I'll skip reading the post above from DiogenesDue. Might glance later.
Because its probably another one of his trolling posts so why feed him?
Its not just replying to such posts as his that is 'feeding' him.
Its reading them too. Feeds him. He wants a toxic atmosphere.
Chess will never be solved, here's why
Sort:
Apr 4, 2025
0
#16462
playerafar wrote:
I'll skip reading the post above from DiogenesDue. Might glance later.
Because its probably another one of his trolling posts so why feed him?
Everyone knows you are reading the posts by how you go out of your way to post a diffusing reply immediately and then try to act nonchalant about reading it at a later time. Stop being so insecure. It's painful to watch.
Apr 4, 2025
0
#16465
DiogenesDue wrote:
playerafar wrote:
I'll skip reading the post above from DiogenesDue. Might glance later.
Because its probably another one of his trolling posts so why feed him?
Everyone knows you are reading the posts by how you go out of your way to post a diffusing reply immediately and then try to act nonchalant about reading it at a later time. Stop being so insecure. It's painful to watch. seriously must have took you time to write
Apr 4, 2025
0
#16467
DiogenesDue wrote:
playerafar wrote:
Not that hard.
Ask any question you like. If its within the rules there's no consequences.
--------------------------------
DiogenesDue will probably not want you to ask questions.
I mentioned 'parallel processing' multiple times and he went nuts over that.
But failed to refute my use of the term. He'd rather troll than be constructive.
And then whines when he's talked back to.
He might accuse anybody joining in of being a 'sockpuppet' account.
He's even got a 'sockpuppets beware' message on his home page.
Translation: He wants people to be afraid of him.
Its correct there was never an 'alliance' between him and me - and he's always seen what he's wanted to see. He often judges - without having enough information. Or unthinkingly.
Just the simple act of somebody speaking up caused him to have delusions of 'sockpuppet' accounts. Later he made a phony apology. Here. Very recently.
Should we care? No. But he does.
----------------------------
Over 16,000 posts here. A lot of those are DiogenesDue feuding with Optimissed.
It wasn't what I thought it was at the time though.
Looks can be deceiving. Something DD doesn't seem to grasp.
Someobdy's not supposed to join in because DiogenesDue might accuse him or her of being a sockpuppet?
---------------------------
But regarding the new guy '7' he had an interesting idea (finally) and I haven't made a complete reply to it yet.
Demonstrate how I have any problem with people just asking questions. I have a problem with posters that are disingenuous, misleading, or making ridiculous claims when they don't know what they are talking about. You fulfill all of these criteria on a regular basis.
Your daft and simplistic usage of "parallel processing" is a case in point. Do you even have an idea of what a "process" means in this context? Do you know anything about hardware/software multithreading vs. multiprocessing vs. distributed computing? Do you know what barrier synchronization is? Interrupts? Virtual machines? Object oriented programming? Anything at all that would allow you to clearly and succinctly distinguish what computers do from magic when they are using various types of parallel processing?
I guess you'll be running off to your AI du jour now, to come back and try to incorporate these terms into your lexicon and failing to do so as you have with parallel processing as a whole. This penchant for making assertions about things you have only shallow knowledge of is something you share with Tygxc.
It's funny how you have been on the thread for years, not saying much of anything of note, then the week after you start fiddling around with AI you suddenly decide you can solve chess and start tossing around the same type of vague notions of massively reducing the problem space without being able to support it or even explain in any detail how you think it will work:
"Its a fraction. Exactly what fraction? Not sure. Would require a lot more math."
' ...different class of positions. Different algorithms to be used. Different handling."
You are saying nothing in all these type of statements, which is all that you have here. It's not even that you haven't taken the first concrete step...it's that you don't even have the first notion of what the first step will entail.
Bro wrote soooooooo much
did it take him an hour to write?
Apr 4, 2025
0
#16468
DiogenesDue wrote:
playerafar wrote:
Not that hard.
Ask any question you like. If its within the rules there's no consequences.
--------------------------------
DiogenesDue will probably not want you to ask questions.
I mentioned 'parallel processing' multiple times and he went nuts over that.
But failed to refute my use of the term. He'd rather troll than be constructive.
And then whines when he's talked back to.
