IO Purchasing Question

On the other hand, if that 0.4% defense is the difference between 44.6% defense and 45% defense, slotting the 25 makes you get hit almost 10% more often. And you might be trying desperately to conserve slots, and thus not be getting hit by the ED cap very much. That's one of the reasons some of us folks plan builds down to the last 0.1% in Mids'.

I believe lots of people slot 30s or 35s to allow for exemplaring without losing set bonuses. And I haven't checked prices for lower level stuff recently, but you can probably also save a fortune that way.

In most situations, the level (past 25 or 30) won't make much difference. And when exemplaring, that level 25 will be a lot better than that level 50. So it just depends on how you plan to play, what your budget is, and what your final numbers end up at.

This statistic does not really represent reality. It's the difference between getting hit 5.4% of the time and 5% of the time, because you will always have a 5% chance of being hit. While that looks like an impressive almost 10% of the time, it really means that you'll get hit 5.4 times out of every 100 attacks, rather than 5 times. That means that you'll get hit less than 1 more time every 200 attacks with a 44.6% defense instead of a 45% defense.

Yeah, it's almost 10% higher hit rate on the percentage. But it's happening so infrequently that it doesn't really make a difference in practice. Ten percent of almost nothing is still almost nothing.

Where it makes a bigger difference is when you're fighting against higher-conning enemies, or when you've been debuffed. When your effective defense is significantly lower than the softcap minute amounts of defense definitely help.

So, yeah. You want to reach the softcap when possible. But it's not a magic bullet. There's always a bigger bad that will sneer at your soft-capped character, like DE with their quartzes, or Radiologists, or level 54 bosses.

The other consideration about what level to get your LotG: +recharge IO at is exemplaring. If you get all level 50s, you won't get that recharge bonus at level 44 and below. That means you won't have it when you run many of the TFs in the game, or when you run most Ouro content.

I'm not saying that the softcap is a bogus thing: I personally try to exceed the softcap. Because of ED, though, you can slot a level 25 LotG: +recharge with two other defense IOs and still get like 50% defense on all three positions on a shield tanker, for example, with the appropriate selections in other IO set bonuses.

If you exercise care in how you slot your lower level LotGs you should still be able to hit the softcap and keep your recharge at lower levels.

Nor does it represent common sense either. It quite literally takes the most extreme example possibly and ignores that no-one would ever do this. It further ignores that I mentioned ED will drain away this difference. I don't believe there is any intellectual honesty given in this example of "up to 10% worse".

For instance on my SD scrapper, electing to take a 25 IO over a 50 IO drops my melee defence by 0.0%. Mid's only puts it to one decimal place so there would be some loss, somewhere.

To me losing 0.0% to gain better examplaring is a no brainer.

except...

and that it is generally considered good practice to build over the soft cap to account for defence debuffs.

It's a lot more than almost nothing for those of us pushing the survivability envelope on a defense toon. Soft-capped toons die all the time. That doesn't seem like "almost nothing". Put yourself in a +4x8 mission and the incoming damage, even at the soft cap, can be very high. 10% MORE incoming damage can mean the difference between death and success.

Right. More enemies, higher conning enemies, whatever brings in more incoming damage. The soft cap doesn't matter if you're running on +0x1. It matters when the going gets rough. So does resistance. So does all damage mitigation. So do regeneration and healing.

No, when your effective defense is significantly lower than the softcap, minute amounts of defense are minute. 44.6% vs. 45% defense means about 10% more incoming damage, while 0% vs. 0.4% defense means about 1% more incoming damage. That last little bit of defense is about 10x as effective as that first little bit.

Quartzes and Radiologists, yes. Level 54 bosses, no. Their to-hit is still floored, and can go no lower. Yes, it's significantly higher than a level 50 minion, but this is also where the soft cap starts getting important - difficult enemies whose to-hit you can still floor.

Right, which is why I said, "I believe lots of people slot 30s or 35s to allow for exemplaring without losing set bonuses" and "it just depends on how you plan to play".


Oh, certainly. You by all means don't NEED to slot level 50 Luck of the Gamblers to hit the soft cap on a Shield Defense, or a Super Reflexes, or whatever. It just makes it slightly easier, and the saved effort can be spent elsewhere in the build for other things. If you exemplar, of course, you're best off keeping your set bonuses. But if you don't, the extra defense can make a difference.

