the last-minute sign up.
Information about the event can be found at:
http://assimilate.int.webmatrixhosting.net/DesktopDefault.aspx?tabid=37

Michael

Hello Zertz folks,
the Zertz+11 league will start at May, 12th.
5 players have already signed up, and 2 slots are still free.
So, take the chance of the last-minute sign up for playing in the first
season. Updated details can be found at:
http://assimilate.int.webmatrixhosting.net/DesktopDefault.aspx?tabid=37

Michael

P.S.: There are still 7 slots free in the Zertz+11 PbeM championship, which
will also start within this month.

Hello Zertz folks,
I'm organizing two Zertz events on the Internet this year.

First of all there will be a new Zertz+11 PbeM tournament on Richard's PbeM
Server with 16 slots (4 groups of 4 players each; established ratings will
be considered in group arrangement).
Every player plays two games against every other player in the same group.
All players winning at least 4 (out of 6) games in the first round will
proceed to the second round.
The second round will (hopefully) consist of 8 players divided into two
groups, played in the same round-robin schedule.
The best two players of both groups will proceed to the final round
competing for the title.
The results of the previous rounds will be used as tiebreakers after the
first round is played.
Every player gets 30 days per round for playing all 6 games simultaneously,
which results in not more than 60 days per round and not more than 6 months
for the whole tournament.

Additionally there will be a Zertz(+11) league at boardspace.net; the league
size is not decided yet (it'll depend on how many players are interested to
participate). The league will be played in seasons lasting some weeks.
Every player plays two games against every other player in the same league
(there will probably be only one league for the beginning) with a 15 minute
time limit per game.
Players will have to organize appointments to play their games within the
given time for the season. Missing games at the end of a season will be
scored as a loss for both players and will give a starting penalty for the
next season (except if it is clear which player caused the game not to be
played; then only this player gets a penalty and the game is scored as a win
for the other player).
It is not decided yet if Zertz or Zertz+11 is played in the first season, so
let me know which one you would prefer.


To sign up for one or both events please send an email to zf1@...
with the following information:
- your name and your userid(s)
- in which events you want to participate
- all your questions, suggestions and comments about the events


Both events will start in May 2004 and more detailed information will become
available during the sign up period.
Just keep an eye on http://zf1.webhop.org -> Events.


Michael Reitz

P.S.: I hope that many players sign up for both events, so that we can find
out how much the strength of particular players differs between PbeM and
realtime play.

Boardspace.net now has zertz+11, and a robot good enough
to win if you're not careful.

I couldn't help myself, I've put up a site where you can
play Zertz against other humans in realtime.  All you need,
right now, is an opponent.  So arrange a game a rendevous
and check it out.

  http://www.boardspace.net

| I just found out that you can get a draw in Zèrtz,
| (well not me really, Zillions did) so I have a question...
|
Such a draw is extremely rare in usual games, but Zillions plays extremely
bad, so I'm not surprised.

| Is the 3-rd repetition rule in effect in Zèrtz?
|
I'm not sure if there is such a rule officially or explicitly, but usually
both players agree to a draw in such a situation.

| If the common pool is empty, and both players just
| feed each other captures with the same marbles...
| (and when there are no free rings to remove)
|
Yes, that is the situation, but it can only be forced by a cooperation of
both players.

Michael

Hello!

I just found out that you can get a draw in Zèrtz,
(well not me really, Zillions did) so I have a question...

Is the 3-rd repetition rule in effect in Zèrtz?

If the common pool is empty, and both players just
feed each other captures with the same marbles...
(and when there are no free rings to remove)

any ideas?

Kindest regards
Rolle

Corrected the draw condition which resulted from depletion of the
draw-pile.  Now when there are no more pieces in the common stock,
just select a vacant spot on the playing field and choose a piece
from the appearing list.  The appropriate piece will be removed from
your hand and placed on the field.

Thanks to Roland Johansson for calling my attention to this problem.