He might accuse anybody joining in of being a 'sockpuppet' account.
He's even got a 'sockpuppets beware' message on his home page.
Translation: He wants people to be afraid of him.
Its correct there was never an 'alliance' between him and me - and he's always seen what he's wanted to see. He often judges - without having enough information. Or unthinkingly.
Just the simple act of somebody speaking up caused him to have delusions of 'sockpuppet' accounts. Later he made a phony apology. Here. Very recently.
Should we care? No. But he does.
----------------------------
Over 16,000 posts here. A lot of those are DiogenesDue feuding with Optimissed.
It wasn't what I thought it was at the time though.
Looks can be deceiving. Something DD doesn't seem to grasp.
Someobdy's not supposed to join in because DiogenesDue might accuse him or her of being a sockpuppet?
---------------------------
But regarding the new guy '7' he had an interesting idea (finally) and I haven't made a complete reply to it yet.
Demonstrate how I have any problem with people just asking questions. I have a problem with posters that are disingenuous, misleading, or making ridiculous claims when they don't know what they are talking about. You fulfill all of these criteria on a regular basis.
Your daft and simplistic usage of "parallel processing" is a case in point. Do you even have an idea of what a "process" means in this context? Do you know anything about hardware/software multithreading vs. multiprocessing vs. distributed computing? Do you know what barrier synchronization is? Interrupts? Virtual machines? Object oriented programming? Anything at all that would allow you to clearly and succinctly distinguish what computers do from magic when they are using various types of parallel processing?
I guess you'll be running off to your AI du jour now, to come back and try to incorporate these terms into your lexicon and failing to do so as you have with parallel processing as a whole. This penchant for making assertions about things you have only shallow knowledge of is something you share with Tygxc.
It's funny how you have been on the thread for years, not saying much of anything of note, then the week after you start fiddling around with AI you suddenly decide you can solve chess and start tossing around the same type of vague notions of massively reducing the problem space without being able to support it or even explain in any detail how you think it will work:
"Its a fraction. Exactly what fraction? Not sure. Would require a lot more math."
' ...different class of positions. Different algorithms to be used. Different handling."
You are saying nothing in all these type of statements, which is all that you have here. It's not even that you haven't taken the first concrete step...it's that you don't even have the first notion of what the first step will entail.
OH BOY
Apr 4, 2025
0
#16469
Elroch wrote:
In the ~80 year history of computing we have seen processing speed increase from about 1 operation per second to about 10^18 operations per second. Just need maybe 1.7 times as much again, on the usual log scale. Assuming constant log improvement, that's 136 years, i.e. 2161, when I will also be celebrating my 200th birthday.
Nice post!
Apr 4, 2025
0
#16472
Elroch wrote:
You are only arguing out of inertia. Here is an algorithm as intended in the definition:
"if you are white, play moves from the white strategy, if you are black play moves from the black strategy"
Earlier I refered to this as a "portmanteau" algorithm and stated that I preferred a definition that referred to it as two separate strategies, rather than an entity containing them.
There is no way that you can claim the INTENDED meaning is the mistaken one you at some point got fixated in your mind. You really should be able to see that now. You would obviously be wrong to think the correct interpretation is inconsistent with what is written - it simply isn't - and there is no reason for anyone to prefer your interpretation (most people wouldn't even think of it).
That being said, I underline my preference for slightly different wording.
That post was a reply from Elroch to Marattigan.
"There is no way that you can claim the INTENDED meaning is the mistaken one you at some point got fixated in your mind. "
Marattigan has been doing exactly that. Constantly.
Even against K+R versus K.
He assigns a new meaning and then attacks his own misinterpretation.
Quite an investment. Or rather mis-investment. Better to not indulge it.
Apr 4, 2025
0
#16473
playerafar wrote:
Elroch wrote:
You are only arguing out of inertia. Here is an algorithm as intended in the definition:
"if you are white, play moves from the white strategy, if you are black play moves from the black strategy"
Earlier I refered to this as a "portmanteau" algorithm and stated that I preferred a definition that referred to it as two separate strategies, rather than an entity containing them.
