Murka 3.0 released

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Günther Höhne
Posts: 164
Joined: Fri Jan 09, 2009 9:48 pm
Contact:

Murka 3.0 released

Post by Günther Höhne » Wed Jul 17, 2013 6:06 pm

Image

Murka 3.0 released (32bit&64bit+sources)
from Igor Korshunov,

http://www.sdchess.ru/download_engines.htm

http://www.sdchess.ru/engines/Murka_3.rar

Regards
Günther

zullil
Posts: 5684
Joined: Mon Jan 08, 2007 11:31 pm
Location: PA USA
Full name: Louis Zulli

Re: Murka 3.0 released

Post by zullil » Wed Jul 17, 2013 6:53 pm

Günther Höhne wrote:Image

Murka 3.0 released (32bit&64bit+sources)
from Igor Korshunov,

http://www.sdchess.ru/download_engines.htm

http://www.sdchess.ru/engines/Murka_3.rar

Regards
Günther
What is known about this engine's "parentage?"

Günther Höhne
Posts: 164
Joined: Fri Jan 09, 2009 9:48 pm
Contact:

Re: Murka 3.0 released

Post by Günther Höhne » Wed Jul 17, 2013 7:43 pm

Hi,

I think this engine is not a derivat. CEGT Blitz already tested precursor.
http://www.husvankempen.de/nunn/40_4_Ra ... n/997.html

Regards
Günther

zullil
Posts: 5684
Joined: Mon Jan 08, 2007 11:31 pm
Location: PA USA
Full name: Louis Zulli

Re: Murka 3.0 released

Post by zullil » Wed Jul 17, 2013 7:48 pm

Günther Höhne wrote:Hi,

I think this engine is not a derivat. CEGT Blitz already tested precursor.
http://www.husvankempen.de/nunn/40_4_Ra ... n/997.html

Regards
Günther
Interesting, since on the download page its rating (blitz-rapid) is given to be 2900. I assumed it was a derivative.

User avatar
Graham Banks
Posts: 33254
Joined: Sun Feb 26, 2006 9:52 am
Location: Auckland, NZ

Re: Murka 3.0 released

Post by Graham Banks » Wed Jul 17, 2013 7:54 pm

zullil wrote:
Günther Höhne wrote:Hi,

I think this engine is not a derivat. CEGT Blitz already tested precursor.
http://www.husvankempen.de/nunn/40_4_Ra ... n/997.html

Regards
Günther
Interesting, since on the download page its rating (blitz-rapid) is given to be 2900. I assumed it was a derivative.
Murka 2.0 was a new engine from Igor and was weaker than his previous engine Wildcat.
Not sure how strong Murka 3.0 is.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz

User avatar
velmarin
Posts: 1600
Joined: Mon Feb 21, 2011 8:48 am

Re: Murka 3.0 released

Post by velmarin » Wed Jul 17, 2013 8:00 pm

zullil wrote:
Interesting, since on the download page its rating (blitz-rapid) is given to be 2900. I assumed it was a derivative.
Engine has several years of development,
A magic like GREKO bitboard (author) Speed KNPS tremendous output.

Its author is relatively active in Russian forums.
The code is in sight, and looks original.

It's a great job.
:wink:

User avatar
reflectionofpower
Posts: 1456
Joined: Fri Mar 01, 2013 4:28 pm
Location: USA

Re: Murka 3.0 released

Post by reflectionofpower » Thu Jul 18, 2013 2:19 pm

velmarin wrote:
zullil wrote:
Engine has several years of development,
A magic like GREKO bitboard (author) Speed KNPS tremendous output.

It's a great job.
:wink:
I would like to see this speed demon with SMP.

User avatar
velmarin
Posts: 1600
Joined: Mon Feb 21, 2011 8:48 am

Re: Murka 3.0 released

Post by velmarin » Thu Jul 18, 2013 2:45 pm

reflectionofpower wrote:
velmarin wrote:
zullil wrote:
Engine has several years of development,
A magic like GREKO bitboard (author) Speed KNPS tremendous output.

It's a great job.
:wink:
I would like to see this speed demon with SMP.
Many authors prefer to develop first in one core and then enter in the SMP.

Perhaps it would not be wrong to develop the SMP with few algorithms, the possibility of less serious problems.
¿Maybe? :?:
is an idea.

As the speed of Formula One, means either too, just as counting,
but it is nice. rum, rum¡¡¡¡

User avatar
Werner
Posts: 2428
Joined: Wed Mar 08, 2006 9:09 pm

Re: Murka 3.0 released

Post by Werner » Fri Jul 19, 2013 6:29 am

Hi,
can anybody please translate the file description text for me.
Google does not - it says it is Vietnam language ??

best wishes
Werner
Werner

User avatar
Ajedrecista
Posts: 1402
Joined: Wed Jul 13, 2011 7:04 pm
Location: Madrid, Spain.
Contact:

Approximated translation of 'description' file.

