An idea to a blindtrainer program

[Sakarie]

This is mainly for threecycling stickers, but could maybe be useful for R2 and M2 too. Also it's mostly for those who uses letters.

This is only an idea, and since I'm not skilled in programming, I have too share it, and ask if any one out there can and want to do it. If you know anyone who's probably capable, but doesn't read the blindforums, please hint them!

One of most important parts of the blindsolve is to as fast as possible know what algorithm to use when you have a pair of letters. What algorithm do I use for corners when I get FR? PF? AL? If I don't have to translate the letters to stickers, and again translate into a commutator, I could gain much time.

My idea would be a letter-generator, that can give me predefined pairs.

The basic would be that the user gives some rules to the program. For my corners it would be "A cannot be followed by B or C" (since they're the same piece), but in some easier form. Then you just take random pairs.

The extra thing you could (Should!) do is to integrate a timer, so that when you press space or something to get the letterpair, you also start a timer, that you should stop when you've solved the threecycle. If you could save those times (the best would be something in the same way as this), you could see statistics over which ones you are best and worst at. Then you could choose to exclude the letters ABCDEFGHIJKL and just do with MNOPQRSTUV, or something.

Perhaps, you can exchange the letter pairs, for your on syllable words or something, but that seems not very necessary.

I actually think this would be a very good way of practising, and you could do it for hours, which most of us couldn't do with just normal blindsolving.

What do you think?

 

[riffz]

I definitely agree that being able to go from a letter pair directly to an algorithm would be the best way to go. I would actually take it further and say that naming each algorithm the word corresponding to its letter pair might even be slightly easier, but I'm sure that's also what you meant.

I don't really know how to code the timer portion of what you're suggesting, but I already use a cue-card program that has my letter pairs on one side and their corresponding word on the other, so I plan on just flipping through these to practice my algs.

 

[amostay2004]

I just do lotsa sighted BLD solves at the moment, but a programme like this would be nice to have

 

[Mike Hughey]

That does seem nice. And you could have the program track the pieces and show the resulting "scramble", so you could check the cube at any point and make sure you haven't made any mistakes. So perhaps you could do 10, then check the cube to see if you got them all right or not, then do another 10 and check again, etc. If you get one wrong that you weren't aware of, you could go back over that last batch and see if you can find your mistake, so you can correct it.

I should probably do more of this sort of practice. Mainly I just do lots of full blind solves. I like that I can do a whole bunch in a row without any problem, but it's probably not the best way for me to improve.

 

[aronpm]

I think this is a good idea. Also, it could have different settings for different pieces. Like, for example, the possible letters on my edges are slightly different to my corners, which are slightly different to my centres.

Also, maybe an option for floating buffers?

 

[Sakarie]

I would actually take it further and say that naming each algorithm the word corresponding to its letter pair might even be slightly easier, but I'm sure that's also what you meant.

That does seem nice. And you could have the program track the pieces and show the resulting "scramble", so you could check the cube at any point and make sure you haven't made any mistakes. So perhaps you could do 10, then check the cube to see if you got them all right or not, then do another 10 and check again, etc. If you get one wrong that you weren't aware of, you could go back over that last batch and see if you can find your mistake, so you can correct it.

I think this is a good idea. Also, it could have different settings for different pieces. Like, for example, the possible letters on my edges are slightly different to my corners, which are slightly different to my centres.

Also, maybe an option for floating buffers?

All of these ideas would be good!

To not have letter pairs, but making words like Cub, soup or something come up instead, representing CB and SU or something, it would be great as an option.

About being able to trace ten algorithms, and see if you were right, would be good. But that makes the program much more complex, wouldn't it? Then every person have to typ in that A=URF, and the biggest part, that the program have to "understand" the cube, and simulate all of the algorithms. It would be good, but maybe it's too complex? I don't really know, since I'm no programmer.

Yeah, different moods for corners, edges, centers and wings would be good! Floating buffers would be good, but does it matter? I mean, does the program "care" what buffer you have? Well, it would if you could be able to use the "tracer" that Mike suggested!

 

[Mike Hughey]

About being able to trace ten algorithms, and see if you were right, would be good. But that makes the program much more complex, wouldn't it? Then every person have to typ in that A=URF, and the biggest part, that the program have to "understand" the cube, and simulate all of the algorithms. It would be good, but maybe it's too complex? I don't really know, since I'm no programmer.

Yes, it's certainly true that this would make it a more ambitious project. If you were going to do this, the easy thing would be to try to get someone else's source for a program (like CCT, for instance) that can do this sort of thing, and then integrate this functionality into it.

I mainly suggested it in case one of the really good programmers on here might happen to get interested and try to take it up.

But your original idea of a program is something a beginner could write fairly easily; perhaps you should try that yourself.

 

[cmhardw]

I personally would love a program like this as well. I currently do this through the use of flash cards. I have one card for every pair of letters, and I go through my stack from time to time to practice recalling my images. I usually just recall the image, I don't really think about the algorithm. I think this algorithm functionality would be pretty neat.

Chris