Fire 8 released

Norman Schmidt has released Fire 8.

We will give it a test.

Ask by Matuiss2 on my channel.
​
"Isn't fire a clone?"

I am not the expert.

Is fire a clone, or a branch of Stockfish, or something else.

Here is what Fire says about itself.

"Fire

is a very strong, state-of-the-art, freeware UCI chess engine, and is designed for modern Windows and Linux systems.



Fire (previously called Firebird) has been in continuous development since Jan. 2010

​​

Over the last 10 years, the computer chess landscape has changed a lot. Previously, programmers had very few rudimentary guides to the process. But now the ideas, techniques, etc. used to create strong engines are well known and extremely well documented...take a look at https://www.chessprogramming.org/



In many ways, developing a strong engine today has become more of an engineering challenge than anything else. One must ask 'What are these techniques that really work well and/or have proven best, and how can one best architect and program them into a functional, strong, and unique entity?'

​

Many open-source projects have contributed enormously to the shared knowledge…serving to educate and guide, like blue-prints.

​

I'm grateful and indebted to many for all this shared knowledge, and would like to specifically credit the open-source

Ippolit, Ivanhoe, Gull, Houdini (https://www.cruxis.com/chess/houdini.htm), and Stockfish (https://stockfishchess.org).

​

I have drawn upon and use the ideas and techniques from all of these engines.

​

If you are interested in learning about my particular testing methodology, I've explained it in some detail here:
http://www.chessdom.com/fire-the-chess-engine-releases-a-new-version/ (or click on the TESTING menu choice above).

​

I want to thank anyone and everyone who has supported my efforts with Fire though these last 10 years.

​

May Peace Be With You!

​

Norman Schmidt"

In other news, Talkchess mods have already decided it is a clone and removed the website URL. Futile effort at censorship, all someone needs is to google it to find it based on the existence of the thread itself.

What a waste of talent.

+ 136 Elo to Fire 7.1 in my tests.

Ray wrote:

+ 136 Elo to Fire 7.1 in my tests.

In the past, I had some interactions with Norman and he seems to be a very nice person, always ready to help, to explain.

I am very confused about all this clone saga. I have no reason to doubt that he used big parts of the SF code, but I think he had the ability to change it. If he did not, it's beyond me. Sometimes, I do not get it, I admit.

Anyway, I just reported that he published a new version of Fire, without opinion about it. I hope one day we will know what it was really all about.

matejst wrote:

In the past, I had some interactions with Norman and he seems to be a very nice person, always ready to help, to explain.

I am very confused about all this clone saga. I have no reason to doubt that he used big parts of the SF code, but I think he had the ability to change it. Sometimes, I do not get it, I admit.

Anyway, I just reported that he published a new version of Fire, without opinion about it. I hope one day we will know what it was really all about.

I am even more confused after reading the CCC post. I do not know...

http://ccrl.chessdom.com/ccrl/404FRC/

I am totally sick of it all. Stefan Pohl made a great post, in which said that he is a tester not a judge, and no-one tells him what to test and what not to test. If they did not like his website, then don't visit it. That was the crux of it, but there was a lot more in it and it was a totally logical and sensible post.

Meanwhile, the witch-hunt on CCC continues. Sometimes it is on legality bases, sometimes on moral foundations, go figure. Depends of the situation, I guess.

Soon, one will not be able to publish a closed-sources engine, nor even to sell an engine without being lynched. In retrospect, Chris was right to react to... wait, wait, wait... Albert Silver's insinuations as harshly as he did.

When I think that all of this mess started when a top engine was decompiled and his sources and, more importantly, his ideas, illegally made available to all...

Hopefully it is a bit calmer here.

I have nothing against people who want to enjoy computer chess programming and don't want to reinvent every wheel and start from an open source. Why should they not? At the same time they should be open about that so that users, TD's, the rating lists guys can make choices to include those engines in their testing or not. Norman and Houdart do not fall in that category. Nobody rants and raves about Sugar, Shashchess and friends because.... they don't hide their origin.

And so (I think) a main ingredient of the current witch-hunt is about honesty vs dishonesty.

Ed,

on one hand, Norman's behaviour is difficult to understand. He does something good, then something stupid, then something... who can understand?

On the other hand, all of this is really too much. A civilized dialogue is always possible, and one should not hurry to make conclusions. You fought yourself for years against the vox populi in the Rybka case. I agree that it was different in essence, but the mob justice was as swift as today.

I see that anybody that published a closed-source engine lately was attacked: Dany (Andscacs), Chris, SMK, before this Naumov and others; everybody who has used some Sf code in some way or another, even the net trainer with original eval (not SF), was denigrated.

Admin wrote:

I have nothing against people who want to enjoy computer chess programming and don't want to reinvent every wheel and start from an open source. Why should they not? At the same time they should be open about that so that users, TD's, the rating lists guys can make choices to include those engines in their testing or not. Norman and Houdart do not fall in that category. Nobody rants and raves about Sugar, Shashchess and friends because.... they don't hide their origin.

And so (I think) a main ingredient of the current witch-hunt is about honesty vs dishonesty.

It’s not possible to credit other sources (etc) and not publish as open source and think all will be okay. Vas verbally and in writing credited everybody, but they still went after him.

