Clearly DBX is the most general technique because it can handle any situation, including jumping from one remote single string to another.
But most people use USX or DSX, presumably because these techniques are faster and more forgiving (less likely to result in a mistake).
So if someone is playing at full speed with USX or DSX, how is it possible to keep time and slip in a slower DBX stroke? What am I missing?
Are you thinking of someone like Martin Miller, who has a DBX technique that (by his own estimate) has a maximum speed of 180bpm (16th notes), but also have other more single-escapey techniques that go 200+?
Well, then you already have an answer 
So the following describing details of one style of DBX? If so, how can it be used at full speed?
The page you’re linking to is a super technical discussion of our research findings from observing certain picking techniques in slow motion. It’s not teaching advice. It’s a technical explainer for people who like Discovery Channel-type stuff. That’s why we moved it to the “reference” section.
Some picking techniques, like the reverse dart technique we teach in the Primer, can mix evens and odds per string. When these techniques are done correctly, the motion mixing/matching is complicated and takes many forms when viewed in slow motion. But it’s all controlled subconsciously, so you don’t have to worry about which joint motions are happening for individual pickstrokes.
Instead, when we help players with this technique in TC, the focus is on the overall form and motion, using a variety of tests and sample phrases to see what’s going on. This is easier said than done, because people do all kinds of idiosyncratic things. But we get great results by making sure these things are working they way they should be.
Sorry for not being clear. My question is amazingly fundamental.
Depends on the definition, and I’m not sure I would call it general in the broad sense. It’s probably the most rarely used in its purest form. The others can handle most situations as well including cross string alternate picking, but possibly with slight artifacts that may not even be audible most of the time. Does this fact detract from them. I would say absolutely not! You are either playing it or you are not, who cares what means you are using to achieve it.
Well no, I don’t think that’s the reason at all. It’s likely way more organic than that. You have to be cognizant going in that you are actually doing them and chose them for this purpose. Most people just learned to play a certain way, a lot of the time it was what physiologically felt most comfortable to them. It’s really like trying to put meaning and intention where there likely wasn’t any. Although with the resources out there, I have no doubt that nowadays some people are doing this and it may be more of a hindrance in the long run.
It depends on how you are defining DBX. In the strict sense or the more realistic “relaxed” mixed escape, skating on strings sense?
It also depends on which range of tempo’s used.
This is something I’m not sure most should worry about. I liken it to a Spike TV fantasy mash up.
There’s a difference between DSX (or USX) plus the occasional helper motion, and full blown continuous DBX.
We did observe some great players having a “DBX technique” that goes pretty fast, but then having also another gear with a technique that is essentially single-escape and goes even faster.
So, for that type of player, they will be able to do the DBX technique only up to a top speed which is fast but inferior to their overall top speed.
The more general question (speaking very loosely): can a dbx technique reach the same top speed as a single-escaped technique? We don’t know the answer yet!
My theory, and this is NOT an expert opinion but it seems to make sense:
This comes down to mechanical efficiency. There does seem to be a slight efficiency penalty to a double-escaped motion, in that even players with very accomplished double-escaped chops tend to switch into a single-escaped motion for their fastest playing.
However, this sort of downplays the fact that our goal here isn’t to do the most mechanically efficient motion possible, but to play music. And, you CAN integrate the occasional less-perfectly-optimized motion into an optimized sequence on a one-off basis with an extremely small efficiency penalty. It’s when you start to stack those motions on top of each other over and over again where the slightly less efficient motion starts to exert a bit more of a toll. That’s why a single-escaped run with the occasional “helper” motion happening every six or eight notes or so can still be played extremely rapidly, while a strict double-espcaped motion tends to top out a little earlier.
That’s a long-winded way of saying that I wouldn’t over-think this - you can logic your way through this, or you can just do it, and in practice the difference between strict single escaped and some sort of helper motion coming in to play every half dozen notes is extremely small. It’s the rapid repetition of a mechanically less efficient motion that causes problems, not doing one in isolation every half second or so.
Indeed! So for fast pieces one can’t mix in slower DBX techniques as one can’t keep uniform time. This means one should think like YJM in these cases and do “legato” notes or other such tricks that work at full speed.
I don’t know, the body is complicated and again we don’t really have an answer on whether “DBX motions” necessarily have to be slower.
It is tempting to draw conclusions like: linear motion is more efficient than curved motion, until we recall that none of our joints can individually produce a linear motion!
@kgk in Yngwie’s case, we could assume that he doesn’t have (yet?) a DBX motion that goes as fast as his USX motion, so your conclusion would be valid for his individual case at this point in time. It’s more of a practical thing dictated by his current personal set of skills — but that does not have the same validity as a “general biomechanics theorem” 
Thats actually the exact opposite of what I was saying. 
My suspicion, and this is purely speculation, is that “less efficient” motions have a mechanical cost that is - on a relative basis - extremely small in isolation, but compounds quickly with fast repetitions as you start firing muscles back and forth rapidly.
So, as long as a suitable amount of time goes by from one “less efficient” - again, relatively speaking, we’re talking very minor differences between single and double escaped motions - motion and another, the actual mechanical cost is pretty negligible.
You listen to some of the MAB stuff in the CtC analysis, or some of Gilbert’s faster runs where he’s using the occasional “helper motion” to facilitate the occasional note with the “wrong” escape, and you’re not hearing timing inefficiencies because it only becomes materially inefficient when you try to do it many times rapidly in a row.
This is just a guess… but it certainly makes sense with what I’ve seen in other players. Lots of people DO play extremely fast runs that are primarily single escaped but with a secondary motion allowing a few notes to escape in the other direction, with no noticable timing inconsistency. You can big-brain it and logically argue that something must be less efficient, but in practice it doesn’t seem to be to any degree that we can see.
When logical reasoning disagrees with observed data, there’s either a gap in the logical process, or the actual impact is smaller than our ability to observe. You could be right and it certainly could be the latter, but my money’s on the former, and in any event if it’s indetectible then it’s kinda, well, irrelevant.
tl;dr - don’t over think it. Just shreeeeeeeeeeed!!! 
Exactly. If I have three techniques that take time
X < Y < Z
And the metronome is at 1/Y then I can only use Z in special places, perhaps if it is followed by a rest, etc.
I am suspicious about DBX because it dramatically slower for me than single-escaped motions, so I should only attempt it if appropriate. For example, DBX is magical at lower speeds.
Fast people only have something like 75ms per note, not a lot of time.