If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
Smallpotato uses lots of ram and always seems to ponder.If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
Turns out, pondering is not a problem. It ponders when I run it from the command line, but somehow when running as a Winboard engine, it gets pondering sorted out correctlySmallpotato uses lots of ram and always seems to ponder.If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
ftp://cap.connx.com/pub/chess-engines/new-approach/smallpotato.exeWhere find an exe? if any
Turns out, pondering is not a problem. It ponders when I run it from the command line, but somehow when running as a Winboard engine, it gets pondering sorted out correctlySmallpotato uses lots of ram and always seems to ponder.If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
Hmmm...Turns out, pondering is not a problem. It ponders when I run it from the command line, but somehow when running as a Winboard engine, it gets pondering sorted out correctlySmallpotato uses lots of ram and always seems to ponder.If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
At least here under WB 4.26 it still uses CPU time of the opponent
when ponder=off!
Moreover it uses more than 132MB Ram from start on!( (almost unusable by now)
Hmmm...At least here under WB 4.26 it still uses CPU time of the opponent
when ponder=off!
Moreover it uses more than 132MB Ram from start on!( (almost unusable by now)
Seems to be OK on my machine. I have to give it a new name though.
"The Rook Wobbler."
I almost busted out laughing the first time I saw it.Hmmm...At least here under WB 4.26 it still uses CPU time of the opponent
when ponder=off!
Moreover it uses more than 132MB Ram from start on!( (almost unusable by now)
Seems to be OK on my machine. I have to give it a new name though.
"The Rook Wobbler."
Do you mean sth like (1. Na3?!) 2.Rb1?! 3.Ra1?? 4.Rb1???)
Pondering is not implemented yet, so ponder is always off, really. The code is there so that one day pondering will do something but right now, it just does a thread yield (which in linux, seems to work fine. when not it's turn it consumes about 0.1% of cpu time on my machine)Turns out, pondering is not a problem. It ponders when I run it from the command line, but somehow when running as a Winboard engine, it gets pondering sorted out correctlySmallpotato uses lots of ram and always seems to ponder.If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
At least here under WB 4.26 it still uses CPU time of the opponent
when ponder=off!
Moreover it uses more than 132MB Ram from start on!( (almost unusable by now)
Regards,
Günther
I almost busted out laughing the first time I saw it.Hmmm...At least here under WB 4.26 it still uses CPU time of the opponent
when ponder=off!
Moreover it uses more than 132MB Ram from start on!( (almost unusable by now)
Seems to be OK on my machine. I have to give it a new name though.
"The Rook Wobbler."
Do you mean sth like (1. Na3?!) 2.Rb1?! 3.Ra1?? 4.Rb1???)
Maybe "The Tempo Toilet" would be even better.
Actually, the code is rather sophisticated. If he added an opening book, I think it would add 200 ELO to his program. Once it gets out of the opening phase, it stops playing like an utter buffoon.
Pondering is not implemented yet, so ponder is always off, really. The code is there so that one day pondering will do something but right now, it just does a thread yield (which in linux, seems to work fine. when not it's turn it consumes about 0.1% of cpu time on my machine)Turns out, pondering is not a problem. It ponders when I run it from the command line, but somehow when running as a Winboard engine, it gets pondering sorted out correctlySmallpotato uses lots of ram and always seems to ponder.If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
At least here under WB 4.26 it still uses CPU time of the opponent
when ponder=off!
Moreover it uses more than 132MB Ram from start on!( (almost unusable by now)
Regards,
Günther
command line option to set hashtable size. The default is 4 million entries, which comes to about 100Mb. I should change it so you can specify memory occupied instead of number of entries
if you do smallpotato --help, you should see all the options
Alejandro
Hi Alejandro,Pondering is not implemented yet, so ponder is always off, really. The code is there so that one day pondering will do something but right now, it just does a thread yield (which in linux, seems to work fine. when not it's turn it consumes about 0.1% of cpu time on my machine)Turns out, pondering is not a problem. It ponders when I run it from the command line, but somehow when running as a Winboard engine, it gets pondering sorted out correctlySmallpotato uses lots of ram and always seems to ponder.If someone with a Linux box can generate a cmdline.h file, I may be able to create a Win32 version. I have one here, but I'm too lazy to go for it.http://members.optushome.com.au/alito/smallpotato.html
Xboard compatiable of course.
At least here under WB 4.26 it still uses CPU time of the opponent
when ponder=off!
Moreover it uses more than 132MB Ram from start on!( (almost unusable by now)
Regards,
Günther
command line option to set hashtable size. The default is 4 million entries, which comes to about 100Mb. I should change it so you can specify memory occupied instead of number of entries
if you do smallpotato --help, you should see all the options
Alejandro
Definitely not the words i would useHi Alejandro,
Forget what I wrote in my first post, SmallPotato appears
to be very strong!
But I still have the problem that it uses 50% CPU time of the
opponent when it' s not to move even with ponder=off under
my WIN98Se.
As I would really like to see it playing in my soon starting tourney,
I beg you for a solution.
I've got a hack in mind which i think would work, but a problem i've got at the moment is that i cannot compile my own version under cygwin, and i want to stop hassling Dann Corbit for them. Once i get that working, i'll post a new version with hopefully this fixed.But I still have the problem that it uses 50% CPU time of the
opponent when it' s not to move even with ponder=off under
my WIN98Se.
As I would really like to see it playing in my soon starting tourney,
I beg you for a solution.
Alejandro
ok, problems solved. i wrote a simple benchmark to run in parallel while i was using smallpotato to check if it was being slowed down at all, and there is no slowdown. i've put the executable on the web page below.I've got a hack in mind which i think would work, but a problem i've got at the moment is that i cannot compile my own version under cygwin, and i want to stop hassling Dann Corbit for them. Once i get that working, i'll post a new version with hopefully this fixed.But I still have the problem that it uses 50% CPU time of the
opponent when it' s not to move even with ponder=off under
my WIN98Se.
As I would really like to see it playing in my soon starting tourney,
I beg you for a solution.
Alejandro
That would be really fine:)
Thanks in advance.
Regards,
Günther
Thanks a lot Alejandrook, problems solved. i wrote a simple benchmark to run in parallel while i was using smallpotato to check if it was being slowed down at all, and there is no slowdown. i've put the executable on the web page below.I've got a hack in mind which i think would work, but a problem i've got at the moment is that i cannot compile my own version under cygwin, and i want to stop hassling Dann Corbit for them. Once i get that working, i'll post a new version with hopefully this fixed.But I still have the problem that it uses 50% CPU time of the
opponent when it' s not to move even with ponder=off under
my WIN98Se.
As I would really like to see it playing in my soon starting tourney,
I beg you for a solution.
Alejandro
That would be really fine:)
Thanks in advance.
Regards,
Günther
Alejandro
no, sorry, analysis is not implemented yet (even worse, the program doesn't know what the PVS is so all it could report would be best score, and best first move for now)I have already tested it and it's all ok now.
(Btw just a cosmetical question: Is it possible by adding
a commandline parameter that it shows thinking under Winboard?)
Return to Archive (Old Parsimony Forum)
Users browsing this forum: No registered users and 14 guests