It’s not about honesty /dishonesty/morality or anything else, it’s about gossip, envy and narcissism. The only difference this time is the “leading complainers” don’t appear to be too heavily infected with psychopathy. Throw psychopathy, coupled with narcissism into the mix and there really is a problem (as we can remember).
Psychopaths are without shame, you can’t stop them with counterfire shame. This current lot actually will shut up if they are called out for going too far.

In this type of discussion it depends how your standing is, just a game, just a hobby or are you competitive. I come from a competitive background and competition means fair play. In a tournament you don't want to play Stockfish 5 times. On a rating list you don't want to be overtaken by a Stockfish clone or derivative.

Ray wrote:

+ 136 Elo to Fire 7.1 in my tests.

Code:

1 Lc0 0.27.0                    410      40    447    371      1      75    408.5  91.4%  16.8%
  2 Stockfish 200221 avx2        373      38    449    360      5      84    402.0  89.5%  18.7%
  3 Stockfish 13                  358      36    448    351      4      93    397.5  88.7%  20.8%
  4 Stockfish 12                  340      36    448    346      9      93    392.5  87.6%  20.8%
  5 Dragon 64bit avx2            301      32    449    326      12    111    381.5  85.0%  24.7%
  6 Stockfish 11 x64 modern      244      30    449    297      25    127    360.5  80.3%  28.3%
  7 Stockfish 10 x64 popcnt      220      31    448    292      41    115    349.5  78.0%  25.7%
  8 Stockfish 9 x64 popcnt        182      29    449    268      52    129    332.5  74.1%  28.7%
  9 Komodo 14 64bit              179      29    448    262      50    136    330.0  73.7%  30.4%
  10 Komodo 14.1 64bit            176      28    448    257      48    143    328.5  73.3%  31.9%
  11 Stockfish 8 x64 popcnt        148      27    448    238      58    152    314.0  70.1%  33.9%
  12 Nemorino 6.00 avx2            136      27    447    233      66    148    307.0  68.7%  33.1%
  13 Slowchess 64 avx2            134      29    448    244      79    125    306.5  68.4%  27.9%
  14 Ethereal 12.75 x64 avx2      133      27    449    234      70    145    306.5  68.3%  32.3%
  15 Ethereal 12.50 x64 popcnt      99      27    449    217      93    139    286.5  63.8%  31.0%
  16 Fire 7.1 x64 popcnt            94      26    449    203      85    161    283.5  63.1%  35.9%

I have not tested Fire 8 yet, as I am still testing with a on going tournament before the release of Fire 8.

But if your rating is correct. It would put Fire 8 between Stockfish 10 and Stockfish 11 on the above list.

And did CCC completely remove the Fire 8 posts? As I could no longer find the thread.

mwyoung wrote:

And did CCC completely remove the Fire 8 posts? As I could no longer find the thread.

Yes, it seems that the witch-hunters, who have set themselves up as self-appointed judge, jury and executioners on behalf of the entire chess community, have removed it.

Now I will run some 40/15 games. If I was sitting on the fence before, their behaviour has made me 100 times more likely to test it, and I''m now totally determined to do so.

Ray wrote:

mwyoung wrote:

And did CCC completely remove the Fire 8 posts? As I could no longer find the thread.

Yes, it seems that the witch-hunters, who have set themselves up as self-appointed judge, jury and executioners on behalf of the entire chess community, have removed it.

Now I will run some 40/15 games. If I was sitting on the fence before, their behaviour has made me 100 times more likely to test it, and I''m now totally determined to do so.

Thank You Ed for creating ProDeo computer chess forum.

And why delete the whole thread! I would like to read all views.
So you are not even allowed to talk about Fire on CCC.

Go here:

http://talkchess.com/forum3/viewtopic.php?f=10&t=76719

It is in "Engines origins" subforum, so you have to log in.

matejst wrote:

Go here:

http://talkchess.com/forum3/viewtopic.php?f=10&t=76719

It is in "Engines origins" subforum, so you have to log in.

I guess that is where it belongs, that is the purpose of that subforum.

I guess mention of it will be allowed in general forums, as long as it doesn't concern its origin. And I'd certainly expect  tournament results would be permitted to be posted in that specific forum.

No, Ray, it is a new thread, where Norman Schmidt gave some answers. The original thread seems to be deleted indeed.

It would be nice if a new rating list saw the light with only original engines. The programmer code be a starting point.

http://rebel13.nl/download/programmer-c-ode-of-honor.html

With reference to the Similarity Report 2019 that would mean:

1. No Fire
2. No Houdini
3. No Shredder 13
4. No NirvanaChess 2.1
5. No Strelka
6. No Equinox 3.3
7. No Gull 2
8. No other Robbolito derivatives, Black Mamba, Bouquet, etc.

9. No FF1 and FF2
10. No other SF and Lc0 derivatives

The OERL can happily exist next to the traditional rating lists.

I think it is easily feasible -- the CCRL could make a list with rankings without derivatives with a few clicks, and label it like "reference rating list", OERL or something like this, and it could be the main page.

But... you do ask a lot of ethical questions by just making this list, and many of them have no answers.

Admin wrote:

It would be nice if a new rating list saw the light with only original engines.

Simply impossible. Who decides what is "original" or not, and on what criteria ? And will everyone agree with those criteria ?

Are you going to disassemble hundreds of engines and investigate them all one by one ?

Even disassembly is not perfect, so would you demand source code for every engine ? And if such requests are declined, as they surely would be, would you disqualify the engine in question ?