Top Law Schools

Going through some logic games and I find myself at times confused as to whether or not I should split up my main diagram into two diagrams, three, or four, etc. Sometimes it's obvious, other times... not sure. What standard do you all go by to asses whether or not it's worth dividing the game up? Thanks!

This is something that I struggled with a lot when I first started learning games a couple months ago. What I have started doing is only splitting the diagram when I think that it is likely that I can visually represent every (or most) possibilities by doing so. For example, if I am doing a sequencing game with a rule like F is on 1 or on 5, I am likely to split my diagram if there are only six slots, but unlikely to do so if there are nine slots.

I hope this is helpful, and I am interested to read what others have to say. I think this is one of the trickiest aspects to learning games.

Which questions and from what PT are you referring to ?

clay7676 wrote:Going through some logic games and I find myself at times confused as to whether or not I should split up my main diagram into two diagrams, three, or four, etc. Sometimes it's obvious, other times... not sure. What standard do you all go by to asses whether or not it's worth dividing the game up? Thanks!

I don't think there's really a standard to go by. After I did enough logic games, I was able to develop a feel for when frames would be useful. I'm not right every time, but I would say about 85% I'm able to spot games that can be made easier by splitting up the diagrams. It's hard to explain, but after practicing enough it does kind of just jump out at you, like the inferences themselves.

Take the 4th game from PT 33, for example. Following the rules, it becomes obvious that store 2 has to be yellow or green. Once I make this inference, I ask myself: Do I have anything to gain by splitting this up? Yes, because if I make it yellow, I have enough information to complete the rest of the board, and if I make it green, my missing pieces, 7 8 9 and 10, only have three possibilities (GYRG, RYGR, or GRRY). How did I know this? I didn't, really. It just seemed like there was a good chance, based on the rather restrictive rules in the game, that making frames would solve the game. Like I said, you get a feel for it as you practice more.

Iroh wrote:

clay7676 wrote:Going through some logic games and I find myself at times confused as to whether or not I should split up my main diagram into two diagrams, three, or four, etc. Sometimes it's obvious, other times... not sure. What standard do you all go by to asses whether or not it's worth dividing the game up? Thanks!

I don't think there's really a standard to go by. After I did enough logic games, I was able to develop a feel for when frames would be useful. I'm not right every time, but I would say about 85% I'm able to spot games that can be made easier by splitting up the diagrams. It's hard to explain, but after practicing enough it does kind of just jump out at you, like the inferences themselves.

Take the 4th game from PT 33, for example. Following the rules, it becomes obvious that store 2 has to be yellow or green. Once I make this inference, I ask myself: Do I have anything to gain by splitting this up? Yes, because if I make it yellow, I have enough information to complete the rest of the board, and if I make it green, my missing pieces, 7 8 9 and 10, only have three possibilities (GYRG, RYGR, or GRRY). How did I know this? I didn't, really. It just seemed like there was a good chance, based on the rather restrictive rules in the game, that making frames would solve the game. Like I said, you get a feel for it as you practice more.

I definitely agree with this. After having done enough games, I believe your subconscious is able to internalize the rules of a newly presented game so well that it can draw some of these key restricting inferences intuitively rather than having to go through the 'calculating' of doing said inferences on paper. Once you are at this intuitive level, you will be able to make likely a quick judgement call on whether to diagram just one main diagram or separate templates.

My rule of thumb is that if I've got any of the following, then I set up an either/or at the beginning of the game:

* a large block (there are H's in the morning and afternoon of two consecutive days)
* a single, prominent conditional or an either/or conditional (if M is in tennis, then stuff; if M is in golf, then stuff; M can only be in tennis or golf)
* a clue relating the columns in an unusual way (twice as many film buffs see the Hitchcock film as see the Fellini film)

There are other times when it just sort of feels right, as noted above, but these three conditions (though they are not common) tend to suggest something is up.

Fortunately, if you miss that you should do this, you can still do the game properly. You just may lose a minute, which is bad but not a disaster.

Previous few posts are dead on. Only other thing I would add is that I don't see a ton of risk in pre-building, even if it ends up not "working," so to speak. Worst case scenario, you build like four fairly open/blank worlds that don't end up giving you much info. But for me at least, even if they don't direct me to the right answer, they give a nice bit of architecture to rely on. Costs no more than a minute, and helps me even if they don't fill.

vuthy wrote:Previous few posts are dead on. Only other thing I would add is that I don't see a ton of risk in pre-building, even if it ends up not "working," so to speak. Worst case scenario, you build like four fairly open/blank worlds that don't end up giving you much info. But for me at least, even if they don't direct me to the right answer, they give a nice bit of architecture to rely on. Costs no more than a minute, and helps me even if they don't fill.

This is very true, erring on the side of trying to do it even when it sometimes doesn't work out with semi filled-in scenarios still provides benefits.
Going through the exercise, even if it doesn't result in two or more scenarios that have a lot filled out in them via deductions, still helps you perform better on the game because during the minute or so you work on scenarios you are going through all the rules of the game looking at the relationships more closely and getting a firmer grasp on the big picture, major structures and deductions. The benefit you get from that when solving the questions easily makes it worth the time spent so it is risk free as long as you don't get out of hand with it.

tomwatts wrote:My rule of thumb is that if I've got any of the following, then I set up an either/or at the beginning of the game:

* a large block (there are H's in the morning and afternoon of two consecutive days)
* a single, prominent conditional or an either/or conditional (if M is in tennis, then stuff; if M is in golf, then stuff; M can only be in tennis or golf)
* a clue relating the columns in an unusual way (twice as many film buffs see the Hitchcock film as see the Fellini film)

There are other times when it just sort of feels right, as noted above, but these three conditions (though they are not common) tend to suggest something is up.

Fortunately, if you miss that you should do this, you can still do the game properly. You just may lose a minute, which is bad but not a disaster.

This is a good list. The only thing I'd add is a single element that shows up in at least two rules will often lead to only a couple possibilities. The strongest example I can think of is the red/green truck one, where Y shows up a few times and ends up being a huge deduction. That came just as easily could have had Y in one of two slots, which would've been just as good to do a scenarios.