papounet,

Well, it doesn't work any worse either. Also, I prefer using it with a single top biner, not two. And besides, it's prettier than the first of yours and simpler than the second.

Papounet - for first picture, the one you dub the asymmetrical 3-way sliding X, I think that if either the center or right placement fails, you'd be looking at possibly a large amount of extension. This depends on the total length of cord you use to tie into an equalette configuration, of course. But unless I'm missing something, I don't see how it would limit extension for either of those 2 pieces failing.
And, same thing with your 2nd picture with the rap rings - it doesn't really limit extension any different than a basic 3-way sliding X w/ no limiters. So, if your 3 placements started out further apart so that you had 1 leg quite a bit longer than the other 2, then if that leg blows, the there's a lot of extension to be caught on the remaining 2.
Plus, both those rigs fail if 1 strand is cut. (theoretical concern only, I know, but still one that I try to safeguard against.)

healyje asked what I meant when I referred to "the obvious 4-placement solution made with three 2-anchor equalelettes." In the spirit of good clean fun and didactic utility I thought I'd post an illustration of an "obvious" 4-placement solution, with a few comments.
http://i1.tinypic.com/rths1x.jpg
* The design is shown with sliding-Xs equalizing pairs of placements (pick pairs closest to each other), although 2-anchor equalettes or troublettes could be used in the alternative. John Long has written here that his measurements show that the sliding-X equalizes within 10%, but in this design each of them gets half the load, so that would mean equalization within 5%. If that's not good enough for you, just substitute 2-anchor equalettes and add carabiners or, for even better performance, use two troublettes (see my nearby post for another illustration).
* This proposal uses three ordinary webbing runners instead of a long accessory cord runner. Select whatever length runners get you close to the configuration you require for a given layout of placed pro. That will speed construction by reducing tuning.
* In this design (using sliding-Xs as shown) the upper strands each support 1/8 of the load, or half that of some other design proposals, such as mhabicht's, in which the upper strands support 1/4 of the load. This is probably of minor importance.
* As with other 4-placement designs, if one placement fails its pair partner's share of the load (particularly the peak load) suddenly doubles, from 1/4 to 1/2 the total. In some other designs, the force on the remaining leg of cord also goes from 1/4 to 1/2. This is the best-case failure scenario.
* As with other 4-placement designs, the limiter knots for the master point must be tied to allow more extension than that allowed by the upper limiter knots, otherwise a failed placement will result in the load being held by a master limiter knot and the entire load will shift to only two of the placements, not three--the anchor will "fail in half." This is the worst-case scenario.
* This general design implemented with 3 troublettes would appear to have the least sliding friction of any 4-placement proposal made thus far.
* This proposal, especially if made with three sliding-Xs or troublettes, would appear to have the most KISS factor of any proposal thus far (based on a criterion, say, of ease in describing it over the phone). "Honey, I'm on my cell phone here. Can you check RC.com and tell me how to set up that anchor?"
* 4-placement designs are rarely used in practice, so experimenting with them may be amusing and educational but not really of practical value unless the design readily adapts to a 3-placement version. 3-placement anchor designs continue to be posted, even recently, that do not actually balance forces (equalize). The most daunting design challenge appears to be an effective, KISS, 3-placement anchor design that easily adapts to a 4-placement version.

Ok, let's start off here with the powerpoint. I don't like the idea of two biners for the powerpoint in the equalette based designs. Too much weight, and it will be a PITA to clip everything into two biners at an anchor. I also don't like only clipping to one strand at the powerpoint because there's no redundancy. So here's a proposal...

According to my informal tests, there is noticeably less friction in the 3 way asymetric sliding x than in any knotted alpine equalizer type.

If nonetheless, one was willing to use a fixed knot systelm, then a butterfly knot would fit the bill.

