Any progress on this front? This is a no-brainer decision, I don’t understand the delay. If I’m not reading the poll wrong, almost 1/4 said analysis should be off by default for even correspondence. 2/3 said analysis should be off by default for live games. This tells me most people are mostly on the same page and want to do the sensible thing, that is make at least live ranked games on OGS as analysis-free as possible.
Is keeping analysis on by default for live games a personal preference by Anoek or a collective one by OGS team? What’s the hold-up?
In the past several days. Against any lower ranked opponent. I allow analyze and don’t use it myself. This form of handicap is certainly less than 1 stone in power. When there is ladder, I just use something else and still win most games. When I’m really not going to lose, I start giving variations on how they needed to move in the past:
[20:18] stone defender[6k]: Variation: 1
[20:18] stone defender[6k]: Variation: 2
[20:18] [7k]: ohhh, yeah I fully ignored that the capture at j1 was possible
[20:19] [7k]: yikes
If I started to do it after game ended, they most likely leave before I start. Without analysis it wouldn’t be possible.
Also I like accepting undo requests. And I don’t wish to waste time by creating unranked games when I could play ranked instead. So analyze mode should be allowed in ranked.
I don’t understand this conclusion at all. Are you saying we should disregard the majority because you like to turn ranked games into a teaching games (unbeknownst to your opponent, I assume)?
I mean it is nice and good of you, but seems like a very weak argument. Maybe I am misunderstanding what you are saying.
I talk to people who absolutely hate analysis. In hope that it doesn’t completely disappears in far future.
And analysis on by default for lower ranked against higher ranked(for higher ranked opponent it will be off) seems like a good idea for me. It seems less weird than giving komi/stones handicaps.
I still like trying very complex variations against higher ranked opponents and I losing anyway.
FWIW, once again I am struck by the observation that being in a tourney where analysis is turned off is doing wonders for my reading skills. I reach for analysis a lot less often in other games as a result.
I wish it was turned off by default. I think that The Universe Would Be A Better Place and It Would Be Best For Go.
Don’t get me wrong, it was just clickbait, but! I have a genuine complaint:
Games with analysis affect rank.
Because OGS allows games with analysis enabled to affect rank…an OGS rank simply does not mean the same thing as any other go server.
It is not an apples to apples comparison. You cannot say “well OGS ranks tend to be 1-2 stones stronger than other servers” and while empirically…data may suggest that…it simply means nothing.
Playing go while cheating…is just not go.
Therefore, I vote that games with analysis enabled not contribute to rank.
You are free to play ranked games without analysis, or unranked games with analysis.
Other users are equally permitted to play ranked games with analysis when both agree to that setting option before the game.
I don’t know about all the other Go servers, but certainly at least Fox Go Server allows analysis during live ranked games.
Ultimately, your main desire (the ability to compare go ranks between servers) is a misguided desire anyway, as ranks sourced from different player pools can not meaningfully be compared in any significant or accurate way. Since at best only a rough comparison of each server’s global community can be made in the first place, I fail to see whether a percentage of any of the populations do or do not use analysis will affect this comparison any more than the fact that we combine everything from 9x9 to 19x19 and correspondence to blitz, whereas most other servers only rank their users based on live 19x19 games.
I still struggle to believe that someone’s using analysis in live games (I wouldn’t have enough time to do that), but it seems to be a fact.
I still think that they’re a minority and their behaviour doesn’t have such big consequences.
What do you base this on? I play on Fox and I don’t have analysis enabled. Perhaps premium members, who pay for it, have it available? It is most certainly not the convention in Fox. Default ranked games don’t have analysis.
I don’t think that’s the main desire. The main desire is to be able to play go in a server with a culture that doesn’t twist the game convention to make it fit a consumer culture. “It’s too hard, let’s make it easier and accesible”…OGS is the “kids table” of Go.
,
Anoek said on english chat that he would change this with the next fix, that default quickmatch ranked games wouldn’t involve analysis tool anymore. That was about 2 months ago. I think it would have still been too late if it was done then, considering the community heavily voted against analysis 7-8 months ago (Make "analysis disabled" the default (poll) - COMING SOON! - #11 by Eugene ) and considering this is a “feature” that should never have existed in the first place(at least as default), yet here we are still waiting on this change that many people deem crucial for the server culture.
Of course, it is up to Anoek. If he had a change of heart, his server, his rules.
I still would like to have a site-wide poll asking people whether they use analysis during live games to come up with better moves. Wording needs to be careful but with good enough wording it would shed a light on what’s actually going on.
I don’t believe analysis has big impact on rank. I use it against higher ranked only and lose anyway.
Against lower ranked opponents I watch anime while playing. This surely affects rank. But you can’t force people to play seriously. Everyone have their own reason why they play Go. I think most people use analysis to just have less stress. Not to find as good move as possible.
So you’re saying ranked analysis being locked behind a paywall is somehow more fair than everyone getting to choose whether both players have access to it or not per game?
I was saying this is simply not true. The fairness can be decided seperately.
He plays more on Tygem, which also permits bot use and even integrates it into their system. Is that the example you were offering, the one server that is infamous for a terrible culture and losing players cuz of it?
Without context perhaps. It is well established that this complaint is about analysis tool not being the convention in Go.
Even if analysis really is helpful enough to be as much as a 1-2 stone difference in strength (which I doubt), then ranks between two servers cannot be compared accurately enough to be within 1-2 stone difference in strength anyways.
Why do I say that? Because we literally have evidence: look at the data of the standard deviation in the last rank survey. Even the most accurate comparison has a 1.5 stone standard deviation, the less accurate ones have a 3 stone standard deviation.