Hello Zèrtz players,
I have an easter egg for all ZF1 users: ZF1 2.6

I just uploaded the setup for ZF1 2.6 to my ZF1 homepage:
http://zf1.webhop.org

This update includes some improvements and the following new features:
- Added new variant tree shortcuts for faster navigation.
- Added variant tree context-menu with two new features:
   - "Show all Captures" adds all possible captures of a particular
     position to the variant tree.
   - "Revise Variant" spawns a subtree with a revised starting move.

Please fill out the registration that shows up after the installation is
finished. I would just like to get an overview who is using ZF1.


Michael

Hello,
|
| > Back to the network play feature. There was a discussion about how a
| > network play feature should be implemented last year on the GIPF
| > yahoogroup.
| > We didn't find a conclusion, because there was no agreement on the
| > following issues:
| > - peer-to-peer vs. server-based solution
| > - plain sockets-based TCP/IP with proprietary protocol vs.
standards-based
| > approach using SOAP
| >
|
| I use a server based solution for the communication, but the server
| itself is a normal peer.
|
"server-based solution" above meant dedicated server, so your solution is a
peer-to-peer solution with a "master-slave" realtionships between the peers.

| My protocol is very proprietary. I use enums in C++ which map names to
| numbers, what means that I have words in the code, but only the numbers
| identifying these words are sent as ASCII Text in a stream. For player
names
| the GUI converts unicode strings to UTF8, so the backend only has to
handle
| traditional 8 Bit strings.
|
So, it is a plain sockets-based TCP/IP communcation on a proprietary
protocol.

| > So, if you are interested in network play / network AI compatibility
| > between Holtz and ZF1, we should bring up the issue again, and try to
find
| > an agreement.
| > I've seen that Holtz is developed in C++, so your opinion about the
above
| > mentioned issues would be of interest (especially a technical statement
| > about using SOAP with C++).
| >
|
| It would be nice to have a common protocol. I should have no problem to
| implement multiple protocols as long as they support all features I have.
|
Okay.
Do you have any experience with or knowledge about using SOAP with C++?

| Can you give me a hint, where I get the description of the PBM format?
| Or we could play a game, so I can see how it works.
|
The official Zertz notation can be found at
http://www.gipf.com/zertz/index.html -> Notation.
The notation used on Richard's PbeM Server is slightly different.

| I have the problem, that holtz displays the field 90° rotated to the way
| everyone else displays it. Do you think this does matter?
|
No, as long as the coordinate system isn't changed.
However, players are used to the standard board orientation and me be
confused by a "rotated" board.

Michael

Hi,

Michael Reitz wrote:

> You should mention on the Holtz project page, that the game is a Zertz
> implementation.
>

I will. But at the moment I am quite busy.

> Back to the network play feature. There was a discussion about how a network
> play feature should be implemented last year on the GIPF yahoogroup.
> We didn't find a conclusion, because there was no agreement on the following
> issues:
> - peer-to-peer vs. server-based solution
> - plain sockets-based TCP/IP with proprietary protocol vs. standards-based
> approach using SOAP
>

I use a server based solution for the communication, but the server
itself is a normal peer. My protocol is very proprietary. I use enums in
C++ which map names to numbers, what means that I have words in the
code, but only the numbers identifying these words are sent as ASCII
Text in a stream. For player names the GUI converts unicode strings to
UTF8, so the backend only has to handle traditional 8 Bit strings.

> So, if you are interested in network play / network AI compatibility between
> Holtz and ZF1, we should bring up the issue again, and try to find an
> agreement.
> I've seen that Holtz is developed in C++, so your opinion about the above
> mentioned issues would be of interest (especially a technical statement
> about using SOAP with C++).
>

It would be nice to have a common protocol. I should have no problem to
implement multiple protocols as long as they support all features I have.


Can you give me a hint, where I get the description of the PBM format?
Or we could play a game, so I can see how it works.

I have the problem, that holtz displays the field 90° rotated to the way
everyone else displays it. Do you think this does matter?

bye

Martin

The Virtual Game Club (VGC) has a Zertz implementation
which is supposed to allow network play (they have
their own Yahoo newsgroup) but I have never been able
to get it to work.  I don't know if anyone else has.

