Well you sure only look at the details....
98% accuracy in this 1 min bullet game!
Sort:
Apr 17, 2020
0
#23
llama44 wrote:
Yeah, but based on what? How'd you come up with 35? Why not 34? How do you determine complexity? What's the average complexity of a position that requires depth 35? What does it mean to be "trustworthy?" What measure of trustworthiness is depth 30 and why?
etc etc
In other words you made it up
No, an IM told me to get atleast 35.
Apr 17, 2020
0
#24
antisunechess wrote:
Well you sure only look at the details....
I don't know what that means, but for example, if the answer was simply "that's what most top players agree on, Hikaru mentions it on stream all the time"
Then I'd say ok, I guess 35 is somehow the good number.
Generally speaking though, the more depth the analysis has the better, and it is good to have at least a depth of 35 moves. The best is 50, in order for the engine to see 50 move rule draws
Apr 17, 2020
0
#26
Asparagusic_acids wrote:
llama44 wrote:
Yeah, but based on what? How'd you come up with 35? Why not 34? How do you determine complexity? What's the average complexity of a position that requires depth 35? What does it mean to be "trustworthy?" What measure of trustworthiness is depth 30 and why?
etc etc
In other words you made it up
No, an IM told me to get atleast 35.
Ok so he made it up 
I'm not saying 35 is bad, I know 35 is a high depth, and to do much more wouldn't be practical, it's just other than that I think it's completely arbitrary.
Apr 17, 2020
0
#27
antisunechess wrote:
Generally speaking though, the more depth the analysis has the better, and it is good to have at least a depth of 35 moves. The best is 50, in order for the engine to see 50 move rule draws
lol.
The "best" would be determined by something like measuring % difference based on further analysis. In other words how likely is the engine to change its mind. In some positions you might not need any depth.
Apr 17, 2020
0
#28
llama44 wrote:
Asparagusic_acids wrote:
llama44 wrote:
Yeah, but based on what? How'd you come up with 35? Why not 34? How do you determine complexity? What's the average complexity of a position that requires depth 35? What does it mean to be "trustworthy?" What measure of trustworthiness is depth 30 and why?
etc etc
In other words you made it up
No, an IM told me to get atleast 35.
Ok so he made it up
I'm not saying 35 is bad, I know 35 is a high depth, and to do much more wouldn't be practical, it's just other than that I think it's completely arbitrary.
There are exceptions e.g. in complex position where you will need a lot of pvs while in simpler positions you will need less pvs.
Apr 17, 2020
0
#29
Well, I'm not an IM, but I feel like any coach (maybe he's not your coach) who tells his student to analyze the games to "at least" depth 35 is a bit of a twit.
Just my opinion.
Apr 17, 2020
0
#30
llama44 wrote:
Well, I'm not an IM, but I feel like any coach (maybe he's not your coach) who tells his student to analyze the games to "at least" depth 35 is a bit of a twit.
Just my opinion.
Not for game analysis but for analyzing specific positions e.g. an opening or an endgame that you don't understand.
Apr 17, 2020
0
#32
LongIslandWoman wrote:
Asparagusic_acids wrote:
llama44 wrote:
Well, I'm not an IM, but I feel like any coach (maybe he's not your coach) who tells his student to analyze the games to "at least" depth 35 is a bit of a twit.
Just my opinion.
Not for game analysis but for analyzing specific positions e.g. an opening or an endgame that you don't understand.
Aren't there many examples of endgames, thematically? Is it very important to look into one specific one or just learn the approach and theory for the different catagories?
Yes, totally agree.
Apr 17, 2020
0
#34
Itude wrote:
llama44 wrote:
Itude wrote:
I mean that they are able to calculate tactics in some kind of "flash manner" as opposed to a more slow ponderous mechanical manner ?!
I have watched very highly rated players discussing some bullet games afterwards.
They can show maybe 5-10 move combinations in their descriptions, ones they presumably saw .
That is not exactly pattern recognition...is it ?
Well... yeah. That's exactly what pattern recognition is
Here's a puzzle
--
Most experienced players can solve this mate in 5 in under 5 seconds.
Mate in 5 that fast? Are they a genius? No. It's a well known pattern. Like you said they don't have to calculate every move, they just notice a few key aspects.
Yes, but pattern recognition can only go so far.
No one can recognise every pattern that arises in every game. can they ?
The pattern needs to be backed up by some level of calculation, simply because each "pattern" will not match exactly those already known.
That back up seems to what is separate, and different.
I dunno man. I've been 2100 bullet on here. I don't think pros are doing anything fundamentally different than me, they've just absorbed more ideas and patterns and played grueling OTB games at a high level.
Spend 5 hours meticulously analyzing an endgame, think you understand it, screw it up in a 5 hours OTB game, go home and analyze it for another 5 hours, get it an yet another OTB game, screw it up again, analyze it for another 5 hours...
And remember this is just one endgame and this analysis - experience - analysis pattern is spread out over months.
Do that and then have some gawker say you're brilliant at that endgame. No you're not, you just put a lot of blood and sweat into it. So much so you can play it blindfolded no problem.
