Andrew Grant:
"Almost exactly one year ago we announced the release of Torch. We were a bit ambitious at the time, and announced it as the #2 engine, after beating Leela in the most recent Bullet event at CCC at the time. Today, that claim rings much more true, although still has some minor points of contention. Torch has beaten Leela in multiple Bullet and Blitz events in a row; Never managed to win one of the Rapid events; But has beaten Leela in a h2h classical event.
Chesscom got into the space out of a need to develop a replacement for Komodo. They needed a strong chess engine which could be used for integrating their other products. Many features on Chesscom, from Game Review, to puzzle generation, to Bots (obviously!) are powered by high quality engines. The goal was always for that. In fact, despite me being the only full time employee working on Torch, a plurality of my time was spent on things not related to Torch's strength.
Its a bit self-aggrandizing, but the Torch team got a lot done, managing to very quickly beat Dragon by large margins, establish itself as the rank 2 engine with some confidence, and take a fair number of games off the king. I personally hoped to achieve more, but considering the sparse nature of the team, we are certainly ahead of schedule and under-budget.
As of now, the development effort strictly for strength is at a close. Unfortunately that also means I have no justification to continue working on OpenBench during company time, so developments there that benefit everyone here will be slowing down. Work in Torch will be rounding out some outstanding issues (Numa, some Syzygy stuff), just generally cleaning up the code where possible, and then harnessing Torch to power tools.
One such example: Torch has a heavily modified alpha beta search, which allows a sort of multi-multi-multi-...-pv. We're able to extract a large number of lines from the search in a single search, and we're going to use that to provide humans a better experience when reviewing their games -- avoiding giving confusing engine lines that are only marginally better than what a strong player would have done. Hopefully one day, Torch can be the new home of all of the Chesscom bots as well.
As time passes, interest in investing in Torch's strength may arise once again. But until then, I want to say thanks to the core team, Finn Eggers, Kim Kahre, Michael Whiteley, and Jay Honnold, as well as those who have contributed to Torch's strength, Dietrich Kappe, Kenneth Wilber, and Morgan Houppin. This ragtag team of guys forever have a place in my heart.
A big thanks to all the fans, testers, viewers, content creators, and general enthusiasts who followed our progress."
Comments
Post a Comment