ICS Interview – Rekursiv

Thanks for agreeing to be interviewed as part of the ICS Interviews series.

REKURSIV

INTERVIEWER

REKURSIV

I always loved chess since I started playing in high school around 1989.  In college I discovered that I could play live chess online.  I don’t remember which server it was, but I started playing on FICS in the late 90s and most of my online chess and atomic chess has been played there.

INTERVIEWER

How did you discover atomic chess? Did you have a teacher to show you the ropes?

REKURSIV

I discovered atomic chess simply by exploring what FICS had to offer. I got sick of playing regular chess badly and atomic offered a nice diversion. I read the rules of all variants FICS had to offer and was immediately drawn to atomic more than any other variation. Before playing my first game, I sensed that the game would be very tactical, very forcing, and easy to analyze–exactly what I wanted. But it took a long time to discover just how deep the game actually was. I didn’t have a teacher per se, but I learned a great deal from a lot of different players. Foxyfiddler and I began playing at the same time, learning the rudiments of the game through trial
and error over dozens of games with each other. Early on, seberg and maciejg would occasionally be kind enough to teach me a few things, usually by beating me. I learned a lot from all of my better opponents, but later when I got better I learned the most from analyzing with yokke, spogorecki, tipau, and especially siggemannen (to whom I owe a great deal). I learned from many other players, but these are the ones that really stick out in my memory.

So what’s with the handle “Rekursiv”?

It refers to the word “recursive,” a math and computer word.  I studied mathematical logic in graduate school.  In particular I specialized in the theory of recursive functions.  In a famous paper written by the Austrian mathematician, Kurt Godel, the author used the spelling “rekursiv.” So, it is an homage to Godel.

Many atomic chess players may know you from the Atomic Chess Book you were working on for a few years. Did you ever complete a working draft of it, or were there still some areas of the Book you wanted to complete? I have an early copy of your website and have discussed this topic with you in the past, but would like to hear about any sections you planned to add to it and were unable to, or if you added any new material to the Book that were never posted online.

I have written a few things that were never put on the website, mostly opening theory, but also a few very deep endgames, and a few
annotated games.

What was your inspiration for starting work on the Atomic Chess Book in the first place?

I simply wanted to catalog the existing atomic knowledge and add to it. Writing and teaching is what I love to do–I teach math by profession. I also wanted to be a better player and thought that writing down what I knew would help. From the start I always felt that all atomic knowledge should be shared to help promote and elevate the game. If I learned something new about the game, I never kept it secret.

Have you played atomic on ICC? What’s your opinion of the lack of check rule that ICC implemented? It’s much like what they did to create loser’s chess because they were unable to implement suicide chess correctly. (On a related note, do you play and enjoy either suicide or loser’s chess?)

I haven’t played atomic on ICC. I think the lack of check rule makes it simpler to implement of course, and it makes it simpler
to communicate the rules, and maybe attract a few extra people to the game. But I think it is a mistake. White has a big advantage
in atomic (although not proven to be a winning advantage by any means). I think the lack of check at ICC gives White an even bigger
advantage. For example, at ICC a king and rook can defeat a lone king if the kings are separated. But at FICS this ending is drawn.
Other endings distinguish the two sets of rules as well, giving White more of an advantage at ICC than at FICS. Also, endings
are just more intricate at FICS (I believe). On the other hand, at ICC you can castle when in check (since there is no check), and
I think it is possible this may have an effect on a couple of openings in Black’s favor, but not enough to overcome the endgame
drawbacks. I think the rules as they are at FICS need to be universally adopted if we want to make any progress legitimizing
this variant. I think schemingmind has the FICS rules, but ICC, buho and other servers don’t.

I do like suicide and losers, but not as much as atomic, so I rarely play them.

You mentioned the deep opening work necessary to succeed in atomic. Do you think atomic chess can be solved? Or at the very least, what’s your opinion on the limited opening tree available in atomic chess?

I think there is less than a 5% chance atomic will ever be solved. Computationally speaking, a move-by-move perfect solution to the
game can never be attained, but there is the possibility of a “human solution,” meaning a way for White to get a winning advantage
no matter how Black responds. The reason I think this is very unlikely is because I think atomic is actually a draw with best play.
I didn’t always think this. Proving a game is a draw is very much harder than proving there is a win.

Openings in atomic are much more limited than in ordinary chess, but there still is a great variety, and new opening lines are being
discovered all the time.

You’ve played some Atomic960, which has been available on schemingmind.com for a while now, although you’ve since left the
site when you quit playing atomic chess. What was your opinion on Atomic960 in general? I particularly appreciated the fact that you
always play a game as both colors in the same position, so that if White had a forced win, the match would level out.

