Mediocre v0.4 released

[Sylwy]

Use Mediocre.bat.

Hi !

Thank you a lot for this new Mediocre !

An .exe compile for Fritzy GUI ? Maybe Mr.Jim Ablett ?

Regards,
SilvianR

[Zlaire]

An .exe compile for Fritzy GUI ? Maybe Mr.Jim Ablett ?

Yepp, awaiting his compile rather than doing a subpar version myself.

His compiles has shown significant improvements in strength in the passed as well.

[mar]

Hi Jonatan.
Good to see that Mediocre continues. I have noticed that you use repetition hash table.
IMHO that is not a good idea. Tried something like that a while ago and it
only caused very rare problems as sometimes you can overwrite an existing repetition already in the table.

Martin

[Engin]

hi jonathan,

right now at the moment is Mediocre 0.4 playing by openwar, i am follow everyday this tournament with TLCV viewer, still seem win against zeta.

you make a right decision with support only UCI and disable any xboard protocol support, that i made with my older engine EnginMax too, because i had many problems under winboard, escpecially with pondering, then i removed all winboard commands from the source, and i tried to fix some of bugs in EnginMax too, the result is now much better, but not what i expected, its still have bugs of course.

UCI is much easier and give you not so big problems with the protocol implementation, especially with ponder.

The UCI protocol is the future, more and more engine are prefer UCI then old winboard/xboard protocol, and i hope all older winboarders jump to UCI in the near future too.

And even i wish for the Winboard GUI thats support 1:1 UCI protocol in the GUI it self instead of using the polyglot adapter.

I am using Winboard only for ICS, because its the best and no GUI can beat the stability and functions of winboard on ICS mode.

But i prefer other GUI thats support UCI directly like Shredder Classic, Chess GUI, or Fritz GUI

i think nobody want make to many effort to configurate engines under winboard, even expert having problems to config engines under winboard, uci is much easier to install and the option can be change und the GUI directly instead of to search any INI file.

[Sven]

The UCI protocol is the future, more and more engine are prefer UCI then old winboard/xboard protocol, and i hope all older winboarders jump to UCI in the near future too.

Hi Engin,

please have a look at the RWBC list, ordered by the used protocol. Last changes are dated about one year ago but in general that list is known to be very accurate. There I count 350 WB engines, 65 supporting both WB and UCI, 109 UCI only (counting some "UCI (WB)" entries for UCI) and one with a different protocol (TMCI). Did that change significantly in the past 12 months? I don't think so. Of all 44 engines listed there with a "first release" date in the year 2009, 24 are WB. For 2010 there were 10 out of 22 supporting WB. From the "top 50" engines listed there, 16 support WB, 21 UCI and 13 both, which is something like "45% WB" in the "top 50" area. So I don't believe that many programmers see it the same way as you do. I think there are simply two different technologies and both have their advantages and disadvantages. Some prefer one, some the other. You prefer UCI, I prefer WB. You want me to convert, I won't

i think nobody want make to many effort to configurate engines under winboard, even expert having problems to config engines under winboard, uci is much easier to install and the option can be change und the GUI directly instead of to search any INI file.

Which difficulties in installation of WinBoard do you have, and which options need to be configured via "winboard.ini" and are not configurable via GUI? Have you asked HGM about these issues?

Sven

[Engin]

hi Sven,
i said UCI is the future, and i hope that more and more will be switch to UCI

yes, i prefer the UCI now, and dont like investing to much time for install winboard engines and change any options in the INI files separate.

Install an UCI engine and change the options directly in the GUI is much easier then WB engines, for WB engines you must be near an expert to let they runable and be sure they have the right setting, i mean hash, tablebase path and such important things , that is fact.

most WB engines are back in the early time where UCI was not so well known, i think todays if an beginner want write a new engine, i nearly sure he will prefer UCI then WB.

[Sylwy]

1.you make a right decision with support only UCI and disable any xboard protocol support, that i made with my older engine EnginMax

2.The UCI protocol is the future, more and more engine are prefer UCI then old winboard/xboard protocol, and i hope all older winboarders jump to UCI in the near future too.



3.i think nobody want make to many effort to configurate engines under winboard, even expert having problems to config engines under winboard, uci is much easier to install and the option can be change und the GUI directly instead of to search any INI file.

WOW !

1.Thats's because I renounced to use EngineMax !
2.Only in your mind ! Take a look how many new chess engines are under WB protocol !!!!!
3.INI files ? A true apotheosis tunning - in a correct way- a WB engine. Take a look to Chessmaster , Rebel , Naum , Frenzee , Gandalf, Gaviota ,Scorpio , The Baron , Diep , Chiron...................................

No effort = laziness >>>> this is an axiom

Your theory seems to be the contrary:



Night.INI file regards,
SilvianR

NR: Just an example:

this young boy from my IASI (Stefan Atirgovitoae-11 years old) just won this year "Ti lascio una canzone"/Italy:



and in 2010 San Remo Junior/Italy. A lot of work !

[mar]

Sorry stupid me i didn't notice you cycle up through entries . However consider the case where a collision would occur: assume A and B map to two consecutive entries. C maps to same index as either A or B. You then remove B. C won't be found as a rep.

[Elorejano]

Not only lazyness, its the GUI comanding our lifes
LIVE WB/XB protocol!!!!!!!!!!!!!!!

[Zlaire]

Sorry stupid me i didn't notice you cycle up through entries . However consider the case where a collision would occur: assume A and B map to two consecutive entries. C maps to same index as either A or B. You then remove B. C won't be found as a rep.

Bit late so might not be thinking clearly , but how would B be removed before C?