Chess Forum

Posted: **Tue Jun 11, 2024 7:37 pm**

UPDATE
2nd League tour results are published, for more details, standings:
https://sites.google.com/site/computers ... -cs-51-2nd

Current Leader: SF-POLY - Congratulations to Tanick Ramz, SF team!

On the other hand,
What is going on ? 14 new entries..just only two managed to qualify:
Stockfish 16.1 and SF-POLY 020624 !!

Some Notes:
Super League is planing later (with the qualified players), but it seems it will
Take some long time..since we have only 2 qualified players for the moment...
To be more clear, in Second League: latest Top engines of 3770+ are played
Vs same/constant opponents, which are minus plus 30-40 Elo weaker..reason,
To reduce the overall draw ratio! You may know, Top engines play as serious
Role in case of appearing high draws! and via current testing way, our tours are
More exciting..otherwise if via more optimal test conditions (as MP, Ponder ON,
Even Blitz etc) then close to 99% draws for sure (I mean too boring for my side)
For right now, the draw ratio is 86%, yes more blunder moves, but if nothing else
More FUN on scene, right? Meanwhile, once more we noticed that unfortunately
There is no much Elo improvement since several months... at least current CS
Confirms this to all of us besides in other CS: SF16.1 performed as best Results
https://sites.google.com/site/computers ... nnue-cs-50
So here I think that it's not as bad 'luck' about that latest SF dev. is again weaker..
Who knows? maybe via weak openings: we may see some Elo improvements?
Or perhaps when HT + LP is enabled and with many concurrent games....or ?)
Btw, just a few more words about the qualification rule (3775+ Elo is required):
In short, any engine better (stronger) than Stockfish 16.1 will be into planning
Super League..if the barrier is less..then no point to organize Super League...

Greetings

Posted: **Thu Jun 13, 2024 1:25 pm**

Breaking News (13.06.2024):
Another BIG update for today: 11 entries but what a pity no one managed..sad!
But anyhow, I have good news as well:
Any high-rated player with 3774 Elo..will be in 'Play Off' test...to be more clear:
In Play Off: 3774 Elo one/s will be newly tested and as only difference in cond:
Unique openings will be used and any Play Off engine, which will be performed
Better points than SF16.1 will be qualified into planning Super League, and via
Play Off: it's expecting to see more qualified players but I admit too that current
CS is one of most hardest that ever seen...at least I can not remember...perhaps
Current CS is easy for programmers but just the used SCCT cond. does not suit
To almost all of entries ?! Who knows ?) or perhaps real SF team plus its based..
Are not so active on working over newer, more various test conditions?any ideas
Opinions over these strange issues are welcomed of course..thanks in advance
If still not so clear...newer vers. should be better/stronger (not weaker ) right ?!)
Btw, soon as possible, C-Dragon 3.0 (as 1st runner-up) will play off...good luck!

The Current Leader: SF-POLY - Congratulations to Tanick Ramz, SF team!
Note: SF-POLY performed 0.5 point better than SF16.1, so current Leader..

For Latest Standings:
https://sites.google.com/site/computers ... -cs-51-2nd

Greetings )

Posted: **Fri Jun 14, 2024 9:16 pm**

Play Off-Update

My Congratulations to Tanick Ramz again!
Via Play Off: C-Dragon in Super League!!

Note: In Play Off match, same cond. are used to play,
Exception, only as difference: Unique Openings used...

And here are the Play Off's Results:

Posted: **Fri Jun 14, 2024 9:36 pm**

UPDATE 2

SF-PRO 140624 (by APOCALYPSE) is the latest new strong entry...
Good luck!

Code: Select all

Rank Name                          Elo     +/-   Games   Score    Draw 
   0 SF-PRO 140624                  39      20     134   55.6%   87.3% 
   1 Fisherov M 1.2                 29      55      12   54.2%   91.7% 
   2 CFishNN 240223                  0       0      10   50.0%  100.0% 
   3 Berserk 13                      0       0      13   50.0%  100.0% 
   4 AI 28.0                         0       0      12   50.0%  100.0% 
   5 Fat Titz 2                    -29      55      12   45.8%   91.7% 
   6 Raubfish X48c3                -58      77      12   41.7%   83.3% 
   7 CatroP 060523                 -58      77      12   41.7%   83.3% 
   8 BlueMarlin 14.4               -58      77      12   41.7%   83.3% 
   9 ShashChess 32                 -64      84      11   40.9%   81.8% 
  10 Kayra 1.0                     -64      84      11   40.9%   81.8% 
  11 Obsidian 12.29                -89     116       8   37.5%   75.0% 
  12 K.Dragon 3.3                 -120     127       9   33.3%   66.7% 

134 of 600 games finished.

