Fat Fritz 2

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Locked
twobeer
Posts: 20
Joined: Sat Feb 22, 2020 5:43 pm
Full name: Leif Aronsson

Re: Fat Fritz 2

Post by twobeer » Sat Feb 13, 2021 3:53 pm

pohl4711 wrote:
Sat Feb 13, 2021 3:32 pm
No. Only the commercial version of FF2 has the strong FF2 net. The ratinglists (CEGT, CCRL and me (SPCC)) got this version for testing. The net, which can be downloaded on the FF2-github website is weaker. Of course. Otherwise nobody would buy FF2 anymore.
On what test have you confirmed it is "weaker"?... I seriously doubt they would produce a "false" additional net just to publish on github.. That would get them into more problems GPLv3 Wise if they try murky stuff like that.

Can you include test with this net for refernce? or is it just an assumption on your part that it is weaker and not equal...

Do you have both bin files? can you post the MD5 checksums? whould be good refrence with the checksum.. How do we even now that SPCC got the same net as CEGT, CCRL etc.. If those are "hand-picked" by mr. A.Silver..

Werewolf
Posts: 1351
Joined: Thu Sep 18, 2008 8:24 pm

Re: Fat Fritz 2

Post by Werewolf » Sat Feb 13, 2021 4:12 pm

Can someone comment on this:

At 31:30 Albert S clearly says Fat Fritz 2 is "genuinely stronger than what's out there right now, without a question and clearly so" And SF Dev is referenced saying Fat Fritz 2 is measurably stronger.

I'm just not seeing this at all on my tests, and I used SF Dev which was released the same day as FF2 came out.

https://en.chessbase.com/post/interview ... -fritz-2-0

I can only think of 4 possible explanations:

a) My test results are inaccurate (Test run ongoing, only 500 games so far, but they are in line with others.)
b) the test criteria is very specific on a large number of cores (I'm testing single-threaded on Cutechess for time's sake)
c) the video is old, perhaps around Christmas time, and doesn't allow for SF Dev improvement
d) a more sinister conclusion.

Can someone comment as I'd prefer to give the benefit of the doubt if possible, I just can't replicate this result.

twobeer
Posts: 20
Joined: Sat Feb 22, 2020 5:43 pm
Full name: Leif Aronsson

Re: Fat Fritz 2

Post by twobeer » Sat Feb 13, 2021 4:17 pm

Werewolf wrote:
Sat Feb 13, 2021 4:12 pm
Can someone comment as I'd prefer to give the benefit of the doubt if possible, I just can't replicate this result.
Option d Obviously ...

gaard
Posts: 336
Joined: Mon Jun 07, 2010 1:13 am
Location: Holland, Michigan
Full name: Martin Wyngaarden

Re: Fat Fritz 2

Post by gaard » Sat Feb 13, 2021 4:39 pm

Werewolf wrote:
Sat Feb 13, 2021 4:12 pm
Can someone comment on this:

At 31:30 Albert S clearly says Fat Fritz 2 is "genuinely stronger than what's out there right now, without a question and clearly so" And SF Dev is referenced saying Fat Fritz 2 is measurably stronger.

I'm just not seeing this at all on my tests, and I used SF Dev which was released the same day as FF2 came out.

https://en.chessbase.com/post/interview ... -fritz-2-0

I can only think of 4 possible explanations:

a) My test results are inaccurate (Test run ongoing, only 500 games so far, but they are in line with others.)
b) the test criteria is very specific on a large number of cores (I'm testing single-threaded on Cutechess for time's sake)
c) the video is old, perhaps around Christmas time, and doesn't allow for SF Dev improvement
d) a more sinister conclusion.

Can someone comment as I'd prefer to give the benefit of the doubt if possible, I just can't replicate this result.
http://talkchess.com/forum3/viewtopic.p ... 09#p882009

"genuinely stronger" is about as honest as him saying that starting from scratch with FF2 was torture. FF2 performed less than 3 Elo better against SF12 than a SF-dev version I tested against SF12 two months ago.

User avatar
Ozymandias
Posts: 1323
Joined: Sun Oct 25, 2009 12:30 am

Re: Fat Fritz 2

Post by Ozymandias » Sat Feb 13, 2021 5:07 pm

Werewolf wrote:
Sat Feb 13, 2021 4:12 pm
Can someone comment on this:

At 31:30 Albert S clearly says Fat Fritz 2 is "genuinely stronger than what's out there right now, without a question and clearly so" And SF Dev is referenced saying Fat Fritz 2 is measurably stronger.

I'm just not seeing this at all on my tests, and I used SF Dev which was released the same day as FF2 came out.

https://en.chessbase.com/post/interview ... -fritz-2-0

I can only think of 4 possible explanations:

a) My test results are inaccurate (Test run ongoing, only 500 games so far, but they are in line with others.)
b) the test criteria is very specific on a large number of cores (I'm testing single-threaded on Cutechess for time's sake)
c) the video is old, perhaps around Christmas time, and doesn't allow for SF Dev improvement
d) a more sinister conclusion.

