What do you want in a ZBLL alg trainer?

[ProStar]

Bump, I'd like to suggest being able to sort by COLL case, in addition to OCLL case. For example, instead of looking at all the pi cases, you could look at all the pi cases where there's a bar on the right. It'd make it way easier to find a specific case.

 

[Hazel]

Bump, I'd like to suggest being able to sort by COLL case, in addition to OCLL case. For example, instead of looking at all the pi cases, you could look at all the pi cases where there's a bar on the right. It'd make it way easier to find a specific case.

I'm sorry, this is a bit late. Right now it's set up like Roman's trainer, where you can expand each OCLL into its COLLs, and then its COLLs into their ZBLLs. Are you saying you'd like options for them to be sorted differently?

ALSO - for anybody who cares about this trainer who isn't in my ZBLL Discord server, I'm still working hard on the trainer I'm still not going to put out a time window for release since there's a bunch of stuff I still need to do (and some of it will be pretty difficult and could take a while), but progress is certainly being made. More frequent updates are in the Discord.

 

[ProStar]

I'm sorry, this is a bit late. Right now it's set up like Roman's trainer, where you can expand each OCLL into its COLLs, and then its COLLs into their ZBLLs. Are you saying you'd like options for them to be sorted differently?

ALSO - for anybody who cares about this trainer who isn't in my ZBLL Discord server, I'm still working hard on the trainer I'm still not going to put out a time window for release since there's a bunch of stuff I still need to do (and some of it will be pretty difficult and could take a while), but progress is certainly being made. More frequent updates are in the Discord.

I haven't checked out Roman's trainer, but how he's doing it is basically what I wanted

 

[Pyjam]

I hope the cases are seen from the top instead of a 3D view.

 

[Hazel]

I hope the cases are seen from the top instead of a 3D view.

They're a top view, don't worry

 

[Hazel]

There's a few important things I need to know for the trainer, and I'd greatly appreciate input from whoever is interested in using it:

In the selection screen, you can right-click on a ZBLL to "flag" it, and it'll have a purple background to show this. Flagging means that the ZBLL will appear twice while going through and practicing which ZBLLs you've selected. this brings up a few questions, though:

1) After you've recapped all the algorithms you've selected (including two of each flagged case), the trainer goes into a mode where it chooses selected cases at random to train you on next. Should the flagged cases continue to come up twice as often as they would otherwise, or should the flagged status be taken away so they occur just as frequently as every non-flagged case?
2) I want the user to be able to un-flag cases while on the trainer screen. When the user un-flags a case, what exactly should happen? Should the duplicate of the case simply be deleted from the list of selected ZBLLs, or what?
3) What exactly should happen if the user flags a previously un-flagged case mid-session? Should a duplicate of the case simply be added on to the end of the list of selected cases?

The way I currently plan to do it is to simply have one to two of each case in the list of selected ZBLLs, depending on if it's flagged or not. So, setting a case to be flagged mid-session would duplicate the case and add it onto the end while in recap mode, and while in random mode, simply double its chances of being selected by the trainer. In either mode, un-flagging a case would delete its duplicate. Does this sound like a good way to go about doing things?

 

[GenTheSnail]

There's a few important things I need to know for the trainer, and I'd greatly appreciate input from whoever is interested in using it:

In the selection screen, you can right-click on a ZBLL to "flag" it, and it'll have a purple background to show this. Flagging means that the ZBLL will appear twice while going through and practicing which ZBLLs you've selected. this brings up a few questions, though:

1) After you've recapped all the algorithms you've selected (including two of each flagged case), the trainer goes into a mode where it chooses selected cases at random to train you on next. Should the flagged cases continue to come up twice as often as they would otherwise, or should the flagged status be taken away so they occur just as frequently as every non-flagged case?
2) I want the user to be able to un-flag cases while on the trainer screen. When the user un-flags a case, what exactly should happen? Should the duplicate of the case simply be deleted from the list of selected ZBLLs, or what?
3) What exactly should happen if the user flags a previously un-flagged case mid-session? Should a duplicate of the case simply be added on to the end of the list of selected cases?