Posted: **Sat Jun 15, 2024 1:51 pm**

Breaking News (15.06.2024):
Current CS is already over, thanks and my congratulations to all book players!
Via final update: many books entered, where almost all of them are not weak..
Exc. I can't say as same words for LCX11-4P book but frankly no BIG problem
Otherwise, if all...will be in same quality/strength then will be hard to determine
Which ones are as best, strongest, right?) Btw, actually if we check more closely
We'll notice also that Top standings are not changed so much and that means
DON, HCAN, Unicornio, Adrian Oliva, ORIONPAX etc ones are super strong,
Born to Win for sure!

Private Book Champion: DON 190524 - Congratulations to Vasid Chouhan!
Public Book Champion Adrian Oliva - Congratulations to Angel Morano too!

For FINAL Standings:
https://sites.google.com/site/computers ... nnue-cs-30

GAMES:
https://mega.nz/file/v1wVVLTB#p4IVRdFIO ... nrPbpFsKOo

Best,
Sedat

Posted: **Sat Jun 15, 2024 5:48 pm**

Breaking News (15.06.2024):
Another new update.. this time only 2 entries and unfortunately no success..
But the good news is that, I've already started Super League tour...sure with
Small number of participants, with only 3, but sometimes not all in my hands!
Be aware too latest Killfish 150624 plus RapTora 3.0 are next new entries..so
After these tests...the current CS Leagues are going to be ended..as reason,
The used conditions + rules do not suit much to the latest Top chess engines,
Not sure exactly..but my 6th sense says that Balsa openings play as BIG role!
I assume that via Unique openings: we'd expect many more qualified players...
In other words, it seems that Balsa openings are too strong...that means also
No much Elo progress with strong opening lines...where Unique suite contains
More weaker and critical lines, sure it's strong but not so much as Balsa suite!
And please stay tuned soon as possible, I'll share Super League's Results too

The Current Leader: SF-POLY - Congratulations to Tanick Ramz, SF team!

For Latest Standings of NNUE CS LI - 2nd League:
https://sites.google.com/site/computers ... -cs-51-2nd

Greetings )

Posted: **Mon Jun 17, 2024 5:55 am**

UPDATE

Congrats to Anton Ces! via Play Off (via Unique openings):
Killfish 150624 engine is managed to be in Super League!
Where unfortunately latest RaspTora 3.0 could not qualify..
But no surprise, since RapTora played with old-dated net..

More Details:
In official 2nd League (via Balsa): Killfish is scored 3774 Elo (57%),
But via Play OFF: Killfish performed with win. percentage: 61.8%!

So once more we noticed that in case of via Unique openings:
The latest Top engines have more chances to gain better points!
Where via Balsa, less chances...as reason Balsa has less holes..
In same that means Weaker engines are safer via Stronger openings!
There is no other explanation of that, right ?)

Ok.. and here are the latest Play Off results by Killfish:

Code: Select all

Rank Name                          Elo     +/-   Games   Score    Draw 
   0 Killfish 150624                83      14     600   61.8%   71.8% 
   1 Fisherov M 1.2                -35      35      50   45.0%   86.0% 
   2 ShashChess 32                 -42      47      50   44.0%   76.0% 
   3 Fat Titz 2                    -49      44      50   43.0%   78.0% 
   4 AI 28.0                       -70      40      50   40.0%   80.0% 
   5 Obsidian 12.29                -92      50      50   37.0%   70.0% 
   6 Kayra 1.0                     -92      54      50   37.0%   66.0% 
   7 CatroP 060523                 -92      46      50   37.0%   74.0% 
   8 Raubfish X48c3               -100      60      50   36.0%   60.0% 
   9 Berserk 13                   -100      56      50   36.0%   64.0% 
  10 K.Dragon 3.3                 -108      49      50   35.0%   70.0% 
  11 CFishNN 240223               -108      49      50   35.0%   70.0% 
  12 BlueMarlin 14.4              -115      50      50   34.0%   68.0% 

600 of 600 games finished.