Lyman

--- Michael Reitz <yahoo@...> wrote:
> | I wanted to play Zertz over a network with friends
> and therfore coded a
> | Zertz implementation myself. The name is Holtz
> because I didn't want to
> risk
> | any copyright problems.
> |
> You should mention on the Holtz project page, that
> the game is a Zertz
> implementation.
>
> | My major fault was not to look for yet available
> realizations.
> |
> Did you ever visit the Project GIPF website
> (http://www.gipf.com)?
>
> | So do you think that there is still a demand for
> Holtz?
> |
> I'm the developer of ZF1 (http://www.assimilate.de).
> It's purpose is more focused on Zertz analysis than
> on network play.
> It is not yet possible to play with ZF1 on a network
> or over the Internet,
> and the next big development step will be the
> support for (third-party) AIs.
> ZF1 is NOT open source, but I'm going to expose
> COM-interfaces with the next
> major release in order to support third-party AI.
> It currently requires Windows, and it will become
> more platform-independent
> when I migrate it to the .Net platform (such a
> migration is not scheduled
> yet).
>
> Back to the network play feature. There was a
> discussion about how a network
> play feature should be implemented last year on the
> GIPF yahoogroup.
> We didn't find a conclusion, because there was no
> agreement on the following
> issues:
> - peer-to-peer vs. server-based solution
> - plain sockets-based TCP/IP with proprietary
> protocol vs. standards-based
> approach using SOAP
>
> So, if you are interested in network play / network
> AI compatibility between
> Holtz and ZF1, we should bring up the issue again,
> and try to find an
> agreement.
> I've seen that Holtz is developed in C++, so your
> opinion about the above
> mentioned issues would be of interest (especially a
> technical statement
> about using SOAP with C++).
>
> Michael
>
>

| I wanted to play Zertz over a network with friends and therfore coded a
| Zertz implementation myself. The name is Holtz because I didn't want to
risk
| any copyright problems.
|
You should mention on the Holtz project page, that the game is a Zertz
implementation.

| My major fault was not to look for yet available realizations.
|
Did you ever visit the Project GIPF website (http://www.gipf.com)?

| So do you think that there is still a demand for Holtz?
|
I'm the developer of ZF1 (http://www.assimilate.de).
It's purpose is more focused on Zertz analysis than on network play.
It is not yet possible to play with ZF1 on a network or over the Internet,
and the next big development step will be the support for (third-party) AIs.
ZF1 is NOT open source, but I'm going to expose COM-interfaces with the next
major release in order to support third-party AI.
It currently requires Windows, and it will become more platform-independent
when I migrate it to the .Net platform (such a migration is not scheduled
yet).

Back to the network play feature. There was a discussion about how a network
play feature should be implemented last year on the GIPF yahoogroup.
We didn't find a conclusion, because there was no agreement on the following
issues:
- peer-to-peer vs. server-based solution
- plain sockets-based TCP/IP with proprietary protocol vs. standards-based
approach using SOAP

So, if you are interested in network play / network AI compatibility between
Holtz and ZF1, we should bring up the issue again, and try to find an
agreement.
I've seen that Holtz is developed in C++, so your opinion about the above
mentioned issues would be of interest (especially a technical statement
about using SOAP with C++).

Michael

I wanted to play Zertz over a network with friends and therfore coded a Zertz
implementation myself. The name is Holtz because I didn't want to risk any
copyright problems.
My major fault was not to look for yet available realizations.

So do you think that there is still a demand for Holtz?
Does anyone like to help?

Holtz is hosted as a GPLed Project by sourceforge:
holtz.sourceforge.net

Martin Trautmann

Hello Zèrtz players,
here are some good news about ZF1.

I just uploaded the final release of ZF1 2.5 on my homepage:
http://www.assimilate.de

I got a bunch of suggestions after releasing the public release candidate.
The most of them are now implemented in ZF1 2.5.25:
- Added "move-by-move navigation" through Move List (use +/- keys).
- Added "Fixed Diagram" and "Move Coloring" options to the "View" menu.
- Added "Load as Puzzle" option to the "Game" menu and "Hint"-support.
- Added support for sending the last move or a show-command to "Richard's
   PbeM Server".
- Changed mouse clicking behaviour on board diagram.

