Chess will never be solved, here's why

The banks are also separate vertices, so I'd say you're both wrong. Looks like 4 vertices all odd.

(And yes, I'm in flat Earther mode for the purposes of the question.)

@Dubrovnik-1950

To recap.

In the post I originally responded to (response here) you defined two types of engine; a type 1 engine which performed a full width search (presumably at each increasing depth) and a type 2 engine as a modernisation, quoting Stockfish as an example. You made no mention of Shannon type A and B engines. In particular Stockfish (or any strong modern engine) is not a Shannon type B engine, so your type 2 doesn't correspond with Shannon type B.

In what I assumed was a response to my first post (you didn't actually specify which morons you were referring to and I think you've applied the term to most people currently posting except yourself), you talk instead about Shannon type A and B engines instead of your type 1 and 2 we'd previously discussed.

In my response to that post I said

"Had you asked the right question, we would probably still be in fairly close agreement. Gronk, the world's smartest AI, being but a superficial intelligence routine, would probably have flatly answered "no", whereas I would answer, "highly unlikely".

I was correct on the first point, but having looked again at Shannon's definition of type B, so was Gronk. (Strictly speaking I was correct on the second also, but not what I had in mind.)

In your latest response to that you say. "It you want it in FULL you got IT!!!!!" [sic] and dump another Grock of sh*te in the thread. No, I didn't want it and didn't ask for it.

Again you answer none of the points in my original response reiterated in my last post. So again:

Would you still maintain that the only way to solve chess is to do a full width search of the whole game tree?

Can you attempt to verify your theory that your type 1 engine would be more successful at finding errors than your type 2 engine by running the tests I suggested and posting the results?

the banks each have two bridges. So they're even.
Each bank is one vertex - (according to the AI) but since they each have two bridges their 'degree' is even.

And again I remind that Dubro doesn't seem to have asked Grok about the endgame tablebases or thought about them himself. Or maybe he did and I missed it.
He doesn't have to talk about that though - of course.
If chess is ever solved would it be done from the rear or the front or both?
(like when the Chunnel teams met in the middle)
and if its ever solved would the idea be invoked of not bothering with sillinesses like all sixteen white pieces against lone black king?
(just have to make sure its not stalemate. If white's on move can one set up a position where he can't avoid stalemate?)

White to move but can't avoid stalemating Black.
Would white have to have blundered into this one?
Maybe it needs a little adjustment. Plus two more white pawns to be added.

In this one - if white's K was at h1 and black played Qh2ch (from g3) then white has to take with Kxh2.

OK - its not Lone King but its still Stalemate!
Point - the chess-solving project can't assign a lopsided position as a win unless its sure that 1) - its not a stalemate and 2) that the winning side can avoid stalemating.

@playerafar

In the version I was shown the banks each had three bridges, two each from one island and one each from the other. Like this:

What does yours look like?

Sounds like you can do the tour in your parallel universe Königsberg.

MAR it may be that the AI presented the problem to me wrong.
I was told that the two islands were side by side not in series.
I was told - three bridges connecting the two islands plus each island had two bridges to its nearby mainland only.
In other words a 2-3-2 arrangement.
But that was before Dubrovnik was mentioning Grok here.
Good that he did.
So I'll see how Grok presents the problem.

Comes of talking to AIs.

I'd say it comes from forums plus AI's and then forums again.
One never has to assume the AI was right about everything.
Adding: the jargon term NPR - 'nodes per second' apparently doesn't mean 'solved nodes per second' which would mean tygxc's nonsense about that was even worse.

Grok stated this just now:
-------------------------
"Northern bank and Southern bank are the two mainland areas.
Kneiphof (the main island) sits in the middle of the river, with the Pregel splitting around it.
Lomse (the second island) is adjacent to Kneiphof, slightly offset, and connected to it and the banks.
The bridges were arranged as follows:

Two bridges connected the Northern bank to Kneiphof.
Two bridges connected the Southern bank to Kneiphof.
One bridge connected Kneiphof to Lomse.
One bridge connected the Northern bank to Lomse.
One bridge connected the Southern bank to Lomse.
------------------------------------
So your diagram looks right Martin - as for the direction of the river that's superflous.
But when going to google maps it doesn't seem to show both islands.
Apparently world war II bombing affected the two islands.
Lomse is supposedly part of the mainland now.
And Kneiphof island has new names one of them being Kant's island it seems.
Yes. That Kant.

