Severe windows 10 problems. What to do?

[Joost Buijs]

I think i will never buy anything again from "Media Markt".
Bought a new computer but I see I paid 99 euro for
"Power Service Altijd aan".
I did not ask for that.
Yes it was included in the package.

Being so foolish to buy that.

Looks like each encounter with a salesman in a computer shop cost you many euros.
All thieves.

Only buy stuff when in economic recession otherwise you pay too much.


Might be person in other computer shop did nothing but only cashed 50 euro.

Maybe better buy online if you can get online but I only had a dumb Phone. Who invented the word smart Phone ?? Many application don't work on smart Phone and sometimes they assume you have fingers as big as a needle. Not to talk about annoying advertisements.

So you bought a new computer.

Only when I'm in a very big hurry I buy sometimes from 'Media Markt'. Media Markt 'Ik ben toch niet gek', last week I needed a 3 meter cat6 patch cable they asked over 20 Euro's for it. At 'allekabels.nl' it costs just 3.99 Euro. You must be mad indeed when you buy something from 'Media Markt'.

Usually I built my PC's from parts, and I look at 'tweakers.net' to find the cheapest seller. It really makes a difference of hundreds of Euros for a standard PC, another advantage it that for each part you can choose the brand that you really like.

I hope Skipper will be back up and running soon.

Programming a chess engine from scratch is not so easy, a few weeks ago I decided to replace my Windows threads/synchronization with C++ threads/synchronization (because I also want to run on Linux), and now I'm experiencing random crashes, I can play like 20 games, no problem and at game 21 it will crash. I've been looking at it for 2 weeks, but I can't find the problem. Maybe I should switch to lazy SMP like many others do, it is not so easy to get YBW working without any data races. Of course you can lock everything, but that hampers performance.
Will be continued.

[Henk]

Maybe best is to go back to a stable state.

Better a worse engine that does not crash and from there try to improve.

Don't know yet how much it cost to get Skipper back and whether I can afford it in my situation.
Maybe better use computer less otherwise it may get broke earlier and I get same horror again.
For I don't trust my backups.

I currently did not manage to get my 'sent mail' back from date less than week ago in microsoft mail. Inbox working normal as it used to.
So that problem has highest priority. So probably settings of outgoing mail in my mail account are not correctly set.

[Joost Buijs]

I just found out at 6.00 AM that I have a sporadic error in my hash key, this has never been before, but it explains a lot.
MS IntelliSense (or whatever they call it) sometimes removes parts of your code when typing, when you don't notice this you are in trouble.
Now I have to search for the error, this can't be too difficult.

[Joost Buijs]

I just found the error, somehow a statement in the hash update disappeared like I expected.
I also added one extra lock in the YBW routine, up until now it seems to be running stable (like it used to be), fingers crossed.

[Sven]

I just found the error, somehow a statement in the hash update disappeared like I expected.

Aren't you using a version control system, e.g. Git? It would be a trivial task to review all changes when using Git.

[Henk]

Last half year I was using GIT too.

That's why i don't trust my backups for I don't know yet how to get back a solution from a repository GIT backup or what is its name.

Wait already found something. Had to go to View -> TeamExplorer in visual studio 2019 community edition.

[Henk]

Ok looks like Skipper alias SchaakMachine working again on my computer.