No one? Have you SEEN the Scrapper forum? Even famous build masters like Shred Monkey have compromised slightly on the soft cap to achieve other goals. Putting those characters in nasty situations where being a little below the soft cap makes a meaningful difference seems almost the rule instead of the exception. A lot of us play that way.

But I didn't say EVERYONE played that way either. Most people don't. I was explaining why "some of us folks plan builds down to the last 0.1% in Mids'", which was brought up by Suspicious Package. I then went on to explain that "lots of people slot 30s or 35s to allow for exemplaring" and "In most situations, the level (past 25 or 30) won't make much difference."

Did you miss where I said, "And you might be trying desperately to conserve slots, and thus not be getting hit by the ED cap very much." I didn't ignore Enhancement Diversification; I mentioned it explicitly, and said why in some (certainly not all) cases it might not be as big a factor as you make it out to be.

And seriously, you're saying that I'm being intellectually DISHONEST? That I'm frickin' LYING? Did you notice ANY of the qualifications I put on what I said? Anything about how what I was saying only applied in certain situations, and didn't apply "in most situations"? Or is accusing someone of lying your default when you don't agree with what they say?

Everything's a trade off. If you want to exemplar, this trade off is obvious - stick with lower level IOs. If you don't, then the trade offs can be much more subtle. As I said, "you might be trying desperately to conserve slots, and thus not be getting hit by the ED cap very much." A common solution on high end defense builds trying to conserve slots is to slot a Luck of the Gambler and one or two Enzymes. For the sake of argument, I'm going to take a Super Reflexes Scrapper, everything from the secondary, Combat Jumping and Weave, all slotted the same (which violates the rule of 5 - no more than five of the same bonus - but it's just an example, and let's assume we're making similar decisions regarding other IOs). Well, Cytoskeletons in the passives since they won't take Enzymes. We get this:

35.66% L25 LotG 1 Enzyme
36.57% L50 LotG 1 Enzyme
39.42% L25 LotG 2 Enzymes
39.69% L50 LotG 2 Enzymes

Not an entirely realistic example, but perhaps somewhere in the ballpark of the numbers you would see following a L25 vs. L50 slotting strategy. Again, obviously go for the L25s if you plan to exemplar much. But if not, we're seeing a 0.91% spread for the fairly unrealistic one Hamio example, and a 0.27% spread for the more realistic two Hamio example. The 0.4% number (which came from YOUR example, so I was just repeating your number as a for instance, not making it up myself) is probably a bit high for the average completed build, but may be a realistic representation of the difference on some builds, particularly those trying to conserve slots on defensive powers.

(Edit: For a real example, my Fire/Shield has 45.05% melee defense. If I use a level 25 Luck of the Gambler in Deflection, my melee defense drops to 44.74%, resulting in an average 5% increase in incoming melee damage in most situations from a single IO change. Tiny enhancement differences can and do have significant effects on real builds. Could I change things around to soft cap again? Of course. But every change is a compromise. Since I "never" exemplar, that level 50 Luck of the Gambler is one less compromise I have to make.)

Almost irrelevant. Quoting myself from a different thread:

I figured I'd jump in here to fix the typo in the above, now bolded, before someone discounts his entire spot-on post for a typo. Also, Werner is, quite frankly, the last person I'd ever think of accusing of "intellectual insincerity".

Thanks, fixed, and thanks.

(Edit: In general, there is certainly a danger of me accidentally misleading people due to my rather singular focus on top end performance. For instance if someone asks about the comparative DPS between two sets, unless they specify otherwise, I'll be giving them the comparison between top end IO'd out level 50 characters dedicated to DPS. I think I often or usually put in appropriate disclaimers, but I may forget, or may think based on context that they can be omitted. So I might accidentally mislead someone if they instead were curious about DPS for the 1-25 level range, or about a level 50 on SOs, and simply didn't say this. Or if they're reading a thread about top end DPS, and think it applies to their level 30 AoE damage output since nobody explicitly says it doesn't.

