Flits Dxp Server

[BertTuyt]

JJ, thanks, that is good news, so it might work ...

I will now explain in more details the patch details.
The basic approach is to use 2 different memory-locations for the max level parameter.
In my implementation the computer_to_move routine uses the old memory location , so basically nothing changes.

For the pondering routine I use another memory location which is not used by the program (sometimes variables are initialized for whatever reason, but never actually used in the program).
Also the ml command is used in the sense that the new maximum level is now stored in the new location.

The 3 patch locations:
- Init of new max level parameter, change the value to 20 (which is the old value, so the program at the start behaves normal).
- pondering routine, change the address of the maximum level parameter to the new address
- command handler, set the address of the max level parameter which receives the new value, into the new address.

In total 4 bytes need to be patched.

That's all folks.
How it is possible to find this in a .exe is another topic

Maybe there is still a direct command which will do the same, but I didn't find it so far...


Bert

[Ed Gilbert]

I added a -p option to the Flits dxp server to disable pondering using "ml 2" and it works exactly as Bert described. I zipped it up along with the .txt file and put it here: http://edgilbert.org/InternationalDraug ... server.zip

-- Ed

[BertTuyt]

Ed, great..
Its a pleasure working together.
Maybe one day I will dig into Truus and try to find for her a no-pondering solution.

Keep you all posted,

Bert

[jj]

Hi Bert, Ed,

Downloaded and tested and working!
Great work!

Jan-Jaap

P.S. Do you know the current status on using the Truus (and Flits) opening book?

[Ed Gilbert]

Hi JJ,

No idea about the opening books. I finished an engine match with flits this morning with pondering off, and kingsrow had fewer wins than usual. It could just be randomness and coincidence, but I'm wondering if pondering was the cause of flits making bad moves in the endings? Maybe you will notice this also.

-- Ed

[jj]

Hi JJ,

No idea about the opening books. I finished an engine match with flits this morning with pondering off, and kingsrow had fewer wins than usual. It could just be randomness and coincidence, but I'm wondering if pondering was the cause of flits making bad moves in the endings? Maybe you will notice this also.

-- Ed

Hi Ed,

Ok, I'll repeat this experiment and see what happens.

In the mean time I created a tool to detect Flits glitches automatically in DXP match PDN files. For each game it goes to the final position, rewinds the last 5 moves (10 ply), and then analyzes (a 10 ply search for instance) the moves till the end of the game. Sudden big and game outcome changing scores are reported as probable glitches. Analysing 3 158-game matches reported 6, 3 and 8 glitches, respectively, and after manual checking 5, 3 and 7 remained.

BTW, Maximus score against Flits is still < 40%. Using the 6 pieces egdb or not does not make a big difference. I guess my evaluation function needs work!

Jan-Jaap

[jj]

I finished an engine match with flits this morning with pondering off, and kingsrow had fewer wins than usual. It could just be randomness and coincidence, but I'm wondering if pondering was the cause of flits making bad moves in the endings? Maybe you will notice this also.

I finished an engine match with Flits with pondering off today. With Flits not pondering and Maximus now using 8 (hyper)threads instead of 6, Maximus now did (6%) better, which makes sense. More glitches than usual, however!

Maximus vs. Flits +20 -23 =113 *2

after correcting 11 glitches and 2 unknowns:

Maximus vs. Flits +9 -23 =126 45.6%

Time controls were: Maximus 5 sec/move and Flits 75 moves in 5 min (equal). Games are stopped by Maximus when it sees a database win or loss, or else when there are 6 pieces or less on the board (database draw), or else after 75 moves (unknown).

All glitches follow the same pattern: they appear in positions with 6 pieces (or less) with Flits to move, after Maximus made a forced (re)capture (i.e. an immediate DXP response). The capture made the position a database draw position, but Maximus first sends this move and stops the game after Flits's next move - and then it often goes wrong. For instance:

Maximus - Flits, game 17

black to move, after 52.36-31 26x37 53.32x41 ...

