Do I play like a computer?

Going over some of my games (mostly live games), I've noticed that my playing style is very mechanical and algorithmic, like a computer of the same rating I am. 

Do you agree?

You are strong mathematics person?

not phenomenally, but yes.

However, that's not the point.

Isn't it? The human mind has neural pathways that are distinct for various types of projects and talents. Exceptions will always be there but Mathematics people tend to be very mechanical and, huh, algorithmic; a computer's logical processes are very mechanical and algorithmic. Therefore, by definition, if you happen to be quite good at math you must be able to think in a logical and mechanical way, which would be the reason it shows up on the chess board.

Hmm... good point.

But I haven't done math in months...

Shouldn't matter. I suspect that while you would be a little rusty, the thought processes behind figuring out math problems have been kept intact simply by playing chess.

In short, I don't think you'd have any issue getting back into and solving a complicated equations.....assuming you remember the rules associated with it lol.

That's the point...

I know...which is why I stated it proving that, because of how your mind works, you will logically think in a way akin to that of a computer.

What's the most obvious difference between human and engine play? With the weak or down-regulated engines I've seen (below 1000, but perhaps specially between 500 and 900) the only thing that striked me as odd was how they seemed to simply give away free queens or other pieces every now and then. Lower than that I think that the most obvious difference was the humans would tend to use the queen a bit more and make more moves with the same piece or pawn early on. But past 1000 (and I didn't go far beyond that, only something like 1200) I couldn't notice anything very weird, inhuman. Except on situations such as a engine vs engine match I saw on youtube, where when they're at a balanced position, they kept just moving the rooks and/or other pieces safely at random and useless positions. The narrator of the video called that an "staring contest", which I think that may have taken something like 30 turns.

Interesting insight into a players thinking processes.

I would likely relate it more to intuitive moves than anything pedantic like a computer type sequence of moves at all.

Consider this idea:

Let's say you know specifically how to outflank and gain opposition against an opposing King in a particular position. You know the concepts behind the most straightforward approach to obtain a winning position as well as what to do in the case of second best replies by your opponent. You don't need to look up the exact sequences on how to accomplish this because you have encountered this type of game hundreds (if not thousands) of times over the years of game play you have had.

Question: Is this an example of playing intuitively based on multiple games experience or is it more of the precise, balanced analytical approach seen by computers?

What do you guys think on that idea?

@polygamous_king

That's interesting! But do I play like that?

@Bill_C

For me, it would be the latter. I don't know or can't recall many patterns, so I rely on calculation, which I'm best at.

The biggest difference lies in a human beings ability to imagine. Imagination is something that may never be gained properly by machines of any type. Sure, they can analyze millions of variations of the board and make it seem as if they have come up with a great inventive plan...but in reality it comes down to statistics.

That said...

I believe it is both, at least at a glance. In order to get to the point of playing intuitively you must first approach prior games and theory in a very balanced and analytical way. Thus you gain the insight to intuitively approach positions only through an approach most often reserved by computers and mathematical problems.

But computers don't learn. How does intuition exist in a computer then?

That is the difference and why it appears to be both at first glance. While there is something somewhat new called the "Adaptive Approach" in software, and Quantum Computing will take this a long way, computers cannot yet truly learn, at least not at the level of human beings and even many animals. Thus, a computer can fake very well the idea of intuition, but this intuition in a human player can only be gained through practice and experience.

Everyone "thinks" like a computer because our brains are really just computers...although some seem to have better operating systems than others.  

The test to see whether you play like a computer is to have someone set up several positions you have never seen before (I'd suggest these be from masters games) and see how your "brain" moves differ from a good chess program move.

The reason for the random positions is to make sure it's not a memorization thing.  For example in my younger days I used to play the computer a lot and memorized how it played several different openings.  I was accused several times of cheating by using a computer (even though I wasn't) just because my first few moves were exactly what a computer would play.

In a computer they make moves based on statistics from hundreds of games while humans that move on instinct make moves that are familiar to them using the cognitive portion of your brain its like riding a bike or driving a car the more you use it the better you become at it.

You caught me I'm not sure what computers base there moves off of whether its addition and subtraction or some other method.

@ Vulpes vulpes -- I don't know, my point was precisely that I don't think I'm good at distinguishing a human player from an engine, perhaps even more at below-grandmaster/whatever levels. At higher levels it seems that sometimes engines will make moves that look outright wrong/impredictable, because they only make sense at a depth of plies that humans don't usually/ever reach.

...

About chess engines learning, there's at least a scientific paper reporting that they used some genetic algorithm scheme to make one engine (gnuchess was used) learn from another (the "tutor", I don't remember the engine). In the end the improved engine was better than the previous version, and even very slightly better than its own tutor.

I don't know if this method warrants to call it "learning," as the genetic algorithm approach sort of suggests something more akin to "evolution"/selection of "individuals", new versions of the program, but I think that perhaps the same principles could be used in an engine that does this selection "in itself", rather than requiring a whole new version (if the method indeed consisting in creating new "versions-individuals", I'm not even sure). The analogy would be then to synaptic reinforcing and "neural darwnism" perhaps.

This (with the addition that "medium length tactics" means "20/30 moves long tactics").

Huh?