There is no way that you can claim the INTENDED meaning is the mistaken one you at some point got fixated in your mind. You really should be able to see that now. You would obviously be wrong to think the correct interpretation is inconsistent with what is written - it simply isn't - and there is no reason for anyone to prefer your interpretation (most people wouldn't even think of it).
That being said, I underline my preference for slightly different wording.
That post was a reply from Elroch to Marattigan.
"There is no way that you can claim the INTENDED meaning is the mistaken one you at some point got fixated in your mind. "
Marattigan has been doing exactly that. Constantly.
Even against K+R versus K.
He assigns a new meaning and then attacks his own misinterpretation.
Quite an investment. Or rather mis-investment.
Let me ask a question; K+R vs K is either a draw or a win for the player with the rook yes? There are only a couple of solutions where K+R wins, while there are endless draws (most being through King takes rook, some being through inability to take, I think), so why not just find every winning position for K+R and then just consider every other position a draw. Of course, you'd still be able to see how it's a draw, but wouldn't that cut the number of games needed to process down? Sorry if this has been said before because these posts contain lots of bulk that makes it harder to read for me.
Apr 4, 2025
0
#16474
@SandStarter
Yes. Your points connect to almost exactly what i've been getting at.
The number of positions where K+R versus K is a win greatly exceed the number of positions where its a draw. Greatly.
But also - K+R versus K is basic to chess theory.
And it defines the minimal force necessary to win.
Its one of the first things that beginner chessplayers learn.
But the application of chess basics to computer solving projects - still not discussed here in this forum that much.
Not yet.
------------------------
And in addition to computer projects - the whole issue of solving chess and considering the entirety of the game as an intellectual subject.
Whether players are 8 years old or 80 years old chess is an intellectual game and subject of discussion.
And the forum subject goes with an academic atmosphere and discussions of the math involved too. Academic atmosphere but not 'stuffy' atmosphere or toxic atmosphere.
K+R versus K is a good starting point.
So is K+Q versus K+B.
And even K+R versus K+R.
There's multiple reasons to not start with pawns.
------------
Apr 4, 2025
0
#16475
playerafar wrote:
@SandStarter
Yes. Your points connect to almost exactly what i've been getting at.
The number of positions where K+R versus K is a win greatly exceed the number of positions where its a draw. Greatly.
But also - K+R versus K is basic to chess theory.
And it defines the minimal force necessary to win.
Its one of the first things that beginner chessplayers learn.
But the application of chess basics to computer solving projects - still not discussed here in this forum that much.
Not yet.
------------------------
And in addition to computer projects - the whole issue of solving chess and considering the entirety of the game as an intellectual subject.
Whether players are 8 years old or 80 years old chess is an intellectual game and subject of discussion.
And the forum subject goes with an academic atmosphere and discussions of the math involved too. Academic atmosphere but not 'stuffy' atmosphere or toxic atmosphere.
K+R versus K is a good starting point.
So is K+Q versus K+B.
And even K+R versus K+R.
There's multiple reasons to not start with pawns.
------------
What I think is that if a computational system were to hypothetically "solve" chess, it would have to run every instance where the game ends in a draw first, because draws are much simpler to find than wins for either side. First K+R vs K, then K+B v K+Q, then K+R v K+B, so on, so forth. A "perfect" game ends in a draw every time, so let the draws happen, and then start looking for imperfections that would lead to a win. Is that sufficient?
I'm not a computer science major by any means, but something like having two computers playing every position and branching out from drawing positions (how could a player each the drawing position) would be possible, yes? Not very simple, but possible?
Apr 4, 2025
0
#16477
There's been a pattern of comparing a view of the game from its opening - to viewing the most simplied possible endings and then considering backwards by adding material.
But there's an idea of keeping both ends under observation.
Like this:
At one end - we have King versus King. All such positions.
At the other end - all positions with 32 pieces on board.
Each of the two has its 'characteristics.'
Projects to solve chess must consider all possible positions of the two Kings.
If they don't - then they're going to miss some positions.
At first I thought there should be three categories of such positions.
Depending on whether the white king is midboard or on an edge or on a corner.
Now - I think there should be more categories.
Like - both kings on corner squares. One on an edge and the other midboard.
How many categories are there?
----------------------
At the other end - positions with all 32 pieces - couldn't have had any promotions.
