Hi all,

I recently acquired a Domino, and have been looking into speedsolve methods. The only one I have found was Stefan Pochmann's, which is as follows:

1) Solve First Layer Edges
2) Solve First Layer Corners (1 at a time)
3) Solve LL Corners
4) Solve LL Edges

This method is fast, and I have been solving with it for some time, but it seems like there are too many moves/steps. I was hoping you all could help me develop a new Domino method?

I think that solving the first layer with blocks could be faster, but what about LL? Maybe a corners or edges first type method for the whole puzzle? Any ideas for another method???

Not sure if this is a speedcubing method... but it's the method I use...

off of Jaap's page - http://www.geocities.com/jaapsch/puzzles/domino.htm

i pretty much solve the Top Corners, then Bottom Corners, then Top Edges then Bottom Edges...

it's got a few cycles too. also note that at the bottom of the page, it also references Stefan's method...

Hope that helps.

Thanks reeeech. How fast are you with Jaap's other method?

Sorry but I'm not a speedcuber, so I can't really give you a time to work with.

But keep working on it, with Jaap's and Stefan's methods, it should be the best solution out there

Just in case you haven't seen it already - Stefan's Domino Solve

I like Stefan's method. I think it works great for speed cubing.


I've gotten a few sub 30 domino times using my 3x3x3 to simulate a 3x3x2.

yeah 14 seconds thats a record!! Plus his long tarantula fingers certainly help out a bit!

Thanks everyone, I think I'll stick with Pochmann's method. My current best time is 19.xx.

Would anyone happen to have a spare split spindle Domino that they would be willing to sell? I've been trying to get one for about a year but eBay hasn't helped at all... (I've only seen groove Dominoes for sale)

*cough* i doubt they make them anymore. I mean the real authentic ones - they were made over 20 years ago. Even those have great ones and not so great ones - check out Stefan's site - he shows how to clean up a domino, and how to lube it. and i think which are the good brands.

My 2x3x3's are modded puzzles by Mark Longridge. He's a member on this site. However, they are not the same as the domino; they are a lot bigger because it's a modded 4x4x4. They do turn great though - i'm having a great time with his Rylox Prism - which is like a domino but with 2 edge pieces splitting up the middle slice (also creating a 2x2 center).

I have seen Mark's cuboids and they look quite nice, but a bit too large for speedcubing... His Rylox Prism also looks really fun. Is it hard to solve? I would assume that the shape-changing property would probably make it a bit harder.

Also, I just got a new PB (non lucky) on Domino of 16.66. I think I might catch up to Stefan soon...

The Rylox is actually fun indeed... In the most basic form, you need a few alg's to solve it.

When in non-cuboid form, it can get trick getting it back to it's solve shape. I really don't know any method other than twisting and turning lol.

Once in cuboid shape - I'll solve the 2x2 centers, then i'll pair the edges, then solve like a domino.

I like to compare it to a 4x4 or 5x5 - whereas I solve it to a 3x3x3 and then solve like a 3x3x3. The Rylox - i'll solve it to so it's a mixed up domino, then solve like a domino.


BTW - if you like cuboid puzzles - learn how to solve the domino!!! hehe i just solved the 3x3x4 for the first time!!!! It was tricky though, I'm stuck on swapping two edge pairs - harder than it is because of the restrictive movements. now i have to solve the 3x3x5 and then i can finally get my hands on the 4x4x5

3x3xNs can just be solved (after returning it to cuboid shape) like a domino, just repeated several times.

In my opinion, the easiest method would be to use square-1 algs for solving the corners, then using setup moves for the 2-cycle algorithm (R2 U2)x3. Really slow, but I'm too lazy to memorize real algorithms.

The main reason I solved corners one piece at a time is that recognition is very hard for this puzzle. You can't just look for two colorwise matching side stickers to connect a corner/edge pair. You have to look for numbers that belong next to each other, which is much harder.

I doubt anyone will be besting your record anytime soon!!!!

((((((I'm not worthy, I'm not worthy, I'm not worthy)))))))

That's why I like the stickered 3x3x2 better than a domino with dots.


Too bad I have neither.

You're right Stefan, recognition on Domino is quite difficult...

Are any puzzle modders making 3x3xN (mainly N>3) cuboids right now for sale? I don't believe Tony Fisher, Okamoto, etc are.

Also, could a 3x3x7 (non cube-shaped cubies) be created from a 5x5x5 (probably Rubik)? This would be as simple as a 2x2x6 if rounding off the centers allows them to turn independently. Does it?

the 3x3x7 was discused here with cube shaped cubies.

Thanks! However, as stated in that thread, the 3x3x7 design looks really unstable...

The 3x3x7 can be built from a 5x5x5, if you use Jin's method twice: the second and sixth layers from the corners and edges, and the first and seventh layers from the centers. But that would be very difficult to do, I would imagine.

Hi :-)

I don't have a domino, but i guess the following (highly tabular) method should be possible:

1)PC - permute corners across the 2 layers > corners solved!
2)PE - permite edges across the 2 layers > edges solved!

Use ACube to find the required algs for all of this

-Per

Well, the Domino has four hundred million combinations, so doing it in two steps would require about 10000 algs for each step (if you allow turns of U), which is too much even for hardcore ZB aficionados.

I considered a few solutions that would require fewer steps than Stefan's method. One was to do the first layer (however you want) and then a PLL alg (I have algorithms for all 40 or so of them). Another was to make two 'Roux blocks' (1x2x3 blocks, one on each layer) and then solve the rest in one step (there are only about 20 possibilities, but I don't have algs for all of them).

Or, since the Domino has parity problems, you could just sort the layers, then do PLL on top layer, x2, PLL on bottom layer (still 42 algs, although you might want to learn another set of algs that are allowed to disturb the bottom layer).