whereagles wrote:Last TCEC needed north of ~10k. The money was used for renting high performance servers.
I'm not sure this business model is adequate because it is expensive and does not capitalize well. That 10k could have been used for buying a reasonable machine for a couple of seasons and chessdom would only have to pay for electricity, which should be much easier.
Indeed! Their "business model" really makes no sense…
And to further save on electricity, as well as make the event more interesting to watch, they should reduce the tc drastically (eg. 60'+30" is more than enough IMO). Just watching a single game of TCEC final takes 8 hours!!! So, apart from a few people who have no life, it's impossible to watch games live: way too slow.
Instead you just watch the replay after, but then you dont need TCEC for that. Graham's tournament are equally interesting for that matter (more in fact). Another problem with ultra long tc is the draw rate skyrockets. But, wait, TCEC folks have found a "solution" for that in S9: they use some biaised openings, giving white a quasi won game. So lame…
It's really a misconception that LTC games are more interesting. On the contrary, blitz games are much more sharp and exciting. You really see engines going for it, and getting into crazy double-edged positions in ultra blitz. Much more fun!
Theory and practice sometimes clash. And when that happens, theory loses. Every single time.