That's right.
If there were no captures - then how could any pawn possibly reach the other back rank?
With more implications for 31 pieces and so on.
--------------------------------------
These considerations impact how the game could be considered in overview.
Apr 4, 2025
0
#16478
StandStarter wrote:
playerafar wrote:
@SandStarter
Yes. Your points connect to almost exactly what i've been getting at.
The number of positions where K+R versus K is a win greatly exceed the number of positions where its a draw. Greatly.
But also - K+R versus K is basic to chess theory.
And it defines the minimal force necessary to win.
Its one of the first things that beginner chessplayers learn.
But the application of chess basics to computer solving projects - still not discussed here in this forum that much.
Not yet.
------------------------
And in addition to computer projects - the whole issue of solving chess and considering the entirety of the game as an intellectual subject.
Whether players are 8 years old or 80 years old chess is an intellectual game and subject of discussion.
And the forum subject goes with an academic atmosphere and discussions of the math involved too. Academic atmosphere but not 'stuffy' atmosphere or toxic atmosphere.
K+R versus K is a good starting point.
So is K+Q versus K+B.
And even K+R versus K+R.
There's multiple reasons to not start with pawns.
------------
What I think is that if a computational system were to hypothetically "solve" chess, it would have to run every instance where the game ends in a draw first, because draws are much simpler to find than wins for either side. First K+R vs K, then K+B v K+Q, then K+R v K+B, so on, so forth. A "perfect" game ends in a draw every time, so let the draws happen, and then start looking for imperfections that would lead to a win. Is that sufficient?
I'm not a computer science major by any means, but something like having two computers playing every position and branching out from drawing positions (how could a player each the drawing position) would be possible, yes? Not very simple, but possible?
When Schaeffer did his famous solution of checkers - he built on drawing ideas.
Chess is much more resistant of course.
What i keep saying is - both drawing ideas and winning ideas can be addressed because of the nature of parallel processing in computers. The more powerful the computer - the more parallel processing potential it has. So - sub-projects can run concurrently.
-----------
But regarding your idea of prioritizing draws - that is why King versus King on a clear board is so key.
It is the most fundamental draw.
Also the two Kings must always be there anyway.
So whether building on draws - or not - or both (I like both) then all positions of the two Kings have to be built from anyway.
By adding material to the position.
And project engineers would have tremendous discretion - as to how that material is to be added.
But so do people discussing the subject!
Apr 4, 2025
0
#16479
playerafar wrote:
StandStarter wrote:
playerafar wrote:
@SandStarter
Yes. Your points connect to almost exactly what i've been getting at.
The number of positions where K+R versus K is a win greatly exceed the number of positions where its a draw. Greatly.
But also - K+R versus K is basic to chess theory.
And it defines the minimal force necessary to win.
Its one of the first things that beginner chessplayers learn.
But the application of chess basics to computer solving projects - still not discussed here in this forum that much.
Not yet.
------------------------
And in addition to computer projects - the whole issue of solving chess and considering the entirety of the game as an intellectual subject.
Whether players are 8 years old or 80 years old chess is an intellectual game and subject of discussion.
And the forum subject goes with an academic atmosphere and discussions of the math involved too. Academic atmosphere but not 'stuffy' atmosphere or toxic atmosphere.
K+R versus K is a good starting point.
So is K+Q versus K+B.
And even K+R versus K+R.
There's multiple reasons to not start with pawns.
------------
What I think is that if a computational system were to hypothetically "solve" chess, it would have to run every instance where the game ends in a draw first, because draws are much simpler to find than wins for either side. First K+R vs K, then K+B v K+Q, then K+R v K+B, so on, so forth. A "perfect" game ends in a draw every time, so let the draws happen, and then start looking for imperfections that would lead to a win. Is that sufficient?
I'm not a computer science major by any means, but something like having two computers playing every position and branching out from drawing positions (how could a player each the drawing position) would be possible, yes? Not very simple, but possible?
When Schaeffer did his famous solution of checkers - he built on drawing ideas.
Chess is much more resistant of course.
What i keep saying is - both drawing ideas and winning ideas can be addressed because of the nature of parallel processing in computers. The more powerful the computer - the more parallel processing potential it has. So - sub-projects can run concurrently.
