100 long games Rybka 4 vs Houdini 1.03a

[Laskos]

Actually, I didn't see any proof. First of all you gave zero information on the test conditions, meaning you might have tested without books, or with. The only information you gave showed you played Rybka in a handicapped 32-bit environment.

As to being stronger... In all honesty I am not at all convinced Rybka 4 is stronger than Houdini 1.03a. I think they are roughly equal.

I tested with Bob's 4,000 opening positions from EPD file, and it's very reliable, as Bob scrambled it from initial ECO sorting. My test was +/- 4 Elo points error 95% confidence, not some 17 Elo points total error 95% confidence. LOS was very close to 100% for IvanHoe. If you deny that at very fast time controls some IvanHoes beat the crap out of Rybka 4, you deny a simple evidence, presented clearly, a guilt punishable even by religious scholars .

Kai

[Laskos]

How come all the ippolites are always so angry? Have they been watching too much Glenn Beck?

How come that Rybka freaks are lying you in face, with clear evidence of opposite? Really, I have to be a monkey not to slap them on their faces. Wake up, guys, the burden is on you to show that Rybka 4 is stronger at chosen by you time control!

Kai

[Gino Figlio]

How come all the ippolites are always so angry? Have they been watching too much Glenn Beck?

I thought it was the rybkaphiles who were predicting the end of the world and finding satan behind every corner...may some of the better engines be some day accepted in your holy testing groups.
Would you risk your soul for that?

[tomgdrums]

How come all the ippolites are always so angry? Have they been watching too much Glenn Beck?

I thought it was the rybkaphiles who were predicting the end of the world and finding satan behind every corner...may some of the better engines be some day accepted in your holy testing groups.
Would you risk your soul for that?

holy testing groups? I don't have any holy testing groups. (now forever known as HTGs....not to be confused with ctgs!)

[Albert Silver]

Actually, I didn't see any proof. First of all you gave zero information on the test conditions, meaning you might have tested without books, or with. The only information you gave showed you played Rybka in a handicapped 32-bit environment.

As to being stronger... In all honesty I am not at all convinced Rybka 4 is stronger than Houdini 1.03a. I think they are roughly equal.

I tested with Bob's 4,000 opening positions from EPD file, and it's very reliable, as Bob scrambled it from initial ECO sorting. My test was +/- 4 Elo points error 95% confidence, not some 17 Elo points total error 95% confidence. LOS was very close to 100% for IvanHoe. If you deny that at very fast time controls some IvanHoes beat the crap out of Rybka 4, you deny a simple evidence, presented clearly, a guilt punishable even by religious scholars .

Kai

Sorry, you failed to publish the results where some Ivanhoes "beat the crap" out of Rybka 4. If you could show them I would be curious.

Also, regarding the test you did publish, as I said, you tested it in a handicapped environment for Rybka, and what is more, an environment that hasn't represented my computer usage for some 5 years or so, which is when I migrated to a 64-bit OS.

Also, it would be nice if you would just state the full match conditions instead of people having to ask for each and every one: CPU? Hash? GUI? Ponder on or off? Single CPU, or not? Were there any time losses? Etc.

[Laskos]

holy testing groups? I don't have any holy testing groups. (now forever known as HTGs....not to be confused with ctgs!)

You see, again hypocritical and lying. You know perfectly that it's about CEGT and CCRL, and maybe SSDF. I am really baffled by you, Rybka freaks, probably this is some sort of ethics, if something costs $60 and you possibly paid for it, it must be better than the free stuff. Still baffled, $60 could make men liars? I am actually afraid of these folks.

Kai

[tomgdrums]

holy testing groups? I don't have any holy testing groups. (now forever known as HTGs....not to be confused with ctgs!)

You see, again hypocritical and lying. You know perfectly that it's about CEGT and CCRL, and maybe SSDF. I am really baffled by you, Rybka freaks, probably this is some sort of ethics, if something costs $60 and you possibly paid for it, it must be better than the free stuff. Still baffled, $60 could make men liars? I am actually afraid of these folks.

Kai

Am I being hypocritical? I don't have any holy testing groups so I am not lying. I don't even pay that much attention to them. I am actually a bigger fan of Hiarcs, Shredder and Junior because I feel that they are better suited to help me get better at chess.

How much Glenn Beck do you watch?

[Laskos]

Sorry, you failed to publish the results where some Ivanhoes "beat the crap" out of Rybka 4. If you could show them I would be curious.

Also, regarding the test you did publish, as I said, you tested it in a handicapped environment for Rybka, and what is more, an environment that hasn't represented my computer usage for some 5 years or so, which is when I migrated to a 64-bit OS.

Also, it would be nice if you would just state the full match conditions instead of people having to ask for each and every one: CPU? Hash? GUI? Ponder on or off? Single CPU, or not? Were there any time losses? Etc.

In the published by me test an IvanHoe beat the crap out of Rybka 4 with >20 Elo points difference even without bitbases, which Rybka lacks. 4 Elo points error 95% confidence, 2 points 68% confidence.

Hardware: AMD Athlon(tm) 64 X2 Dual Core Processor 4800+ Dual 2600 MHz with 2,048 MB Memory

OS: 32 bit XP SP3

Engines: single core, ponder off

Hash: 32MB

Time Control: 1'' base + 0.1'' increment/move

No losses on time or any extra-chess reasons (only mates). Draw after 200 moves.

GUI: LittleBlitzer 2.5, an excellent program for short time controls.

Rybka used 8% more time without losing a single game on time.

The hadicap from 64 bit to 32 bit is the same for Rybka 4 and for Ippo* engines within 4 Elo points error 95% confidence. Stating again and again the same 32 bit problem is only whining, from you, Majd and Uri.

Now, publicly, do you want to bet $1,000 for a match of 10,000 games at 10'+10'' TC (medium TC, longer is not practical), 64 bits, 8 cores? I will pick an IvanHoe and you Rybka 4. Fine?

Kai

[Laskos]

Am I being hypocritical?

Yes, you are.

I don't have any holy testing groups so I am not lying.

You lied and lie again. You know what Gino meant by "holy" testing groups. If you deny again, you are a lier for the third time.

Kai

[tomgdrums]

Am I being hypocritical?

Yes, you are.

I don't have any holy testing groups so I am not lying.

You lied and lie again. You know what Gino meant by "holy" testing groups. If you deny again, you are a lier for the third time.

Kai

How do you know I am being hypocritical? I have never praised a testing group. (HTGs!)

And actually I use some free software when it is the best choice for me. In chess software the best choice for me happens to be Shredder, Hiarcs and Junior.

But I use open office for word processing. And I use e-sword for Bible study. Both of those are free. So I don't "need" to buy stuff to like it. I use the software that best serves my needs in any given endeavor...as long as it is legal software.

Have you read Glenn Beck's book??