Komodo wins TCEC

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Modern Times
Posts: 3550
Joined: Thu Jun 07, 2012 11:02 pm

Re: Komodo wins TCEC

Post by Modern Times »

lkaufman wrote:
They primarily run blitz games, whereas you run games at reasonably "serious" time controls. Komodo is not (yet) as strong as current dev. Stockfish at blitz, but at longer time controls it's a different story, as the CEGT and CCRL rating lists make clear. The drawscore is also an issue, but in my opinion the longer time control is much more important for Komodo than using zero drawscore.
I'm curious, so I've started a 3+2 match on 6 cores. I'll let you know the result in due course. Early days, but so far Komodo is doing just fine.
beram
Posts: 1187
Joined: Wed Jan 06, 2010 3:11 pm

Re: Komodo wins TCEC

Post by beram »

Modern Times wrote:
lkaufman wrote:
They primarily run blitz games, whereas you run games at reasonably "serious" time controls. Komodo is not (yet) as strong as current dev. Stockfish at blitz, but at longer time controls it's a different story, as the CEGT and CCRL rating lists make clear. The drawscore is also an issue, but in my opinion the longer time control is much more important for Komodo than using zero drawscore.
I'm curious, so I've started a 3+2 match on 6 cores. I'll let you know the result in due course. Early days, but so far Komodo is doing just fine.
hi Ray,

Happy new year to you and everyone
I suppose you have sufficient games now
Can you share the results with us Ray ?

grts Bram
User avatar
Graham Banks
Posts: 41455
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Komodo wins TCEC

Post by Graham Banks »

beram wrote:
Modern Times wrote:
lkaufman wrote:
They primarily run blitz games, whereas you run games at reasonably "serious" time controls. Komodo is not (yet) as strong as current dev. Stockfish at blitz, but at longer time controls it's a different story, as the CEGT and CCRL rating lists make clear. The drawscore is also an issue, but in my opinion the longer time control is much more important for Komodo than using zero drawscore.
I'm curious, so I've started a 3+2 match on 6 cores. I'll let you know the result in due course. Early days, but so far Komodo is doing just fine.
hi Ray,

Happy new year to you and everyone
I suppose you have sufficient games now
Can you share the results with us Ray ?

grts Bram
I haven't heard from Ray for the past couple of days, but I'm sure he'll let you know his results once he gets the chance.
gbanksnz at gmail.com
Modern Times
Posts: 3550
Joined: Thu Jun 07, 2012 11:02 pm

Re: Komodo wins TCEC

Post by Modern Times »

beram wrote: hi Ray,

Happy new year to you and everyone
I suppose you have sufficient games now
Can you share the results with us Ray ?

grts Bram
KOMODO 1339 64-BIT 6CPU v STOCKFISH 221214 64-BIT 6CPU

AMD X6 1090T using 6 cores
Shredder GUI
Hash 1GB per engine
Syzygy 5-men tablebases on SSD
Time control 3' + 2"
200 opening positions from swcr-fq-openings-v4.1.pgn
each played reversed colours
* Komodo playing with drawscore 0 *

Currently 145.0 - 155.0 (+43, =204, -53)

So it is very close. 100 more games to go.

I have no idea what the openings are like or the range, please feel free to be critical when I post the pgn on completion....
shrapnel
Posts: 1339
Joined: Fri Nov 02, 2012 9:43 am
Location: New Delhi, India

Re: Komodo wins TCEC

Post by shrapnel »

Only 1 GB hash and 3 minute games ??
DEFINITELY NOT ideal conditions for Komodo !?
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis
Modern Times
Posts: 3550
Joined: Thu Jun 07, 2012 11:02 pm

Re: Komodo wins TCEC

Post by Modern Times »

shrapnel wrote:Only 1 GB hash and 3 minute games ??
DEFINITELY NOT ideal conditions for Komodo !?
Acccording to Larry, no.
beram
Posts: 1187
Joined: Wed Jan 06, 2010 3:11 pm

Re: Komodo wins TCEC

Post by beram »

lkaufman wrote:
Modern Times wrote:
beram wrote:If Tom Casanovas or me could get a hand on latest Komodo 1339 as now tested by Ray Banks I think our testresult for latest Stockfish 221214 against that Komodo latest dev, would still be about 53-55%, as for now latest SF dev scores about 56-59% against Komodo 8
What are your test conditions ? I might have time to run a match here.
They primarily run blitz games, whereas you run games at reasonably "serious" time controls. Komodo is not (yet) as strong as current dev. Stockfish at blitz, but at longer time controls it's a different story, as the CEGT and CCRL rating lists make clear. The drawscore is also an issue, but in my opinion the longer time control is much more important for Komodo than using zero drawscore.
Dear Larry, two statements you made need correction
1. I have shown that there is no difference between latest SF dev an K8 when going from TC 3m2s by 10m15s to 30m20s which seems to me reasonable ''serious'' time controls.
2. In the latest LTC matches run by Graham and Ray Banks on 8 cores between latest SF dev(from dec) and latest K1331 and K1339, SF came out as a clear winner

When it comes to drawscore, this can be a serious issue
Maybe Ray will rerun his 400 games blitz match with K1339 at default drawscore. In the first match after 300 games SF was only winning by a small margin of 51-52%
stagehand
Posts: 14
Joined: Fri Oct 04, 2013 1:43 pm

Re: Komodo wins TCEC

Post by stagehand »

[quote="
M ANSARI i remember you from the rybka board many years ago; glad youve kept this hobby alive as ive always enjoyed your posts about it very much.[/quote]

Yes always had interesting posts, sorry to say I can no longer post there myself (but visit!) due to being accused of 'trolling'. Should anyone disagree
with a certain person there about Aquarium and it's faults (even with the new version)you are banished :lol:
However great work from Komodo. The subscription idea is interesting having used/purchased the engine right from its start.