SedatChess

[OrgZ]

Breaking News (02.11.2023):
Via latest 6 (six) engine entries...it's time to announce who are first Champions:
Ladies and Gentlemen, we have two Champs...interesting.. rarely happens two..
Yes.. their learning is so effective that non-learning engines are hopeless under
These cond..e.g all Top 5 ranked engines are able to learn... so picture is clear!
And very likely after playing more games, their Elo difference may be increased..
About current win. percentage stats, Whites: 55%, Blacks 45%..so quite normal..
Frankly stats are important..and not all of opening suites produce these values!

Champions: SunLight / Tactical - Congrats to E. Nemeth, A. Ces and SF team!

For Latest Standings:
https://sites.google.com/site/computers ... t-nn-cs-44

Greetings

were you running in gauntlet style or straight up round robin?

if gauntlet then Hazard didn't give u problems i take it?

and congratulations to the winners

Well, the tour is started with round-robin system
That means mostly games are played in round-robin,
Where the latest entries are played in gauntlet..

Btw, just in case of new entries, which are able to learn...
I plan to test in gauntlet but with concurrent games: 32
In this way, I expect the learning engines to gain more...

About latest Hazard version,
No any crash...all stable so far..
Same words are going for the rest engines too !

Greetings )

you would want to check your mail

Thanks a lot... and good luck )

[Sedat Canbaz]

Breaking News (03.11.2023):
Current championship ended..congratulations and many thanks to all authors!
Via latest entry..the standings are changed exc.for Tactical, Leading as before..
In same time managed to be win the current CS.. really amazing performance!
2nd rank belongs to two engines: DarkSister+SunLight, 3rd to new Raid version
Normally the plan was to work a little bit more over this tour...but I think that it's
Not required..because latest entries are starting as without any pre-learned exp..
Where constant Top players are already playing with own pre-learned exp files!
In other words, not so fair... e.g in Gauntlet: if the entry is learning engine then
Its learning is not so effective..due to matches are playing at high concurrency..
For solving all these issues (for more fair and better conditions) I have to run all
These CS in Round Robin systems...and organizing via two separate contests..
Learning / NON-Learning League, sure in NON-Learning tour, Gauntlet can be
Btw, as I stated before, Learning is very affective with weak positions and after
Serious number of games plus with small number of openings..so under these
Cond. Unique suite contains not so weak lines..plus close to 400 openings, so
Here Learning seems to be as advantage, but not so much...and who knows?
Maybe after 5.000 - 10.000 games (per player) the learning will play more role?
Anyhow, even about 1.000 games the learning helps, sure not much but helps!

XLIV's Champion: Tactical 281023 - Congratulations to Anton Ces, SF Team!

FOR FINAL STANDINGS:
https://sites.google.com/site/computers ... t-nn-cs-44

GAMES:
https://mega.nz/file/WwQyTDLa#45qwPQw9i ... P59jXJkMC4


Best,
Sedat

[OrgZ]

Breaking News (03.11.2023):
Current championship ended..congratulations and many thanks to all authors!
Via latest entry..the standings are changed exc.for Tactical, Leading as before..
In same time managed to be win the current CS.. really amazing performance!
2nd rank belongs to two engines: DarkSister+SunLight, 3rd to new Raid version
Normally the plan was to work a little bit more over this tour...but I think that it's
Not required..because latest entries are starting as without any pre-learned exp..
Where constant Top players are already playing with own pre-learned exp files!
In other words, not so fair... e.g in Gauntlet: if the entry is learning engine then
Its learning is not so effective..due to matches are playing at high concurrency..
For solving all these issues (for more fair and better conditions) I have to run all
These CS in Round Robin systems...and organizing via two separate contests..
Learning / NON-Learning League, sure in NON-Learning tour, Gauntlet can be
Btw, as I stated before, Learning is very affective with weak positions and after
Serious number of games plus with small number of openings..so under these
Cond. Unique suite contains not so weak lines..plus close to 400 openings, so
Here Learning seems to be as advantage, but not so much...and who knows?
Maybe after 5.000 - 10.000 games (per player) the learning will play more role?
Anyhow, even about 1.000 games the learning helps, sure not much but helps!

XLIV's Champion: Tactical 281023 - Congratulations to Anton Ces, SF Team!

