Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fritz unleashed = 3123?

Author: Robert Henry Durrett

Date: 07:16:13 06/06/02

Go up one level in this thread


On June 06, 2002 at 05:16:21, Uri Blass wrote:

>On June 06, 2002 at 00:32:11, Joseph Merolle wrote:
>
>>No , In  blitz and bullet chess computers can hit 3400 check out the high
>>ratings on icc.
>
>What is the results that the best humnans get in blitz and bullet?
>
>If they get higher rating in blitz or bullet relative to their rating at
>standard time control then the numbers seem to mean nothing.
>
>Uri

On June 06, 2002 at 05:16:21, Uri Blass wrote:

>On June 06, 2002 at 00:32:11, Joseph Merolle wrote:
>
>>No , In  blitz and bullet chess computers can hit 3400 check out the high
>>ratings on icc.
>
>What is the results that the best humnans get in blitz and bullet?
>
>If they get higher rating in blitz or bullet relative to their rating at
>standard time control then the numbers seem to mean nothing.
>
>Uri

Unclear.

At least is makes it possible to compare the chess engine to the top players at
icc, both human and computer.  The ratings are, to a first approximation, an
indication of playing strength although, as already noted by Hyatt, it's
possible to run an engine's rating up artificially [by "cheating"].

The blitz ratings are based on a pool which includes many more humans than
computers.  Perhaps the same is true for standard.

Do computers excel over humans at Blitz?  Probably not.  More precisely, don't
humans perform relatively poorly at blitz?  Probably so.

Standard time limits make it too easy for the human to "cheat."  Also, humans
may do better at standard even if they don't cheat.

Another consideration which may be a contributing factor is that there are many
more blitz games played on that particular server than standard.

Interesting to note is that the strongest human players almost never play
standard time limit chess at icc.  Perhaps they don't like standard?  More
likely, they know that it's too easy for weaker players to cheat at standard.
They may also be reluctant to reveal their latest novelties.

The ratings do serve a useful role, at least.  I hate playing someone who is
going to slaughter me.  By limiting the ratings of those who can challenge me, I
have more fun.  [Or did, when I used to play chess.  Now, I do analysis
exclusively.]

Another point:  Blitz ratings being higher than standard may just be a
reflection of the popularity of blitz.  Anything that causes rating creep in the
ratings will occur much more frequently in a more active pool.

Anyway . . .

Bob D.







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.