Post by Ajedrecista » Fri Jul 19, 2013 12:15 pm

Hello Werner:
Werner wrote:Hi,
can anybody please translate the file description text for me.
Google does not - it says it is Vietnam language ??

best wishes
Werner
I used my imagination for make a plausible, decent translation. I started with the word 'PROTOCOLS', easily recognizable just before UCI and XBoard; then I replaced characters that matched with the corresponding cyrillic characters (thanks to Google Translator). Later I tried to complete words by the context and replaced hanging characters with more Cyrillic characters... it was a very long task, at least Notepad replaces character very fast. :)

Finally I cooked the translation into something more reader-friendly (at least I hope it). After some hours of spare time (I really did not have better to do):

Code: Select all

Murka 3



ПаРамеТРЫ КОмаНдНОй сТРОКи
--------------------------

-h N (размер табЛицы перестаНовок в мегабайтах, по умоЛчаНие 128)
-l (вкЛючить Лог)


ПРОТОКОЛЫ
---------

- UCI
- XBoard


сТРуКТуРЫ
---------

- магические битборды


геНеРаТОРЫ хОдОв
----------------

- все ходы
- взятия и превращеНия в ферзя
- защиты от шаха
- тихие ходы
- шахи (с маской дЛя искЛечеНия уже сгеНерироваННых взятий)


ПеРебОР
-------

- fail soft PVS
- сортировка: хеш, хорошие взятия (SEE>=0) по MVV/LVA, 2 киЛЛера, история, пЛохие взятия
- табЛица перестаНовок (стратегия простой замеНы)
- пустой ход
- продЛеНие шахов, ответНых хороших взятий в ПЛ
- коНтроЛь повтореНия позиций
- иНкремеНтаЛьНая геНерация ходов
- отсечеНия по дистаНции до мата
- особо обрабатывется фв под шахом
- отсечеНия На маЛой гЛубиНе (в сЛишком хороших и в сЛишком пЛохих позициях)
- сокращеНие посЛедНих (по сортировке) ходов с верификацией
- поиск первого хода перебором На меНьшуе гЛубиНу в ПЛ (IID)


РасПРедеЛеНие вРемеНи
---------------------

- очевидНый ход
- есЛи Не быЛо существеННого падеНия оцеНки На посЛедНей итерации
- есЛи Лучший ход Не меНяЛся
- есЛи Лучший ход Не помеНяЛся На посЛедНей итерации


фв
__

- рассматриваем тоЛько поЛезНые взятия и превращеНия в ферзя
- шахи На входе в фв
- аЛьфа-бета отсечеНия
- сортировка MVV/LVA


Оф
--

- материаЛ с коррекцией
- цеННость поЛей
- НичейНость материаЛьНого соотНошеНия
- Ладьи На открытых/поЛуоткрытых ЛиНиях
- мобиЛьНость
- атака На короЛя
- пешечНый щит
- возможНость рокировки
- открытые/закрытые пешки
- изоЛироваННые пешки
- отстаЛые пешки
- проходНые пешки
- Нет своих На пути проходНой
- Нет чужих На пути проходНой
- На пути проходНой Нет поЛей атаковаННых тоЛько соперНиком
- дистаНция от проходНой до короЛей
- каНдидаты в проходНые
- поймаННый сЛоН
- запертый сЛоН
- запертая Ладья
- разНоцветНый сЛоНовый эНдшпиЛь
- очередНость хода


КОНТРОЛи
--------

- гЛубиНа
- время На ход
- время На НескоЛько ходов
- время На партие
- иНкремеНт
- бескоНечНый аНаЛиз
- позиций На ход (дЛя теНиНга)
There must be some character slips in this Russian text (I do not know Russian) and there is also a mixture of uppercase and lowercase, but the approximated translation should be something similar to the following text:

Code: Select all

Murka 3



Command-line options
--------------------

-h N (the size of the transposition table in megabites, by default 128)
-l (include Log)


PROTOCOLS
---------

- UCI
- XBoard


PATTERN
-------

- Magic bitbords


MOVE GENERATION
---------------

- All the moves
- Capture a queen and transformation
- Protection from checks
- Quiet moves
- Checks (with a mask to exclude the already generated combined)


BUSTING
-------

- Fail soft PVS
- Sort: hash, taking good (SEE> = 0) in MVV / LVA, two killers, history, taking Poor
- The transposition table (simple replacement strategy)
- Empty the course
- Extension of checks, the response takes good in PL
- Recurrence control positions
- The generation of incremental moves
- Cut-off by the distance to the mat
- Especially iterated PV in check
- Cut-off at a shallow depth (in too good and too bad positions)
- Reduction of the latter (for sorting) moves with verification
- Brute force search of the first move on meNshue depths in the PL (IID)


DISTRIBUTION OF TIME
--------------------

- Evidence of progress
- If there has been a significant drop in evaluation of the last iteration
- If the best move does not change the
- If you have not changed the best move at the last iteration


PV
--

- Consider only useful promoting into a queen
- Checks are at the entrance to the PV
- Alpha-beta pruning
- Sort MVV / LVA


Of
--

- Material with correction
- The value of squares
- No Man's material relation
- Rooks on open / semi-open lines
- Mobility
- The attack on the King
- Pawn shield
- Castling ability
- Open / closed pawn
- Isolated pawns
- Backward pawns
- Passed pawns
- No. On the way are giving their
- No other people on the road are giving
- On the way there are no squares are giving only attack the opponent
- The distance from the walk-through to King
- Candidates in the pass
- Catch the Bishops
- Trapped Bishops
- Trapped Rook
- Multi-colored bishop ending
- The next move


CONTROLS
--------

- Depth
- Time for move
- Time for a few moves
- Time of the game
- Increments
- Infinite analysis
- Positions on the course (for a Teningen)
There are sentences that are not properly translated (you will note it by weird sentences). But the general idea is clear enough IMHO. Good luck with the testing!

Regards from Spain.

Ajedrecista.

Post Reply