RomiChess vs. Oli's 5.29 & 5.30

[Michael Sherwin]

RomiChess96 - Olithink64529 : 3.0/3 3-0-0 (111) 100% +1200
RomiChess96 - Olithink64530 : 2.0/3 2-1-0 (101) 67% +123

I know that it is early, but, I am very excited about a new idea that I am trying! And Oliver did ask for an update between our programs. I hope that it will be interesting.

[OliverBr]

More results to come?

[Michael Sherwin]

Yes, more results to follow.

However, I had a tree explosion bug that showed up in quite a few positions that was causing way too many losses.

That seems to be fixed now and I have restarted the match.

p.s. I have a new beta tester now that is currently testing P3k to establish a baseline. When that is done I do have a beta to send to him and if it passes beta testing there will be a new release version.

[Michael Sherwin]

More results to come?

RomiChess96 - Olithink64529 : 5.0/10 3-3-4 (1010====10) 50% ±0
RomiChess96 - Olithink64530 : 8.0/10 6-0-4 (==11=11=11) 80% +241

These results really have me stumped. Must just be randomness because of not enough games.

Still:

Oliver Reports;

olithink530 - crafty20.14 : 460.0/1000 374-172-454 46%

So, Romi's 80% is quite unexpected!

More results to come!

[OliverBr]

My experience says, 10 is by far too few games. That's why it's 1000 games against Crafty 20.14.

[Michael Sherwin]

I don't know what I am doing anymore. I lost the last tournament results. Fighting off a head cold. Dealing with my mom's Alzheimer's. It is pandemonium around here.

New test is not bad, but seems sub par for Romi at this stage.

RomiChess96 - Olithink64529 : 22.5/39 21-15-3 (0110=111001010101010=01111001101010111=) 58% +56
RomiChess96 - Olithink64530 : 19.5/39 16-16-7 (=0111101==10101=01=000000=01100101=0111) 50% ±0

I need to add Oli 5.22 to see if progress still shows.

At least Oli 5.29 is not destroying Romi anymore.

[michiguel]

I don't know what I am doing anymore. I lost the last tournament results. Fighting off a head cold. Dealing with my mom's Alzheimer's. It is pandemonium around here.

New test is not bad, but seems sub par for Romi at this stage.

RomiChess96 - Olithink64529 : 22.5/39 21-15-3 (0110=111001010101010=01111001101010111=) 58% +56
RomiChess96 - Olithink64530 : 19.5/39 16-16-7 (=0111101==10101=01=000000=01100101=0111) 50% ±0

I need to add Oli 5.22 to see if progress still shows.

At least Oli 5.29 is not destroying Romi anymore.

Michael,

I think you are doing way too few games. Also, testing against one engine only might not be good. Include as many as you can that are of similar (+/- 100 elo) strength (Hint: Gaviota is very close to Romi ).

Miguel
PS: I hope you situation with your mother is as less rough as possible. I lived with my grandmother and her sister (grand aunt?) in the 80's. Both of them went into a senile condition within year. I understand you pretty well because it is a very sad situation.

[Michael Sherwin]

I don't know what I am doing anymore. I lost the last tournament results. Fighting off a head cold. Dealing with my mom's Alzheimer's. It is pandemonium around here.

New test is not bad, but seems sub par for Romi at this stage.

RomiChess96 - Olithink64529 : 22.5/39 21-15-3 (0110=111001010101010=01111001101010111=) 58% +56
RomiChess96 - Olithink64530 : 19.5/39 16-16-7 (=0111101==10101=01=000000=01100101=0111) 50% ±0

I need to add Oli 5.22 to see if progress still shows.

At least Oli 5.29 is not destroying Romi anymore.

Michael,

I think you are doing way too few games. Also, testing against one engine only might not be good. Include as many as you can that are of similar (+/- 100 elo) strength (Hint: Gaviota is very close to Romi ).

Miguel
PS: I hope you situation with your mother is as less rough as possible. I lived with my grandmother and her sister (grand aunt?) in the 80's. Both of them went into a senile condition within year. I understand you pretty well because it is a very sad situation.

Thanks for the understanding!

I have already downloaded the latest version of Gaviota and I will add it to the tournament.

[Michael Sherwin]

Code: Select all

1: RomiChess96   92.0/160 ········································ 
2: Gaviota64     20.5/40  01001101010101==0110=0==01=0=0011=0111=1 
3: Olithink64530 20.0/40  =1000010==01010=10=111111=10011010=1000= 
4: Olithink64529 17.0/40  1001=000110101010101=10000110010101000== 
5: Olithinkwin32 10.5/40  ==00=001010=0=00=00001=00100=00000=100== 

Gaviota and Romi are very close!

Olithink is getting stronger. The good news for Romi is that she is no longer getting killed by the newer Olithink versions!

There are some more things that I would like to test so I am going to end this test here and turn this version over to Romi's beta tester.

[OliverBr]

Aren't Gaviota and RomiChess supposed to be much stronger?

OliThink still has the feature, that it's evalation is only mobility. There isn't any further stratetigal information!
If you consider the HUGE eval function of crafty 20.14, I am very happy that it is very close to it on fast blitz games already.

But, on the other site, whatever change on search, hashtables, eval immediately let it drop in strength. It's quite an unstable system...