FOR FINAL STANDINGS:
https://sites.google.com/site/computers ... t-nn-cs-44

GAMES:
https://mega.nz/file/WwQyTDLa#45qwPQw9i ... P59jXJkMC4


Best,
Sedat

Mr Sedat Thank you, for the wonderful tour and your time.
Congratulations to Anton and Edward
I see no complaints about New Raid, means its fixed hey. Super . Hard luck for him though my boy joined latebut am impressed either way. Its a rapid rise.
Thank you again Mr Sedat.
As for the current leaders good job guys, next tym.

[Sedat Canbaz]

Breaking News (03.11.2023):
Current championship ended..congratulations and many thanks to all authors!
Via latest entry..the standings are changed exc.for Tactical, Leading as before..
In same time managed to be win the current CS.. really amazing performance!
2nd rank belongs to two engines: DarkSister+SunLight, 3rd to new Raid version
Normally the plan was to work a little bit more over this tour...but I think that it's
Not required..because latest entries are starting as without any pre-learned exp..
Where constant Top players are already playing with own pre-learned exp files!
In other words, not so fair... e.g in Gauntlet: if the entry is learning engine then
Its learning is not so effective..due to matches are playing at high concurrency..
For solving all these issues (for more fair and better conditions) I have to run all
These CS in Round Robin systems...and organizing via two separate contests..
Learning / NON-Learning League, sure in NON-Learning tour, Gauntlet can be
Btw, as I stated before, Learning is very affective with weak positions and after
Serious number of games plus with small number of openings..so under these
Cond. Unique suite contains not so weak lines..plus close to 400 openings, so
Here Learning seems to be as advantage, but not so much...and who knows?
Maybe after 5.000 - 10.000 games (per player) the learning will play more role?
Anyhow, even about 1.000 games the learning helps, sure not much but helps!

XLIV's Champion: Tactical 281023 - Congratulations to Anton Ces, SF Team!

FOR FINAL STANDINGS:
https://sites.google.com/site/computers ... t-nn-cs-44

GAMES:
https://mega.nz/file/WwQyTDLa#45qwPQw9i ... P59jXJkMC4


Best,
Sedat

Mr Sedat Thank you, for the wonderful tour and your time.
Congratulations to Anton and Edward
I see no complaints about New Raid, means its fixed hey. Super . Hard luck for him though my boy joined latebut am impressed either way. Its a rapid rise.
Thank you again Mr Sedat.
As for the current leaders good job guys, next tym.

Dear Mr. Tanick,

You are welcome...

Yes, no complain ) because no crashing..

Actually my target is simply to help...
If not today Sedat...I am sure sooner or later
Other tester/s would 'complain' as you say ))

Btw, currently I run new strength testings of close to 3700+,
Which released during mainly 2021 year...e.g some of them
Are capable to learn...and no any crash via 64 concurrency...

So as conclusion...sometimes, even newer 2023 versions can be worse !
Yes... not only testers..even programmers work is not perfect too...
Besides anyone can make mistakes...but blaming testers just because
X program is crashing...this is one of worst thing in Computerchess..
A small note: I am not meaning you...but I assume you know who...

Actually here...
Many have to be happy and thankful... for all these testings...
Otherwise if sharing no info, games etc..then I am afraid to say that
We can not make much useful steps in Compterchess's progress..!!
In other words, we are all stronger together !)


Greetings )

[OrgZ]

Breaking News (03.11.2023):
Current championship ended..congratulations and many thanks to all authors!
Via latest entry..the standings are changed exc.for Tactical, Leading as before..
In same time managed to be win the current CS.. really amazing performance!
2nd rank belongs to two engines: DarkSister+SunLight, 3rd to new Raid version
Normally the plan was to work a little bit more over this tour...but I think that it's
Not required..because latest entries are starting as without any pre-learned exp..
Where constant Top players are already playing with own pre-learned exp files!
In other words, not so fair... e.g in Gauntlet: if the entry is learning engine then
Its learning is not so effective..due to matches are playing at high concurrency..
For solving all these issues (for more fair and better conditions) I have to run all
These CS in Round Robin systems...and organizing via two separate contests..
Learning / NON-Learning League, sure in NON-Learning tour, Gauntlet can be
Btw, as I stated before, Learning is very affective with weak positions and after
Serious number of games plus with small number of openings..so under these
Cond. Unique suite contains not so weak lines..plus close to 400 openings, so
Here Learning seems to be as advantage, but not so much...and who knows?
Maybe after 5.000 - 10.000 games (per player) the learning will play more role?
Anyhow, even about 1.000 games the learning helps, sure not much but helps!

XLIV's Champion: Tactical 281023 - Congratulations to Anton Ces, SF Team!