The way I currently plan to do it is to simply have one to two of each case in the list of selected ZBLLs, depending on if it's flagged or not. So, setting a case to be flagged mid-session would duplicate the case and add it onto the end while in recap mode, and while in random mode, simply double its chances of being selected by the trainer. In either mode, un-flagging a case would delete its duplicate. Does this sound like a good way to go about doing things?

1. Keep the flagged case coming up twice as often. I would un-flag it if I didn't want to keep specifically training that case.
2. Yes, just delete the duplicate case from the list of selected ZBLLs.
3. Add the duplicate in a random position in the last 90% of remaining selected cases. I don't want it to happen right after I flag the other case, and I don't want to know that it's the last case, especially if I flag a couple different ones, because then I'll know in what order and which cases will be coming.

Also, can random mode have some sort of not-so-random aspect to it? Something like the old tetris randomizer, where if a case is the same as the one before it, it selects a new case, even if that case is the same one as before, except maybe it could do this in blocks of 10?
Like, it would select 10 cases (or 10% of a set with a max of 10, minimum of 3. ya know, whatever), and then it would select another set of 10 cases. If any of those cases in the second set were the same as the ones in the first set, it would chose again, but it could still choose the same cases.
This would reduce back to back cases, but not eliminate them, and make sure that the cases get a little bit more spaced out, or something.

I don't know what sort of randomizer you're using, and I have no idea if this would be compatible.

 

[Hazel]

1. Keep the flagged case coming up twice as often. I would un-flag it if I didn't want to keep specifically training that case.
2. Yes, just delete the duplicate case from the list of selected ZBLLs.
3. Add the duplicate in a random position in the last 90% of remaining selected cases. I don't want it to happen right after I flag the other case, and I don't want to know that it's the last case, especially if I flag a couple different ones, because then I'll know in what order and which cases will be coming.

Also, can random mode have some sort of not-so-random aspect to it? Something like the old tetris randomizer, where if a case is the same as the one before it, it selects a new case, even if that case is the same one as before, except maybe it could do this in blocks of 10?
Like, it would select 10 cases (or 10% of a set with a max of 10, minimum of 3. ya know, whatever), and then it would select another set of 10 cases. If any of those cases in the second set were the same as the ones in the first set, it would chose again, but it could still choose the same cases.
This would reduce back to back cases, but not eliminate them, and make sure that the cases get a little bit more spaced out, or something.

I don't know what sort of randomizer you're using, and I have no idea if this would be compatible.

Thanks for the suggestions!
With the current system, after all the selected cases are recapped, it just starts selecting cases entirely at random. I could look into trying that not-so-random aspect, where after 10 (or how ever many) cases, each new case would look at the last 10 and if it showed up in that last 10, it would get one re-roll.
As for your point #3, I could certainly look into making that possible.

 

[Lazy Einstein]

Woah! Am I too late to this party?

If I could dream up ZBLL trainer, I'd want bestsiteever.ru but with:
- customizable colourways
- alg inputs for checking algs
- statistics pre case. (avg time, # of times practiced.)

Hardmode:
- version for mobile for the road

Honestly, bestsitever.ru, Tao's site, and stuff are getting pretty close to good enough.
Cubing community needs a JARCS replacement that is equal or better than the original. HARCS is okay-ish but good resources shouldn't have big learning curves. Should be login and go.

 

[Hazel]

Woah! Am I too late to this party?

If I could dream up ZBLL trainer, I'd want bestsiteever.ru but with:
- customizable colourways
- alg inputs for checking algs
- statistics pre case. (avg time, # of times practiced.)

Hardmode:
- version for mobile for the road

Honestly, bestsitever.ru, Tao's site, and stuff are getting pretty close to good enough.
Cubing community needs a JARCS replacement that is equal or better than the original. HARCS is okay-ish but good resources shouldn't have big learning curves. Should be login and go.

You're a bit late yeah, I've been mostly out of the cubing scene for a little while now xD I'd recommend going over to Roman's thread for his trainer and suggesting these things.

 

[Lazy Einstein]

You're a bit late yeah, I've been mostly out of the cubing scene for a little while now xD I'd recommend going over to Roman's thread for his trainer and suggesting these things.

Oh! Don't worry. I am all over that. It's a difficult thing to find an area of cubing online that I haven't infiltrated. XD Thanks for the suggestion though.

If you find yourself back in cubing and dealing with ZBLL, let me know.
Hope all is well with whatever you are doing.