But in my original post, I was responding to some specific points, and I feel that I included adequate disclaimers about when what I was saying did and didn't apply, and what made sense when "in most situations" what I was saying did NOT apply. So I didn't feel like my post would even accidentally mislead people (much less purposefully do so), as long as they read the whole thing and not just the first three sentences.)

This here is exactly what I was talking about intellectual honesty. What you have written may technically be true but it is expressed in such a way as to bamboozle anyone who doesn't have a respectable grasp of statistics and to me that is not honest at all.

Let's see why. The problem has been shown before by Rodion but I will highlight it once more.

Compared versus actual received damage this is true. However, against incoming damage, it is false. 0.4% defence provides the same quantity of mitigation wherever and however it is applied. If you are going from 0% to 0.4%, it mitigates 0.4%. When from 44.6% to 45%, it mitigates 0.4%. If you were to undergo any sustained attack and calculated the mitigation provided, they are equal. Losing 0.4% defence means you are hit by an additional 1 in 250 attacks, whichever way you may choose to express it.

Firstly, here, I'd like us to move away from 0.4%. I expressed in my first post that this is not a logical number but one arrived at only in the total absence of ED. It is the most extreme difference possible, but it is not a rational difference. Why?

It is a dominant strategy to build to at least the ED cap, as it provides more returns/slot than any other way. If you want defence, build to the ED cap. Being under the cap is a dominated strategy because you will be expending more slots elsewhere to receive the defence you could have just by slotting it 'normally'.

Now to provide some examples. I'll use a Shield Defence scrapper here, with Deflection (their main melee defence toggle) the power of choice.

2 Enzymes + LOTG 50: 17.9%
2 Enzymes + LOTG 25: 17.9%

LOTG set (Def, Def/End, Def/Rech, Global/Def) 50: 17.8%
LOTG set (Def, Def/End, Def/Rech, Global/Def) 50: 17.7%

In the first example, there's no difference to 1 decimal place. On my /SD scrapper, that's the way I slot mine.

In the second example, there's a 0.1% difference. Now, we need to make some bold assumptions here before we explore what differences are even felt. The reason they are bold assumptions is that your own argument attempts to highlight the significance of the difference and then goes ahead by making no attempt to close this difference. To me, that's not a logical argument at all. If you were truly concerned about the impact, you would presumably make this up elsewhere - the difficulty of which is going to be highly build dependent and also will rely upon your own personal views of the trade off. Because we can't view each others personal wants from a build, let's pretend that it was impossibly to find another 0.1% in a build. What really happens?

Let's assume there's a continual stream of attacks that so happen to attack that particular defence you have sacrificed the 0.1% on. For sake of argument, it's 500 damage per second (DPS). Now we are looking at some serious damage if there's no mitigation at all in play. A squishy will die in very short order from this (250 DPS with no defence, so.. 4 seconds = 1000 damage, 8 seconds = likely death).

What happens if you can reduce this incoming damage to a 5% hit chance, versus the 5.1% hit chance?

At 45% defence, you are taking 25 DPS.

At 44.9% defence, you are taking 25.5 DPS.

Over the course of a full 10 minutes of attack, this sums to a paltry 300 damage. To put that another way, of the original 500 damage you would take, the difference is 0.5, so you mitigate a grand total of 0.1% with 0.1% more defence - as I have elaborated above.

This is where I believe the "10%" is not an intellectually honest answer. It does not display the reality (that it is significantly less than 0.4%, as I have written since the very beginning), nor does it reflect the real terms of difference. It is, to restate, also inconsistent throughout your argument when you say that 0.4% mitigates less than 0.4%.


Disclaimer: This was hard to write without sounding rude, take my apologies on it first up...

I'm not interested in you putting words into the mouth of others here. Perhaps they feel this way, perhaps they don't, but it really is irrelevant as it doesn't validate an argument. As a side tangent, this game is so spectacularly easy and so utterly skill deprived that it generally is just the (loudest/most prolific posters/most prone to present their achievements to others) that achieve such 'titles'. For all you know, you could be talking to anyone who has done exactly the same but hasn't felt the urge to tell the forums about it. Even if you were or are speaking to someone like that, it doesn't necessarily make them right!

