TCEC Season 9

[S.Taylor]

(I was being sarcastic).

But i would NEVER!!!!!!! wish for quicker time cointrols. This would be upsetting.
They need as much core power and time as possible, to mazimize level of play as much as practically possible.
Neither SF nor Komodo are yet as great as humans, positionally. I'm still wishing for a lot better.
Reducing TC or core power would not help this.

[fenchel]

(I was being sarcastic).

But i would NEVER!!!!!!! wish for quicker time cointrols. This would be upsetting.
They need as much core power and time as possible, to mazimize level of play as much as practically possible.
Neither SF nor Komodo are yet as great as humans, positionally. I'm still wishing for a lot better.
Reducing TC or core power would not help this.

(That's why I said you were being critical; your use of sarcasm indicated this issue really bothered you.)

I definitely am not suggesting shorter TC, I'm just trying to keep people interested. Martin runs this as a spectator event, it is much different from tournaments run by Graham, Ray, Clemens, etc.

Personally, I wish we could really trust engines' use of 0.00. If they both say it early on, for a few moves early, then fine let's end the game, regardless of pawn moves etc. If these engines can crush GMs with handicaps, I similarly thing they should be brilliant at offering early draws.

Anyway, just an opinion. I'm a newcomer here.

[S.Taylor]

But since it is ongoing 24/7, and no one has the time to hang around anyway, they can go away for longer times and still come back, before missing too much.
I mean, making it shorter tc's does not mean that people will then be watching games from beginning to end in real time.
(Although everyones life setups, and preferences are different)

[APassionForCriminalJustic]

(I was being sarcastic).

But i would NEVER!!!!!!! wish for quicker time cointrols. This would be upsetting.
They need as much core power and time as possible, to mazimize level of play as much as practically possible.
Neither SF nor Komodo are yet as great as humans, positionally. I'm still wishing for a lot better.
Reducing TC or core power would not help this.

Both Stockfish and Komodo are a lot better than humans both positionally and tactically. Heck even on a cell phone the games would be a lot better than what a bunch of super grandmasters could produce. Awesome dual-socket hardware only increases ELO, and level of play further. Perhaps what you see is a level of play too good for you to properly judge. It is in many ways too complicated for human comprehension (more so amateur comprehension) except in simple, obvious scenarios.

The only real area where engines still struggle with is closed positions. They seem to struggle with understanding that some positions are unbreachable. But hey it is code.

Computer-chess isn't exactly a popular hobby here. I wouldn't be expecting a mass number of people. It doesn't matter that engines could stomp on super grandmasters like little ants in the dirt. The majority of people still take most of their interest in human chess - such as the case with the World Chess Championship aka Magnus.

[APassionForCriminalJustic]

But since it is ongoing 24/7, and no one has the time to hang around anyway, they can go away for longer times and still come back, before missing too much.
I mean, making it shorter tc's does not mean that people will then be watching games from beginning to end in real time.
(Although everyones life setups, and preferences are different)

Honestly to even have 500 people or 400 people or whatever number in the hundreds continuously logged in for long, draw-like games like this is a huge feat for Martin. Perhaps there is simply too much expectation.

[Modern Times]

Honestly to even have 500 people or 400 people or whatever number in the hundreds continuously logged in for long, draw-like games like this is a huge feat for Martin. Perhaps there is simply too much expectation.

Yes. Combination of very big hardware, the two strongest engines, long time control, live match accessible through the browser, lots of information with PVs etc, and live chat. The chat is particularly important I think, and more so when the authors themselves sometimes contribute.

[S.Taylor]

One improvement i feel strongly about, and i thought i read that someone else mentioned it but i can't find it right now, is,

that the reverse sides of the same opening should immeadiately follow the the first time of that opening (e.g. when it was 8 moves for each side).

This is when it is still fresh in peoples minds and they can immeadiately compare how the reverse engines handled what their counterparts did.
If there are other games in the middle, it is almost impossible to still have the same interest as when it was fresh in everyones mind. Why distract us by mixing us up?

[Graham Banks]

The improvement I'd most like to see is watching the engines think as it's their move.

[pohl4711]

For a lower number of draws, I recommend the SALC-openings, which I use for my testwork.
They can be found on my website (Download & Links- section), or just click:

http://spcc.beepworld.de/files/salc.zip



Stefan (SPCC)

[JJJ]

For a lower number of draws, I recommend the SALC-openings, which I use for my testwork.
They can be found on my website (Download & Links- section), or just click:

http://spcc.beepworld.de/files/salc.zip



Stefan (SPCC)

Yeah it could be interesting because the game would be a lot more agressive with opposed king side.