Tomcass wrote:Damir's versions are stronger every day. Look at these results:
Gui: Fritz 12
Ponder: Off
Time control: 4 min + 2 sec.
Book: HS Masterbook 2.0 DRK56 and Mishas12T using ONE CORE.
Deep Rybka4, Fire 1.31 and Naum 4.2 using FOUR CORES.
120 games.
DRK 56 has got an impressive 33,89% against Deep Rybka4, Fire 1.31 and Naum 4.2, with such a big handicap in cores. (Current tests -going on- are good but not that excellent).
On the other hand, look at this new, powerful MishasMauler12T. Perhaps the best engine today in the world using one core. (Not enough games, I know, but... please, rely on the instinct of this old tester).
Tom.
It is worth mentioning that the results aren't really that incredible. If one goes by the CEGT blitz list for example, Rybka 4 single CPU is about 80 elo less than Rybka 4 4CPU, but Damir is scoring about 120 Elo less (33%), which makes it about 40 elo weaker than Rybka 4 Single-core.
Yes, but DRK is free, so the price of each elo is about $2.
The point is that it is not stronger than the other Ippos. It may even be weaker.
That may be true, but the only "point" i see in your post is that it is about 40 elo weaker than Rybka 4 Single-core.
Then all I can say is that you missed the point. Tom posted this with the claim that it was the top single-core engine around, based on these results. My "point" (as you put it) is that his results don't show this at all.
It's no big deal, but nowhere in Tom's post did he claim that DRK was the top single-core engine. This was his "claim" ...
"On the other hand, look at this new, powerful MishasMauler12T. Perhaps the best engine today in the world using one core."
And for the sake of accuracy, "perhaps" doesn't mean it is, it means "perhaps it is".
I'm not sure about the point of your line of argumentation. My initial reply states the posted results are not incredible. You asked what my point was, and I replied that his results not only showed Damir as being weaker than Rybka 4, but weaker than other Ippos. So the question then becomes: what is impressive about the results?
That is my question and point. What is yours?
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
Tomcass wrote:Damir's versions are stronger every day. Look at these results:
Gui: Fritz 12
Ponder: Off
Time control: 4 min + 2 sec.
Book: HS Masterbook 2.0 DRK56 and Mishas12T using ONE CORE.
Deep Rybka4, Fire 1.31 and Naum 4.2 using FOUR CORES.
120 games.
DRK 56 has got an impressive 33,89% against Deep Rybka4, Fire 1.31 and Naum 4.2, with such a big handicap in cores. (Current tests -going on- are good but not that excellent).
On the other hand, look at this new, powerful MishasMauler12T. Perhaps the best engine today in the world using one core. (Not enough games, I know, but... please, rely on the instinct of this old tester).
Tom.
It is worth mentioning that the results aren't really that incredible. If one goes by the CEGT blitz list for example, Rybka 4 single CPU is about 80 elo less than Rybka 4 4CPU, but Damir is scoring about 120 Elo less (33%), which makes it about 40 elo weaker than Rybka 4 Single-core.
Yes, but DRK is free, so the price of each elo is about $2.
The point is that it is not stronger than the other Ippos. It may even be weaker.
That may be true, but the only "point" i see in your post is that it is about 40 elo weaker than Rybka 4 Single-core.
Then all I can say is that you missed the point. Tom posted this with the claim that it was the top single-core engine around, based on these results. My "point" (as you put it) is that his results don't show this at all.
It's no big deal, but nowhere in Tom's post did he claim that DRK was the top single-core engine. This was his "claim" ...
"On the other hand, look at this new, powerful MishasMauler12T. Perhaps the best engine today in the world using one core."
And for the sake of accuracy, "perhaps" doesn't mean it is, it means "perhaps it is".
I'm not sure about the point of your line of argumentation. My initial reply states the posted results are not incredible. You asked what my point was, and I replied that his results not only showed Damir as being weaker than Rybka 4, but weaker than other Ippos. So the question then becomes: what is impressive about the results?
