Author: Christophe Theron
Date: 12:50:22 04/23/00
Go up one level in this thread
On April 23, 2000 at 04:53:14, Pete Galati wrote:
>On April 23, 2000 at 04:12:12, Christophe Theron wrote:
>
>>On April 23, 2000 at 02:31:28, Pete Galati wrote:
>>
>>>On April 23, 2000 at 02:12:19, Chessfun wrote:
>>>
>>>>On April 23, 2000 at 01:53:55, Pete Galati wrote:
>>>>
>>>>>On April 23, 2000 at 01:38:20, Chessfun wrote:
>>>>>
>>>>>>On April 23, 2000 at 01:31:17, Pete Galati wrote:
>>>>>>
>>>>>>>On April 23, 2000 at 00:48:44, Chessfun wrote:
>>>>>>>
>>>>>>>>On April 22, 2000 at 14:46:28, Pete Galati wrote:
>>>>>>>>
>>>>>>>>>On April 22, 2000 at 15:02:19, Chessfun wrote:
>>>>>>>>>
>>>>>>>>>>On April 22, 2000 at 14:49:59, Pete Galati wrote:
>>>>>>>>>>
>>>>>>>>>>>On April 22, 2000 at 14:20:57, A.L.Mourik wrote:
>>>>>>>>>>>
>>>>>>>>>>>>Hello dear CCC friends,
>>>>>>>>>>>>
>>>>>>>>>>>>Although earlier reports from e.g. Jouni Uski, suggest an enormous increase in
>>>>>>>>>>>>strenght for Crafty 17.10
>>>>>>>>>>>>Nuun 2 match result Fritz6 against Crafty 17.10 ends in a very clear
>>>>>>>>>>>>29,5-10,5!! victory for Fritz6.
>>>>>>>>>>>>Played on PII 400 8mb for HT, Timecontrol 5 min + 3 sec per move.
>>>>>>>>>>>
>>>>>>>>>>>There isn't generally an enormous increase in strength from version to version
>>>>>>>>>>>of any program, that's unrealistic to expect there to be.
>>>>>>>>>>>
>>>>>>>>>>>Your time controls are a bit short there.
>>>>>>>>>>>
>>>>>>>>>>>Pete
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>Why didn't you say or ask that of Jouni when he posted?
>>>>>>>>>>as all he said was blitz, that may even have been faster
>>>>>>>>>>than this.
>>>>>>>>>>Thanks.
>>>>>>>>>
>>>>>>>>>Sorry, I don't read all posts, this one caught my eye because it was at the top
>>>>>>>>>of the board.
>>>>>>>>
>>>>>>>>
>>>>>>>>You had posted in the thread I referred to.
>>>>>>>>Which was Sensation Crafty 17-10 beats F6a in nunn1 .
>>>>>>>>
>>>>>>>
>>>>>>>You'd need to post the html of the post where I responded to that, setting the
>>>>>>>filters at 7 days and doing a search for "Sensation Crafty 17-10" does not turn
>>>>>>>anything up. I don't recall responding in such a thread. But if you're
>>>>>>>attempting to put me on trial for some comment that I may or may not have made
>>>>>>>about Crafty, then you simply have too much time on your hands.
>>>>>>
>>>>>>
>>>>>>It is fairly simple to set search for author.
>>>>>>You made no comment for a trial, it was the lack of request
>>>>>>for the time control of the blitz games then the statement in
>>>>>>this thread that it needs longer time controls.
>>>>>>
>>>>>>But, since your last comment obviously means you take every question
>>>>>>of your postings as an offense and feel the need to take a stab at
>>>>>>anyone who questions anything you write, even the obvious questions
>>>>>>posed by Christophe that you fail to understand, then don't worry I
>>>>>>won't bother to reply in future.
>>>>>>
>>>>>>Thanks.
>>>>>>
>>>>>
>>>>>Ok, I searched for "Jouni" and found what you're talking about, but you left out
>>>>>a ":" when you quoted the name of the thread so my search didn't turn up
>>>>>anything.
>>>>
>>>>
>>>>Wouldn't matter what thread I quoted. A simple search by author
>>>>would give you all posts you had made.
>>>>
>>>>
>>>>>First of all, in that thread, I was responding to a comment that Fernando made,
>>>>>not anything that Jouni said.
>>>>
>>>>You posted in a thread about crafty beating F6 at nunn 1 in blitz.
>>>>You never questioned the time controls as you did in this thread.
>>>>The difference Crafty was alledged to have won in the other thread.
>>>>It lost in this one.
>>>
>>>You're still trying to ignore that I was responding to Fernando's post, and not
>>>to Jouni's, I don't see that bothering anyone else.
>>>
>>>>
>>>>>Second, I didn't take back anything.
>>>>
>>>>Who asked you to take back. A simply asking the same question in the
>>>>other thread was apt.
>>>>
>>>>>Third, Christophe was not being clear what he was asking "why" about.
>>>>
>>>>The question he asked is simple enough. You said these blitz times were
>>>>too fast and that they are too fast to judge the strength of a program.
>>>>He asked, why and can you explain. Since you had made these statements
>>>>you must have evidence that these statements are correct.
>>>>Seems simple to me.
>>>
>>>No, _you're_ asking that question, Christophe didn't, he wasn't clear what he
>>>was asking about. Blitz games don't allow a program to search deep enough,
>>>that's obvious enough that the question does not need to be asked.
>>
>>
>>This is a very frightening answer.
>>
>>Is the earth flat? Yes it is, that's obvious enough that the question does not
>>need to be asked.
>>
>>Pete, I fear you are not on the right track anymore.
>>
>>
>>
>> Christophe
>
>Are you saying that Blitz games are an adaquate test of a program's strength?
Several points to think about:
1) What is the only authorized time control to use to decide program's strength?
Who decided this?
2) What is the time control most used in reality by chess programs users?
3) Where is your data about the difference in playing strength between blitz
and, say, 40 moves in 2 hours, of any program, on a significant number of games?
4) When you play a blitz on a very very fast computer, it is like playing the
same game at slow time controls on a much slower computer. What does "blitz"
mean in a world where you can find very fast and very slow computers?
5) What is the difference in search depth between a blitz game and a slow time
control game? Why do you expect that one program will benefit more than its
opponent from this deeper search?
6) What evidence do you have that the curve of "relative strength vs time
control" is monoton? I mean it is possible that prog A is better than prog B at
blitz, then prog B is better at 40 moves in 2 hours, then prog A is again better
at move in 1 day. If you take for granted that prog A wins at blitz, but would
lose at 40 moves in 2 hours, you cannot reject my proposition so easily. And
which program is stronger in this case?
The assumption that the relative strength of chess programs changes with the
time controls used is one of the many legends that people like to believe in.
It might or might not be true for a given pair of programs, but it is NOT a
general rule. You can assert something like this only if you have done a careful
and long experiment.
The best guess when you have not checked it carefully is to assume that the time
control does not affect the relative strength of 2 programs. By assuming this,
you will be right most of the time.
Take a look at Chessfun's current experiment. While each match is not long
enough to get a good statistical measure, what does it suggest? If you want more
data, it's easy. Take some SSDF matches, and replay them at home at blitz time
control.
In short, I'm not saying that blitz is always an adequate test of program's
strength. I'm just saying that it is very unwise to assume it is not, and I
don't see any reason to reject blitz matches results.
Christophe
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.