Apr 17, 2020
0
#35
LongIslandWoman wrote:
llama44 wrote:
Itude wrote:
llama44 wrote:
Itude wrote:
I mean that they are able to calculate tactics in some kind of "flash manner" as opposed to a more slow ponderous mechanical manner ?!
I have watched very highly rated players discussing some bullet games afterwards.
They can show maybe 5-10 move combinations in their descriptions, ones they presumably saw .
That is not exactly pattern recognition...is it ?
Well... yeah. That's exactly what pattern recognition is
Here's a puzzle
--
Most experienced players can solve this mate in 5 in under 5 seconds.
Mate in 5 that fast? Are they a genius? No. It's a well known pattern. Like you said they don't have to calculate every move, they just notice a few key aspects.
Yes, but pattern recognition can only go so far.
No one can recognise every pattern that arises in every game. can they ?
The pattern needs to be backed up by some level of calculation, simply because each "pattern" will not match exactly those already known.
That back up seems to what is separate, and different.
I dunno man. I've been 2100 bullet on here. I don't think pros are doing anything fundamentally different than me, they've just absorbed more ideas and patterns and played grueling OTB games at a high level.
Spend 5 hours meticulously analyzing an endgame, think you understand it, screw it up in a 5 hours OTB game, go home and analyze it for another 5 hours, get it an yet another OTB game, screw it up again, analyze it for another 5 hours...
And remember this is just one endgame and this analysis - experience - analysis pattern is spread out over months.
Do that and then have some gawker say you're brilliant at that endgame. No you're not, you just put a lot of blood and sweat into it. So much so you can play it blindfolded no problem.
Brilliance is a thing. Both hard work and talent can be means to the same end but I do think hard work will ABSOLUTELY own almost all talent when talent doesn't work hard
Sergei karjakin became a gm at 12 by studying chess 9 hours a day.
Apr 17, 2020
0
#36
Sure, top 10 players are chess geniuses.
But I don't think their brains work fundamentally differently.
Physiologically they're probably somehow optimized for some kind of fundamental chess process. Like being 7 ft tall physically optimizes you for playing in the NBA. But even if you're 7ft tall you run and jump and move the same way a 5 ft tall person does.
Apr 17, 2020
0
#37
LongIslandWoman wrote:
Asparagusic_acids wrote:
LongIslandWoman wrote:
llama44 wrote:
Itude wrote:
llama44 wrote:
Itude wrote:
I mean that they are able to calculate tactics in some kind of "flash manner" as opposed to a more slow ponderous mechanical manner ?!
I have watched very highly rated players discussing some bullet games afterwards.
They can show maybe 5-10 move combinations in their descriptions, ones they presumably saw .
That is not exactly pattern recognition...is it ?
Well... yeah. That's exactly what pattern recognition is
Here's a puzzle
--
Most experienced players can solve this mate in 5 in under 5 seconds.
Mate in 5 that fast? Are they a genius? No. It's a well known pattern. Like you said they don't have to calculate every move, they just notice a few key aspects.
Yes, but pattern recognition can only go so far.
No one can recognise every pattern that arises in every game. can they ?
The pattern needs to be backed up by some level of calculation, simply because each "pattern" will not match exactly those already known.
That back up seems to what is separate, and different.
I dunno man. I've been 2100 bullet on here. I don't think pros are doing anything fundamentally different than me, they've just absorbed more ideas and patterns and played grueling OTB games at a high level.
Spend 5 hours meticulously analyzing an endgame, think you understand it, screw it up in a 5 hours OTB game, go home and analyze it for another 5 hours, get it an yet another OTB game, screw it up again, analyze it for another 5 hours...
And remember this is just one endgame and this analysis - experience - analysis pattern is spread out over months.
Do that and then have some gawker say you're brilliant at that endgame. No you're not, you just put a lot of blood and sweat into it. So much so you can play it blindfolded no problem.
Brilliance is a thing. Both hard work and talent can be means to the same end but I do think hard work will ABSOLUTELY own almost all talent when talent doesn't work hard
Sergei karjakin became a gm at 12 by studying chess 9 hours a day.
How big of a deal is that? Are all GMs seriously competitive at the highest level?
He almost beat magnus in a world championship.
Apr 17, 2020
0
#38
Karjakin almost beat Carlsen?
No way. Caruana was a lot closer. I don't care that Karjakin drew first blood in that fluke win. His play is garbage compared to Caruana, and I'm not saying that as some moronically nationalistic American, I'm saying Karjakin was playing scared. Anand put up a better fight and Carlsen crushed Anand.
Apr 17, 2020
0
#39
knightscape007 wrote:
I find it amazing Hikaru can play so well in 1 min games to a much higher standard than i can with days of thinking. In this game you can see by clicking the link below he got 98% accuracy
https://www chess.com/club/champions-of-the-world1/join?key=f3us28
do y’all agree?
And look what all that bullet has done to his OTB rating. He isn't even in the top 10 anymore.
Yeah, but based on what? How'd you come up with 35? Why not 34? How do you determine complexity? What's the average complexity of a position that requires depth 35? What does it mean to be "trustworthy?" What measure of trustworthiness is depth 30 and why?
etc etc
In other words you made it up