Will a human ever make 2900?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

We a human ever make 2900

Poll ended at Mon Mar 16, 2015 1:41 am

Yes
47
90%
No
5
10%
It's not possible
0
No votes
 
Total votes: 52

User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Will a human ever make 2900?

Post by reflectionofpower »

If they allowed Stockfish 6, Komodo 8 or Houdini 4 in tourneys what do you think its rating would be. I would would venture to say it would be at least 2900 if not 3000 against humans. In fact you can not even do a rating calculation on FIDE website with anything over a 2900, you can 2900 but once you try 2901 it says invalid rating.

A human will probably never reach 2900 although Carlsen is at 2863 with Caruana at 2802 & Nakamura at 2798 respectively. SO only 2 players in the world are 2800 at stc but the one you want to watch is Nakamura. I have been watching him and his style is very active and he plays offbeat stuff. He will be WC sometime soon. He strength is growing still and he is tempering some of his traits that in the end would give him a 0. His blitz is insanely psychotic but powerful.

Even Nakamura admitted and the clip is on YouGoob where he is holding an open forum at the St. Lousi Chess Center that the chess software would destroy him.

I have pic of Transwarp(C) getting over the 4000 mark. You know how hard that is. You win one game agaist a 3300 play and you get 1 point!

Here's the snap of that moment:

Image

I was sincityandmarv at that time on ICC and I mentioined how seeing 4000 was like seeing a 3 breasted women which was weird but good.

They had a scene like this is "Paul" (2011) excellent comedy by the way.


Two British comic-book geeks traveling across the U.S. encounter an alien outside Area 51.
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie
supersharp77
Posts: 1242
Joined: Sat Jul 05, 2014 7:54 am
Location: Southwest USA

Re: Will a human ever make 2900?

Post by supersharp77 »

First they have to make 2800 first.......(and I am not talking about "ratings inflation') If anyone thinks this current crop is better than Fischer...Kasparov...Tal...Botvinnik or Petrosian.....they are nuts!!
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: Will a human ever make 2900?

Post by reflectionofpower »

supersharp77 wrote:First they have to make 2800 first.......(and I am not talking about "ratings inflation') If anyone thinks this current crop is better than Fischer...Kasparov...Tal...Botvinnik or Petrosian.....they are nuts!!
That's not an easy one to differentiate. Looking at chess a hundred years there is a definite demarcation of strength. Ironically if you look at boxing when you see them slugging it out (no such thing as a TKO) for 15 rds. in the pre 1950's you realize they were tougher then the fighters of today, baseball is the same way. I personally feel that of the 5 you mentioned Petrosian would have gotten his azz handed to him in a basket against today's top GM's.
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie
syzygy
Posts: 5566
Joined: Tue Feb 28, 2012 11:56 pm

Re: Will a human ever make 2900?

Post by syzygy »

supersharp77 wrote:First they have to make 2800 first.......(and I am not talking about "ratings inflation') If anyone thinks this current crop is better than Fischer...Kasparov...Tal...Botvinnik or Petrosian.....they are nuts!!
The Carlsen of now would easily outplay the Fischer of then.

Not that this is a fair comparison; Magnus has the benefit of having much better access to much more knowledge as well as to computer analysis.

If you take Elo as a measure of playing strength relative to your contemporaries (which strictly speaking it is according to its definition), then there has clearly been rating inflation.

But if you take Elo as a measure of absolute playing strength, then I'm pretty sure there has been rating deflation.
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: Will a human ever make 2900?

Post by reflectionofpower »

syzygy wrote:
supersharp77 wrote:First they have to make 2800 first.......(and I am not talking about "ratings inflation') If anyone thinks this current crop is better than Fischer...Kasparov...Tal...Botvinnik or Petrosian.....they are nuts!!
The Carlsen of now would easily outplay the Fischer of then.

Not that this is a fair comparison; Magnus has the benefit of having much better access to much more knowledge as well as to computer analysis.

If you take Elo as a measure of playing strength relative to your contemporaries (which strictly speaking it is according to its definition), then there has clearly been rating inflation.

But if you take Elo as a measure of absolute playing strength, then I'm pretty sure there has been rating deflation.
I am not so sure of Carlsen manhandling Fischer. You always hit someone who knows your Achilles Heel and the styles clash. I will say though that Carlsen vs. Fischer would've been more exciting than Carlsen vs. Anand.
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie
syzygy
Posts: 5566
Joined: Tue Feb 28, 2012 11:56 pm

Re: Will a human ever make 2900?

Post by syzygy »

reflectionofpower wrote:I am not so sure of Carlsen manhandling Fischer. You always hit someone who knows your Achilles Heel and the styles clash. I will say though that Carlsen vs. Fischer would've been more exciting than Carlsen vs. Anand.
But the Carlsen of now knows the style of the Fischer of then. And I'm afraid Carlsen understands the games Fischer has played better than Fischer ever did himself. Even if we could somehow take that advantage away from Carlsen, Fischer would still be lucky to get past the opening.

I agree, we should let them play Fischer Random :-)
Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Re: Will a human ever make 2900?

