Bugs in Arena 3.5.1

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Harvey Williamson, bob

EN
Posts: 30
Joined: Sun Nov 30, 2014 1:21 am

Re: Bugs in Arena 3.5.1

Post by EN » Wed Mar 14, 2018 3:08 am

rcmaddox wrote:Arena 3.5.1 is not without bugs but I haven't experienced the three mentioned.
Well, I don't know what to say. Like I said in initial post; Deep Rybka 3 x64, with same settings, Perfect2017.abk book, same engine configuration and everything else - works in Arena 1.1. I don't know why DR3 won't play in Arena 3.5.1.

Here is the message from Arena's Window (or Tournament pane):

Pairing 1/1: Deep Rybka 3 x64 - IvanHoe999970_PrefetchSS_x64_FT
Round 1 of 10: Deep Rybka 3 x64 - IvanHoe999970_PrefetchSS_x64_FT
Game end: 0-1{0-1 Arena Adjudication. Illegal move!}
Last game won by IvanHoe999970_PrefetchSS_x64_FT with Black in 0 moves
Deep Rybka 3 x64 - IvanHoe999970_PrefetchSS_x64_FT:
1 B 2 1.0:0.0 9:28:05PM 0.00 0 0:0 0-1-0
+++++++File not found+++++++
+++++++File not found+++++++
+++++++File not found+++++++


It does this over and over - with the wins counting up - but no move.

EN
Posts: 30
Joined: Sun Nov 30, 2014 1:21 am

Re: Bugs in Arena 3.5.1

Post by EN » Wed Mar 14, 2018 3:11 am

JohnW wrote:I have used Arena 3.5.1 a lot and have never had those issues. The only big issue I ever see is it crashes frequently and restarts during a tournament, but it restarts right where it left off.
I think what Erin Dame suggested might fix it.

I've seen that exact option but never bothered to modify it to my needs. Usually, after few games with 2 engines, I may opt to try another different pair of engines. All for entertainment. So I never encountered what you're experiencing. But in my case, when I run Deep Rybka 3 x64 in Arena 1.1, DR3 starts to play! No problems whatsoever. This is the message from Arena 3.5.1 when attempting to play IvanHoe with Deep Rybka 3:

Pairing 1/1: Deep Rybka 3 x64 - IvanHoe999970_PrefetchSS_x64_FT
Round 1 of 10: Deep Rybka 3 x64 - IvanHoe999970_PrefetchSS_x64_FT
Game end: 0-1{0-1 Arena Adjudication. Illegal move!}
Last game won by IvanHoe999970_PrefetchSS_x64_FT with Black in 0 moves
Deep Rybka 3 x64 - IvanHoe999970_PrefetchSS_x64_FT:
1 B 2 1.0:0.0 9:28:05PM 0.00 0 0:0 0-1-0
+++++++File not found+++++++
+++++++File not found+++++++
+++++++File not found+++++++


Over and over...ad infinitum.

Modern Times
Posts: 2141
Joined: Thu Jun 07, 2012 9:02 pm

Re: Bugs in Arena 3.5.1

Post by Modern Times » Wed Mar 14, 2018 6:59 am

The worst bug I see frequently is that Arena continues a game after an engine has lost on time
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

SzG
Posts: 2424
Joined: Fri Mar 10, 2006 6:20 am
Location: Szentendre, Hungary

Re: Bugs in Arena 3.5.1

Post by SzG » Wed Mar 14, 2018 8:56 am

Modern Times wrote:The worst bug I see frequently is that Arena continues a game after an engine has lost on time
If my memory serves me well there is a setting by which it can be enabled to continue a game after it has ended.
Gabor Szots

CCRL testing group

mike angel
Posts: 50
Joined: Sun Jan 30, 2011 12:08 pm

Re: Bugs in Arena 3.5.1

Post by mike angel » Wed Mar 14, 2018 10:49 am

Surely it could happen - for unknown reasons - that an engine did not run under Arena 3.51 but with Cutechess-Gui 1.00. But also vice versa.

In that case I take the other Gui.

Yes I know in Cutechess-Gui there is actually no possibility for restart a tournament - thats a big problem. But I hope that will be solved soon.

User avatar
CMCanavessi
Posts: 550
Joined: Thu Dec 28, 2017 3:06 pm
Location: Argentina

Re: Bugs in Arena 3.5.1

Post by CMCanavessi » Wed Mar 14, 2018 1:15 pm

mike angel wrote:Yes I know in Cutechess-Gui there is actually no possibility for restart a tournament - thats a big problem. But I hope that will be solved soon.
It's implemented on the TCEC fork of Cutechess, which is also public and on Github.
Follow my tournament and some Leela gauntlets live at http://twitch.tv/ccls

mike angel
Posts: 50
Joined: Sun Jan 30, 2011 12:08 pm

Re: Bugs in Arena 3.5.1

Post by mike angel » Wed Mar 14, 2018 1:38 pm

have you been successfull in compiling it?

SzG
Posts: 2424
Joined: Fri Mar 10, 2006 6:20 am
Location: Szentendre, Hungary

Re: Bugs in Arena 3.5.1

Post by SzG » Wed Mar 14, 2018 2:04 pm

CMCanavessi wrote:
mike angel wrote:Yes I know in Cutechess-Gui there is actually no possibility for restart a tournament - thats a big problem. But I hope that will be solved soon.
It's implemented on the TCEC fork of Cutechess, which is also public and on Github.
I haven't yet been able to find an executable with that feature.
Gabor Szots

CCRL testing group

Milos
Posts: 2984
Joined: Wed Nov 25, 2009 12:47 am

Re: Bugs in Arena 3.5.1

Post by Milos » Wed Mar 14, 2018 3:44 pm

SzG wrote:
CMCanavessi wrote:
mike angel wrote:Yes I know in Cutechess-Gui there is actually no possibility for restart a tournament - thats a big problem. But I hope that will be solved soon.
It's implemented on the TCEC fork of Cutechess, which is also public and on Github.
I haven't yet been able to find an executable with that feature.
Well you should learn to perform trivial tasks such as compiling yourself, so that you don't have to go around and look and ask for executables.
Here is the source:
https://github.com/gvreuls/cutechess

SzG
Posts: 2424
Joined: Fri Mar 10, 2006 6:20 am
Location: Szentendre, Hungary

Re: Bugs in Arena 3.5.1

Post by SzG » Wed Mar 14, 2018 5:41 pm

Milos wrote:
SzG wrote:
CMCanavessi wrote:
mike angel wrote:Yes I know in Cutechess-Gui there is actually no possibility for restart a tournament - thats a big problem. But I hope that will be solved soon.
It's implemented on the TCEC fork of Cutechess, which is also public and on Github.
I haven't yet been able to find an executable with that feature.
Well you should learn to perform trivial tasks such as compiling yourself, so that you don't have to go around and look and ask for executables.
Here is the source:
https://github.com/gvreuls/cutechess
Thanks Milos. I did not know it was trivial. Then the fact that some programmers do not provide executables to their engines must be due to some other reason.
Gabor Szots

CCRL testing group

Post Reply