Fire 5 is out!

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

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
User avatar
velmarin
Posts: 1598
Joined: Mon Feb 21, 2011 8:48 am

Re: What for a bullshit clone discussion!

Post by velmarin » Fri Nov 18, 2016 6:26 pm

No, Colin is right.
the explanation is the following, my explanation.

Damir knows that I was right.
He is intelligent and has seen the explanation in origins of the engine.
Explanation you typed Stephan, but yet to find controversy mixed Stephan with me.
You can see the origins of the engine and giving opinions.
Is more comfortable say here and insult.
You can continue using this fascinating engine, miraculous and testing and testing.
The time will give or take away the reason

Alexander Schmidt
Posts: 1086
Joined: Thu May 10, 2007 12:49 pm

Re: What for a bullshit clone discussion!

Post by Alexander Schmidt » Fri Nov 18, 2016 7:09 pm

pohl4711 wrote:And the conclusion is: Fire 5 code contains the same endgamestrings as Stockfish and in the exact same sequence...So, there can be no doubt about, that Fire 5 has taken code from Stockfish.
It is really a shame what you guys do.

You found one sequence of very common strings in two engines. Yes, it is indeed possible that Norman copied it from Stockfish, but it is also possible that he looked how it was implemented and did it in the same way. This strings are not very innovative. Maybe this sequence was even published somewhere else. Take a look into fora, programmers talk about how to implement tablebases and share codes. Maybe it was taken from such a place, maybe even from the Stockfish developers.

What you are doing is to blame someone of a crime based on pure speculation. What you have is definitively not enough as an evidence.

Please remember that we are talking about is a little sequence in the implementation of endgame knowledge or tablebases. Mr. V. claimed Fire is a dumped down clone of Stockfish and that is definitively not true.

User avatar
velmarin
Posts: 1598
Joined: Mon Feb 21, 2011 8:48 am

Re: What for a bullshit clone discussion!

Post by velmarin » Fri Nov 18, 2016 7:31 pm

Alexander Schmidt wrote:
pohl4711 wrote:And the conclusion is: Fire 5 code contains the same endgamestrings as Stockfish and in the exact same sequence...So, there can be no doubt about, that Fire 5 has taken code from Stockfish.
It is really a shame what you guys do.

You found one sequence of very common strings in two engines. Yes, it is indeed possible that Norman copied it from Stockfish
Of course, hehe
And the arrays, and the material.cpp, bitboards.cpp, eval.cpp., search.cpp, uci, cpp, ect, this is C++ and all are related,
It is no coincidence.
You will yield has suddenly of that this is true.

You already was critical with Norman before. This is not new.

You know that this Fire is not clean, but it is between the sword and the wall.

Kill the message or the Messenger,

Death to the Messenger. :twisted: :twisted: :twisted:

rainhaus
Posts: 143
Joined: Sun Feb 01, 2009 6:26 pm

Re: Fire 5 is out!

Post by rainhaus » Fri Nov 18, 2016 7:53 pm

Dr. Torsten Schoop wrote:

Obviously you are a non-programmer and you don’t know the difference between clone, derivative and taken ideas. Let me try to explain it.
Example: the Mercedes formula one car from Nico Rosberg.
Clone: just steal the car, paint it in another color and participate in the next formula one grand prix with “your” car.
Derivative: take the Mercedes formula one car from Nico Rosberg and try to modify it to make it faster. But you are not in expert in engineering a formula one car and at Mercedes several hundreds people are working at the formula one car to make it faster. So you will have a big chance to make the car SLOWER (a chess program weaker).
taken ideas: today it is not a secret how to design a car. But the way form that theoretical knowledge to a physical formula one car is gigantic.


After the thread now flows into an endless loop and will go down the gurgler, soon, for sure, I would like to share my insights on the occasion of this post.

I like such kind of vivid didactics.:!: So I got inspired now with the theory that somebody of Lewis Hamilton's crew has ported the software of his car into assembler! For how else he could win the last races against team-mate Nico Rosberg? :idea:

Also following this analogy: what would the consequences be, when Mercedes would release all the Formula One construction plans and setups as open source after the rules of GPL? I must confess, it exceeds my imagination
-
Thread viewImage
flat view
is a bad view
without thread view

PaulieD
Posts: 205
Joined: Tue Jun 25, 2013 6:19 pm

Re: Fire 5 is out!

Post by PaulieD » Sat Nov 19, 2016 1:01 am

Why is everyone so dead set against Norman Schmidt (Firefather)!

What the hell did he do to you to deserve this kind of condescending attitude?

Let he who never sinned cast the first stone. You hippocrites.

BrendanJNorman
Posts: 1358
Joined: Sun Feb 07, 2016 11:43 pm
Full name: Brendan J Norman

Re: Fire 5 is out!

Post by BrendanJNorman » Sat Nov 19, 2016 1:19 am

PaulieD wrote:Why is everyone so dead set against Norman Schmidt (Firefather)!

What the hell did he do to you to deserve this kind of condescending attitude?

Let he who never sinned cast the first stone. You hippocrites.
Probably jealousy and insecurity.

When a strong muscular guy walks past a fat guy and his girlfriend, what does the fat guy say to his girlfriend?