Yes, I enjoy philosophy, but don’t really have the time to pursue it (or many of my other interests). I think the best way to understand
the world is through a combination of mathematics, science, and philosophy. We don’t have much of a mathematical or philosophical
library here (we’re only a two-year college). But the neighboring University of Illinois does, and I sometimes get books through inter-library loan. But again, not a lot of time to pursue my academic interests.

What chess books do you read? I personally sold off my chess book collection, retaining only the variant chess books that I had, plus My
Memorable 60 Games and My System. What are the best chess books you’ve read? For me, it’s probably Seirawan’s No Regrets, which I now regret selling, along with his Five Crowns as well.

I’m not a very strong chess player (about 1700-1800 blitz on FICS). I haven’t read a lot of chess books. My favorite used to be
The Ideas Behind the Chess Openings by Reuben Fine. I also liked Jeremy Silman’s books. My first endgame book ever was a little book
by Pandolfini that was a great introduction to endgames. I’ve bought some specialized opening books over the years (and studied a few
chapters of each before quitting), including one on the Smith-Morra and another on the Closed Spanish. A few months ago, I bought
Chess Openings for White, Explained by Lev Alburt and two other authors. It’s an excellent book that I would highly recommend. There’s a
companion book for Black I want to get eventually. Oh, one of the most entertaining books I own is Test Your Endgame Ability by Livshits and Speelman. It’s just full of extremely fascinating (yet often not practical) endgame studies collected from various authors.

You’ve played against several atomic players and engines. Who are the strongest human players you’ve played and which engine has been
the best overall in atomic chess? Here’s a brief (and not at all complete list) of engines I’ve played in the past: Sjaak (TrojanKnight on FICS),
pulsar, Atomix, and Sordid.

The strongest human players I’ve ever played are tipau and siggemannen. I think it’s common knowledge in my generation of atomic players that they are the best. Other top players that I have rarely played are gmonubense, wildkiller, moltenthinker, and a few others that escape my memory. Currently, GrandLapin is a very good player. He reached a 2500+ rating a while ago and was the FICS champion. Over the years, other strong players that I have played include yourself, yokke, SPogorecki, hooby, mrundersun, and others. The one I fear the most is maybe mrundersun. He’s not as good as many of the other players, but he’s very unconventional and very creative. He plays losing openings and wins them. He’s very tenacious. He’d be a top player if he would sit down and study the openings. As far as engines go, I think the best one that has ever made an appearance on FICS was TrojanKnight. It was very strong (maybe 2700+). If tipau would write it’s opening book it would be unbeatable. But it hasn’t played on FICS for a long time. I’ve never personally played it (but I’ve seen some of its games). The next best engine on FICS is Sordid. Sordid is about 2500+, but it has weaknesses. It needs an updated opening book and it plays some endgames poorly. I like to say that the way to beat Sordid is to arrive at a drawn endgame against it. Anyway, Sordid has been going strong on FICS (playing only unrated in the last few years) ever since I started playing atomic. Its owner, MightyByte, should really be thanked for tirelessly providing it to us.

You also mentioned playing chess against your engines. What’s your opinion on computer chess in general and of computer cheating? Have you ever read the classic website (C)heating on ICSes by Doctor Unclear? It used to be hosted on Geocities, but will be archived on this website as soon as I’m able to clean up my archive of it.

It gets boring playing computer engines, but they can be great training partners and can help you build an opening repetoir. I know that
computer cheating exists on FICS (and other servers), but it doesn’t bother me. When I get an opponent, they have a rating. Whether they’ve achieved that rating through cheating or through their own skill, I should have approximately the same probability of beating them (unless they’ve just started cheating and are on their way up the rating ladder). But over the last several years, I’ve played atomic almost exclusively and I don’t think there’s much cheating going on (if any) in that variant. As far as I know, the only engine widely available is pulsar, and last time I checked pulsar isn’t strong enough to beat me anyway. I’ve never read the (C)heating website you’re referring to.

Atomic chess grabbed your interest immediately. Did you enjoy any of the other variants available on FICS such as crazyhouse or bughouse?

I enjoy suicide and losers a bit. I never liked bughouse. Playing as a team never appealed to me. Crazyhouse is ok, but it’s too complicated
for me. In a blitz game, I would have to rely too much on intuition and making my best guess. But I’ve only played about a dozen games, so I’m not a good judge. But I prefer games that are easier to analyze.

Do you enjoy other chess variants that aren’t playable on FICS such as progressive chess or extinction chess?

I love extinction chess, but haven’t played it in a long time. I think it may have been on brainking that I played it. I also like three checks
chess a lot. Both of these variants are very tactical and forcing, but still have a positional aspect.

What interface do you use to play online chess? What interfaces have you used over the years?

The only interface I have ever used is Winboard. When I started playing at FICS 6 or 7 years ago, I stumbled upon Winboard quite randomly, installed it and have been happy with it ever since. It serves my purposes. I don’t know anything about the other interfaces, but people have told me to switch to something else that they say is much better. But I just don’t care.