Can someone comment as I'd prefer to give the benefit of the doubt if possible, I just can't replicate this result.
He's not providing any data to sustain his claim, that should give you a good idea as to where the wind is blowing.

nio
Posts: 6
Joined: Sun Jan 17, 2021 6:27 pm
Full name: Adi Fischet

Re: Fat Fritz 2

Post by nio » Sat Feb 13, 2021 5:10 pm

He also said evaluation is changed to old fat fritz style. I cant find that in the source.

User avatar
Ozymandias
Posts: 1323
Joined: Sun Oct 25, 2009 12:30 am

Re: Fat Fritz 2

Post by Ozymandias » Sat Feb 13, 2021 5:30 pm

He's just talking about the use of mcts/nn data rather than SF's nnue, but don't worry, he's been saying questionable things as far back as the original AlphaZero article he wrote for ChessBase:
A new paradigm

On December 5 the DeepMind group published a new paper at the site of Cornell University called "Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm", and the results were nothing short of staggering. AlphaZero had done more than just master the game, it had attained new heights in ways considered inconceivable. The test is in the pudding of course, so before going into some of the fascinating nitty-gritty details, let’s cut to the chase. It played a match against the latest and greatest version of Stockfish, and won by an incredible score of 64 : 36, and not only that, AlphaZero had zero losses (28 wins and 72 draws)!

Stockfish needs no introduction to ChessBase readers, but it's worth noting that the program was on a computer that was running nearly 900 times faster! Indeed, AlphaZero was calculating roughly 80 thousand positions per second, while Stockfish, running on a PC with 64 threads (likely a 32-core machine) was running at 70 million positions per second. To better understand how big a deficit that is, if another version of Stockfish were to run 900 times slower, this would be equivalent to roughly 8 moves less deep. How is this possible?
The amount of BS in these two paragraphs alone are enough for a whole thread, and he didn't even have any other stake in the matter other than having the article published.

pohl4711
Posts: 1544
Joined: Sat Sep 03, 2011 5:25 am
Location: Berlin, Germany
Contact:

Re: Fat Fritz 2

Post by pohl4711 » Sat Feb 13, 2021 6:00 pm

twobeer wrote:
Sat Feb 13, 2021 3:53 pm
pohl4711 wrote:
Sat Feb 13, 2021 3:32 pm
No. Only the commercial version of FF2 has the strong FF2 net. The ratinglists (CEGT, CCRL and me (SPCC)) got this version for testing. The net, which can be downloaded on the FF2-github website is weaker. Of course. Otherwise nobody would buy FF2 anymore.
On what test have you confirmed it is "weaker"?... I seriously doubt they would produce a "false" additional net just to publish on github.. That would get them into more problems GPLv3 Wise if they try murky stuff like that.

Can you include test with this net for refernce? or is it just an assumption on your part that it is weaker and not equal...

Do you have both bin files? can you post the MD5 checksums? whould be good refrence with the checksum.. How do we even now that SPCC got the same net as CEGT, CCRL etc.. If those are "hand-picked" by mr. A.Silver..
A.Silver said, that the github net is weaker. And it is obvious, that this must be true. If not, nobody needed to buy FF2, so it would be crazy, if Chessbase would provide the strong commercial net on github for free.
I have both nets,,of course, the commercial net and the github net. They are definitly different. As I said before, the bench is different, if FF2 uses one or the other net. And I looked into them with a HEX Editor. The nets are different.
A 3000 games testrun is running on one of my machines (a bullet RoundRobin with FF2 commercial, FF2 with free github net and Stockfish 210111 (the StockfishDev, FF2 was forked of)) - result in 2-3 days on my website.
The net is not part of the Stockfish sourcecode, so it is not under gpl, since the net is a separate file. Only the FF2 sourcecode is under gpl. Thats why the FF2 update separated the net from the binary.

User avatar
Sylwy
Posts: 3600
Joined: Fri Apr 21, 2006 2:19 pm
Location: IASI (Romania) - the historical capital of MOLDOVA

Re: Fat Fritz 2

Post by Sylwy » Sat Feb 13, 2021 6:46 pm

A 512 neurons net game (engine & net from Mr.Michael Taktikos archive). Weaker ? :lol:


MTaktikos
Posts: 16
Joined: Fri Oct 25, 2019 5:58 pm
Full name: Michael Taktikos

Re: Fat Fritz 2

Post by MTaktikos » Sat Feb 13, 2021 7:42 pm

nio wrote:
Sat Feb 13, 2021 3:26 pm
Hi, thanks: Whats the difference between FatFritz2.exe and FatFritz2mycompile.exe.
Whats the version of the included FatFritz2_v1.bin? Is it official or the one from github that some say is older version?
thanks
The FatFritz2mycompile was compiled one day earlier, but it's src code of Github/DanielUranga was broken and the NNUE file could not be read, i.e. the FatFritz2mycompile.exe is worthless and can be deleted.
Only FatFritz2.exe can work with the weights file FatFritz2_v1.bin.
This weights file is surprisingly strong. I have my doubts if the Chessbase version is really stronger, not impossible that CB has mixed up the NNUE files and now sells the weaker one embedded in the engine, while the stronger is open src in Github :laughing:

Locked