Also not to forget that even the old-time lifters used to use shock routines here and there, like "railroads" (up and down the db rack), drop sets, restpause and others. But even back then they knew not to overdo it. It was often suggested that a shock on a muscle should only be done once a week, maximum.
I'm just getting ready to study up on MS, but as I understood it, it required a percentage of your max, not your maxes. That, along with the M-time fatigue management I can see how it can be done consistently, as a cycle.
Clustering, OTOH, is done at the max state and IMO should be regarded as intense as shock training, and done sparingly. QP did his usual good job of describing the futility of something (failure training) as a hypertrophy benefit, but we also know that it will produce muscle for a little while.
The original question was for the difference between continuous reps or MS and clustering; but those two are different from each other, so we're talking about 3 separate systems, which leads to the question, "what do you mean by continuous reps?" Reps all the way to failure, we know are disadvantageous. HST has continuous reps to submax, which we know works due to SD, progression and the like. So it appears to me that we're really comparing (again) HST to MS. The clustering principle fits in with shock routines - a supplement to any system you're using. HST and MS can be done consecutively, clustering continuously would eat your CNS as quickly as HIT I would think. It would become HIT.