Flits now played 53. ... 21-43? The other 10 glitch cases are, except for the position, identical.

What I can do is try to stop the games earlier when approaching a database draw. (Maybe an extra delay before sending an immediate MOVE?) See if they go away then. Pondering off seems to make the problem worse, but it can also be randomness.

I'm interested in how many glitches and when they occur other programmers who played matches against Flits can report.

Jan-Jaap

BTW Ed, there is a tiny omission in the Flits DXP server: when the DXP partner acts as match initiator and sends the GAMEREQs, the Flits opponent names in ...\Flits DXP server\dxpgames.pdn are empty. As Maximus also stores the games this is not a problem for me.

[Ed Gilbert]

Hi JJ,

Well, so much for my theory about fixing the bad flits moves. I guess the best defense is to try to adjudicate the games before they simplify down to 6 pieces.

Thanks for reporting the missing engine name. It is fixed now.

-- Ed

[jj]

Well, so much for my theory about fixing the bad flits moves. I guess the best defense is to try to adjudicate the games before they simplify down to 6 pieces.

Hi Ed,

I changed Maximus so that it stops the game immediately when there are 6 (or less) pieces on the board, not giving Flits the opportunity to move in a db position. The glitches disappeared completely!

Also, Maximus is down to

Maximus vs. Flits +6 -32 =120 41,8%

1-0 for randomness? I never experienced such big differences in subsequent matches against Dam2.2.

Jan-Jaap

[Ed Gilbert]

I put an updated Flits DXP server zip file on my web site here: http://edgilbert.org/InternationalDraug ... server.zip With this update the server now has the ability to correctly play when the remote engine sends a 3-move start position. I also updated the writing of the dxpgames.pdn file to conform to the new PDN 3.0 standard.

-- Ed

[Ed Gilbert]

It's been a few years since I tried to use Flits or the Flits DXP server. Someone asked me for help running an engine match with it, so I started it up to refresh my memory on how it works. I noticed that Flits is now making a Windows sound whenever a command is entered in the "Uw zet" window. I can't seem to turn this off, and I don't remember it doing this in the past. This is the first time I've tried it on Windows 10. I've turned off "overig, bel" setting in Flits, but it still makes the sound. Can someone confirm that this is the way Flits always works on Win10, or tell me how I can configure Flits to not emit the sound? Thanks.

-- Ed

[Sidiki]

Hi Ed, Bert and others.

It's was a long dream to be able to run a dxp match against Flits, but unfortunally, it don't work for me. Only the beginning FEN position stay on the board of Flits and the game don't beginning.
Can you help me please.
Thank

--Sidiki

[Krzysztof Grzelak]

And what operating system you have.

[igorlemasson]

It's been a few years since I tried to use Flits or the Flits DXP server. Someone asked me for help running an engine match with it, so I started it up to refresh my memory on how it works. I noticed that Flits is now making a Windows sound whenever a command is entered in the "Uw zet" window. I can't seem to turn this off, and I don't remember it doing this in the past. This is the first time I've tried it on Windows 10. I've turned off "overig, bel" setting in Flits, but it still makes the sound. Can someone confirm that this is the way Flits always works on Win10, or tell me how I can configure Flits to not emit the sound? Thanks.

-- Ed

Hi Ed,
Flits 3.02 works properly on Win10. The option to enable sound for new moves is "Overig > Bel"

[Ed Gilbert]

It's was a long dream to be able to run a dxp match against Flits, but unfortunally, it don't work for me. Only the beginning FEN position stay on the board of Flits and the game don't beginning.
Can you help me please.

There are some very specific requirements for setting up the flits windows to allow the dxp server to work. Read the instructions in flits_dxp_server.txt carefully, and pay attention to the requirements for the sizing of the "Uw zet:" and "Zetverloop" windows.

If you post or email me a screen shot of your desktop showing both the command line console and flits windows, I can probably see the problem with your setup.

-- Ed