Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Ruffian better than Shredder 7.04??

Author: Frank Quisinsky

Date: 03:27:24 01/22/04

Go up one level in this thread


On January 22, 2004 at 04:51:25, Uri Blass wrote:

>On January 22, 2004 at 04:25:25, Frank Quisinsky wrote:
>
>>On January 22, 2004 at 02:41:13, José Carlos wrote:
>>
>>>  I know your opinion: pick the money and disappear for a while.
>>>
>>>  José C.
>>
>>Jose,
>>
>>sorry, but I have a lot to do and you have to wait of it.
>>Maybe 20 years ... hope so!
>>
>>In the next year I have too many ideas to make a little bit.
>>
>>I know that you are now disappointed (unkown the reason, maybe you are jealous?)
>
>The reason is clear.
>
>People deserve to get correct information when they buy a new program and nobody
>told them that Ruffian2 is less tested than Ruffian Leiden so they made wrong
>assumption and tested only Ruffian2.
>
>Ruffian2 is not called something like Ruffian 23.06.2003 or beta ruffian or
>ruffian experimental version that suggest that it is not tested enough so people
>could not know that it was not tested enough.
>
>Situation with free versions is different because people do not pay for them.
>professional behaviour is first telling people correct information about what
>they get.
>
>I can be only sorry that instead of trying to learn for the future to give
>correct information you disagree with Jose.
>
>Uri

Hi Uri,

sorry, but I cann't saw only one point which is interesting to disucss.

We have two Ruffian versions on the CD.

Ruffian Leiden (the version won the tournament in Leiden).
Ruffian 2.0.0
and the free Ruffian versions ...

For me it's absolutly clear that every programmer try to make his program
stronger and after compiling you cann't know (in the most cases) ... is my new
version stronger or not. This is normaly, right or not?

After that the programmer and beta tester have to test a new version. In my
opinion is this clear too. With other words ... we are speaking about absolutly
clear things!

All is not a big secret!

If you or others search a version which is stabil and strong we must test in a
small group a program vs. x other programs more as 6 months with different time
controls. And this is not possible!

We test Ruffian now one year (different versions) and the result by Per-Ola is
Ruffian Leiden which won the Dutch-open 2003 in front of Rebel, Tiger, Sjeng,
King and others. This results are available on the commercial CD-Rom and the
programmer closed version 1.x with version 2.0.0!

Version 2.0.0 with small changes are on the CD available too.
Ruffian go in production after Leiden and we can wait one year and test Ruffian
2.0.0 ... Ruffian will win the tournament in Leiden 2004 and we start a new test
of one year and as result Ruffian is available in 10 years!

The way Per-Ola go is right.

At the moment Per-Ola is working on an update for Ruffian 2.0.0 and maybe we
should test this update six months before we give this update free? The most
Ruffian customers are not very happy about it if we wait a half year :-))

The complete disucss about it is in my opinion a discuss without a result
because all is clear and normaly.

You can be sure that we test newer Ruffian versions but you can be sure too that
we don't know after a test the results vs. x available engines. And all is
possible, possible that a newer version is not stronger as an older version.
It's possible but normaly ... I am sure ... we all can see that newer comemrcial
versions are stronger.

Nimzo 2000 to Nimzo 7-32 (3 or 5 days different)
Not stronger
Nimzo 7.32 to Nimzo 8 (not stronger)

Chess Tiger 14 to Chess Tiger 15 (not a big new step)

Shredder 6 to Shredder 7.04
A very big step

But in fact ...
All programmers try to make his programs stronger and this is great and I can
live with it if a new version not directly stronger as more as 50 ELO compare to
a new version.

We all have a main problem:

We all (better 95% of the readers here, I am sure) have no chances vs. chess
engines. Meaning the most of the readers losed 95% of the games vs. number 150
from the freeware engine rating list. And this is a point which forget the power
user. But we all are experts and must give comments to programs which are 1.000
ELO stronger.

And to test a commercial program vs. other commercial programs alone is boring
because we have not enough commercial programs which are compatible to
standards, speaking from WinBoard as main engine protocol and UCI as second free
engine protocol. I have no interest to test Ruffian under other GUIs which don't
supported standards, this is clearly not interesting for myself.

Hope this is enough bad English by myself for today :-)

Best
Frank



This page took 0.01 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.