They say: "Obviously on steroids...It's disgusting how to roid freaks abuse their bodies!" while inside knowing that they'd die to have the same physique and their girlfriend would die to...

In all areas of life there are bitter people who pull down others to prop up their own fragile ego...the chess world is packed with these people.

I really get the feeling that a lot of people on this forum don't even play/enjoy chess anymore, but rather, spend 90% of their time running B.S tests to see whose engine is "original".

Do you guys even go over the games the engines play?

Or just leave the "test" on overnight and export the crosstable/engine "comparison" the following day.

There is little chess beauty involved at all (you've removed all the aesthetics)...it's just a big ego-gratification game.

"OMG! Fire/Strelka/Vitruvius/Gull/Thinker/Rybka/(insert almost any 3000+ engine) is a clone/derivative! :o Call the police! What a terrible thing! I ran a 10,000 game match between (engine 1) and (engine 2) and (engine 1) is CLEARLY a clone of..."

Jeez...If I go into the forum history, some people have been doing this for YEARS and YEARS.

Guess what? Nothing is going to change.

I learnt even at 10 years old: "Don't complain about something you cannot change - it's wasted energy you can use elsewhere"

Why not get back to enjoying chess?

Here's how:

1.Delete all of your "latest" SF1313116 builds.

2. Run a SF8/Fire5/Shredder 13/Houdini 5 gauntlet against Fritz 7, Junior 7, Hiarcs 8 Bareev, Shredder 6PB, Ktulu 7.1 and ChessTiger 14 at 5/2 Time Control with a book of sharp lines.

3. Grab a glass of wine.

4. Enjoy the bloodbath. :lol:

mehmet karaman
Posts: 142
Joined: Tue Jan 28, 2014 7:37 am
Location: TURKEY

Re: Fire 5 is out!

Post by mehmet karaman » Sat Nov 19, 2016 4:40 am

I tested some positons with Fire 5 and Stockfish 6/7.
I didn't see much similarity between two engines.

For now I'm sure that Norman took some ideas from Stockfish code, but it's not very much. I dont' think Fire is a derivative of Stockfish.
Fire 5 is 40 elo stronger than Fire 4. 40 elo increase for 2 years isn't a big elo jump.
Stockfish 8 is 169 elo stronger than Stockfish 6 according to Cegt 40/20 (1 core)

corres
Posts: 1507
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: What for a bullshit clone discussion!

Post by corres » Sat Nov 19, 2016 12:11 pm

[quote="Frank Quisinsky"]
I am very happy with Fire and the work Norman do.

In my opinion one of the most talented programmer we have ... and again ... we lost enough good programmers in the past. Different lost the interest with all the clone discussion.

[b]Enough is enough![/b]

The knowledge available today is in most of cases that what programmers give us in the last 40 years. Today the programmer try to make it better again and again (nice example the work the Stockfish Team do). We are on the right way today ... and don't need spectaculare clone discussion.

[b]Time is over for it![/b]

Best
Frank
[/quote]

I agree with you.
But I have some note:
1, There is a method to avoid any accusation: Norman makes the source of Fire 5 for open. I suppose he does not want to sell it. So what is the obstacle? I should be proud if someone "steal" one of my tricks.
2, There are lot of peoples who mix up "to make a clone" with "to make a derivative".
To make a clone of a program this means by that to make an exact copy of all parts of the program's source code what are essential to working of the original program. So for e.g. to change the name of an engine only the "new" engine is a clone obviously. Also obvious that to decide an engine being a clone or being not a clone you have to know the original FULL source code and the FULL source code of that engine accused to be clone.
Without this you ought to speak about "to be suspicious" that engine.
Naturally to make a comparison between source codes needs to be a programmer expert. Any other methods like reverse engineering or to run a program of "similarity" or make a comparison between the two engines how they respond to commands of UCI - these do not give established evidence but a suspicion only, if any.
Btw. Who is a programmer expert in this sub-forum?
To make a derivative of a program this means by that to use a kind of sketch of the original source code. In this case there is always an issue that how many part of that sketch is copied only from the original source code. I think this is a practical and an ethical question, too.

Modern Times
Posts: 2366
Joined: Thu Jun 07, 2012 9:02 pm

Re: What for a bullshit clone discussion!

Post by Modern Times » Sat Nov 19, 2016 12:20 pm

I don't see why people should have to prove their innocence. That goes against everything we stand for in society, which is innocent until proven guilty.

Norman does not need to do anything. But if he wanted to submit his source code under a non-disclosure agreement to a trusted, unbiased third party (probably outside of the chess community, because you can see the bias that exists here) then he could do that. But why would such an expert give their time free of charge ?

I have no problem testing Fire 5 personally, and indeed I'm currently running it for the CCRL chess960 list. Expect to see it there within about a week.
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

corres
Posts: 1507
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: What for a bullshit clone discussion!

Post by corres » Sat Nov 19, 2016 1:38 pm

[quote="Modern Times"]
I have no problem testing Fire 5 personally, and indeed I'm currently running it for the CCRL chess960 list. Expect to see it there within about a week.
[/quote]

You do it well.

Post Reply