The decisions about what to aim for are going to be the individual preferences about what someone wants, so an argument along these lines is difficult to resolve. However, you would have seen by now what a 0.1% difference actually means in the defense score.

You wrote something that is to me highly misleading. You furthered that by writing something that is untrue, that 0.4% applied to 0% defence provides less mitigation than adding 0.4% to 44.6%. They in fact mitigate the same quantity of damage. A lot of your other argument appeals to this concept of 10% damage.

Regardless of numbers you use, it provides 0.1% mitigation. It was for illustrative purposes only, feel free to submit your own numbers and see that the difference in mitigated damage is not 10%. If you survived with ~0.1% of your life at the end ( so uh... for a 2000 hp brute, that's 2hp), then you can thank using the 50 IO. Otherwise, it didn't matter. Assuming the damage is distributed continually.

I must be bad at math, because I'm showing a much higher damage mitigated % than .1%, but I'm easily confused.

Are comparing 44.6 to 45? Or are we comparing 44.9 to 45?

As I said, I'm not very bright, so clarification is important I guess

I used 44.9% to 45% as the example as that's what you got from just using LOTG as a set. As I showed earlier there is 0.0% difference with Enzymes.

Let's say there's a 1000 damage hit coming at you.

1000 * (50-defence score%)/100

Your expected damage with 44.9% vs 45% defence:

1000 * 0.05 = 50
1000 * 0.051 = 51

Your expected damage with 0% vs 0.1% defence:

1000 * 0.5 = 500
1000 * 0.501 = 501

As you can see, adding 0.1% defence to "near soft cap" and "to absolutely no defence" both provides the same amount of mitigation. In this example, it is a mere 1 point of damage. Now you can say that 51/50 is 2% more damage, and that 501/500 is 0.02%, and so it is more important for the first example... but... would you feel that this is misleading when they mitigate the same damage?

May I enquire what defence your /SR brute has? Is he/she over the 45% soft cap? Because if so, it's quite probable that in the absence of debuffs there is quite literally no difference to performance in mitigation between a 50 and a 25 LOTG Global/Def IO.

If people read it just like I said it, that you'll take almost 10% more damage, that the last percentage point of defense is almost 10x more important than the first percentage point, then they're reading it RIGHT. I'm not BAMBOOZLING people with no grasp of statistics. I'm trying to FIX their incorrect perception. While true that losing 0.4% defense means you're hit by an extra 1 in 250 attacks, that extra hit matters MUCH more when you're barely being hit by any attacks at all. If you're already being hit by 125 in 250 attacks, an extra 1 makes almost no difference. If you're only being hit by 13 of every 250 attacks, that one more hit WILL make a difference, at least if you're playing at a difficulty level where being hit by 13 of every 250 attacks is a challenge.

I am very explicitly saying that you're looking at it wrong. I'm not tossing out weird numbers to try to mess with people with a poor grasp of mathematics. I am very explicitly saying that yes, the way you're reading what I'm saying IS what I'm saying. That is exactly what I am trying to say, and it is correct. Even if you don't believe it is correct, it is not intellectually dishonest for me to say it. I'm being as honest as I can be about the truth as I understand it.

I'm sure you'd like to move away from 0.4%, but I'm not going to. It is not out of question, and it is the number I was responding to. I realize you then said it would be lower because of ED, but you didn't state another number. Had you stated 0.1% from the start, I might not even have responded. But that's neither here nor there at this point.

Anyway, let's have a look at Enhancement Diversification for defense. You get full value to 40%. You get 90% of the value for the next 10%. You get 70% of the value for the next 10%. You get 15% of the value after that. So I'll agree that on a typical build, you'd want to go to the ED cap, by which you should mean the 60% mark or so, picking up 56% enhancement for your efforts. Anything above that is getting pretty wasteful. It might still be a good idea for some powers in some builds, but it wouldn't be the rule of thumb.

So your rule of thumb is to be AT 60%. Less, and you're not taking best advantage of the power. More, and most of the enhancement is wasted. If you're at 60%, that 4% difference in the luck of the gambler is a 4%*70% = 2.8% difference. Yes, it's hit by ED, but it's not removed by ED. That 0.4% difference becomes a 0.28% difference, and I believe you were talking about a single power, not all powers across the build.