FOR FINAL STANDINGS:
https://sites.google.com/site/computers ... t-nn-cs-44

GAMES:
https://mega.nz/file/WwQyTDLa#45qwPQw9i ... P59jXJkMC4


Best,
Sedat

Mr Sedat Thank you, for the wonderful tour and your time.
Congratulations to Anton and Edward
I see no complaints about New Raid, means its fixed hey. Super . Hard luck for him though my boy joined latebut am impressed either way. Its a rapid rise.
Thank you again Mr Sedat.
As for the current leaders good job guys, next tym.

Dear Mr. Tanick,

You are welcome...

Yes, no complain ) because no crashing..

Actually my target is simply to help...
If not today Sedat...I am sure sooner or later
Other tester/s would 'complain' as you say ))

Btw, currently I run new strength testings of close to 3700+,
Which released during mainly 2021 year...e.g some of them
Are capable to learn...and no any crash via 64 concurrency...

So as conclusion...sometimes, even newer 2023 versions can be worse !
Yes... not only testers..even programmers work is not perfect too...
Besides anyone can make mistakes...but blaming testers just because
X program is crashing...this is one of worst thing in Computerchess..
A small note: I am not meaning you...but I assume you know who...

Actually here...
Many have to be happy and thankful... for all these testings...
Otherwise if sharing no info, games etc..then I am afraid to say that
We can not make much useful steps in Compterchess's progress..!!
In other words, we are all stronger together !)


Greetings )

No harm done Mr Sedat all is well, i just wanted to confirm if what i changed worked or not , Thats why i said seems theres no complaints (i see you interpreted it in another way).
you see, you helped me see something that i couldn't see . i.e giving my engines stability

Lets be happy, and Thanks again

[Sedat Canbaz]

Breaking News (03.11.2023):
Current championship ended..congratulations and many thanks to all authors!
Via latest entry..the standings are changed exc.for Tactical, Leading as before..
In same time managed to be win the current CS.. really amazing performance!
2nd rank belongs to two engines: DarkSister+SunLight, 3rd to new Raid version
Normally the plan was to work a little bit more over this tour...but I think that it's
Not required..because latest entries are starting as without any pre-learned exp..
Where constant Top players are already playing with own pre-learned exp files!
In other words, not so fair... e.g in Gauntlet: if the entry is learning engine then
Its learning is not so effective..due to matches are playing at high concurrency..
For solving all these issues (for more fair and better conditions) I have to run all
These CS in Round Robin systems...and organizing via two separate contests..
Learning / NON-Learning League, sure in NON-Learning tour, Gauntlet can be
Btw, as I stated before, Learning is very affective with weak positions and after
Serious number of games plus with small number of openings..so under these
Cond. Unique suite contains not so weak lines..plus close to 400 openings, so
Here Learning seems to be as advantage, but not so much...and who knows?
Maybe after 5.000 - 10.000 games (per player) the learning will play more role?
Anyhow, even about 1.000 games the learning helps, sure not much but helps!

XLIV's Champion: Tactical 281023 - Congratulations to Anton Ces, SF Team!

FOR FINAL STANDINGS:
https://sites.google.com/site/computers ... t-nn-cs-44

GAMES:
https://mega.nz/file/WwQyTDLa#45qwPQw9i ... P59jXJkMC4


Best,
Sedat

Mr Sedat Thank you, for the wonderful tour and your time.
Congratulations to Anton and Edward
I see no complaints about New Raid, means its fixed hey. Super . Hard luck for him though my boy joined latebut am impressed either way. Its a rapid rise.
Thank you again Mr Sedat.
As for the current leaders good job guys, next tym.

Dear Mr. Tanick,

You are welcome...

Yes, no complain ) because no crashing..

Actually my target is simply to help...
If not today Sedat...I am sure sooner or later
Other tester/s would 'complain' as you say ))

Btw, currently I run new strength testings of close to 3700+,
Which released during mainly 2021 year...e.g some of them
Are capable to learn...and no any crash via 64 concurrency...

So as conclusion...sometimes, even newer 2023 versions can be worse !
Yes... not only testers..even programmers work is not perfect too...
Besides anyone can make mistakes...but blaming testers just because
X program is crashing...this is one of worst thing in Computerchess..
A small note: I am not meaning you...but I assume you know who...

Actually here...
Many have to be happy and thankful... for all these testings...
Otherwise if sharing no info, games etc..then I am afraid to say that
We can not make much useful steps in Compterchess's progress..!!
In other words, we are all stronger together !)


