Before I begin, understand that I have not done any of this yet. Just so we're clear.
I understand I'm not supposed to use engines to give me moves during a game in progess. Simple enough. I also understand that it IS legal to use a games database to search for positions that have been played by others (or even myself). That's also simple. Finally, I understand I can't use an endgame database to assist me in how to play perfect endgames once the material is down far enough that an endgame database would be able to provide perfect play.
All of this I understand.
So, I want to get better as a player (I'm around 1600 here ATM). I want to do two things:
I want to create a database of my own games for study.
I want to use fritz to do post-mortem analysis of my games and annotate them.
Here's where I get fuzzy. If I take all of the games in my personal games database and use fritz to analyze them and put the proper continuation into the analysis it seems that I'd be getting an unfair(?) advantage because playing the lines that I do I'm very likely to have a lot of repeated positions coming up. Take that and combine the reccomendations from fritz for how to continue from a given point isn't that getting an advantage from software? Is this perfectly ok or is there something I missed in my understanding of what is allowed?
Thanks.
-VAR
Originally posted by varoadstterThere was a thread about this before started by Dragon Fire.
Before I begin, understand that I have not done any of this yet. Just so we're clear.
I understand I'm not supposed to use engines to give me moves during a game in progess. Simple enough. I also understand that it IS legal to use a games database to search for positions that have been played by others (or even myself). That's also simple. Finally, I is there something I missed in my understanding of what is allowed?
Thanks.
-VAR
Another consideration is that if you use repertoire books like I do, many of the authors analyse with Fritz & use this in their work on the key lines. Thomas Johansson in The Fascinating Réti Gambit states so quite openly in his introduction.
The mods must totally discount opening lines (some very deep to 20+ moves) & focus on endgame or strategic anomolies for obvious reasons.
Also, if you can prove your source - ie a quote from a book or database reference - you will be fine, so don't worry!
🙂
Originally posted by SquelchbelchAs prolific a poster as he is, could someone point me at the discussion?
There was a thread about this before started by Dragon Fire.
...
I just want some clarification as to how the interplay of personal databases and post-mortem analysis (using Fritz) is conducted so I don't mistakenly violate the spirit of the rules.
I don't think there is any problem in remembering and using better lines found from post mordem analysis. I mean that is what it is for to - to play the better line the next time you get in the same position - if you are fortunate enough to keep in your memory. Just do not open the engine and ask for its advice during actual play. Not using what you learned in post mordem analysis for Future games would make post mordem analysis kind of useless no?
Originally posted by onehandgannYes, I understand. But it's not a question of memory as I described in the OP. I am specifically asking if taking the reccomended continuation from Fritz and using that as the annotation of a specific position in my database of games is ok. There's nothing in the ToS that says I can't refer to a game database during play.
I don't think there is any problem in remembering and using better lines found from post mordem analysis. I mean that is what it is for to - to play the better line the next time you get in the same position - if you are fortunate enough to keep in your memory. Just do not open the engine and ask for its advice during actual play. Not using what you lea ...[text shortened]... ed in post mordem analysis for Future games would make post mordem analysis kind of useless no?
I believe if you generate the game study in PM analysis and store it, it is fair game (ha, ha). The one wrinkle someone might claim is when they are in a situation where the next five half moves are fairly clear. They could throw in the three or so candidate moves from the future position into an engine and store the analysis. Then when the game gets there, they just call up their analysis and continue playing. I believe this would violate the spirit of the competition, but it might be within the letter of the law. Also, except for treating this the same as outright engine cheating, I don't see how it could be prosecuted. It appears to me we just have to have a certain amount of trust, and if we don't think our opponent is being honest, we just don't play them again. I don't know what to do if they enter a tournament you want to participate.
Originally posted by varoadstterThread 84528
As prolific a poster as he is, could someone point me at the discussion?
I just want some clarification as to how the interplay of personal databases and post-mortem analysis (using Fritz) is conducted so I don't mistakenly violate the spirit of the rules.
Originally posted by SquelchbelchIn the other thread DragonFire was asking a different question - what he described is using an endgame table-base to find positions to aim for. If someone did calculation by hand and then used an engine on the final positions it would still be cheating - and probably detectable using the games mods' standard methods as someone doing that would probably end up choosing the same lines as the engine would have had it been set running at the base of the search tree. Doing what he said is doing just the same but working backwards from the answer rather than forwards.
Thread 84528
What the original poster here is talking about is engine analysis on past games where the position is likely to come up again. The games mods have access to all your games and if they are investigating will notice that the line came up before. The mods have made clear that this type of pre-game analysis is fine.