That is my question and point. What is yours?
I'm not argueing, but i didn't ask what your point was, i merely pointed out that your point was wrong (misquoted Tom). But i see it's useless to converse reasonably with you, you misquote and change the subject. I have better things to do with my time.
Tomcass wrote:Damir's versions are stronger every day. Look at these results:
Gui: Fritz 12
Ponder: Off
Time control: 4 min + 2 sec.
Book: HS Masterbook 2.0 DRK56 and Mishas12T using ONE CORE.
Deep Rybka4, Fire 1.31 and Naum 4.2 using FOUR CORES.
120 games.
DRK 56 has got an impressive 33,89% against Deep Rybka4, Fire 1.31 and Naum 4.2, with such a big handicap in cores. (Current tests -going on- are good but not that excellent).
On the other hand, look at this new, powerful MishasMauler12T. Perhaps the best engine today in the world using one core. (Not enough games, I know, but... please, rely on the instinct of this old tester).
Tom.
It is worth mentioning that the results aren't really that incredible. If one goes by the CEGT blitz list for example, Rybka 4 single CPU is about 80 elo less than Rybka 4 4CPU, but Damir is scoring about 120 Elo less (33%), which makes it about 40 elo weaker than Rybka 4 Single-core.
Yes, but DRK is free, so the price of each elo is about $2.
The point is that it is not stronger than the other Ippos. It may even be weaker.
That may be true, but the only "point" i see in your post is that it is about 40 elo weaker than Rybka 4 Single-core.
Then all I can say is that you missed the point. Tom posted this with the claim that it was the top single-core engine around, based on these results. My "point" (as you put it) is that his results don't show this at all.
It's no big deal, but nowhere in Tom's post did he claim that DRK was the top single-core engine. This was his "claim" ...
"On the other hand, look at this new, powerful MishasMauler12T. Perhaps the best engine today in the world using one core."
And for the sake of accuracy, "perhaps" doesn't mean it is, it means "perhaps it is".
I'm not sure about the point of your line of argumentation. My initial reply states the posted results are not incredible. You asked what my point was, and I replied that his results not only showed Damir as being weaker than Rybka 4, but weaker than other Ippos. So the question then becomes: what is impressive about the results?
That is my question and point. What is yours?
I'm not argueing, but i didn't ask what your point was, i merely pointed out that your point was wrong (misquoted Tom). But i see it's useless to converse reasonably with you, you misquote and change the subject. I have better things to do with my time.
How am I misquoting and changing the subject?
Tom wrote: "DRK 56 has got an impressive 33,89% against Deep Rybka4, Fire 1.31 and Naum 4.2, with such a big handicap in cores."
to which I replied: "It is worth mentioning that the results aren't really that incredible."
and you replied: "Yes, but DRK is free"
Now, he said his results were impressive, I said they weren't, and you entered by explaining it was free.... The relevance of which completely escapes me.
As far as I'm concerned the result is not impressive, and is subpar for an IPPO. All IPPOs are free, so mentioning this one also being free, as per your interjection, changes nothing.
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
For me these DRK56 results are really impressive. Even 40 elo below DR4 is an excellent result. The fact of being free is for me really marginal.
Albert, this is obviously a subjective opinion. Also when I said that Mishas Mauler12T is PERHAPS the best engine today in the world I openly ask the CChess community to rely a little bit to the opinion of a very experienced -although not always scientifically precise- tester.
I test for my own enjoyment. But I have no problem to share my results with the community. And watching the games of engines such as the two mentioned is for me a great source of pleasure.
The main difference between commercial programmes -I own and pay for all of them- and Ippo sons is that commercials behave in a DISCRETE way (one verson every year, average) and IPPO in a CONTINUOUS way (new versions every week and dozens of testers helping to improve). In the near future non commercials engines, supported by a strong co-ordinated chess community, will be always better than commercials.
If you really love computer chess, Albert, please don't miss this fantastic new version of IVAN (57b) and the new Houdini 1.03 that will appear in three weeks time.
