Interesting read about 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.
willmorton
Posts: 13
Joined: Thu Sep 17, 2020 7:19 pm
Full name: William Morton

Re: Interesting read about Fat Fritz 2

Post by willmorton » Sat Feb 20, 2021 7:22 am

Collingwood wrote:
Sat Feb 20, 2021 4:02 am
Someone on lichess pointed to this article: https://frederic-38110.medium.com/scam- ... a683203283
I have said it before: I am basically a scammer. An amateur who has never made a penny off his scams, but quite talented, you must admit.

The art of scamming includes seizing the moment ... Opportunities arise on a daily basis, and if you make use of them you can impress people no end and earn a big reputation.
Maybe the second sentence needs updating.
F. Friedel and A. Silver make a perfect couple :lol:

willmorton
Posts: 13
Joined: Thu Sep 17, 2020 7:19 pm
Full name: William Morton

Re: Interesting read about Fat Fritz 2

Post by willmorton » Sat Feb 20, 2021 7:47 am

amchess wrote:
Fri Feb 19, 2021 3:05 pm
CEGT, in the past, told me they don't accept any stockfish derivate...
http://www.cegt.net/40_40%20Rating%20Li ... liste.html
clearly, the clones yes...
Image

dkappe
Posts: 817
Joined: Tue Aug 21, 2018 5:52 pm
Full name: Dietrich Kappe

Re: Interesting read about Fat Fritz 2

Post by dkappe » Sat Feb 20, 2021 5:37 pm


Cornfed
Posts: 169
Joined: Sun Apr 26, 2020 9:40 pm
Full name: Brian D. Smith

Re: Interesting read about Fat Fritz 2

Post by Cornfed » Sat Feb 20, 2021 5:59 pm

Funny...perhaps people were slamming them for ditching their second rate engine under this GUI because one does not seem to be able to 'comment' on the articles on their website anymore.

amchess
Posts: 157
Joined: Tue Dec 05, 2017 1:42 pm

Re: Interesting read about Fat Fritz 2

Post by amchess » Sat Feb 20, 2021 6:06 pm

willmorton wrote:
Sat Feb 20, 2021 7:47 am
amchess wrote:
Fri Feb 19, 2021 3:05 pm
CEGT, in the past, told me they don't accept any stockfish derivate...
http://www.cegt.net/40_40%20Rating%20Li ... liste.html
clearly, the clones yes...
Image
The most shocking is they seem unable to distinguish a clone from an interesting derivative:
no need of particular programming skills because it's enough a compare program like WinMerge and similar...
Don't forget, for example, our derivatives (BrainLearn and ShashChess) are totally open source.
The same for others interesting derivatives.
I'm a computer science engineer, not a thief: IMHO, so, this testing strategy has nothing serious and professional.
Last edited by amchess on Sat Feb 20, 2021 6:10 pm, edited 1 time in total.

dkappe
Posts: 817
Joined: Tue Aug 21, 2018 5:52 pm
Full name: Dietrich Kappe

Re: Interesting read about Fat Fritz 2

Post by dkappe » Sat Feb 20, 2021 6:09 pm

dkappe wrote:
Sat Feb 20, 2021 5:37 pm
Albert’s response:

https://www.facebook.com/Albert.W.Silve ... 6059995567
Almost too predictable. Some person or persons started spamming it. He had to make it private. Reminds me of Gamer Gate.

connor_mcmonigle
Posts: 107
Joined: Sun Sep 06, 2020 2:40 am
Full name: Connor McMonigle

Re: Interesting read about Fat Fritz 2

Post by connor_mcmonigle » Sat Feb 20, 2021 6:26 pm

dkappe wrote:
Sat Feb 20, 2021 5:37 pm
Albert’s response:

