How to disprove Fernando's proposition ... ?

[Tony Thomas]

It would be. I look forward...

My best
Fernando

You still havent given him the name of an engine. Why dont you tell him to test it against Chess Alex?

[gerold]

It would be. I look forward...

My best
Fernando

You still havent given him the name of an engine. Why dont you tell him to test it against Chess Alex?

Chess Alex would lose.

[Peter Fendrich]

A good start would be to explain what CC7 is...
Then it would probably be easy to select an engine.
/Peter

[smrf]

My assumption to this is: CC7 == Chess Challenger 7

[Peter Fendrich]

of course! It was soooo long ago....

I thought SSDF full list could help you but it is not there:
http://web.telia.com/~u85924109/ssdf/rlwww071.txt

Let's say its rating is about 1300 following SSDF level.

/Peter

[hgm]

see at: http://216.25.93.108/forum/viewtopic.ph ... 53&t=13472, where Fernando stated on SMIRF: ... It is even weaker than CC7 ...

Which engine of minimum strength has to be beaten by SMIRF to refute his claim?

Regards, Reinhard.

The problem is that you cannot decide on the strength of an engine from a single game, or from a set of games against a single opponent. The strength of engines is determined just as much by how efficienly it finishes off opponents that are 300 rating points weaker, as by its 'surprise rate' against engines that are 300 points stronger.

At the very minimum, you should have SMIRF play a total of at least 100 games, against at least 10 different opponents, spread reasonably over the mentioned ELO range. Otherwise you wouldn't have proved a thing.

[smrf]

see at: http://216.25.93.108/forum/viewtopic.ph ... 53&t=13472, where Fernando stated on SMIRF: ... It is even weaker than CC7 ...

Which engine of minimum strength has to be beaten by SMIRF to refute his claim?

Regards, Reinhard.

The problem is that you cannot decide on the strength of an engine from a single game, or from a set of games against a single opponent. The strength of engines is determined just as much by how efficienly it finishes off opponents that are 300 rating points weaker, as by its 'surprise rate' against engines that are 300 points stronger.

At the very minimum, you should have SMIRF play a total of at least 100 games, against at least 10 different opponents, spread reasonably over the mentioned ELO range. Otherwise you wouldn't have proved a thing.

Well, but if there would be some clear victories of SMIRF it would be the task of others to prove CC7 to be better, what I intend not to do.

Reinhard.

[fern]

I used CC7 as metaphor. I doubt it plays better than Smirf. But Smif does play moves a la CC7. I refer, h5 without reason, things like that as I have explained several times to date.

My best
Fernando

[Uri Blass]

I used CC7 as metaphor. I doubt it plays better than Smirf. But Smif does play moves a la CC7. I refer, h5 without reason, things like that as I have explained several times to date.

My best
Fernando

Maybe the reason that it played strange moves is simply the fact that you did not make a donation

"The donationware version of SMIRF is not restricted at all in its operation. In contrast to this the use of the optimized bonus version is reserved for people having made a serious donation"

Maybe your hardware is the problem:
"The author is pointing out, that this software has not been tested at all imaginable hardware configurations"

Note that testing it I do not find that it plays strange moves and it even plays by its own known gambit with black.

1.e4 d5 2.exd5 Nf6

Uri

[fern]

Even a limited chess software -due to no payment- does not play h5 just like that.
My hardware has a fast 3 Gig Intel processor, two threads, etc. XP Pro.

My bst
Fdo