Post by Terry McCracken »

In Capablanca's day they thought no one could ever reach much more than ~2600-2700 playing strength so I see no real reason why 2900 would be an impossible or even improbable barrier. The human limit is likely closer to 3000 elo, maybe even a bit more.
Terry McCracken
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: Will a human ever make 2900?

Post by reflectionofpower »

syzygy wrote:
reflectionofpower wrote:I am not so sure of Carlsen manhandling Fischer. You always hit someone who knows your Achilles Heel and the styles clash. I will say though that Carlsen vs. Fischer would've been more exciting than Carlsen vs. Anand.
But the Carlsen of now knows the style of the Fischer of then. And I'm afraid Carlsen understands the games Fischer has played better than Fischer ever did himself. Even if we could somehow take that advantage away from Carlsen, Fischer would still be lucky to get past the opening.

I agree, we should let them play Fischer Random :-)
Knowing a style and being able to overcome is 2 different worlds. Fischer get past the openings? Fischer would pour over openings again and again. His opening prowess was universally feared whereas Carlsen really does not focus to much on the openings which I think is a sound pracitcality.
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie
Dann Corbit
Posts: 12542
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Will a human ever make 2900?

Post by Dann Corbit »

To paraphrase Inigo: "That number you keep using... I do not think it means what you think it means."

An Elo of 2900 is not some sort of absolute measure of chess strength. It represents a relative value compared to a standardized pool of talent. If the talent pool changes, the number moves, even though the real strength did not change.

You can run most Elo calculation programs and get them to parrot back any number that you like by simply setting a different baseline (or starting Elo for the members of the pool).

A "recalibrated" max rating list has RJF at 2881 for a period of one year:
http://www.chessmetrics.com/cm/CM2/Peak ... 0000010100

and a total of 14 over 2800
  • Player Name Average Rating 1 year peak range
    #1 Bobby Fischer 2881 1972-Jan through 1972-Dec
    #2 Garry Kasparov 2879 1990-Jan through 1990-Dec
    #3 Mikhail Botvinnik 2871 1946-Jan through 1946-Dec
    #4 José Capablanca 2866 1919-Jan through 1919-Dec
    #5 Emanuel Lasker 2863 1894-Jan through 1894-Dec
    #6 Alexander Alekhine 2851 1931-Jan through 1931-Dec
    #7 Anatoly Karpov 2842 1989-Jan through 1989-Dec
    #8 Viswanathan Anand 2828 1998-Jan through 1998-Dec
    #9 Vladimir Kramnik 2822 2002-Jan through 2002-Dec
    #10 Siegbert Tarrasch 2818 1895-Jan through 1895-Dec
    #11 Géza Maróczy 2815 1906-Jan through 1906-Dec
    #12 Harry Pillsbury 2813 1901-Jan through 1901-Dec
    #13 Viktor Korchnoi 2803 1978-Jan through 1978-Dec
    #14 Wilhelm Steinitz 2802 1886-Jan through 1886-Dec
User avatar
reflectionofpower
Posts: 1610
Joined: Fri Mar 01, 2013 5:28 pm
Location: USA

Re: Will a human ever make 2900?

Post by reflectionofpower »

Dann Corbit wrote:To paraphrase Inigo: "That number you keep using... I do not think it means what you think it means."

An Elo of 2900 is not some sort of absolute measure of chess strength. It represents a relative value compared to a standardized pool of talent. If the talent pool changes, the number moves, even though the real strength did not change.

You can run most Elo calculation programs and get them to parrot back any number that you like by simply setting a different baseline (or starting Elo for the members of the pool).

A "recalibrated" max rating list has RJF at 2881 for a period of one year:
http://www.chessmetrics.com/cm/CM2/Peak ... 0000010100

and a total of 14 over 2800
  • Player Name Average Rating 1 year peak range
    #1 Bobby Fischer 2881 1972-Jan through 1972-Dec
    #2 Garry Kasparov 2879 1990-Jan through 1990-Dec
    #3 Mikhail Botvinnik 2871 1946-Jan through 1946-Dec
    #4 José Capablanca 2866 1919-Jan through 1919-Dec
    #5 Emanuel Lasker 2863 1894-Jan through 1894-Dec
    #6 Alexander Alekhine 2851 1931-Jan through 1931-Dec
    #7 Anatoly Karpov 2842 1989-Jan through 1989-Dec
    #8 Viswanathan Anand 2828 1998-Jan through 1998-Dec
    #9 Vladimir Kramnik 2822 2002-Jan through 2002-Dec
    #10 Siegbert Tarrasch 2818 1895-Jan through 1895-Dec
    #11 Géza Maróczy 2815 1906-Jan through 1906-Dec
    #12 Harry Pillsbury 2813 1901-Jan through 1901-Dec
    #13 Viktor Korchnoi 2803 1978-Jan through 1978-Dec
    #14 Wilhelm Steinitz 2802 1886-Jan through 1886-Dec
Good point!
"Without change, something sleeps inside us, and seldom awakens. The sleeper must awaken." (Dune - 1984)

Lonnie