GM and Rybka vs. Stockfish

[Sedat Canbaz]

Note that fide rating is not about blitz of 3 minutes per game.

I think that no human can win in your challange of 3 minutes per game
but it does not mean that they cannot do it at longer time control.

90+30 is the normal time control that I play for fide rating.

Ok...if 3 minutes per game is too fast...

What about 3 min + 2 sec per game ? I can agree with this time control too...

Note that,
This time control (3m+2s) is one of the most used in the latest World Blitz Championships !

Actually I like more slow time controls than blitz...
For example my favorite slow time control is 40/120

Frankly I announced a blitz challenge due to,
I have no much patience to see the final results

Plus if we run a gauntlet with slow time control...
then we will need a lot of free time...it requires months....sorry I have other jobs to do

Another reason about why I preferred Blitz due to,
We can run many games, more than 100 games...then the played results will give us more accurate Elo measuring

For example with slow time controls,
Of course GMs will perform better....
But however, Stockfish performance is expecting much higher than your 3200 Elo estimation!
Probably Stockfish will perform at least +3400 Elo even at slow time control !!


And here is FIDE Blitz Rating:

.

[Sedat Canbaz]

instead of writing CHALLENGE everywhere you could start sending email to some of the top players and trying to get some sponsor.

it could be a very important match

To be honest,
I don't care too much about this challenge...and I have no much free time make announcement in other places too

That's why I give a 10 % commision for that person who will find me a GM (due to I have no much free time )


Have a nice weekend

[M ANSARI]

I think it would be more interesting if they give the GM the 8 core and Rybka 3 and let SF5 play on single core. On quite a few moves it is obvious that SF5 could simply see much deeper than R3 and tactically a tiny mistake can lead to an immediate lost position when playing such strong engines ... although a human GM might have excellent positional awareness, it cannot cover for lack of tactical awareness. In this case the human was depending on R3 to give him a tactical safety net ... which was silly when you consider it was up against a tactically superior engine with dramatically stronger hardware. Again I think the human is the weak link in this setup.

[Leto]

I think it would be more interesting if they give the GM the 8 core and Rybka 3 and let SF5 play on single core. On quite a few moves it is obvious that SF5 could simply see much deeper than R3 and tactically a tiny mistake can lead to an immediate lost position when playing such strong engines ... although a human GM might have excellent positional awareness, it cannot cover for lack of tactical awareness. In this case the human was depending on R3 to give him a tactical safety net ... which was silly when you consider it was up against a tactically superior engine with dramatically stronger hardware. Again I think the human is the weak link in this setup.

Rybka 4 on 8 core is very close in strength to Stockfish 5 on single core. But that essentially means that if you give Rybka 4 an 800% time handicap it would be close to the strength of Stockfish 5 running on one core.

[carldaman]

In principle, it was interesting to see whether the human GM could close the gap between Rybka 3 and Stockfish 5. This GM could not, as he admittedly made some rash choices at some key moments.

However, a GM (or lesser player) with a more suitable anti-computer style might have performed better. One is left wondering what would have happened if GM Naroditsky had played all 4 games the way he played the last. Taking into account Rybka's tactical input, he might have drawn all games had he settled for boring, closed games, and especially if he had opted for KID-like structures that could also offer some winning chances (as in the last game).

[Sedat Canbaz]

Just my 2 cents more over this issue,

Of course with Slow time controls,
It is hard to estimate and to be beaten Stockfish (even if Stockfish is playing without book)...

But in Blitz (it is not required especially human to be a GM)
In case of Stockfish is playing without book, I can use Rybka + very strong book
And then I expect to see completely different results...
I mean there is a big possibility to see several wins in favor for Rybka

For example,
Normally the Blitz Elo difference with short neutral books between Rybka 3 and Stockfish 5 is around 200 Elo,
but with the above special setup... I expect to see the Elo difference to be approx. 50-100 Elo (in favor for Stockfish)
Even this setup can be improved to be around in equal Elo performance between both engines

[Sedat Canbaz]

Just I'd like to add (Rybka 3 vs Stockfish 5),

With the above special conditions and even with Slow time controls,
I can reduce the Elo difference to be around 50 Elo and even there is possibility to see almost equal in Elo points

Note: normally the Slow Elo difference (with short neutral books) between Rybka 3 and Stockfish 5 is around 150 Elo

In other words,
All this depends on the superiority of the opening lines !

Hopes this helps too!