In practice, some builds will be comfortably over the 60%, and will only be hit for 4%*15% = 0.6% difference, which will be fairly negligible, and seems to be what you're assuming. That could add up across an entire build sitting right at the soft cap, but it'll probably be rare. Some builds might have some powers low enough to get hit by 90% or 100% of the difference.

Again, since I was RESPONDING to the mention of 0.4%, and since I was explicitly stating that that's what I was talking about, and why a number like that would make a difference, I'm going to use 44.6%, and not ED cap it down. Again, this isn't completely off base for some builds, and the point I was making was specifically that 0.4% matters, not that 0.1% matters. Obviously 0.1% matters much less.

So at 45% defense, you are taking 25 DPS. At 44.6% defense, you are taking 27 DPS. Over 10 minutes of attacks, that's 1200 damage. Some of the fights that some of us care about last over 10 minutes. For some of those fights, that 1200 damage is the difference between success and failure.

I didn't say that 0.4% mitigates less than 0.4%. They'll mitigate the same AMOUNT of damage. But mitigating that amount of damage will be almost 10x as effective at boosting your survivability when you're at the soft cap than when you're at 0%. And this is where people might get a little confused by the difference between mitigation and survivability, and about semantics, and about what the forum commonly means by various words.

So saying that a lot of Scrappers on the forum play this way is putting words in the mouth of others??? Also, it IS relevant to the PART of the argument I was responding to, which was "no-one would ever do this". So here's this part of the argument from my perspective:

You: What you say makes no sense because no one would ever do this.
Me: Actually, some people do this.
You: Don't put words in other people's mouths. Also, that's irrelevant.
Me: ???

Agreed that this game can be almost facerolling on the keyboard easy. Fortunately, the difficulty settings among other things allow us to find a challenge if we want one. The points I'm making are only relevant to the people doing so. You may be one of those people, and you might be able to wipe the floor with AVs, solo pylons at the top of the charts, run +4x8 with no problem, and so forth. I'm aware of that and trying to make no assumptions about your capabilities as a builder and a player. It doesn't mean I have to agree with your points any more than you have to agree with my points. You are correct that such accomplishments mean nothing in this sort of discussion.

I feel that what I wrote was highly leading, and honest. I did not say that 0.4% defense mitigated a different amount of damage depending on your defense. I agree that it mitigates the same quantity of damage. I'm sorry if you misread that, and hope it didn't mislead anyone else. My point is that if you're already being hit 125 times out of 250 times, an extra hit makes little difference. When you're being hit 13 times out of 250 times, an extra hit makes a big difference. Almost ten times as much difference.

(Edit: No, actually, I WILL make the stronger point that you're accusing me of having already made. I'll say that in practical situations, where the Scrapper is trying to get some sort of challenge rather than facerolling the keyboard through the mission, 0.4% extra defense DOES prevent more damage on the higher-defense Scrapper.

Your numbers start with both Scrappers facing the same enemies, the same attempted damage. That's not particularly realistic unless one Scrapper is woefully underchallenged or the other is killed in seconds. That's not how either will play in practice.

The Scrapper with 0% defense regenerating 25 hit points per second can only survive 50 DPS of attempted damage. Most of us would just say 25 hit points of damage that would hit a 0% defense 0% resistance Scrapper, but either convention works out the same. For a challenge, they'll be running missions at a setting that does about 50 hit points per second of attempted damage. Adding another 0.4% defense will prevent 0.4% of this damage, or 0.4% * 50 DPS = 0.2 DPS.

The Scrapper with 44.6% defense regenerating 25 hit points per second can survive 463 DPS of attempted damage. For a challenge, they'll be running at a setting that does about 463 DPS of attempted damage. Adding another 0.4% defense will prevent 0.4% of this damage, or 0.4% * 463 DPS = 1.852 DPS, or about 10x as much incoming damage.

Still overly simplistic, leaving out resistance, pretending that average damage regenerated is what you can survive and what you'll tune your missions to, but more representative of reality, I think, than both Scrappers facing exactly the same enemies with vastly different levels of mitigation.)