Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Hiarcs v Fritz opening books

Author: Steve Lopez

Date: 22:11:48 07/01/99

Go up one level in this thread


On July 01, 1999 at 22:28:02, Melvin S. Schwartz wrote:

>I posted the move order for you to evaluate - did you?

Yes, and did so in another post in another thread, after Terry was good enough
to send me the correct move order. The Sicilian line you keep talking about is
*not* in Hiarcs' original opening book. If it is in yours, you somehow added it,
regardless of your protestations to the contrary. You mentioned a move being
shown in red -- the original opening book ends three or four ply before that
move. How, then, did that line get in there? Who put it in there -- to use
another of your favorite expressions: you? Me? The man in the moon?

>In my two games at 40/2 against Hiarcs, I have a plus score from the opening on.
>My point of posting the two openings played by Hiarcs was to show how it can
>play bad openings.

I have yet to see the correct move order for the Bishop's Opening game. The line
you posted in another thread contains the illegal White move f5-f6.

>The two games are not finished but it is not very relevant
>because I am not a Grandmaster. If I can get the better of Hiarcs in some
>openings, something is wrong with the opening book - and that is my point! Do
>you understand what I am talking about?

Yes, but my point is that in the Sicilian game, the line you quote as being part
of the opening book is not in there. Terry concurs.

>I notice an angry tone in your message and probably it is in fact due to my
>hasty reply to you.

No, it's due to this repeated "Confusionbase" stuff. You called me on the
telephone *repeatedly* with the same three questions: opening book choice (which
one is "best), hash table size, and tablebase use. All have been addressed in
T-Notes. I stated in a past issue (I forget the date, but I think it was the
9/27/98 issue) that Hiarcs6 uses only about two-thirds of the available hash
tables when the Fritz formula was used. I see nothing to indicate that it's any
different with Hiarcs7.32, and comments I've seen in CCC (and elsewhere on the
Interrant) indicate that this is indeed the case. I've also not received any
feedback to the contrary from ChessBase in Germany (and believe me, they *do*
correct me when I'm wrong and I *do* print their corrections in later issues of
T-Notes).

The fact is, Mel, while there is some "confusion" about a couple of the features
in Hiarcs7.32, the Fritz manual supplied on the CD is sufficient for explaining
the vast majority of the features in the program since the interfaces are
identical. The remaining features are covered in the Hiarcs7.32 booklet that's
packaged in the jewel case. Your near-constant sarcastic characterization of the
company as being slipshod and uncaring is grating to say the least, especially
in light of the *hours* that you and I have spent on the phone and the
additional *hours* I've spent in e-mail exchanges with you in trying to answer
your questions. It's *doubly* grating in light of the tendency you have to play
"both ends against the middle". After I would answer one of your questions, you
would e-mail Andre to ask it again. If he gave you a different response, you
would contact me again to tell me that the response differed. I would tell you
to listen to Andre's suggestion instead (since he has direct face-to-face
contact with the programmers) and you would respond that I originally told you
something different than what Andre said. Then, when I would tell you that I
sometimes base my replies on evidence from other users that's presented in CCC
and rec.games.chess.computer, you would tell me that this "isn't very
scientific" -- yet you later had no problems in telling me that I was wrong in
one of my responses to you because "so-and-so in CCC" said something different.

I think we've done a lot to attempt to alleviate your concerns about the
program. I'm sorry that the answers have not been to your liking, but I've given
you all of the information I have. The main problem I have with our
phone/e-mail/CCC exchanges is that you keep "moving the goalposts" -- the recent
posts on resetting the opening book weights being a case in point. You asked a
question, I answered it (as well as gave a followup or two to other posters on
the same subject), got a "thank you" from another CCC reader, and your reply was
"thanks for what?" I'm sorry you didn't like/understand the answer, but that
doesn't change the fact that it *is* the answer.

So the fact is that from my point of view your questions can't be answered to
your satisfaction. That's fine and I understand this. Either you don't
understand the answers or they're not complete/accurate/authoritative enough for
you. I freely admit that I'm not the divine oracle of ChessBase and that my
answers to customers are sometimes not sufficient or correct. But in light of
the sheer amount of time and effort I've spent in attempting to answer your
inquiries, I find your characterization of ChessBase as uncaring to be grossly
unfair.

Consequently, my only recommendation to you at this point is for you to return
the product for a refund since it evidently doesn't meet your needs and
requirements.

> I didn't have much time to reply and so quickly composed the
>message while frustrated at your response. I get a lot of frustration from
>ChessBase. You recommended you know who to ask questions...well...I told him of
>my problem registering Hiarcs and he asked for my address. I sent him my address
>and haven't heard anything in over a week.

I can't answer for Andre, nor would I presume to. However, *my* questions to
Andre and the programmers have also gone unanswered, so my assumption is that
they're presently heavily involved with the various computer-computer and
computer-human events presently being held.

>Also, what about the formula for hash
>tables on the Hiarcs program? Is it correct? I asked you know who and he said
>that Hiarcs doesn't need as much as Fritz - this is an answer?

It is. See above.

>Did anyone from
>ChessBase use the Hiarcs program before it was released?

Again, I can't answer for the programmers. However, my information in T-Notes is
based *heavily* upon my own personal experience and experimentation with the
programs (e.g. I'm not *fed* the information by the programmers -- I discover
most of it myself and when I'm wrong, they correct me). Mathias Feist reads
T-Notes religiously. If I write something and don't receive a correction from
him, then it follows that the information is correct.

>Do you see a Chevrolet
>manual in a Ford?

You make remarks like this and then wonder why my last reply seemed a bit curt?

>Why do consumers have to be running around with their tail
>between their legs trying to get answers to the program?

Sure, Mel, and I guess the *hours* of phone calls you made to me (at ChessBase's
expense) don't count for much in the area of customer service.

>You're a nice guy, Steve, and I appreciate the work you do to help people.
>However, you do work for ConfusionBase, err, I mean ChessBase,

See above regarding "Confusionbase".

>Do you or anyone at ChessBase know the hash table
>formula for Hiarcs? Hmmm?

See above. Two-thirds of the Fritz formula. 66.67%. Yes, this is from personal
experience -- it did not come straight from the lips of Mark Uniacke. *HOWEVER*
at the 40/2 time controls you're playing at, I can *guarantee* you that Hiarcs
is filling the hash tables, no matter what size you have them set.

>All this has gotten off the issue I raised about some opening book lines in
>Hiarcs. If you like, I can e-mail the present position where I have a plus score
>in both games due to bad opening lines by Hiarcs.

You can do that or else post them here. I've already addressed the Sicilian line
in another thread. The Bishop's Opening post I saw contained an illegal move.

-- Steve Lopez



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.