And very soon, very likely today,
Killfish will be participated in SL

GOOD LUCK )

Posted: **Mon Jun 17, 2024 9:41 am**

Breaking News (17.06.2024):
Both League CS are ended... congrats and many thanks to all programmers!
And as I stated earlier: via stronger opening lines (e.g via Balsa..) the weaker
Engines are safer, gaining better points for sure! at least current tour confirms..
And not only in this tour, I can share many other tours, where we'll see again..
But via Unique suite, which is not so strong as Balsa: then the weaker engines
Fall more often in trouble...not always of course.. but situation is more critical!
Note: As reference, for comparing... latest Play Off results can be checked too!
Btw, since many months SF team is more busy over weak opening lines, right ?
And what about eng improvements over strong opening theory? project dead?
Anyhow no problem, just if future tours 3770+ I hope to use weaker openings..
Be aware too, in a few hours, Final Super League's Results will be shared too

2nd League Champ: SF-POLY - Congratulations to Tanick Ramz, SF team!

For Final 2nd League Standings:
https://sites.google.com/site/computers ... -cs-51-2nd

Greetings

Posted: **Mon Jun 17, 2024 12:18 pm**

Breaking News (17.06.2023):
Both Leagues are ended..congratulations and thanks to all programmers!
And will be not so bad if Mr Tanick to take a bead worn against the evil eye !)
Yes, a man who's engines managed to be Number One of both hard Leagues!
On the other hand, as we see Killfish performed very good too..just 1 point
Behind C-Dragon and frankly in my eyes, both, actually even SF-POLY too
Are almost identical in strength and not via papers of course but via facts!
Ok friends that's all for now. here is summer so I need to take a good rest..
You may know, running similar competitions.. are not so easy as it looks!
However I hope to organize new tours..but not so frequent..see you later )

Super League Champ: C-Dragon - Congrats to Tanick Ramz, SF team!

Code: Select all

Rank Name              Elo   +    -  games score oppo. draws 
   1 C-Dragon v3.0    3797    5    5   600   51%  3792   93% 
   2 Killfish 150624  3796    5    5   600   51%  3792   96% 
   3 SF-POLY 020624   3795    5    5   600   50%  3793   94% 
   4 Stockfish 16.1   3785    5    5   600   48%  3796   92%

GAMES:
https://mega.nz/file/C4ZxBbAZ#OgzJ3fs0E ... z-vC3dgqLk

More Details:
https://sites.google.com/site/computers ... r-league_1

Greetings

Posted: **Sat Jun 22, 2024 9:46 am**

Hello there,

1st of all,
Very soon I hope to organize another new CS,
And again with players mainly 3770+ Elo, but
Before starting the planning official CS,
There will be candidate tours..as reason:
One engine (per author) will be allowed...
Otherwise (sure I refer for older tourneys),
Some of authors participate with 1 engine,
Some with 2...even some author/s participate
With many engines and usually the One/s with
Many... have more chances to be Number One!

In other words,
Via new rule (one engine per author),
The CS will be played with more fair cond.!

And here is 1st Candidate tour results (by OrgZ Engines),
Where only the Winner will be qualified into planning CS..

Note also that (in Candidate CS),
Minimum 500 games will be played (each participant)
I'm aware that not enough..may appear error margin,
Since the engines are playing with Unique openings...
But at least, in official CS: it's planning each player
To contain over than 1500+ games (per player) so best,
Strongest ones are expecting to be one of Top ranks!

By the way,
For the current test, Good Luck 'only' to Tanick Ramz

Code: Select all

Rank Name                          Elo     +/-   Games   Score    Draw 
   1 Private 19Sf                    7      14     258   51.0%   88.8% 
   2 C-Dragon v3.0                   3      16     256   50.4%   85.9% 
   3 SF-POLY 020624                  0      17     257   50.0%   84.4% 
   4 C-Dragon v3.1                  -1      13     259   49.8%   91.1% 
   5 Marauders 3.6                  -1      13     256   49.8%   90.2% 
   6 JigSaw v5.8                    -7      15     260   49.0%   88.1% 

773 of 1500 games finished.

Conditions:
2x Epyc 7B12, CuteChess, 1 Core, Ponder OFF, 30s+0.6s, Unique, 64 Hash, 4-MEN

Meanwhile, if you are going to ask:
Why two C-Dragon vers. are entered...? simply because
Both engine versions are used to play with different nets..,
You know even single influence we may expect diff. results

Best,
Sedat

Chess Forum

Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz

Re: Sedat Canbaz