-----------
But regarding your idea of prioritizing draws - that is why King versus King on a clear board is so key.
It is the most fundamental draw.
Also the two Kings must always be there anyway.
So whether building on draws - or not - or both (I like both) then all positions of the two Kings have to be built from anyway.
To "solve" chess is to know every position and every singular win, loss, and draw, but we don't need that to necessarily solve chess. To "beat" chess is to know wins and losses on feasible positions. Having 9 queens is possible, but the probability is too low. I think solving chess should be more on solving based on probability of positions. Kind of like tablebase; look at every common move o every position and continue until draw, then find a way to win, they find a different way, so on and so forth. Then go to less common positions, do the same, repeat. This way, we can solve basic-intermediate chess while also actually SOLVING chess as a whole, in the manner I described at the very beginning. That's what I think, at least. Would help with being able to actually see chess solved to an extent while I'm still alive.
Apr 4, 2025
0
#16480
Regarding nine queens on board - I don't think it should be skipped.
But it should get separate processing. To whatever degree.
And if its five queens versus four - that's going to be much harder and require different processing.
Nine queens for one side looks frivolous because how/why would that happen in a game?
But that's another reason for separate processing.
------------
Nine queens on board means a minimum of 11 pieces on board but would it mean a maximum of 25 pieces on board?
For each promotion does there have to be at least one capture along the way?
I'm thinking yes but am not 100% sure.
---------------------------
"Would help with being able to actually see chess solved to an extent while I'm still alive."
You might. But I won't. I haven't got that kind of time left.
You might. But they'd have to be going at the solving differently though.
Not that hard.
Ask any question you like. If its within the rules there's no consequences.
--------------------------------
DiogenesDue will probably not want you to ask questions.
I mentioned 'parallel processing' multiple times and he went nuts over that.
But failed to refute my use of the term. He'd rather troll than be constructive.
And then whines when he's talked back to.
He might accuse anybody joining in of being a 'sockpuppet' account.
He's even got a 'sockpuppets beware' message on his home page.
Translation: He wants people to be afraid of him.
Its correct there was never an 'alliance' between him and me - and he's always seen what he's wanted to see. He often judges - without having enough information. Or unthinkingly.
Just the simple act of somebody speaking up caused him to have delusions of 'sockpuppet' accounts. Later he made a phony apology. Here. Very recently.
Should we care? No. But he does.
----------------------------
Over 16,000 posts here. A lot of those are DiogenesDue feuding with Optimissed.
It wasn't what I thought it was at the time though.
Looks can be deceiving. Something DD doesn't seem to grasp.
Someobdy's not supposed to join in because DiogenesDue might accuse him or her of being a sockpuppet?
---------------------------
But regarding the new guy '7' he had an interesting idea (finally) and I haven't made a complete reply to it yet.
Demonstrate how I have any problem with people just asking questions. I have a problem with posters that are disingenuous, misleading, or making ridiculous claims when they don't know what they are talking about. You fulfill all of these criteria on a regular basis.
Your daft and simplistic usage of "parallel processing" is a case in point. Do you even have an idea of what a "process" means in this context? Do you know anything about hardware/software multithreading vs. multiprocessing vs. distributed computing? Do you know what barrier synchronization is? Interrupts? Virtual machines? Object oriented programming? Anything at all that would allow you to clearly and succinctly distinguish what computers do from magic when they are using various types of parallel processing?
I guess you'll be running off to your AI du jour now, to come back and try to incorporate these terms into your lexicon and failing to do so as you have with parallel processing as a whole. This penchant for making assertions about things you have only shallow knowledge of is something you share with Tygxc.
It's funny how you have been on the thread for years, not saying much of anything of note, then the week after you start fiddling around with AI you suddenly decide you can solve chess and start tossing around the same type of vague notions of massively reducing the problem space without being able to support it or even explain in any detail how you think it will work:
"Its a fraction. Exactly what fraction? Not sure. Would require a lot more math."
' ...different class of positions. Different algorithms to be used. Different handling."
You are saying nothing in all these type of statements, which is all that you have here. It's not even that you haven't taken the first concrete step...it's that you don't even have the first notion of what the first step will entail.