[pgn] [Event "Computer Chess Game"] [Site ""] [Date "2019.07.21"] [Round "-"] [White "SchaakMachine"] [Black "Stockfish 10"] [Result "0-1"] [TimeControl "40/300"] [Annotator "1. +0.03 1... -0.05"] 1. e4 {+0.03/11} e6 {-0.05/22 16} 2. d4 {+0.15/10 6} d5 {+0.05/22 0.1} 3. Bd3 {+0.39/9 6} dxe4 {+0.37/21 7} 4. Bxe4 {+0.21/9 6} Nf6 {+0.49/22 0.1} 5. Bg5 {+0.16/9 6} Nbd7 {+0.58/19 7} 6. Qf3 {+0.57/7 6} c5 {+1.43/17 5} 7. dxc5 {-0.28/8 5} Nxc5 {+1.73/19 2.8} 8. Bxf6 {-0.52/10 5} gxf6 {+1.63/22 1.6} 9. Nh3 {-0.74/11 5} f5 {+2.10/21 11} 10. Qc3 {-0.12/11 5} Rg8 {+1.95/20 1.0} 11. Bf3 {-0.50/9 5} Bd7 {+1.75/21 3} 12. O-O {+0.16/8 5} Ne4 {+2.68/21 6} 13. Qb3 {+0.37/8 5} Bc6 {+2.13/23 19} 14. g3 {+0.25/7 5} Qc7 {+2.08/24 5} 15. Rd1 {+0.02/7 5} Bc5 {+2.21/23 0.1} 16. Rd3 {-0.12/7 5} Rg6 {+3.61/21 8} 17. c4 {+0.22/7 5} Rh6 {+5.47/20 3} 18. Kg2 {-2.70/10 5} Rxh3 {+7.37/26 6} 19. Kxh3 {-2.20/11 4} Nxf2+ {+7.54/26 0.1} 20. Kg2 {-2.44/11 4} Qe5 {+7.71/26 1.3} 21. Bxc6+ {+1.09/10 4} bxc6 {+7.86/26 0.4} 22. Qb7 {+1.61/11 4} Qe4+ {+25.91/24 4} 23. Rf3 {+2.07/11 4} Ng4 {+99.75/24 0.2} 24. Qxa8+ {+2.09/9 4} Ke7 {+99.85/39 5} 25. Qb7+ {+1.58/8 4} Kf6 {+99.89/51 6} 26. Kh3 {+0.30/8 4} Qxf3 {+99.91/64 6} 27. Kh4 {+0.18/8 4} Ne3 {+99.93/71 6} 28. Qxf7+ {-10.29/6 4} Kxf7 {+99.95/103 2.4} 29. Kg5 {-2621.38/12 4} Qg4+ {+99.97/127 0.1} 30. Kh6 {-2621.40/13 4} Qg6# {+99.99/127 0.1} {Xboard adjudication: Checkmate} 0-1 [/pgn]

[Joost Buijs]

I just found the error, somehow a statement in the hash update disappeared like I expected.

Aren't you using a version control system, e.g. Git? It would be a trivial task to review all changes when using Git.

No, I never did. I stem from an age that all these things didn't exist. I make automated backups on a daily basis though, so I can always fall back to an earlier version.

The stupid thing is that sometimes when you type a space in the middle of a statement the Visual Studio editor removes the whole statement (I think it is a bug), usually I notice this, but I missed it on a few occasions. It has something to do with IntelliSense, the automatic statement completion can be switched off somewhere in the editor settings, recently I switched to VS2019 and now it seems to happen again.

I was really fighting with C++ threads and synchronization objects. my engine was perfectly stable with Windows threads, SRW locks and condition variables, after I switched to C++ threads, mutexes/lock_guards and condition variables I experienced random crashes, sometimes just 1 in 40 games and not reproducible. I got the impression that a thread somehow tried to lock a mutex twice without releasing it, and that this generated an exeption. After adding an extra lock at a location where it doesn't seem necessary my engine looks stable again, but I still don't understand it.

[Henk]

I hate code I don't understand. So probably I would even go back to a worse previous version with code I do understand.

[Joost Buijs]

I hate it too, I understand what the code does because I wrote it myself, but I don't understand why the extra lock is necessary. Maybe there is still a very subtle bug in Microsofts threading library, it would not be the first, the last year they removed several bugs in this subsystem. Pretty soon I will have a Linux system up and running and I just want to see if it also crashes when I compile my engine with GCC.

I assume there will be an HGM tourney next weekend, and I hope you are completely back up and running with Skipper by that time.