More optimised Petrus is pretty much APB, where APB starts very similar with a 2x2x3 block but instead of EO after 223 APB inserts the BR pair then does EO which decreases the edges that need to be looked at, as well as increasing ergonomics. Then LXS helps to increase the fluidity and ergonomics of APB even more, then they both end with ZB.
I think that APB has a lot of potential, having a very similar move-count to ZB and Roux. Another method that might have potential is Epidote which is similar to APB except you insert the DFR corner while doing EO, this is EOLC, then E2CP where you solve the FR and DR edges while permuting corners which leaves you with 2GLL and it has a slightly lower move-count than APB, 46 compared to APB's 48. I think either method could be viable since the current top method ZB uses a good amount of algs so it leaves intuition at a minimum, and APB and Epidote are also very algorithmic methods aside form 223, which means that they can become very quick if every alg is used and it increases the TPS cap especially compared to a method like Roux where you are using almost entirely intuition which leaves room for pauses.
I would like to mention that Mehta method still has potential to go mainstream and become good for OH and 2H yo.
Epidote is actually less efficient by 1.25 moves
docs.google.com
Excuse me, but Petrus is already OO
If that is the case ZBLS wouldn't exist because there would be no reason to orient edges on the last pair, it would just be solving the standard F2L case and getting solved EO. ZBLS exists because you solve F2L minus one then orient all of the LL edges while solving the last pair, it is described as such on the wiki. What you described is just ZZ while also doing ZBLL for LL which no top cubers use ZZ, but they use ZB which includes ZBLS and it's purpose is to orient all of the LL edges which would be redundant if EO-cross/EO-line is used. I have linked both wiki pages that describe ZB and ZBLS in case you need to see what the method actually is and how EO-cross/EO-line are never mentioned.
It makes sense that top cubers would try to preserve any already oriented edges because that just makes the solve easier. If they knew an alg that preserved part of EO and an alg that didn't, I'm sure every top cuber would use the alg that preserved EO because otherwise they are just undoing something that they would need to fix again for no benefit. Yiheng influences his next F2L pairs while solving one, is he no longer doing what we call CFOP because he is making his next step easier? It's not listed on the wiki either, but it doesn't change the method. It doesn't make sense that you're saying that just because the try to make the next step easier it is no longer ZB, of course they would want to preserve any EO that they have it would be counterproductive not to. Just because it doesn't fit the textbook definition of the method doesn't mean it is no longer the same method, and making the rest of the solve easier is actually better, if they can preserve or improve EO then they should do that but it doesn't change the method they use it just means they are using a more advanced and situational version that applies in the moment. They wouldn't choose to force a dot ZBLS when solving 3rd pair if they had the option to improve EO because it would make the solve worse. They would choose a different alg that orients more edges to make the ZBLS easier and if their pair solutions end up solving all of EO then that is still ZB, it just makes the rest of the solve better.
