No +100 Elo for SF on Windows for me

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

Moderator: Ras

AndrewGrant
Posts: 1945
Joined: Tue Apr 19, 2016 6:08 am
Location: U.S.A
Full name: Andrew Grant

Re: No +100 Elo for SF on Windows for me

Post by AndrewGrant »

Graham Banks wrote: Sat Jun 01, 2024 1:09 am
AndrewGrant wrote: Sat Jun 01, 2024 1:05 am
Graham Banks wrote: Sat Jun 01, 2024 12:59 am We are way too quick in this community to ostracize people who've made valuable contributions to computer chess.

Forgiveness seems to be a very rare commodity.
Fool me once, shame on you. Fool me twice, shame on me.
How many times will you be taken a fool, Graham? Norman has gotten you 4+ times now lol.
Norman might have done some dubious things in the past, but my understanding is that the latest versions of Fire are fine.
Both Norm and I are retired gentlemen who just want to enjoy computer chess without getting involved in any pettiness or nastiness.
It was only a couple years ago that Fire illegally contained code taken from Houdart.
Viz
Posts: 223
Joined: Tue Apr 09, 2024 6:24 am
Full name: Michael Chaly

Re: No +100 Elo for SF on Windows for me

Post by Viz »

This is pretty funny.
Just because someone contributed in some way to something at some point of his life doesn't mean he can't be a fraudster.
Remember guy called Lance Armstrong? He contributed a lot to popularising his sport and making money for cancer charity, yet he is definitely a fraudster.
Same goes for people like albert silver, although I have no idea what exactly he contributed to chess engine world. Where is part of code from him (he can't code anything) or any idea that actually was implemented after he proposed it? Can't recall anything probably because they don't exist.
User avatar
Graham Banks
Posts: 43483
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: No +100 Elo for SF on Windows for me

Post by Graham Banks »

Viz wrote: Sat Jun 01, 2024 3:33 am This is pretty funny.
Just because someone contributed in some way to something at some point of his life doesn't mean he can't be a fraudster.
Remember guy called Lance Armstrong? He contributed a lot to popularising his sport and making money for cancer charity, yet he is definitely a fraudster.
Same goes for people like albert silver, although I have no idea what exactly he contributed to chess engine world. Where is part of code from him (he can't code anything) or any idea that actually was implemented after he proposed it? Can't recall anything probably because they don't exist.
https://www.chessprogramming.org/Albert_Silver

Albert Silver,
a Brazilian independent information services professional, chess player [2], chess and computer chess journalist, aficionado, tester and expert, from 1999 - 2002 affiliated with Convekta as software designer for Chess Assistant, and editor and writer at ChessBase News [3] [4]. He is author the Silver Openings Suite [5] and the Silver Gambit Suite [6] [7] meant to be used in test matches between engines, and served as moderator in CCC.
gbanksnz at gmail.com
Viz
Posts: 223
Joined: Tue Apr 09, 2024 6:24 am
Full name: Michael Chaly

Re: No +100 Elo for SF on Windows for me

Post by Viz »

Yeah, as I said, don't exist.
I would guess he wrote this article himself because somehow and some way it forgots to mention that fat fritz in it marketing used match result vs sf12 while using +35 elo patches from search stockfish dev version accumulated by this point - and thus indeed was able to beat sf 12 convincingly to literally no one surprise.
But I'm actually wrong. He contributed. His main contribution is that now stockfish make new releases not every 50 elo as it did for almost a decade by that point but releases 15.1 14.1 16.1 versions which are like 20 elo stronger - just so scammers like this wouldn't be able to do exactly what I described in previous paragraph.
User avatar
Graham Banks
Posts: 43483
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: No +100 Elo for SF on Windows for me

Post by Graham Banks »

Viz wrote: Sat Jun 01, 2024 3:50 am Yeah, as I said, don't exist.
I would guess he wrote this article himself because somehow and some way it forgots to mention that fat fritz in it marketing used match result vs sf12 while using +35 elo patches from search stockfish dev version accumulated by this point - and thus indeed was able to beat sf 12 convincingly to literally no one surprise.
But I'm actually wrong. He contributed. His main contribution is that now stockfish make new releases not every 50 elo as it did for almost a decade by that point but releases 15.1 14.1 16.1 versions which are like 20 elo stronger - just so scammers like this wouldn't be able to do exactly what I described in previous paragraph.
I think you're being too unkind.
There is also this:

Albert also designed the first fully Windows interface for online play ever. It was the groundbreaking novelty in Chess Assistant 6.0 and was made in collaboration with ICC whose service was the one it accessed. Until then all games searched or created were done through command line commands. The concept of a list of games to click on with a column showing the player, the time control and others, all standards today, were first designed and implemented by Albert. The possibility of then just dragging and dropping the game just played to be analyzed by an engine. Albert's idea. The ability to use an openings tree to accompany games being broadcast? Albert's idea. All introduced in CA6. The credits of CA 6.0 reads: Thanks to Albert Silver who gave us so many ideas it would take centuries to implement them all.
gbanksnz at gmail.com
Viz
Posts: 223
Joined: Tue Apr 09, 2024 6:24 am
Full name: Michael Chaly

Re: No +100 Elo for SF on Windows for me

Post by Viz »

Yeah, I'm unkind towards person who first stole leela work, then stole stockfish work, and then proceeded to fraudulently market it with bullshit claims of incredible innovations only that kind of genius is capable of - while needing help of multiple people for weeks to be able to set up training and even to compile an engine. And as I said, did it not only once, but twice.
And yes, I'm pretty upset that somehow and someway talkchess has some decent opinion about this type of people because 20 years ago they wrote some articles and whatever.
User avatar
Graham Banks
Posts: 43483
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: No +100 Elo for SF on Windows for me

Post by Graham Banks »

Viz wrote: Sat Jun 01, 2024 4:13 am Yeah, I'm unkind towards person who first stole leela work, then stole stockfish work, and then proceeded to fraudulently market it with bullshit claims of incredible innovations only that kind of genius is capable of - while needing help of multiple people for weeks to be able to set up training and even to compile an engine. And as I said, did it not only once, but twice.
And yes, I'm pretty upset that somehow and someway talkchess has some decent opinion about this type of people because 20 years ago they wrote some articles and whatever.
I think he's done more than just write articles.
Some might argue that he has contributed more towards computer chess than you have.
gbanksnz at gmail.com
AndrewGrant
Posts: 1945
Joined: Tue Apr 19, 2016 6:08 am
Location: U.S.A
Full name: Andrew Grant

Re: No +100 Elo for SF on Windows for me

Post by AndrewGrant »

Graham Banks wrote: Sat Jun 01, 2024 4:17 am
Viz wrote: Sat Jun 01, 2024 4:13 am Yeah, I'm unkind towards person who first stole leela work, then stole stockfish work, and then proceeded to fraudulently market it with bullshit claims of incredible innovations only that kind of genius is capable of - while needing help of multiple people for weeks to be able to set up training and even to compile an engine. And as I said, did it not only once, but twice.
And yes, I'm pretty upset that somehow and someway talkchess has some decent opinion about this type of people because 20 years ago they wrote some articles and whatever.
I think he's done more than just write articles.
Some might argue that he has contributed more towards computer chess than you have.
But no one would argue that Viz has stolen the work of as many people as Albert has :)
User avatar
Graham Banks
Posts: 43483
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: No +100 Elo for SF on Windows for me

Post by Graham Banks »

AndrewGrant wrote: Sat Jun 01, 2024 4:18 am
Graham Banks wrote: Sat Jun 01, 2024 4:17 am
Viz wrote: Sat Jun 01, 2024 4:13 am Yeah, I'm unkind towards person who first stole leela work, then stole stockfish work, and then proceeded to fraudulently market it with bullshit claims of incredible innovations only that kind of genius is capable of - while needing help of multiple people for weeks to be able to set up training and even to compile an engine. And as I said, did it not only once, but twice.
And yes, I'm pretty upset that somehow and someway talkchess has some decent opinion about this type of people because 20 years ago they wrote some articles and whatever.
I think he's done more than just write articles.
Some might argue that he has contributed more towards computer chess than you have.
But no one would argue that Viz has stolen the work of as many people as Albert has :)
Where has he been convicted of theft and by whom?
gbanksnz at gmail.com
User avatar
Graham Banks
Posts: 43483
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: No +100 Elo for SF on Windows for me

Post by Graham Banks »

What I'm about to post is already well-known and has been posted elsewhere:

The only mistake, which I (Albert) will gladly take ownership over, was to bundle the NN and the binary into a single executable. It was a genuinely innocent mistake, but one that led to all the subsequent problems.
gbanksnz at gmail.com