https://www.facebook.com/Albert.W.Silve ... 6059995567
Empirically, from the innumerable comments on the Lichess forum, in which average chess players stated they felt duped by Albert's articles on ChessBase and ChessBase's misleading advertising, we can deduce that everything wasn't nearly as "crystal clear" to consumers as Albert suggests.
Albert Silver is the head coach of Fat Fritz, and its successor. His path to inventing the 2.0 engine is worth talking about, because once he was almost ready to publishing it he came across a new neural network technology from Japan – "NNUE". Silver reconsidered his close-to-finished project, and dedicated his mind and soul to the efficiently updatable neural network. It paid off! Fat Fritz 2.0 is out and ready now, defining opening theory with a "human" touch, and being as precise as can be in endgames.
This paragraph is highly misleading. It is exceedingly evident to me that this paragraph would lead the average reader to believe that Albert is responsible for porting HalfKP input features/NNUE code from Shogi to chess.
Fat Fritz 2 is trained and developed using this NNUE technology, running inside the open source Stockfish binary, but with a completely new and different neural network
Finally, where would any of this be without the fabulous Stockfish team and their legion of contributors, as well as Yu Nasu for his groundbreaking NNUE work, and Tanuki (nodchip) for his translation of the Shogi code for use in chess
The notion that these two sentences are sufficient is utterly ridiculous. The former is buried in the middle of a lengthy article. The latter sentence is positioned at the very end of a lengthy article, framed as a question, after the allocation of two paragraphs to the significance of Daniel Uranga and Dietrich Kappe's contributions to the project.

glennsamuel32
Posts: 115
Joined: Sat Dec 04, 2010 4:31 pm
Location: 223

Re: Interesting read about Fat Fritz 2

Post by glennsamuel32 » Sat Feb 20, 2021 7:38 pm

dkappe wrote:
Sat Feb 20, 2021 5:37 pm
Albert’s response:

https://www.facebook.com/Albert.W.Silve ... 6059995567

This Content Isn't Available Right Now
When this happens, it's usually because the owner only shared it with a small group of people, changed who can see it or it's been deleted.


:shock: :shock: :shock:

connor_mcmonigle
Posts: 107
Joined: Sun Sep 06, 2020 2:40 am
Full name: Connor McMonigle

Re: Interesting read about Fat Fritz 2

Post by connor_mcmonigle » Sat Feb 20, 2021 7:46 pm

For reference:

Image

gonzochess75
Posts: 95
Joined: Mon Dec 10, 2018 2:29 pm
Full name: Adam Treat

Re: Interesting read about Fat Fritz 2

Post by gonzochess75 » Sat Feb 20, 2021 7:56 pm

For those who didn't get to see the FB post before it was hidden:
"(picture) Albert Silver,
3h

This has been making waves and while I am finishing a lengthy description for an article in ChessBase, I will make a few brief comments here.

The article that was released by me (Best of Both Worlds), linked a bit below, gives a very clear description of what is what. You will notice that the lynch mob claims, among other things, that I 'suggest' I authored the NNUE integration in SF, yet the article explicitly says it was not me, but the Japanese programmers. There is no suggestion, it is in black and white.

The rest of the smear campaign is summed up by the very language used: a plethora of 'likely', 'suggested', 'probably', 'seems to' and more. It is innuendo after innuendo. A lot of things have taken place they fail to mention, such as all credits were clarified for Lc0 via a joint statement on Chessdom, in which they welcomed Deus X in TCEC. Two weeks later they posted a scathing attack on my character in the Leela blog. Of course (sarcasm mode = on) it is sheer coincidence this 'change of heart' took place the exact same day Deus X defeated Leela in TCEC (sarcasm mode = off).

Over the months in the public Discord of Stockfish I was asked if I planned to develop a Fat Fritz 2 using SF and NNUE. I always replied yes. There were no protests, just nods and declarations of curiosity and interest.

When it came out, I clearly and explicitly stated what was what and that I claimed authorship of the NN only. By all means see for yourself.

I quote:

"Fat Fritz 2 is trained and developed using this NNUE technology, running inside the open source Stockfish binary, but with a completely new and different neural network."

How unclear is that?

No one is being tricked or hoodwinked into thinking I did anything other than train a powerful unique neural network that runs inside Stockfish.

In the very same article, timed with the official release of Fat Fritz 2, I also wrote at the very end:

"Finally, where would any of this be without the fabulous Stockfish team and their legion of contributors, as well as Yu Nasu for his groundbreaking NNUE work, and Tanuki (nodchip) for his translation of the Shogi code for use in chess."

Again this is all crystal clear. I can only surmise the authors of that piece of trash manifesto all graduated Summa Cum Laude from the Trump School of Truth.

Addendum: I will not comment on the Hitler memes, Hitler videos and anti-Semitic comments I saw when someone suggested I was of Jewish origin.

<link to lichess.org article "Fat Fritz 2 is a rip-off">
"

Post Reply