Some recent lower level engines

[Graham Banks]

With my Amateur Series, I'm always looking at new engines to see whether any are strong enough to be included.

My cutoff point is that they must be 2000+ ELO based on CCRL ratings.
As Reger 0.09 is around that level, I use it as a test engine against which to measure the newbies.

I tried out six new engines - Betsabe II 1.0, Bubble 1.5, JaksaH 1.10, Purplehaze 2.0.0, Revati 0.4c and Supra 1.1.

Three of the engines were unsuitable due to operating issues.
I couldn't get JaksaH working, but suspect that it might still be below the 2000 threshold.
Supra 1.1 didn't seem to understand either time controls or correct usage of time, moving almost instantly each move.
Revati didn't seem to understand secondary time controls.

Of the three that seemed functional, both Betsabe II 1.0 and Bubble 1.5 are stronger than the 2000 cutoff and will play in my lowest division.
Purplehaze isn't quite strong enough yet, which was disappointing because I like the name and logo. However, for anybody interested in using it, I can tell you that it's stable.

Hope this information might be of use to somebody.

[Matthias Gemuh]

Useful infos. Thanks.


Matthias.

[vincent]

Hi Graham,

I'm the author of Purple Haze and yes there is still a lot of work to be above 2000. For example the evaluation function is too simplistic and there is almost certainly hidden bugs in the search. My current goal is to end up with something stronger than TSCP...

But I almost finished polishing the version 2.0.1 which will be my first "official" release. And I think it could be useful for other developers looking for opponents in weak but complete chess engines using Xboard protocol and supporting GNU/Linux.

[Graham Banks]

Hi Graham,

I'm the author of Purple Haze and yes there is still a lot of work to be above 2000. For example the evaluation function is too simplistic and there is almost certainly hidden bugs in the search. My current goal is to end up with something stronger than TSCP...

But I almost finished polishing the version 2.0.1 which will be my first "official" release. And I think it could be useful for other developers looking for opponents in weak but complete chess engines using Xboard protocol and supporting GNU/Linux.

Hi Vincent,

as I mentioned, great name and great logo.
Really looking forward to the progress of Purplehaze. Thanks for your efforts.

Cheers,
Graham.

[Evert]

With my Amateur Series, I'm always looking at new engines to see whether any are strong enough to be included.

I'm working on releasing both of my engines over the weekend, Jazz and Sjaak (I may need to tweak the latter's name slightly). Jazz is considerably stronger than Sjaak, which is a more general "chess-like" engine. According to my own tests, Jazz is a bit weaker than OliThink, Sjaak is comparable to or a bit stronger than Micro-Max.

Both are single processor, no-ponder engines. Jazz is UCI or XBoard, Sjaak is XBoard only. When I release them, both of them will be GPL. I can make a Win32 compile if I can borrow the Windows netbook for a few minutes, but both engines would really benefit from a 64 bit compile (they use bitboards).

[Graham Banks]

With my Amateur Series, I'm always looking at new engines to see whether any are strong enough to be included.

I'm working on releasing both of my engines over the weekend, Jazz and Sjaak (I may need to tweak the latter's name slightly). Jazz is considerably stronger than Sjaak, which is a more general "chess-like" engine. According to my own tests, Jazz is a bit weaker than OliThink, Sjaak is comparable to or a bit stronger than Micro-Max.

Both are single processor, no-ponder engines. Jazz is UCI or XBoard, Sjaak is XBoard only. When I release them, both of them will be GPL. I can make a Win32 compile if I can borrow the Windows netbook for a few minutes, but both engines would really benefit from a 64 bit compile (they use bitboards).

Hi Evert,

I'll include the stronger of the two engines because I limit participation to one engine per author.
As an example, I include Spark, but not Bright (even though both play differently).

Looking forward to testing Jazz for you.

Cheers,
Graham.

[Dr.Wael Deeb]

With my Amateur Series, I'm always looking at new engines to see whether any are strong enough to be included.

I'm working on releasing both of my engines over the weekend, Jazz and Sjaak (I may need to tweak the latter's name slightly). Jazz is considerably stronger than Sjaak, which is a more general "chess-like" engine. According to my own tests, Jazz is a bit weaker than OliThink, Sjaak is comparable to or a bit stronger than Micro-Max.

Both are single processor, no-ponder engines. Jazz is UCI or XBoard, Sjaak is XBoard only. When I release them, both of them will be GPL. I can make a Win32 compile if I can borrow the Windows netbook for a few minutes, but both engines would really benefit from a 64 bit compile (they use bitboards).

Hi Evert,

I'll include the stronger of the two engines because I limit participation to one engine per author.
As an example, I include Spark, but not Bright (even though both play differently).

Looking forward to testing Jazz for you.

Cheers,
Graham.

Bright plays stronger than Spark and it's rated higher in my rating list....
Dr.D

[Graham Banks]

Bright plays stronger than Spark and it's rated higher in my rating list....
Dr.D

Spark 1.0 (after 940 games) is roughly 90 ELO stronger than Bright 0.5c (after 1539 games) in the list that I use, which is why I choose it over Bright in my Amateur Series.
I guess we can agree that both are great engines though.

Cheers,
Graham.

[hgm]

Have you already tried Daniel Shawul's Nebiyu 1.1? I have not tried that at normal Chess (but it says it supports it, if you don't click the Checkers version by mistake ), but it crushes Fairy-Max at Spartan, (even with 6-fold time odds), so it must be well over 2000 CCRL for normal Chess as well.

[edit] Oh, I guess you don't want that either because of Scorpio...

[Graham Banks]

Have you already tried Daniel Shawul's Nebiyu 1.1? I have not tried that at normal Chess (but it says it supports it, if you don't click the Checkers version by mistake ), but it crushes Fairy-Max at Spartan, (even with 6-fold time odds), so it must be well over 2000 CCRL for normal Chess as well.

No I haven't, mainly because I use his strongest engine, Scorpio.

Cheers,
Graham.