You're trying very hard not to answer #19606

Dubro - of course O is a 'fraud' - and much worse too.
But MARattigan is not.
Unlike O - Martin does have a preference if not a care also that his posts are accurate.
(how do I know? It doesn't take long to become obvious.
And Martin might know more about solving chess projects than anybody here.
Martin makes accurate posts.
Except when he's kidding around.
Which does happen.

Nothing to do with refuting Grok. That's a side issue that can wait till I'm ready.

On the other hand a number of people in addition to yourself have asserted that the only way to solve chess is to consider (in some way) all the nodes in the game tree, so the question is very much at the heart of the discussion.

I have given you reasons why the statement is false (ridiculous in fact) but as of your last post you are still asserting it as fact. So can you please answer at least the first of my questions in #19606 (preferably with some substance rather than a silly ChatGPT diagram).

Really? I wasn't aware of it.

Must be old age creeping up.

Please show me where.

I would in fact say, and have said, that chess is definitely solvable with with some meanings of the ambiguous words "chess" and "solvable". I would also say, and have said, that chess is definitely not solvable with with some other meanings of the ambiguous words "chess" and "solvable".

But, of course, anyone who is not a moron might have difficulty understanding the nuances.

You're still trying very hard not to answer #19606.

By the way, I was updating the post you responded to while you made your last post. You may like to see the rest.

Dubro - you and MAR just disagree about AIs.
---------------------------
Just now - I asked chatggpt to contrast inflation and recession and depression.
Just to brush up a bit on the qualifications.
Stagflation and federal banks were also discussed.
And aspects of politics not allowed here.
--------------------------------------------
The reading speed of AIs is astronomical!
A gigabyte in a billionth of a second.
Notepad can do that? Not in 'comprehending'.
------------
But another big thing about the AIs is 2 way conversation.
When looking up a word - you can use an online dictionary.
But with the AIs you can ask followup questions and discuss the word much further.
Grok seems to have an edge over chatgpt which has an edge over copilot.
But if you look at the rate of improvement of the top AIs since their inception in 2018 ...
its fast.
The AI's a year from now could be much better than those of now.
---------------------------------------
Contrast with the eighties .... and the Commodore 64.
The internet existed. There was news service on the net (America Online for example. AOL.)
There was email. Compuserve or whatever.
Now - a midscale laptop is about 5000 times faster and more powerful than a C64 pc from 40 years ago.
In the 70's ... a Xerox Alto pc would have cost you the equivalent of about $500,000 today.
In other words - the internet and the pc revolution is about 40 years old.
Affordable laptops started appearing in the late 90s.
Source: Just for you Martin. Chatgpt.
-------------------------------------
No AI on this next - just moi.
AIs have been going about 5 years now.
What might AIs of 2030 look like?
AI in your vaccum cleaner?
Hey VC - what's the weather for tomorrow? How's the traffic right now?
And VC - don't forget to wash the dishes.
VC - 'OK boss. But they're done already.
By the way the Williams are coming up the street boss. I think they're coming here.'
You: 'Whaat? Wilma - do you know about that?'
Wilma: 'Of course. For lunch. And Barney. You forgot?'
You: 'VC - will chess be solved soon?'
VC: 'Of course not boss. You believed that tgyxc character and the Cloud stuff?'
You: 'Stop rubbing it in. Okay you win.'

And apparently are running from showing me where. (Perhaps you could ask Gronk instead.)

(Any progress on answering #19606?)

You think there is only one version of chess. You think there is one meaning of "solvable". You talk about the solution and the tablebase, so you apparently believe those are also unique. You think refuting something is simply saying it's false or sidestepping and letting Gronk post a pathetic picture.

You are definitely not a moron like the rest of us here.

But now that you've shown everybody what a fraudster I am, is there the slightest chance you might stop running away from answering #19606 with no sidestepping or silly pictures and get back to the topic?

Not really on topic, but out of interest; where did you learn your punctuation? It's consistently peculiar.

What a clever answer! You must be not a moron.