Testing LazySMP

[Uri Blass]

I see that carlos test all the weak engines not for CCRL and publish the results.

viewtopic.php?t=84433

I guess you can wait for his next tournament when LazySMP can start from the lowest league and promote(because the lowest league has players that are clearly below rating 1800).

[LazySMP]

4)I do not see how the stockfish community members relevant...

Initial messages deleted. see this message viewtopic.php?p=969466#p969466

1 hour after I created the topic, two members of Stockfish (lucametehau and noobpwnftw) started insulting and hating.

2 hour after I created the topic, lucametehau continued his disrespect and my topic was locked for 2 day....

Does this have anything to do with the Stockfish community?

[Sylwy]

Man, three versions are tested by me (at TC=4'+2") with opponents of the same class. Why don't you have patience? A tournament lasts. From time to time I will post intermediate results in the matches and tournaments section. You have no patience. That's the problem. Your engine is OK.



Opponents have the ratings (CCRL Blitz) below:

[supernova]

For the second time, you did not even mention and reply to my message. This is the short test with the last version. Results are similar to previous versions. Engine is around 2000-2100 CCRL.

Sorry, I decided to leave chess programming forever. The behavior of CCRL members was so disappointing that I see no point to continue. They did not test my engine for no reason and it's not their fault, it's the Stockfish community members who made them pessimistic and destroyed my discussion:

[/quote]

I regret to say that your constant obsession with the CCRL discourages others from trying your engine. Due to the illogical nature of this conversation, I must close it again.

Regards.

[LazySMP]

Dear Gabor, Dear Graham, Dear Basti,

As I’m sure you know, you will find testing in CCRL very different from testing in private GUI. For example, you cannot begin testing engines on day one as you did at PC. Engines that do not use a Transposition Table tend to be very fast and this is not a good reason not to test it.

In any event, TT is not necessarily an engine to greater efficiency. I humbly request you to tell me your reasons for not testing my engine. It is a widely believed myth that you accepted the words of the Stockfish community and that's why you don't want to test my engine.

In an age of artificial intelligence, weaker engines are desperately needed to fill the ranks. Complying with complex testing rules makes it slow and difficult to test an engine, and there is little social prestige to saying you work for the chess community.

Under these conditions, you cannot find your way to excellence. I hope you will realize soon, then you find why some people get frustrated.

Yours Sincerely,

Daniel Pierce

[Graham Banks]

I've already explained why.

Here is the current waiting list of engines waiting for 1CPU gauntlets for the 40/15 list:

Caissa 1.21 64-bit (+20)
Elixir 3.0 64-bit (3500?)
Velvet 8.1.1 64-bit (+25)
Lizard 11.1 64-bit
WhiteCore 0.2 64-bit
Sirius 8.0 64-bit (+140)
Quanticade Electra 0.9 64-bit (+90)
Calvin 5.0.2 64-bit (+80 over 4.3.0)
Priessnitz 2.0 64-bit (+80)
Arcanum 2.3.1 64-bit
Heimdall 1.1.1 64-bit (+30)
Cadie 2.0 64-bit
Jackychess 24.06 64-bit (+25)
Yakka 1.1 64-bit (+40)
CT800 1.46 64-bit
Lozza 3.0 64-bit
Tornado 9.0 64-bit (+200)
Aramis 1.4.0 64-bit
Schoenemann 0.3.0 64-bit (+40)
SSCG13 64-bit
Aurora 1.22.0 64-bit (2500)
Tcheran 4.0 64-bit
Fatalii 0.8.0 64-bit (+60)
Euwe 1.3.0 64-bit
Prophet 4.4 64-bit
Molybdenum 4.1 64-bit
Rengar 2.1.1 64-bit (+100)
Ynode 0144 64-bit
Zangdar 2.31 64-bit (+30)
KnightX 4.2 64-bit
Barbarossa 0.7.0 64-bit
Astra 4.0.1 64-bit
Knur 20240903 64-bit
ChessGM 0.7.0 64-bit
Sentinel 1.02 64-bit
ChessKISS 1.8d 64-bit
Camel 1.6.0 64-bit
Stellar 1.4.3 64-bit

Due to this, I currently don't give attention to weaker engines. Andreas is helping me to reduce this list.
As an aside, our lists are full of weaker engines anyway.

I can't speak for the Blitz testers though, but they've probably given you reasons as well.

When your engine is strong enough to include in my Amateur Series, I'll add it to the list. That is where the 2400 I mentioned came from.
Let's forget about hash size, although I can assure you that when all other engines in my tournaments are using 256mb, yours is likely to struggle.

The other concern I have is that you're using an obviously fake name.

[LazySMP]

Thank you for your response to my letter.

I can't speak for the Blitz testers though, but they've probably given you reasons as well.

No, they didn't give me any reason. About 2 months ago, Mr. Gabor said he would test my engine if I fix its bugs.

I did not refuse to test your engine. Only, just because you release an engine you cannot expect it will be tested immediately. But now, based on your conversation with Alex, I am certainly postponing the test until your engine can make legal moves.

A few days ago, Rebel confirmed that my engine bugs were fixed.

I can mention progress, no more crashes and time forfeits with version 5

Code: Select all

No. Name         Win Draw Loss Unf.  Score Games       %
--------------------------------------------------------
  1 lazysmp-5.0 +549 =179 -272   *0  638.5  1000   63.9%
  2 Drofa_1.0   +272 =179 -549   *0  361.5  1000   36.1% CCRL ELO 1988

So, according to Mr. Gabor he was supposed to test my engine but what happened he didn't?!

Let's forget about hash size, although I can assure you that when all other engines in my tournaments are using 256mb, yours is likely to struggle.

Even all other engines in the tournament are using Transposition Table, I prefer don't use it.

The other concern I have is that you're using an obviously fake name.

To be honest, I'm not the main author of this engine. This engine was developed by one of my friends at my request. He asked me not to mention his name. Anyway, I hope the author's name is not an excuse for not testing engine.

[RubiChess]

A troll is a troll is a troll.

[mar]

A troll is a troll is a troll.

his sole goal seems to be stalking Graham

a troll with false identity, spammer and stalker - how is this not against the charter?

[LazySMP]

A troll is a troll is a troll.

As I said earlier, Stockfish members (lucametehau, noobpwnftw, RubiChess, shawn, Viz, ...) try to spoil/ruin my discussion so that my engine wouldn't be tested. It seems HGM was right:

It is in no way relevant for what happens here, and I already know that Stockfish community is populated by a bunch of savages.