Author: Vine Smith
Date: 00:36:46 05/14/02
Go up one level in this thread
On May 14, 2002 at 01:29:59, Harald Faber wrote:
>On May 13, 2002 at 22:30:32, John Merlino wrote:
>
>>On May 13, 2002 at 20:42:55, Terry Ripple wrote:
>>
>>>On May 13, 2002 at 20:36:44, John Reynolds wrote:
>>>
>>>>
>>>>
>>>> Firstly, I have heard reports of Phony Hiarcs being passed around, I suspect
>>>>there are some people testing the Psuedo Hiarcs. My own results have shown
>>>>Hiarcs8 to be exceptionally strong, comparable to Fritz for sure. Number 2: It
>>>>may be that the book testers are using with Hiarcs is Bad, Personally I am using
>>>>a 500 mb book I downloaded from the net. There is no other Logical explanation
>>>>for all these negative Posts. Also it is possible that the program is not being
>>>>tested at longer time controls, all my games are atleast game in 60. Any
>>>>thoughts?
>>>---------------
>>>Well this is interesting news, as one of the testers shows Hiarcs 8 not
>>>performing very well at game in 60 minutes! Maybe some of the problems are due
>>>to Hiarc's opening book!
>>>
>>>Regards,
>>> Terry
>>
>>Perhaps, but you shouldn't have to download a 500MB opening book for the program
>>to play well.
>>
>>jm
>
>Hiarcs 8 comes with an own, dedicated (! maintained by Eric Hallsworth in >1
>year) opening book which is already large enough (280MB), there is absolutely no
>need to download a 500MB book. I have not seen Hiarcs playing badly. Hiarcs
>sometimes sees a desaster to the own king later than the attacking programs
>(there are some smart examples I have seen during my running Hiarcs8-GambitTiger
>2 aggressive match but Hiarcs can keep the edge, at the moment 3.5-3.5) but this
>is no issue of the opening book.
>
>In this regard, where does this 500MB book come from, thinking of copyright? Did
>someone create it or is this an old powerbook?
This game was posted a couple of days ago:
[Event "60 10"]
[Site "?"]
[Date "2002.05.12"]
[Round "?"]
[White "Nimzo 2000b"]
[Black "Hiarcs 8"]
[Result "1-0"]
[ECO "C47"]
[Annotator "PIII1200-Athlon1333,Ponder=On"]
[PlyCount "37"]
[TimeControl "3600+10"]
{288MB, Hiarcs8.ctg, PIII1200
} 1. e4 {[%emt 0:00:00]} e5 {[%emt 0:00:00]} 2.
Nf3 {[%emt 0:00:06]} Nc6 {[%emt 0:00:00]} 3. Nc3 {(Lb5) [%emt 0:00:03]} Nf6 {
[%emt 0:00:00]} 4. d4 {[%emt 0:00:04]} exd4 {[%emt 0:00:00]} 5. Nd5 {
(Sxd4) [%emt 0:00:03]} Nxe4 {[%emt 0:00:00]} 6. Qe2 {[%emt 0:00:04]} f5 {
[%emt 0:00:00]} 7. Ng5 {[%emt 0:00:04]} d3 {[%emt 0:00:00]} 8. cxd3 {
[%emt 0:00:04]} Nd4 {[%emt 0:00:00]} 9. Qh5+ {[%emt 0:00:04]} g6 {
[%emt 0:00:00]} 10. Qh4 {[%emt 0:00:03]} c6 {[%emt 0:00:00]} 11. dxe4 {
[%emt 0:00:04]} cxd5 {[%emt 0:00:00]} 12. exd5 {[%emt 0:00:03]} Nc2+ {
[%emt 0:00:00]} 13. Kd1 {[%emt 0:00:03]} Nxa1 {[%emt 0:00:00]} 14. Qd4 {
[%emt 0:00:03]} Rg8 {[%emt 0:00:00]} 15. d6 {[%emt 0:00:04]} Bxd6 {
[%eval 186,13] [%emt 0:29:41]} 16. Qxd6 {[%emt 0:00:05]} Qe7 {
[%eval 191,12] [%emt 0:01:10]} 17. Qd5 {(Lf4) [%emt 0:00:05]} Rf8 {
[%eval 565,13] [%emt 0:11:30]} 18. Bb5 {[%emt 0:00:38]} Nc2 {
[%eval 562,12] [%emt 0:00:16]} 19. Kxc2 {[%emt 0:00:06]} 1-0
If you look at when it left book, and check the opening references, you'll see
that it willingly entered a known trap in the Belgrade Gambit analyzed by Keres
about 50 years ago, and is immediately lost upon exiting book. How can you claim
there is no need for an alternative book?
Two more examples I can recall -- Hiarcs as Black, 1.e4 e6 2.Qe2 Nc6 3.Nf3 Be7?.
The program must have left book at move 2 or 3, because the only reasonable
followup to 2...Nc6 is 3...e5. Now the Chigorin French with 2.Qe2 isn't wildly
popular or anything, but omitting it entirely from the book is inexcusable, and
Hiarcs lost horribly because of this. Especially when you consider all the
weirdness that DOES make it into the book, like 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6
4.Bc4?!, which Hiarcs played in another game, when Black can just answer
4...Nf6, and be playing a Two Knights Defense with the extra tempo ...a6 thrown
in, so has probably already equalized.
Then today, I saw Hiarcs as Black had played 1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4
Qc7, and was thrown out of book by 5.Nb5, even though this reply had been played
in at least one master level game before. Why include a trick line like 4...Qc7
(because that's all it is) in the book to have it exit at move 5, thus saving no
time on the clock versus the opponent?
This book seems just dreadful to me, and if I owned Hiarcs (fortunately, I held
off, pending reviews), I probably would spend the countless hours necessary to
download the Lunsen book, since anything else would have to be better, possibly
even not using a book at all, which would be interesting to test.
Regards,
Vine Smith
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.