Everytime I visit the site, I have hope that when I go to the Rybka section, I will find Vas' post say 'Rybka 4 release date', unfortunately not.
By the way, where is Vas? :)
I don't find it gentle.
I hope Vas will need me for something, be sure he will wait toooo long without any any explanation :)
with that said, if it's not out by the end of the month -- we grab the torches, find him & lynch him.
I think that Rybka 3 was so strong that Vas will have problems to create another version with at least 80 ELO more than R3. I think the 80 ELO limit will be reached around February or so, so April sounds a logical date of release before it gets cleaned up.
A Rybka 4 version less than 60 ELO for example would bring the first turbulences on the hard core Rybka fans and if this continues with newest Rybka's(i believe ELO improvement for Rybka from now on will be more and more difficult to achieve, not because Rybka 3 has no room for improvement, on the contrary, but because i think we are on a point(like in F8,S8 time) where one can not improve on Rybka 3 too much until some new good ideas arise) then Rybka's fame and publicity will start to fall rapidly.
So i think Vas is too clever and will avoid this, so he would release the new Rybkas only when he has considerable improvement and that would take more than a year for sure.
> i think we are on a point(like in F8,S8 time) where one can not improve on Rybka 3 too much until some new good ideas arise) then Rybka's fame and publicity will start to fall rapidly.
Same thing was said with Rybka 2.3.2a and before Rybka 3 arrive. A 100 elo improvement over 2.3.2a version was unimaginable!
Now, Rybka 3 is 14 months old, and although I know that each time is more difficult to improve Rybka, I still believe that an improvement of about 80 or 100 points over Rybka 3 is perfectly possible. Vas practically works full time in Rybka, so please don't underestimate the great Vas now! :-)
I think Rybka's fame will NOT fall while it continues being the number one, not even a little. Rybka's reign seems safe, its throne doesn't wobble easily.
1. Rybka 4 is being developed and will be out when it's ready. Nobody knows when that is
2. Vas has given up on Rybka for a bit and is having a break. There is nothing to suggest that is the case.
In the meatime, what is it about Rybka 3 that people are so unhappy with that they must have Rybka 4? If everyone else has Rybka 3 there is no disadvantage by Rybka 4 not being out? Rybka 3 is perfectly good for analysing games i.e. what's the problem with waiting?
> 1. Rybka 4 is being developed and will be out when it's ready. Nobody knows when that is
1.a) Rybka 4 is being developed and it will come very soon. (F12 and S12 are out, R4 waiting for what?)
1.b) Rybka 4 is not being developed and will be out when it's ready. Nobody knows when that is.
(I don't know)
Could this be one of the reasons for the delay, and the Frankenstein of course?
Vas had that one as a promise "not too far away" early this year.
What went wrong?
I as a customer want to know now!
Please Vas clarify a little at least.
You owe all rybka users an explanation, do it quick.
To keep us all holding our balls, really suks, Vas
I share your frustration - but in the end, no news is hopefully good news & Vas will stun us all once again. We could always use the waiting time to improve our own chess?!
> let Rybka 4 be released when Vas thinks its ready!
And if there are real problems, I whish that these problems are cleared, when R4 comes out.
It's Vas who has to earn less Rybka-Money today. So he will have thought about how to do best.
But perhaps he really could have had time to give us R3+ several months ago.
That would have given a nice toy to us impatient kids.
Maybe you know that good old German proverb: Gut Ding will Weile haben. ;)
> richtig :-)
Wow! this sounds like Afrikaans "rigtig" :)
( the first one was way too easy ;) )
> There are no real problems :-)
Lukas, at least if we hear words like these from you we can be calm. At least the words come from the reliable source, unlike waiting in the darkness. :)
There are no real problems if you have nine octal i7 computers working in a cluster. :-D
> There are no real problems
Nice to hear this. :-)
> Gut Ding will Weile haben
I remember, when Rybka 2.3x came out.
The last version was very good. But he first seemed to be not ready developed.
It's a good Idea not to do again in this way.
R3 didn't solve all whishes, and these whishes aren't solved until now. Not fine, but OK.
But R3 was a very strong an stable version right from the first day on. That was really good!
> It wont be far fetched to assume Vas is no different to the suicide bomber, both extremist!!!
Should we give him until month end? Maybe he should not hang too long in the forum next time. He must only hang in the forum for only 2 months after the release of R4. After 2 months he must start working on Rybka 5. I wonder if he will give us the crazy personality he was talking about :)
Sidserious and Jessica must start posting again :)
I'm only interested in analysing games. Though I do not have a lot of concrete evidence (yet),
I have the feeling that R2.3.2a was way better in analysis (blunder check). R3 seems to give (perhaps regularly)
"better" moves where "you" (i.e. Rybka!) end up losing a minor piece after only 2 plies or so...
Also, R3 gives a comment ("better move" annotation) on about 30 to 60 % of the moves in a game whereas R232a gave
comments only on about 10 to 30¨% of the moves, with the same threshold.
I'm a bit suspicious, and at the same time a bit disappointed... Vas promised that rybka was going to be the best
analysis engine ever... but seems to have sold his soul to the Elo devil....
At the end of my club's current tournament I will try to make an evaluation and probably switch back to R2.3.2a for analysing games :-(
> but seems to have sold his soul to the Elo devil....
I will say one thing though:
I have made many comparisons between the versions in analysis of many different games, and Rybka 3 gives better, more consistent, and more accurate analysis than any of the 2.3.* versions in over 95% of games I've analyzed.
My impression is that Rybka 3 IS the best analysis engine ever made.
However, I am curious, and wonder if you could post the game analysis of the same game by Rybka 2.3.something and by Rybka 3, where you believe Rybka 2.3.*'s analysis was better. That would be easier to comment on.
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill