Thanks Bram and Jesus for your interest and input. Usually when a person is testing a version that has not been released to the public yet- that is it. But I have 2 engine versions running against each other- and "neither" has been released. So to honor Yuri as well as Don and Larry- the less I say is probably for the best. The only thing I don't mind saying is these are 2 damn strong engines- and both are chasing a difficult foe. It seems by the time you get there- he has come and gone. [referring to Houdini as the foe]
Thanks Dmitri for your nice comment. I wanted to wait until an update to respond because when I do, it pulls all the threads back to the top. I try not to do it any more than necessary out of respect for other testers. Tho I may be the only thread viewer left.
90 more games have been played, and Komodo has increased his lead a slight bit. On 2nd look- the increase maybe should be called more than "slight."
I am afraid that any chance Strelka had of catching Komodo has gone by the wayside. What he can do is make a fairly deep cut in the lead- I suppose that is always a possibility. However slim it may or may not be.
Let me just say that Komodo 4536 is definitely a beta version. As for Strelka 5.6- it hit me that I don't know if this is the version he will release or not. It is something I never considered asking him. That is assuming there "is a release." Another little item I overlooked asking as well.
I would imagine with 150 games remaining- there will be 2 more updates: the next-to-last and the last. But that will not be the last testing of either. We need games with time controls that are a bit longer.
In fact, I have a result I will post for you very soon between these 2 engines here in a 40/40, 50-game match. Very interesting result!
A change of pace here, with the final results from k-4536 and Strelka 5.6 at 40/40 repeating for a 50-game match. I meant to post these results last night and got busy and forgot.
XPS 8500 Intel i5-4 True Cores
Fritz 13 gui
1CPU/64-bit
128MB hash
Bases=NONE
Ponder_Learning=OFF
Perfect 2012b.ctg w/12-move limit 40/19 Repeating (Benched to adapt to 40/40)
Match=50 games
I am going to tread real lightly here when I say that from my POV, the Komodo beta in question, k-4536- is slightly stronger than Strelka 5.6 or Strelka 5.6 beta- depending here on what you prefer to call it.
There are a couple things here that need to happen now. (1) Run a 40/40 match, again 50 games- between Komodo 4536 and Critter 1.6a. (2) Get a comparison in terms of elo between Strelka 5.5 and Strelka 5.6. I have many matches to use for the comparison, but if a few more are needed I will run them as well. Then I will give you the stats and you can see if you agree with me or if you are wrong.
1 more update thread in 20 or so hours to finish out this 1000 game match- and I will then let these threads drift into oblivion. I will start a new post out of respect to other testers. This is now way too many threads to keep pulling back to the top of Page 1. Just stay with me for one more update.
This posting finishes out the 1000 games, and at 5'+3" you got some games of high quality that 1+1 "might" not generate. Anyway, we got 2 good versions here, and the nice thing is they very well may "both" be betas when the dust settles. Remains to be seen.
I enjoyed this one, as 1000 at 5+3 can give you some reasonably good ideas that might be applicable in dealing with some possible strength issues- at certain points in time.
