How many rating points is a photographic memory worth?

I think if you tried to take it a step further, understanding rather than just memorizing, improvement would be enormous. Thouroughly analyze 1,000 master games and you might just become a master yourself.

 Ask Fischer...

If you run "Searching for Bobby Fischer" in reverse it says  "Bobby is dead man, Bobby is dead man."

I think anyone should be able to get to 2200 by photographic memory alone

Slyusarchuk for example became a top GM purely by memory tricks

127

I read where Dan Heisman says there are 2000 common tactical patterns - so someone with a photographic memory who could learn all those would seem to have a big advantage in the middle game.

Yes, but assuming that most people have  some analysis skills, if you have numerous games memorized you would see patterns emerging.

Reading through this thread, I begin to fear I invest way too much time into opening theory ... but then again, I especially work through the books where opening lines are conveyed via annotated master games, so I hope this helps in other ways too.

Regarding eidetic memory ... in a way, everyone has it, but unfortunately it lasts only 250 ms. I think no cases of eidetic long term memory are really scientifically verified until now.

As I have said over and over, calculation of the best move is the most important skill. That proves it in a nutshell. I do think if you can memorize a bunch of material it can free up part of your mind for calculation though.

Oh, totally, I think being able to pick the best move in every situation would be worth at least a couple rating points.

How to improve memory?

It is literally priceless at any level. Thats why playing the computer on the hardest level is nearly impossible. It always picks the best move, no matter the position.

Well, what if you could pick a better best move than the computer?  That's what I'm talking about.

It is not theoretically impossible...but good luck...from what I have gathered, the level they can currently play at is basically humanly impossible to match, let alone defeat. Kasparov's 2nd go around with deep blue pretty much proves this. The computers are much better since then.

You would literally have to have an opening completely calculated to have a chance.

You mean, like, the best best move?

You have to have it in some way! Visualizing the board is what chess players do -- you imagine the changes made in the position with each move in your head, starting from your cerebral copy of the current position. I usually use the board in front of me of course, and pretend that the pieces are moving on the board as I calculate.

Strong players do it all the time. I'd randomly guess that the best humans understand at least 10% of chess positions better than computers.

Even if I'm wrong, it seems you're putting way to much faith in engines. The fact that they, and centaurs, still lose to eachother shows that they don't also know the best moves.

Right. It's a skill that really does get easier as you continue to do it every game you play. It used to seem really awkward for me, but now, it doesn't feel so strange to imagine pieces changing their habitats on a static board position.