Syzygy / egbb discussion

Daniel Shawul · Post by **Daniel Shawul** » Thu Oct 24, 2013 8:34 pm

Yes i want to see how you butchered the code. Send it to dshawul@yahoo.com

Ryan Benitez · Post by **Ryan Benitez** » Thu Oct 24, 2013 8:49 pm

Houdini wrote:Daniel, I don't have time for your nonsense.
As I told you before, Houdini 4 will use Syzygy bases which is a far superior solution in many respects.
egbb are good, but Syzygy are excellent. Deal with it...

I call BS. It is not the system used but the probing code that is the issue. Saying that you have bitabases working but they are only good for analyzing but not games is avoiding the fact that they are not implemented correctly.

syzygy · Post by **syzygy** » Thu Oct 24, 2013 9:22 pm

Ryan Benitez wrote:
Houdini wrote:Daniel, I don't have time for your nonsense.
As I told you before, Houdini 4 will use Syzygy bases which is a far superior solution in many respects.
egbb are good, but Syzygy are excellent. Deal with it...
I call BS. It is not the system used but the probing code that is the issue. Saying that you have bitabases working but they are only good for analyzing but not games is avoiding the fact that they are not implemented correctly.

I don't think you at all know what you're talking about. Both systems come with probing code. But you're just another one of those ... persons ... that seem to bear a grudge so this is just not interesting.

Daniel Shawul · Post by **Daniel Shawul** » Thu Oct 24, 2013 9:29 pm

Ryan Benitez wrote:
Houdini wrote:Daniel, I don't have time for your nonsense.
As I told you before, Houdini 4 will use Syzygy bases which is a far superior solution in many respects.
egbb are good, but Syzygy are excellent. Deal with it...
I call BS. It is not the system used but the probing code that is the issue. Saying that you have bitabases working but they are only good for analyzing but not games is avoiding the fact that they are not implemented correctly.

Houdart's ignorance has been established with Kai's tests already. Ofcourse Toga and others implemented it correctly as they should be used and don't need any other DTM/DTZ to win games, which has shown the same ELO as others with DTM+bitbases. That is like an 100Gb solution to what you can do with bitbases alone, go figure ...

Ryan Benitez · Post by **Ryan Benitez** » Thu Oct 24, 2013 9:32 pm

syzygy wrote:
Ryan Benitez wrote:
Houdini wrote:Daniel, I don't have time for your nonsense.
As I told you before, Houdini 4 will use Syzygy bases which is a far superior solution in many respects.
egbb are good, but Syzygy are excellent. Deal with it...
I call BS. It is not the system used but the probing code that is the issue. Saying that you have bitabases working but they are only good for analyzing but not games is avoiding the fact that they are not implemented correctly.
I don't think you at all know what you're talking about. Both systems come with probing code. But you're just another one of those ... persons ... that seem to bear a grudge so this is just not interesting.

Grudge? I don't know you. I simply do not accept claims on insufficient evidence. Comparing the whole system based on who makes to easiest example code to cut and paste into their robbolito based engine seems absurd to me. In practical terms I expect the same results with both systems given a competent implementation of each for the engine they are going into. Please show me data if you wish to claim otherwise.

Daniel Shawul · Post by **Daniel Shawul** » Thu Oct 24, 2013 9:38 pm

In practical terms I expect the same results with both systems given a competent implementation of each for the engine they are going into.

It is just a load of crap. Kai has done tests with all bitbases and shown the same ELO for all with same error bars.
http://www.talkchess.com/forum/viewtopi ... 72&t=49684

Houdini · Post by **Houdini** » Thu Oct 24, 2013 9:54 pm

syzygy wrote:I don't think you at all know what you're talking about. Both systems come with probing code. But you're just another one of those ... persons ... that seem to bear a grudge so this is just not interesting.

Agreed, the nonsense that has been written above is beyond even the usual standards of internet forums, we're really hitting very deep lows.

This discussion becomes even more insane when you consider that probably nobody else on this forum has the same EGTB implementation experience as I have. Is there another engine with successful Nalimov, Gaviota, Scorpio and Syzygy bases implementation?

What a freaking waste of time...

P.S. Daniel, if you want a copy of the code please send your e-mail to houdini@cruxis.be before tomorrow evening, after that I will be off-line until early November.

Daniel Shawul · Post by **Daniel Shawul** » Thu Oct 24, 2013 10:03 pm

Are you for real? You butchered Scorpio bitbases implementation and thought they should be removed until Ferdinad showed you how it should be done. Here is a reminder.

http://www.talkchess.com/forum/viewtopi ... 05&t=49684

I already gave you my email address but didn't get anything ....

syzygy · Post by **syzygy** » Thu Oct 24, 2013 10:08 pm

There is no reason to expect the "same" results as the technical details are vastly different.

Houdini · Post by **Houdini** » Thu Oct 24, 2013 10:11 pm

Daniel, stop spreading FUD about the egbb probing in Houdini 3. It works very well for analysis.

Note that the egbb probing WILL be removed from Houdini 4, there is no point any longer now that Syzygy is available. Did you really think that I changed my mind?

And for the third time, send an e-mail to houdini@cruxis.be with your request for egbbdll.cpp so this can be handled tomorrow by the Houdini support.

Can this stop now?

Syzygy / egbb discussion

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?

Re: Where are you Houdart?