Epsilon Rose

I just wish the defense things didn't feel so pointless. You get nothing from them, not even loot, since the scene resets right after you beat them.

Given their dearth of active abilities, I'm not sure combining two of their primary actives and then reducing them to once per encounter (or twice with a talent) is a good idea. When the flames/hands are out, the paladin loses a lot of its flavor and even a large boost to attack will lose it's relevance, at later levels, when compared to abilities that can be used more frequently and/or against more targets.

Which is why it can't be generalized to other parts of game play. All you have shown is that you can kill an ogre without being hit when using interrupts. You have not shown what happens without interrupts, though you have asserted it many times, nor have you shown what it likely to happen in other situations. No I/they wouldn't. We're talking about wolves here. They'd disintegrate on first contact and, even if they didn't, most of the classes have plenty of per encounter abilities I'd use long before spells. That was the second set of tests from AFTER you complained about him using cheats to hold the other stats constant and demanded "normalized" stats. So, yeah, at this point you're a liar and a troll.

We already know from ruminate's testing and general experience that the 100% won't hold up. Your incredibly brief "test" does nothing to show otherwise. You have yet to produce a single screenshot that shows what you need to and that is a comparison of situations, not you simply killing a single enemy. I could kill a pair of wolves and show there's absolutely no difference between interrupts or no interrupts and that wouldn't be compelling either, but it would be the exact same as what you're doing. Actually, not only did I read them, but I double checked them before posting last time. Don't worry, this time I've actually quoted the posts and helpfully color coded and emphasized the stats. He did change the stats between sets of trials, but he wasn't directly comparing results from different sets, so that's not actually a problem. So, are you being incompetent or purposely lying?

How does going from 100% to 17% not matter? First, none of those effect attack speed, which is not d100 based, and that's all we care about. Second, if you run enough tests random variables smooth out. That's basic statistics. Third, the quote is cute, but not terribly relevant seeing that you have yet to prove it works in practice. A large enough stat change will easily override the d100. If any result, from 1 to 100, results in an interrupt than the d100 doesn't matter. Also, I'm not sure where you got that he was changing all of his stats, but it's patently false. He listed the stats he used on the previous page and the only one he changed was per. At this point, either you aren't reading what's been written or you're purposely lying. Regardless, even if he was changing all of the stats, that wouldn't force you to change all of the stats in your tests. You have yet to show this in an even moderately convincing fashion and you have consitently dismissed all evidence to the contrary without even addressing it properly, except for the instance where you addressed it in a blatantly incorrect and possibly dishonest manner.

A) Showing that you can interrupt something and kill it before it gets an attack is not useful because it doesn't give us information on how many attacks it would get without that interrupt. This is incredibly important information that you don't seem capable of comprehending. Allow me to illustrate: Lets say, without you interrupting, an enemy can normally get in 1 attack for every 5 of yours. With constant interrupts it gets 1 attack for every 6 of yours. That's an improvement, but over the course of a long fight it's only ~17% less incoming damage. However, if you can kill your test monster in only 6 hits then what you'll see is 1 incoming attack in the first case and 0 in the second. Now your test reads a whopping 100% reduction, far more than that piddling 17% a longer fight would yield. Now, if you could guarantee every fight would be over in 6 hits or less, this would be accurate enough and your hypothesis that interrupts==lock-down would be a good approximation. Unfortunately, stronger enemies and bad rolls exist, so you won't always be able to guarantee a kill between 5 and 6 hits (at <=4 hits or 7 hits interrupting and non-interrupting will look exactly the same), which means your results won't accurately describe many situations and interrupts won't act as lock-down. B) Seeing as this is a video game, it's actually very easy to generate a "Parallel" world where something different happened. A quick save/load before your test would be more than sufficient, particularly if you ran more than two trials. C) Similarly, it is also trivial to force interrupts/non!interrupts. Interrupts can be forced by using the cheat console to massively increase your per, as it has no other effects on your attacks. Non!interrupts can be forced by either using the console to drop your per to zero or simply standing there and doing nothing (if you're measuring enemy attacks in a time frame, rather than enemy attacks per your attacks).

That would be a good test. You'd need to be able to ensure that you don't die in the first test or kill it in the second and both trials would need to go on for a substantial amount of time.

Interrupt isn't lock down. That's not how it works in this game. It just adds a very small delay to attacks. That is what it is, a delay and you posted screenshots of it acting as a delay. They key detail is how much of a delay it is. It could be that you only prevented one attack from the ogre, you'd have seen just as many interrupt messages either way, but a single extra attack probably wouldn't have been a terrible drain on your resources. For someone who was railing on about the scientific method and controlled variables, you are doing a terrible job of constructing your proofs. A single set of screenshots has no persuasive power and cannot prove anything, because any combination, even a single interrupt, could support your case equally well, regardless of if a single interrupt actually does anything of merit. Get off your high horse and conduct a proper experiment or stop debating in bad faith and bow out. Right now, you are arguing, very poorly, in the face of much more persuasive evidence without doing anything to counter that evidence beyond loudly exclaiming how little you like it.

Ok? That doesn't really tell us anything though, because we can't see what would have happened if you didn't interrupt him. All you've shown is that you can somewhat reliably interrupt an ogre. If interrupts represented a full reset or cancellation of actions (where you would have stopped 6 attacks), that would be pretty impressive, but that's not how interrupts work in this game, so it's not at all persuasive.

Uh? He wasn't interrupting the bear indefinitely in either of those scenarios. It was still getting in hits, it was just missing. Doesn't the 97-point swing favor the half of the experiment you support? Also, in this experiment the 97-point swing IS the independent variable (i.e. the variable you change for the experiment) and the only effect it would have in this encounter is changing the rate of interrupts.

That would likely help to make the amount of power a mage can bring to bear in a single fight more predictable. That said, I'd think it would be REALLY hard to balance properly, to the point that you'd be better off losing the "HP" portion of it and just making it a per encounter limiter.

This is precisely the type of trade-off that must be avoided in a well-designed game, in line with what I mentioned above. With the current game design, players are being put in a lose-lose situation, while developers are unable to balance encounters properly due to their inability to anticipate how rested the party is. I know. You'll find I've stated both points several times, possibly in this thread.

Actually, I don't think that addresses any of the problems with rest. You still have a rest mechanic that has you trading out-of game frustration for in-game power, still does not actually constrain you, still discourages using many of your more interesting abilities, still asks you to plan on the per rest timescale but doesn't give you anything to effect that scale, and still doesn't really have the wide breadth of powerful abilities that can pretend to warrant it.

Hmm. I've heard Monks attack really quickly and barbs hit everything around them. I wonder how a monk+barb team-up focused on interrupts would work out. Maybe throw in a a rogue with an arbalist to support them.

I don't think this will give you much information beyond "Who does auto attack better if that's all they're doing", because it ignores all of their other abilities and ignores how they interact with the rest of the party, both of which are incredibly important.