A potential cheat may think they could use an engine in this area of the game in positions they'd reached before, but there's several problems:
Unless there is an immediately winning line the few moves where the window of opportunity is present for engine use aren't going to help all that much.
If the hapless victim makes a bad move where the payoff is obscure then a cheat will follow the engine line. Someone like varroadster, who has simply done work on an earlier game, won't know the engine refutation as the move is way down the engine's move list and so he probably won't have recorded any information about what to do. This won't happen all the time, but probably enough for the games mods to distinguish the two cases.
Cheating behaviour escalates. I think that a lot of cheats start off in a small way, maybe only using an engine in the occasional game. Then, because there they're rewarded with wins, they start doing it more until they are caught. Really this is the biggest problem - anyone who lacks the self-discipline not to cheat in the first place isn't going to have enough self-control not to let the behaviour escalate.
Pre-prepared analysis is fine as far as the rules of this site go. Clause 3b allows "previously existing research materials". If you really want to be paranoid about it why not upload your database to some webspace so there is a record of your research. Provided you can't alter timestamps it would provide sufficient proof that you knew about a line before the game started.
Originally posted by DeepThoughtThank you for your thoughtful response. In the end I want to not just avoid any legal missteps but further wish to make sure I am adhering to the spirit of the rule as well. Thanks to your and others kind responses I have a better feel for where "the line" exists. More importantly, I am confident that I will stay far clear of that line.
In the other thread DragonFire was asking a different question - what he described is using an endgame table-base to find positions to aim for. If someone did calculation by hand and then used an engine on the final positions it would still be cheating - and probably detectable using the games mods' standard methods as someone doing that would probably would provide sufficient proof that you knew about a line before the game started.
Originally posted by SquelchbelchThere are lots of potentially grey areas, some of which I tried to address in the above thread but the fundamental seems to be that your current research, whatever it may be, should not be designed to help you achieve an advantage in any of your current games.
Thread 84528
Research using a tablebase as I was discussing should be avoided if it is designed to assist is where a current game is going. Similarly analysis of your past games is totally aceptable unless you take a past game that has followed a line you are currently playing and that this analysis could be used or even assist in making move choices in your current game(s) when it should be avoided.
It becomes problematical if you are a regular player of (say) a Morra Gambit (as I am) and have half a dozen on-going games with this gambit (as I do). Should you avoid analysing your past games until you have no more on going Morra Gambits? A difficult question to resolve that has been discussed before. The simple answer is to avoid such analysis until you have no ongoing games in that variation but perhaps the correct answer is that it is OK to analyse them as long as such analysis does not impact in any way on any current game and that is easier saif than done.
Originally posted by Dragon FireA disclaimer before I start on this post. What I am about to mention, I have never done either (don't think it would be very effective anyway).
There are lots of potentially grey areas, some of which I tried to address in the above thread but the fundamental seems to be that your current research, whatever it may be, should not be designed to help you achieve an advantage in any of your [b]current games.
Research using a tablebase as I was discussing should be avoided if it is designed to ...[text shortened]... such analysis does not impact in any way on any current game and that is easier saif than done.[/b]
Would it be against the rules if someone was to use previous lines (generated from an engine) on a completed game, and apply this to the same position in an in progress game?
After all, your opponent can deviate from all of these lines easily anyway, but these lines can still give some insight into what moves to make next. Wouldn't it be equivalent to insight gained from previously engine analysed games anyway? Technically, I cannot see it against the rules, as you wouldn't be using an engine to calculate what to do next from your current position, just what was calculated from an equal position from before. This will be like looking up a previously generated database of engine moves.
EDIT: Although I can understand it will be ill-advised, as the game mods might believe an engine was used to calculate from that point. However, the probability of an equal position from a previously played game later in the game will come up again is minute.
Originally posted by lauseyThis is the same issue I raised when repertoire book authors look through the key lines, analyse with Fritz looking for improvements, then publish their findings.
A disclaimer before I start on this post. What I am about to mention, I have never done either (don't think it would be very effective anyway).
Would it be against the rules if someone was to use previous lines (generated from an engine) on a completed game, and apply this to the same position in an in progress game?
After all, your opponent can deviate ...[text shortened]... equal position from a previously played game later in the game will come up again is minute.
Take Nigel Davies Gambiteer books:
"Extensive use was made of Shredder 8 & Fritz 9"
or Thomas Johansson in the Fascinating Réti Gambit:
"Anyhow, I decided to collect the few sources I could find and then do some digging on my own, accompanied by Fritz 9."