Greetings )

No harm done Mr Sedat all is well, i just wanted to confirm if what i changed worked or not , Thats why i said seems theres no complaints (i see you interpreted it in another way).
you see, you helped me see something that i couldn't see . i.e giving my engines stability

Lets be happy, and Thanks again

Yes my dear friend...do not worry...it is all right )

And if I cause a little bit on improving.. how happy I am

On other hand,
Stability is most important...
If I can't test or produce properly standings...I'm affecting...I assume many too
So as a tester...when there is a bug...my responsibility to inform all of you...
Sure... the computer systems are so various over world-wide...
So I give understanding to programmers...


Greetings

[Homayoun]

Many thanks. Is there any chance for Abbystein to participate in next tour updates?
Greetings

[Sedat Canbaz]

Many thanks. Is there any chance for Abbystein to participate in next tour updates?
Greetings

Not at all...unfortunately I do not have it.. sure just in case..
I may include it as well...

Greetings

[STUDENT_MASTER]

Breaking News (03.11.2023):
Current championship ended..congratulations and many thanks to all authors!
Via latest entry..the standings are changed exc.for Tactical, Leading as before..
In same time managed to be win the current CS.. really amazing performance!
2nd rank belongs to two engines: DarkSister+SunLight, 3rd to new Raid version
Normally the plan was to work a little bit more over this tour...but I think that it's
Not required..because latest entries are starting as without any pre-learned exp..
Where constant Top players are already playing with own pre-learned exp files!
In other words, not so fair... e.g in Gauntlet: if the entry is learning engine then
Its learning is not so effective..due to matches are playing at high concurrency..
For solving all these issues (for more fair and better conditions) I have to run all
These CS in Round Robin systems...and organizing via two separate contests..
Learning / NON-Learning League, sure in NON-Learning tour, Gauntlet can be
Btw, as I stated before, Learning is very affective with weak positions and after
Serious number of games plus with small number of openings..so under these
Cond. Unique suite contains not so weak lines..plus close to 400 openings, so
Here Learning seems to be as advantage, but not so much...and who knows?
Maybe after 5.000 - 10.000 games (per player) the learning will play more role?
Anyhow, even about 1.000 games the learning helps, sure not much but helps!

XLIV's Champion: Tactical 281023 - Congratulations to Anton Ces, SF Team!

FOR FINAL STANDINGS:
https://sites.google.com/site/computers ... t-nn-cs-44

GAMES:
https://mega.nz/file/WwQyTDLa#45qwPQw9i ... P59jXJkMC4


Best,
Sedat

more wonderful games to analyse and pick the best ones. Thank you sir Sedat . i see you also have Awol, i use that baby to analyse. the lines that he picks i like

greetings

[Sedat Canbaz]

Breaking News (03.11.2023):
Current championship ended..congratulations and many thanks to all authors!
Via latest entry..the standings are changed exc.for Tactical, Leading as before..
In same time managed to be win the current CS.. really amazing performance!
2nd rank belongs to two engines: DarkSister+SunLight, 3rd to new Raid version
Normally the plan was to work a little bit more over this tour...but I think that it's
Not required..because latest entries are starting as without any pre-learned exp..
Where constant Top players are already playing with own pre-learned exp files!
In other words, not so fair... e.g in Gauntlet: if the entry is learning engine then
Its learning is not so effective..due to matches are playing at high concurrency..
For solving all these issues (for more fair and better conditions) I have to run all
These CS in Round Robin systems...and organizing via two separate contests..
Learning / NON-Learning League, sure in NON-Learning tour, Gauntlet can be
Btw, as I stated before, Learning is very affective with weak positions and after
Serious number of games plus with small number of openings..so under these
Cond. Unique suite contains not so weak lines..plus close to 400 openings, so
Here Learning seems to be as advantage, but not so much...and who knows?
Maybe after 5.000 - 10.000 games (per player) the learning will play more role?
Anyhow, even about 1.000 games the learning helps, sure not much but helps!

XLIV's Champion: Tactical 281023 - Congratulations to Anton Ces, SF Team!

FOR FINAL STANDINGS:
https://sites.google.com/site/computers ... t-nn-cs-44

GAMES:
https://mega.nz/file/WwQyTDLa#45qwPQw9i ... P59jXJkMC4


Best,
Sedat

more wonderful games to analyse and pick the best ones. Thank you sir Sedat . i see you also have Awol, i use that baby to analyse. the lines that he picks i like

greetings

You are welcome...

Awol's start was very good..was leading engine,
But it seems was not so lucky in the last rounds

On other hand,
That's all right... no one can stay on Top forever.!

Greetings )