Oh God, I'm old...what's a tablebase?
Tablebases correspondence legal?
Sort:
I'm not completely sure how it works or if it qualifies as an engine, but it can be used to find winning moves. Databases are legal; engines are not.
Oct 3, 2010
0
#4
After looking it up on Wikipedia: I'd say not good. It seems like a gizmo to solve a position by use of tree searches. Putting a position into a computer to get a solution is like asking a human for advice. Whereas databases are passive and act like a really huge book (with statistical outcomes), the tablebase seems to be doing the analysis.
I'm glad. I've expirimented a bit with tablebases, and they seem to take all the playing out of chess.
I've never used them in a game. 
If I'm not mistaken, USCF deals entirely with face-to-face games, so I'm pretty sure the answer to the specific questions posed is no.
I believe that the ICCF, on the other hand, does allow them.
I know with certainty that Chess.com does not.
Oct 3, 2010
0
#8
TheGrobe wrote:
If I'm not mistaken, USCF deals entirely with face-to-face games, so I'm pretty sure the answer to the specific questions posed is no.
I believe that the ICCF, on the other hand, does allow them.
I know with certainty that Chess.com does not.
USCF also deals with games on computer and postal (with actual post cards) chess as well. Makes you want to sign up, huh?
Is that right? I didn't realize -- I thought that was the domain of the ICCF alone.
Oct 3, 2010
0
#10
The USCF Correspondence Chess entry form has "e-mail rated events" listed. In general, though, USCF missed the boat by a couple of years on computer chess and will probably never regain the lost opportunity.
Estragon wrote:
In correspondence chess in all its forms, all major entities (including USCF, ICCF, and Chess.com among others) agree: databases are allowed, tablebases are NOT.
Tablebases are databases of certain endgames (those with six or fewer total pieces on the board) which have been absolutely solved, and contain the best play against any opposing line of play.
Because they represent perfect play of these endings, they are considered illegitimate as resources. They are in effect perfect engines for these endings, and engines are never allowed.
Exactly.
And if by some stroke of misfortune they are ever allowed, you'll have to assume that both players have access to them, at which point you may as well terminate the the game at the tablebase evaluation.
Are there any endgame databases that function only as databases, that is to say, just giving statistical information on moves with no analysis?
SoilentGreen wrote:
Are there any endgame databases that function only as databases, that is to say, just giving statistical information on moves with no analysis?
I don't think that would be practical. There are a gazillion possible endgame positions, of which the vast majority has never been reached, so it would be largely incomplete.
I'm glad of that also. This means once out of the opening, it's a true test of skill. I think I'll sign up for a correspondence match tomorrow.
Jan 18, 2020
0
#15
Frank_Taylor wrote:
This is an old thread, but in looking at what has happened more recently, the International Correspondence Chess Federation, with which the USCF correspondence program is affiliated (http://www.iccfus.com/links.htm), now allows use of both computer engines and tablebases. Chess.com does not. I suspect the reasoning of the ICCF/USCF is that the old rules were hard to enforce and essentially ended up being an unrealistic honor system. Chess.com has claimed that it can detect cheating, which I suspect is often true.
A different consideration: Despite some rhetoric to the contrary on this old threat and elsewhere online, Tablebases are not "engines." They are databases, just like opening databases. They literally have cataloged possible positions for certain numbers of pieces and the paths from those positions that constitute conclusions by draw or victory by white or black. There is no analysis. If you put a cataloged position into a tablebase, it looks up the possible resulting positions for each move and each derivative move. In many cases, this duplicates what old-fashioned endgame reference books (already allowed) have always done. But in this case, the database is extremely thorough, accurate and broad, though only for the small number of positions with very few pieces remaining. One should abide by the rules of the platform in which one is playing, but my viewpoint is that Chess.com ought to reconsider allowing tablebases for daily games. I'm not in favor of allowing engines that conduct actual analysis, but I don't think tablebases are the same thing.
Pretty certain USCF rated correspondence does not allow engine use.
For a little while, chess.com was allowing tablebases, but reversed the decision.
Sep 10, 2020
0
#16
USCF correspondence chess does not allow engine use. I'm not sure about Tablebases. The Walter Muir quad I'm playing in (no engines, but hosted on ICCF web site) says in the tournament instructions that you can make claims based on the 7 piece tablebases. But I don't know if that holds for all USCF correspondence games. They don't seem to address it. My Electronic Knights section instructions I got today just says no engines. I'm going ask about it. For US-based players, basically USCF and CCLA don't allow engines, but CCLA has has severe restrictions about the use of databases, too.
Sep 10, 2020
0
#17
douglas_stewart wrote:
USCF correspondence chess does not allow engine use. I'm not sure about Tablebases. The Walter Muir quad I'm playing in (no engines, but hosted on ICCF web site) says in the tournament instructions that you can make claims based on the 7 piece tablebases. But I don't know if that holds for all USCF correspondence games. They don't seem to address it. My Electronic Knights section instructions I got today just says no engines. I'm going ask about it. For US-based players, basically USCF and CCLA don't allow engines, but CCLA has has severe restrictions about the use of databases, too.
A table base enables an engine to play perfect chess, and see checkmates 549 moves ahead.
As it's an engine ofc it would be banned
Sorry for two correspondence threads; I just came up with this question, and it wouldn't let me modify the title of my other thread. Just wanted to know if the USCF allows tablebases.