This is an overview of the changed mouse clicking behaviour:
- Moving a marble: left-click on a pool/marble and drag it on a ring/pool
- Removing/Restoring a ring: left-click on a ring / removed ring
- Toggle active player: right-click / press Enter key

So, compared to ZF1 2.0 you'll have to use a left-click for removing/restoring
a ring (instead of a right-click) and the
double-click is replaced by a right-click.

The enhanced support for "Richard's PbeM Server" should be useful for all
Zertz+11 eMail tournament players.
You should import a PbeM game before using the "Send Last Move ..."-feature,
because the ZF1 game/file name must follow a particular format.

Please feel free to ask questions and send suggestions to zf1@....

Michael

First to clarify dates.  It should read:

Last date to enter: Wednesday, January 8, 2003
Tournament begins:  Friday, January 10, 2003

I wanted to give people on vacation a chance to
respond.

Michael Reitz has made some useful suggestions
(particularly as they entail less wear and tear on the
tournament director).

There will be a 30 day/player total limit on play time
(it was pointed out that this number is kept track of
by the server and therefore easy to track), rather
than a per move limit.

With the number of respondents, there will certainly
need to be (at least) two rounds.

If feasible, we will try to schedule it so that
everybody plays everybody else once as PLayer 1 and
once as Player 2 to avoid asymmetry.

More details to follow.  Please feel free to email
suggestions.

Cheers,

Lyman

PS I will message everyone via the server who has
indicated that they want to partiucipate.  You only
need respond again if you change your mind :-).

I have uploaded a couple dozen new ZERTZ puzzles to the File section
of this group site.

I've attempted to create some which are both challenging and
interesting.

Please let me know if there are any errors or omissions.

There will be a Zertz+11 (i.e., 48 ring) Zertz
tournament on Richard's PBEM server starting Friday
January 8, 2003.  All entries must be in by January 6
to give me time to make the pairings.

The format will be round-robin.  There are currently 7
people who have expressed interest.  If I get
significantly more people, I will break into pools and
have two rounds.  I will announce the final pairings
and format sometime between January 6 and the January
8.

Players are requested to make moves within five days
(maximum, although faster would be better) including
forced moves (i.e., whose turn it is should shift
within five days).

More details as the date approaches.

Cheers and hoping everyone is having a good holiday
season,

Lyman Hurd

I found an error created by the "scan-board" macro. It allowed
illegal isolation captures. The fix was simple. I just doubled up
on the macro, and everything is now copacetic.

I also specifically stated the draw-condition of stalemate. This
seems to make the engine respond better. I actually achieved a draw
game.

Sorry for the inconvience.

Hello ZF1 users,
the changed mouse clicking behaviour (allowing a left-click for
removing/restoring a ring) in the ZF1 release candidate interferes with the
double-click (for executing a move / toggle active player).

Therefore I plan to change the mouse clicking behaviour a little bit more in
the final release of ZF1 2.5:
- Moving a marble: left-click on a pool/marble and drag it on a ring/pool
- Removing/Restoring a ring: left-click on a ring / removed ring
- Toggle active player: right-click / press Enter key

So, compared to ZF1 2.0 you'll have to use a left-click for removing/restoring
a ring (instead of a right-click) and the double-click is replaced by a
right-click.

Alternatively I could drop the option to use a left-click for
removing/restoring a ring, restoring the original behaviour of ZF1 2.0.

So, please let me know if you prefer the original behaviour of ZF1 2.0 or the
new behaviour described above.
I'll implement the new behaviour if there isn't a significant number of
"protests" against it.
Any feedback is received at zf1@....


Michael

P.S.:
For users of ZF1 2.5.24: Avoid double-clicking over a removed ring!

Hello Zèrtz players,
here are some good news about ZF1.

I just uploaded the public release candidate of ZF1 2.5 on my homepage:
http://www.assimilate.de

I want to thank all people that participated in beta testing; I received
important feedback and I'm going to do closed beta testing again for the next
major development step (to be announced next year).

I decided to publish this release candidate now, because there are no critical
to-do's left and I want to share the bunch of new features with all interested
Zèrtz players.

