Just lookin for some opinions…
I play a Deathless Moze, with 200+% wpn dmg, 100+% Splash, and 150/50 rad anoints…
-Does Selfless Vengeance have any use? I reload often, so is it better to get some elemental damage or get dmg I might not need in Armored Infantry?
-I hear about “diminishing returns” when you stack too much of one dmg type, like weapon damage. But does that apply if you have boosts from splash and radiation 
I have a build that uses both SV and AI. But it’s a rocketeer not a bloodletter. I actually think SV is pretty useful on a reloading Moze. You would need StE to get any real use out of it.
Oh yea…I use Stoke the Embers with Experimental Munitions, Cloud of Lead, FitSD, w/ 13% more elemental dmg from Guardian rank…but Im still on the fence vs AI
Well in this rocketeer build, which I’m doing an M10 playthrough on, I dropped DiB for SV+AI. So I could get to Experimental Munitions. I don’t go any further in red. It certainly works, but I’m backing it up with some pretty well-rolled gear. DiB would usually be the better option in most cases.
I hate the term “diminishing returns” as it is very misleading IMHO.
Basically, when you stack two types of a boost onto one type of damage they say you get diminishing return.
For instance, lets say your radiation damage. You do 1000 radiation damage and get a boost of +20% radiation damage. you now do 1200. If you put another boost of +20 radiation damage you now do 1400. They call it a diminishing return because it only stacks against the base damage and not the boosted damage. You still get the full 20% boost of the base damage from each source.
If it stacked against the boosted damage you would receive 1440 instead of the 1400. They say it is a diminishing return because it doesnt stack up. So if you had a boost to weapon damage, a boost to radiation damage, a boost to SMG damage, etc. Each of the seperate types would give a full boost from the initial damage and some of them will actually work on the stacked damage. I can’t recall which ones stack and which ones dont but there are threads out there to micro manage all of that.
Yes, those little boost differences can add up if you have enough of them but, all in all, bonus damage is still better than no bonus damage.
Hopefully I have helped to clear it up a tad.
@Xolos is right but I’ll try to expand on what he said a bit.
People also use the terms “additive” and “multiplicative” in referring to the same thing, somewhat sloppily.
In his example the two +20% damage buffs are “additive” with each other but would be “multiplicative” with e.g. a splash damage buff.
The reason people say “multiplicative“ buffs are better is because in this game if you have a choice between stacking buffs of a single type and getting a few different “types” of buffs, the numbers generally favor the different-types approach.
Building on the above example, say you have a rad gun that deals 1000 damage.
Which is better: 3 different buffs of 20% to gun damage, or a 20% buff to rad, a 20% buff to gun, and a 20% debuff on an enemy?
One might think it shouldn’t matter, It’s three 20% damage increases either way. But in fact the different-types option will be unambiguously better because of how BL’s engine calculates the damage dealt:
Damage = Base x (1+gun buffs) x (1+elemental buffs) x (1+debuff)… (Note, greatly simplified).
Damage with gun buffs = 1000 x 1.6 = 1600 (The three gun buffs are “additive”).
Damage with gun, rad, debuff = 1000 x 1.2 x 1.2 x 1.2 = 1728 (these buffs are “multiplicative”).
Because the additional gun buffs aren’t as effective as stacking 1 gun buff with a rad buff and an enemy debuff, people would say the additional gun buffs have “diminishing returns.”
So the point here is that it is generally better to spread damage buffs out across different types - elemental, splash, gun, enemy debuffs, etc. If you already have a ton of gun damage buffs, probably go get more elemental or splash bonuses instead of still more gun damage.
Of course, if your choice is between a really big “additive” buff and a small “multiplicative“ one, the math may well tilt in favor of the big additive buff.
For your purposes, SV probably doesn’t have a major “diminishing returns” or “additive” problem if you’re already reasonably balanced between gun, elemental, and splash buffs.
I admit I don’t use SV much, but that’s mostly down to personal preference. It may well be right for your build and gear.
Thank you for expounding upon my explanation there. As there are several threads and videos abounding about the subject I didn’t want to get too far into it but I think you did it very well and succinctly. Well done. 
I would think the term compounding would be more apt than multiplicative but it has became a part of the lexicon now so no real chance of changing it.
Technically speaking, the term diminishing returns is really only applicable to stats where you are trying to reduce something, like fuel drain for Iron Bear or shield recharge delay. Although there is also reload speed and action skill cooldown rate which are stats you can increase, they function based on duration and ‘increases’ in this context are just an alternative way of saying you’re reducing the time spent reloading or waiting for your AS while on cooldown.
When it comes to stacking one type of damage multiplier more than any other, no doubt it’s less effective than trying to balance the other multipliers. But in that instance, I would simply say that the former is a less commensurate approach than the latter, rather than label it as diminshing returns. Buffing a splash gun with bonuses of 50% gun damage and 50% splash damage, will obviously yield a higher total than 100% of just one bonus. But the inferior option is still a 100% bonus - the base damage of that splash gun is double, so no returns from that 100% bonus gun damage are being diminished. It’s simply less commensurate than diversifying the multipliers.
For example, The Pearl of Ineffable Knowledge adds 90% gun damage to the formula, while the stacks add their own multiplier of about 16% - each stack is multiplicative according to research carried out by others so it’s like 16.09% instead of 15%. Unless you’re a Spiritual Driver Amara who can seamlessly activate elemental projector 24/7, it’s the BIS artefact when it comes to the damage formula. But because the PoIK comes with a limited range of passive bonuses, you then have to get splash damage and weapon type damage passive bonuses from your class mod so their respective multipliers are put on the board.
After that point, there’s only so much that can be done with the character’s skill trees, special skills like Short Fuse being a prime example, and whatever else is on offer from class mods like the Blast Master’s bonus 100% splash. My point is that even though we should all try and take the commensurate approach when it comes to our builds, our options are rather limited at the same time for optimal DPS when you account for using the best gear out there.
Thanks everyone…that def cleared it up
Spreading your multipliers is as simple to explain as why 3 x 3 is better than 4 x 2.
No diminishing returns. It’s just math. It isn’t more complex than that.
Let’s say in your case you are at 200% total bonus gun dmg and 100% total bonus splash. You are at a 3 x 2 situation.
Suppose you are trying to decide whether you should go for 6/3 Desperate Measures (+100% gun damage) or Blast Master com (+100% splash damage).
Desperate Measures:
4 x 2 = 8
Blast Master:
3 x 3 = 9
There is your answer.