Mitsurugi wrote:
Losers whine about their best. Winners go home and watch Dragonball Z with the prom queen.
Now that was truly funny.
An Idea That Could Change Everything (in chess)
Sort:
trysts wrote:
Christ you write fast, Moon_Knight!
One can achieve great feats in order to justify being a loser
TeslasLightning wrote:
Mitsurugi wrote:
Losers whine about their best. Winners go home and watch Dragonball Z with the prom queen.
Now that was truly funny.
+1
trysts wrote:
"I beleive this can fail."(sic)
Yes. I too believe this(?) can fail. Just like Christopher Columbus failed in having any moral sense whatsoever, this grand and original(?!) quest to play like a computer, makes no sense whatsoever. But I will show my friends your unique interpretation of a question mark, so they too can see what a self-absorbed sermon about pretty much nothing, looks like
Excuse my "I'm 15" insensitivity
What do you mean? Columbus wrote back to Isabella--describing the indigenous peoples of the "West Indies", something along the lines of: "They are a beautiful people, well-built and strong. With twenty men we could enslave them all."
If that isn't kinder and gentler, I'll be damned if I know what is!
Moon_Knight- Please stop posting. I don't want to know how you're going to feel when you come back a year or two later and re-read what you wrote.
luisfelipe wrote:
Not only that, it is a very expensive statistic to compute, as it requires a lot of dedicated cpu time to come up with a reasonable set of 'bests' for every single move. And there's also the fact that different programs (rated >2800) come up with different 'bests'.
This is the main reason why a "potential" rating fails. If there is no absolute order of best moves (which is not possible to achieve with the speed of today's computers, since it may require a depth of anywhere between 50 and 100 moves from the start of the game, or even more - and if the best move entails not exchanging any of your pieces, the calculation of a single move at around 15-20, when all pieces are out, is going to take a LONG time), there is no basis of comparison. It's like measuring your height with a tape measure from K-Mart only to find out that the measure from Sears makes you an inch taller.
What could work is the institution of the separate "Rybka potential rating", "Shredder potential rating", etc. based solely on the algorithms of those engines, which are good enough to beat humans but still aren't "perfect". Still, doing the necessary calculations from a database of even only 100 games would take weeks or months on a single modern CPU. (Of course, parallelization is entirely possible, reducing the required time.)
Conclusion: for a very patient and dedicated person with at least one spare CPU, finding out his/her personal engine-specific potential rating is possible. Finding out a general potential rating, however, is not. As for the aforementioned "creativity", there is simply no way of quantifying it - this potential rating would be exclusively a comparison to computer moves.
I think that the potential rating-system is a potentially good idea. But potentially it would be hard to make potentially. I think it would be potentially good if a computer analysed your games and potentially showed how good your moves are and potentially showed you how well you played that game, potentially speaking. Potentially x2.
Actually, it's not a bad idea. You just have it the wrong way round.
The strength of a player is not determined by the quality of his best moves, but by the quality of his worst moves.
Kasparov probably has a potential low of about 2400, while mine is about 300 :(
I'm done. Not enough of you are even willing to give me a chance. The rest of you didn't read the whole thing so you didn't get that I changed what I mean in the little. And the best answer truly came from "luisfelipe". The rest of you had good answers and thank you for helping me reach this conclusion.
Personally I don't see why a bunch of grown people would keep coming back here to be assholes. It's pathetic. Your that bored dipshits? Go use some of that money of yours to get laid. It might calm you down and keep you from coming to the forums as much for "entertainment". Hopefully it'll give you herpes too. ;)
To all of you who helped; thank-you so much. This will stop bothering me now. And lol at the Christopher Columbus thing.. He had to prove what no one else would even contemplate. He used the calculations of a muslim in a christian empire to try and prove it! That's outrageous.
lol I'm not coming back a year later to reread this. I'm going to be really busy next year getting my grades high as possible for college.