Please feel free to ask questions and send suggestions to zf1@....
And I'd like to receive a short notification eMail from all ZF1 users, so that
I get an impression how many people are using it.

Michael

One of the primary hurdles with alignment and accumulation games is
the ability of the program to either visualize or anticipate these
conditions.  Zillions is a plodder, it merely runs through each and
every possible move.

So, if its Strength is set too low it will not be able to calculate
certain positions or conditions.

I stated that Zillions plays ZERTZ 'fairly well', and it can.  If the
player has the patience and is willing to allow the engine to run for
a while.  I suggested Strength=Expert, Variety=Large and Average
Time=1-3 minutes.  In fact, you can set the engine for harsher
conditions, Variety=Small and Average Time=Forever.  Make your move
then let the Zillions engine run for as long as you can stand.  Force
the engine to move when you are ready.

The first moves will seem frustrating, but in the mid-game the engine
will start catching up to the potentials.  In fact, the end-game will
become a disappointment, especially when Zillions declares that it
can win in a certain number of moves.  Of course, if Zillions
considers that it has lost you still might make an error and allow it
to win.

Remember, Zillions considers two moves a turn.  So when it is
displaying the level of search, multiply that by two and you will
have the depth of moves calculated.  Moves consist of placement,
removal, jumps and isolation capture.  Each jump is a single move and
isolation captures are taken in one.

One area that I've noticed, the engine doesn't try to create or take
advantage of isolation captures in the early part of the game.  I am
working on some coding which will encourage this behaviour.  When
I've finished I'll post the result.

I hope that this rant has aided others in enjoying this
implementation.  I did test the engine with the two problems which
are available at this site, it was able to solve both.

BTW, those two problems are in error.  They both have too many discs
removed for normal play.

| I am hopeful that between ZF1 and this implementation
| we will someday in the not too distant future be able
| to have a tournament of a battle of the Zertz AI's!
|
ZF1 will allow the development of third-party AI as soon as I introduce the AI
interface which is planned for next year.
For programmers it should be interesting to know that ZF1 will offer ActiveX
interfaces; so an ZoG-AI won't be compatible.
The ZF1 AI interface will be discussed on this group as soon as I'm ready to
start on developing it.

BTW, I'm currently preparing the ZF1 2.5 public release candidate to be
launched on my ZF1 homepage. So, watch out an announcement here.

Michael

I have to concur on the level of play, which I am
afraid was inevitable given the logical depth of
Zertz.  However, the Zillions framework allows for
independent AI DLL's and I am hopeful that someone
will leverage the great work on the UI to implement a
stronger back-end.

I am hopeful that between ZF1 and this implementation
we will someday in the not too distant future be able
to have a tournament of a battle of the Zertz AI's!
Believe me if pesky responsibities like work and
family didn't keep creeping in, I would be spending
much of my time writing one!

Someday, I promise...

Cheers,

Lyman

--- Michael Reitz <yahoo@...> wrote:
> | I've uploaded to the Files section a ZRF which
> will play ZERTZ fairly
> | well.
> |
> Sorry, but ZoG doesn't play Zèrtz fairly well, in
> fact it plays really worse.
> Okay, it depends on the playing level of the human,
> but even simple exchanges
> are not found.
>
> Don't get me wrong, I don't want to blame your rules
> file and it is not your
> fault! The ZoG engine is just not appropriate for
> finding good Zertz moves.
>
> Michael
>
>
>


__________________________________________________
Do you Yahoo!?
Yahoo! Web Hosting - Let the expert host your site
http://webhosting.yahoo.com

| I've uploaded to the Files section a ZRF which will play ZERTZ fairly
| well.
|
Sorry, but ZoG doesn't play Zèrtz fairly well, in fact it plays really worse.
Okay, it depends on the playing level of the human, but even simple exchanges
are not found.

Don't get me wrong, I don't want to blame your rules file and it is not your
fault! The ZoG engine is just not appropriate for finding good Zertz moves.

Michael

That's great! Thanks.

--- In zertz@y..., "interrupt27" <llsmith@e...> wrote:
<< I've uploaded to the Files section a ZRF [...] >>

(I went ahead and tucked it into the Program folder of the Files section.)

-Walt