Solving chess? With no BS. (moderated)

another question is does the chess world really want to solve chess. I feel like  so many of the people who have the most competence to try solving chess dont actually want to solve for their own good. Not to mention its extremely difficult to do anyways

As for the encoding a table base on a single atom, what would this premise be based on?  If you go to the quark level, you are adding what, 1 order of magnitude to just storing on electrons?  So, for even the most unstable atoms, still less storage for a single atom than a QR code.

Blueemu brought up what in computer/hardware parlance would be called "destructive reads" i.e. when the process of reading the data destroys the usable data.  Just because particles have N properties does not mean we can store N bits of data thereby, nor read them back without disrupting the data.

Like trying to use the Bat Thermos to store a Riddler's riddle left in alphabet soup and feed it into the bat computer later, there are serious issues involved.  I'll be surprised if anyone knows that reference .  

If unlimited levels of smaller particles (heretofore undiscovered) useful for storage is the notion, then I would consider anything much past quarks currently to be in the realm of positing magic.

Chess players as a group won't really have a say in it .  As this thread and Ponz's thread and the much larger "Will computer solve chess?" thread, and many threads on Quora and elsewhere prove, there's more than enough interest in solving to motivate people, if the technology were available.

You don't need to keep track of each node in the game tree to solve chess. You need at most keep track of the corresponding positions associated with each node (as is done in EGTBs).

Tromplin puts an upper bound of 10⁴⁷ on the total number of positions, which would accommodate a game of any length, so you are overestimating the number of things to be tracked by a factor of more than 10⁷³.

(Reality check: That's more than 100,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 times too big.)

And I think with 10⁸⁰ you may be excluding  dark matter from consideration.

10^47 and 10^120 are only different in this context in that one number is larger than the number of particles in the universe, and one isn't, which leads to some nice one liners and some discussions about how impossible "impossible" really is...but it doesn't matter in any real discussion of humanity solving chess.  Both numbers are currently impossible to deal with and will remain impossible for any realistically foreseeable future.  And 10^46 *is* identified as the number of positions that need to be traversed in the OP.  The Shannon Number is useful, however.  If we didn't talk about it, then people would be saying "but we can just prune the positions down to a number we can calculate" even more.  The number is already "pruned" as far as current understanding of chess allows and it sits at roughly 10^46.  

In order to be in reasonable reach of solving, that number would have drop another 20 orders of magnitude or the technology would have to improve the same amount to allow for it.  Quantum computing cannot bridge that gap, nor can any feasible forecasted technology expected to be developed in the far future.

Just two quick points: 1) we have been talking number of particles in the universe when the practical question is how many can be harnessed from earth for computing purposes? 2) In order to prove  a solution 100%, per the original post, would not every single possible chess combination have to be tried?....like in the traveling salesman problem?

Point 1:  Yes, that's my point...we can talk about particles in the universe etc. as a fun exercise, but it terms of what humanity can feasibly achieve, we can't really go past our own solar system currently, even in conjecture.  The notion of trying to build a computer to solve chess using multi-generational spaceships that take centuries to reach another star system is pointless .  Someone can posit that we'll just keep discovering smaller subatomic particles and find some way to manipulate them well enough to do the job, but anything beyond the particles we have already proven exist or are pretty sure must exist would just be a "call to magic".  Just like saying that we will have FTL travel one day...posit a viable approach, or it's just fantasy.

If you read The Fountains of Paradise (Clarke), that's hard science fiction.  If you read...Piers Antony, for example, that's fantasy, not real science fiction.  Just magic in a future, science-y setting.  Same thing applies here, we should talk about hard science fiction, not fantasy/magic . 

Point 2:  If you read the definition of solved games from the first page, we're "only" going for a "weak" solution with 10^46.  The "strong" solution would require every possible combination.  Sort of in the same way that you don't have to check every single number between 1 and the largest prime number to prove a prime, you can prove it with roughly the square root of that number of operations.

Are you familiar with Miguel Alcubierre's work? Taken together with some further development by Sonny White (USA) and by a team at Pulkovo (Russia), it was realistic enough that NASA is investigating it.

References on request.

Human beings traveling FTL unharmed?  Or matter in some form traveling FTL?

I've seen this:  https://en.wikipedia.org/wiki/IXS_Enterprise

...but given that the main 2 deliverables from this article are artist's renditions of what such a ship *might look like*, I feel safe in waiting for something more concrete.  If it goes anywhere significant, the buzz in the scientific community will be quite apparent...

NASA and the US government are prone to investigating anything at all that smacks of "we might get left behind if we don't research this...America might lose it's #1 ranking in XYZ".  It's a well known way to garner funding for dubious efforts .

Please note, I am not saying Alcubierre's work is dubious, I'm just saying I won't really dive into this until it makes a much bigger radar blip.

It's all quite speculative at the moment, but is being actively worked on by a team at JPL, another at Pulkovo (headed by Sergiy Krasnikov, IIRC) and another at NASA headed by Sonny White.

Here's an article in Science and Nature: NASA Admits Alcubierre Drive Initiative: Faster Than The Speed Of Light - Science And Nature (sci-nature.vip)

Interesting, but a few comments about the article (not the concept):

- Not at all the first time people have posited this type of drive...it's essentially the old "fold the cloth of space like this and you bring two distant points together" idea.

- Don't say "we've discovered 3 super earths that are most probably teeming with life"...either there's some significant evidence, or there isn't.  Since there aren't a gazillion other front page articles about this, I'd say it's the latter.

- Linking a UFO video is not a good way to increase credibility.  Pandering for hits is not a good way to show your publication has integrity.

Meanwhile, on the other Lovecraftian "don't look at this too closely or you will lose your sanity" thread, people are are arguing that testing 2 identical engines playing each other to reach win or draw is somehow the best test for solving chess ...sigh.

"If we place there two high school band members in the orchestra pit, and they do *not* produce Mozart, then we know music is XYZ" would be about the same level of premise.

YOU ARE USING THE LOGICAL FALLACY OF 'STRAWMAN'   .

Nice impression.  Note how the professed medical condition causes the person in question to have to reach over and hit the all caps key, rather than just posting in all *small* letters, which would save keystrokes .  It's a very quirky medical condition. 

Perhaps it's an eyesight issue. Still, dipping into that thread from time to time it appears that over the course of 7 years mental health can deteriorate. 

A simple browser font size adjustment takes care of that, though .  Admittedly, that adjustment is possibly beyond the Ponz level of computer savvy.

If the only acceptable solution to chess is playing every possible variation out, then no, it will never be solved. 

If we are only searching for the answer to whether chess is a draw with best play, we have pretty good data and evidence showing that yes, it is a draw with best play.

Yes, the article was just my first google hit. Perhaps you would prefer to read the write-up by NASA themselves?

There are of course several articles. Here's one by NASA's Sonny White:

NASA Technical Reports Server (NTRS)

and a paper by the team at Pulkovo Observatory:

arXiv:gr-qc/0207057v3 19 May 2003

It's a nice theory, but still just 4.5 pages (after discarding the preamble and summary), mostly theoretical math that explains why anyone should even entertain the idea, without even an attempt to discuss practical progression towards the goal (there is discussion of how to test the field once one is made, but not how to attempt to make one).  When they start funding a launch to get far enough away to attempt a practical "warp field" experiment, I'll give it another look...and getting a tiny lunchbox-sized experiment up to the ISS doesn't count .

NASA has run experiments which yielded positive (non-zero) results... but still within the error limits of a null hypothesis. In other words, inconclusive but not discouraging.

White–Juday warp-field interferometer - Wikipedia