I have to say a bunch of you (Charles, Finger, papounet, Knud) obviously are enamored with this [separate] upper W-shaped structure that results in two strands - one of which you use - to support the lower powerpoint mechanism. In the end I can't quite put words to it, but I don't like either this upper structure or any of the design elements where multiple strands are created but only one is clipped. Again, not sure exactly what bugs me about it all beyond using two parts, but every time I see it it rubs me entirely the wrong way and I wouldn't swap either the equalette or mhabicht's rig for any of the ones based on these structures. I just don't see enough advantages to be had from the complexity within the ellipse compared to an equalette (blue legs). Or if I am going for complexity in that area I'd go for mhabicht's rig every time. Again, it's a matter of personal preference and taste and these types of two-part rigs don't work for me even though they exhibit interesting ideas.

Healyje, a close inspection of the b---- will reveal it is in fact a 2 separate structures design joined by 2 biners to the lower powerpoint mechanism.

“….I just don't see enough advantages to be had from the complexity within the ellipse compared to an equalette….”. If you call for better equalization, you need the complexity of the b----. In such a case, the difference in complexity lies in how the slack cord (controlling extension) is being used. In b----’s, the slack cord is used vertically while in TIA it’s used horizontally.

.

http://i49.photobucket.com/...59/carlosjmm/TIA.jpg

very intersting setup process: extremely straightforward.
But it ends up doubling the connecting cord going to each pro and using single strand to provide the sliding x balancing function in the upper parts.

the end results looks very much like two small equalette for the 2 sets of 2 anchor but you clip only one center strand to be used by the lower sliding X.

Yes, but you're on a single biner , the other two are either one biner across two x'd strands or two biners on two strands and there is no way to do that without some additional friction

I just happen to prefer the equalette and mhabicht's designs to yours, papounet's, and knudenoggin's.

Healyje, a close inspection of the b---- will reveal it is in fact a 2 separate structures design joined by 2 biners to the lower powerpoint mechanism.

Maybe I don’t understand what you mean by “….multiple strands are created but only one is clipped…..”, but in the b---- multiple strands have been created but only one is hitched. Is there any difference?

If you call for better equalization, you need the complexity of the mhabicht. In such a case, the difference in complexity lies in how the slack cord (controlling extension) is being used. In mhabicht’s, the slack cord is used vertically while in TIA it’s used horizontally.

One point of importance:
Ina equalette design, the inner "box" that limits the sliding around can be small because the outer strands are clovehitched. In a mhabicht mode, the inner "box" should be larger to accomodate for the changes brought by the failure of a pro to the relative position of the limiter knot.

Both of those contraptions look INSANE!!! how about just slap a screamer on it if it looks sketch and climb the damn thing!!

glowering, I came to the same conclusion that you did about the powerpoint. I don't like needing 2 biners there either. The added weight/bulk is one drawback, but I also take into consideration that after setting up an anchor, I usually belay up the second with a reverso, and I can't fit 2 lockers into the little loop on the reverso. So, I did the same thing as you, and tied it so one strand is longer between the 2 limiting knots, so it works well with a sliding X between the knots.

However, for friction, I found that using a cordalette-length dyneema sling (400cm x 10mm by wild country) was even better than standard cord.
Plus, it's long enough that I could inline figure 8s for the 2 limiter knots without making it too short. the inline 8s allow the rig to switch between a standard equalette and the mhabricht (or astroglide) style rig without having to put biners through the "butt" of a regular fig. 8.

healyje-The beauty of mhabicht's design is that you have complete control to fine tune extension.

I´m trying to visualize the 2 equalettes you perceive, but I only see 2 Vs on top supporting another V at the bottom. Besides, 1 pair of anchors of an equalette will not equalize on their own, they need to be connected to another arm (1 or 2 anchors) to achieve equalization; in contrast, one pair of anchors of TIA will in fact equalize on their own. Also, the strand going from the middle anchor to build the master point has enough slack so that no additional load is inflicted to that anchor.

I have just realized that in a three points Equalette, the collapse of the unpaired anchor will result in a 2 point fixed knot anchor which provides no equalization; therefore, the Equalette would in fact turn into a non equalizing Cordelette; will this resulting configuration handle the load that induced the collapse of the individual anchor?. In the case of TIA, b----, Astroglide and maybe other designs I forget, the collapse of the unpaired leg will result in an equalizing subsystem, a very important trait to possess.