George Speight wrote:I have converted Ant to UCI, and it runs perfectly in engine matches. But the problem is it is supposed to use Nalimov TB. As with many converted winboard engines, it doesnt show that it is using them in the analysis lines during a game. But with the other engines, I have a couple of foolproof 3 and 4 man position setup tests. When you set up the position, they all auto show #44. So you know they are accessing the bases, even tho they dont show in analysis lines during a game. But with Ant, when you set up the position, and click on black or white to move, depending on the position you are using, Ant just sits there and does nothing. Shows no thinking lines at all. And this is the only way i have of knowing if it is accessing the bases or not. Is there anyone out there that can help with why it wont show ANYTHING when you set up a position and ask it to move. Thanks very much for any help you can give on this.
George
Hello George,
This message is first for you and other hobbyists.
The bewildering number and types of amateur engines affords computerchess hobbyists endless hours of tinkering and fun.
There are bound to be engines that present challenges but there are some standard procedures that may be employed before a complaint is launched.
Leo (he who needs no last name!) has done a phenomenonal work on his website at his engine details dropdown pages.
A quick check there would have answered virtually all your questions. Ant does not support analysis mode so your test would fail. You could alllow the engine to play out the position and verify the moves with Chest UCI, another wonderful piece of software.
In particular, if you are going to convert winboard engines to UCI via the use of an adapter, you should know that there are prepared files for precisely this purpose freely available on the net. Try these first.
Assuming that you are already proficient at that, where you notice a
bug, please try the engine in its native gui (Winboard or Arena (set as a Winboard engine)) first. Should it work properly there then the problem is in the set up somewhere. A bit of rational experimentation is not an unreasonable requirement.
When all else fails - perhaps not even all but when you have at least consulted Leo's excellent pages - then feel free to post here with the details of your problem. It is just a wee bit disconcerting to see post about a problem that could be answered in a second.
That is why I posted this, to assist all the WB2UCI fans out there.
I hope this helps.
Later.