Sorry I missed the other valid conclusions.. "Azukikuru" you're right.. This could work; it'd be difficult and there would be systems for whichever program. Thanks; I might try and push the idea one day... But all this hate has worn me down...
I have to say though.. You haters drove the best conlusion out of me. Way better than the original... WAY better. And to that even though I didn't enjoy the process I jave to say thank-you.
The worst moves thing is interesting.. -falls on the floor dead tired- But that would be for another day! xD
You know...if your feelings get hurt, you should be able to take it without using foul language, as others haven't posted bad language here. You posted a bad idea and people ripped it apart. Get used to it. That is the life of coming up with ideas. Most of them fail. But, seriously, maybe you should stay off the forums until you are older.
Oct 13, 2010
0
#73
Moon_Knight wrote:
Personally I don't see why a bunch of grown people would keep coming back here to be assholes. It's pathetic. Your that bored dipshits? Go use some of that money of yours to get laid. It might calm you down and keep you from coming to the forums as much for "entertainment". Hopefully it'll give you herpes too. ;)
I think the computer would have played a different move here, Moon_Knight.
trysts wrote:
Moon_Knight wrote:
Personally I don't see why a bunch of grown people would keep coming back here to be assholes. It's pathetic. Your that bored dipshits? Go use some of that money of yours to get laid. It might calm you down and keep you from coming to the forums as much for "entertainment". Hopefully it'll give you herpes too. ;)
I think the computer would have played a different move here, Moon_Knight.
Only potentially... But Vegetta totally agrees with Moon_Knight
Moon_Knight wrote:
I have to say though.. You haters drove the best conlusion out of me. Way better than the original... WAY better. And to that even though I didn't enjoy the process I jave to say thank-you.
It's called constructive criticism, and it's the basis of all creative work (although it isn't usually communicated with such sarcasm). Don't take it too hard - you'll encounter plenty of it as you grow older. 
The problem is that you're trying to make something complicated out of something that starts out simple. The current rating system is a simple generalization of your model: it is expected that whoever has the potential to make the best moves also has a comparatively similar potential to win games. For all intents and purposes, this is enough - ratings are used primarily to find suitable opponents, and experience has proven that people of similar ratings are a good match for that purpose. Your higher-rated opponent is going to be less impressed if you can make several good moves, and more aggravated if you're suddenly going to make a stupid mistake, ruining the level of the game. Thus, a "potential" rating isn't going to help when deciding against whom to play.
There really is no other need for a rating, which is why making the system more complicated is unnecessary. So, even though it might be possible to do, you have to ask yourself this: why should it be done? We adults like things simple.
I guess this is what people were trying to tell you, only with less tact.
Just to inject a bit of balance here - and I also feel kind of sorry for Moon_Knight - the concept is an interesting one. Potential is always going to be difficult to measure and there's usually something to be gained from discussing how best to do so. Please don't be too discouraged or angered by people's responses, Moon_Knight - it may be that one day you come up with an absolute winner. It's my opinion that today's not that day, but I genuinely wish you all the best.
I'm pretty sure my rating would would be OVER NINE THOUUUUSSSSAAAAND!!
(for those who don't know) http://www.youtube.com/watch?v=SiMHTK15Pik&feature=related
This topic has inspired me to visit the forums more
Oct 13, 2010
0
#79
Measuring potential skill would only make your actual rating look even more sad by contrast.
This kind of measurement is used all the time as a curiosity to see how close top players match computers' 'best moves'. It was popularized during Kramnik/Topalov with all the allegations of computer cheating flying around. In fact, I believe some other thread mentioned that chess.com uses something similar to catch cheaters, by comparing their OTB correlation with the computer's first and second suggested moves versus the chess.com play correlation.
It is an interesting statistic, but as others have pointed out, it has no practical application when playing others and should not be used for ratings/pairings.
Not only that, it is a very expensive statistic to compute, as it requires a lot of dedicated cpu time to come up with a reasonable set of 'bests' for every single move. And there's also the fact that different programs (rated >2800) come up with different 'bests'.