I think you could make the argument that 5-star mythic epics with three good substats has roughly the same value as a 6-star normal epic with three good substats. The mythicness essentially compensates for the weaker substat/primary stat by providing a free substat roll.

However, it's a pain to filter your artifacts each time to say "sell only 5-star items that are not mythic epic or lego". So, that alone will make those mythic items go straight to the vendor.

Well, when the promised UI upgrades come, you won't have that excuse.

I'm not holding my breath :)

5* mythic equivalent to 6* uncommon? This doesnt seem accurate, as usual I did not do the math... but that conclusion doesnt pass the sniff test.  Certainly not for top row items, at least?  What would a 5star mythic with speed/res/acc/hp% subs compare to a uncommon 6* of same?

ok, just looking quickly at my gear, it seems that:

spd is 4-5 vs 5-6

acc/res is 8-9 vs 9-11

hp is 4-6% vs 5-7%

So it looks like my 75% guesstimate was a bit off. It's more like 80-85% depending on which substat. If we take 85%, then we have -1.35 instead of -2.25 substats, so I guess it's equivalent to rare.

And yes, I was considering bottom row items, with the -1.5 substats penalty on the main stat. That's half of the difference, hmm...

On a weapon, the difference between 5* and 6* is 40. Given a base attack of ~1000-1500, that's around 3%. On the helmet, the difference is 600 hp, which is around 3-4%. So -0.5 substats would be fair. For speed boots, it's also only -1 substat instead of -1.5.

That means a top-row 5* mythic is equivalent to a 6* epic. Boots are halfway between rare and epic. Does that sound more reasonable?

Overall, this just strengthens my point even further, specially regarding 6* epics. You're spending a lot of silver for a much higher rate of bad rolls at +16.

I don't think I'm following your conclusion at all. Are you trying to argue that 5* mythic is better than 6* epic? If so, I can't say I agree at all - except if the 5* mythic has all four substats being good.

A level-12 5* mythic with three "good" rolls should be roughly equal to a level 12 6* epic with three "good" rolls. But rolling that 5* to 16 and hitting the "bad" roll is going to be more punishing than doing the same for the 6* epic and getting a "bad" roll, because the primary stat is significantly weaker.

You can surely make exceptions, yes, but the hassle of finding and filtering items just doesn't warrant the effort, I don't think.

Yes, that's exactly what I'm arguing. One important thing is that I'm less focused on the specialized triple/quad roll pieces, and am looking at the raw amount of stats instead.

I agree that hitting the bad roll on the 5* is worse than on the 6*. But the 5* only has a 25% chance for a bad roll, while the 6* has a 75%!

That's not the right conclusion at all. The 6* can't hit a bad substat on the first three rolls, because there are no bad substats. There are only good substats. So, the only chance for it to hit a bad substat is on the 4th roll - and what defines "bad" for that last roll is pretty subjective. Like, even if I were running a nuker champ, is it the end of the world if that last roll hits HP%, DEF%, or even RES? Probably not. Realistically you probably will only be upset if you hit flat substats, which is a 33% chance of happening.

So - if you start from level 1, the mythic with a single bad substat has a 68% chance to hit a bad substat by level 16, vs the epic having only a 33% chance. If you start from level 12, the mythic has a 25% chance vs 33% chance for the epic.

If you have a tool that lets you auto-upgrade all your items to 12, sure, there's logic in then taking the resulting mythics and selling the ones that hit the bad roll, and at that point, sure, the mythic has a marginally better chance of being good than the epic. But if you don't, it's just such a slog to work through, that the *very* small improvement on the mythic is simply not worth the effort.

If the 5* mythic rolls bad at +8 or +12, I just sell it and no harm done. A much larger proportion of 6* epic gear will look good at +12 (because when it doesn't, you just sell it at +0). But the point I'm making is that this is a trap. You'll end up with a lot of mediocre +16 pieces.

An important point is that my analysis puts a lot of weight on silver efficiency. I'm always running short on silver, and I used to have hundreds of unrolled items lying around. Nowadays I try to sell much more aggressively, and actually roll more pieces.

We're not comparing gear in terms of "would I be happy if a dungeon removed 1% from 6* epic drops and put that 1% into 5* mythic instead?". That's not the question that's on the table here. Rather it's: "I don't have enough silver to roll everything. How aggressively should I sell partially rolled gear, and at what levels?". The conclusion I'm reaching is that 6* epics need to be extremely impressive (basically the exact 3 substats you want) to merit being rolled at all, because the 4th substat will be a disappointment more often than not.

I basically used to look at epic pieces and be like "does this have a chance to look impressive at +12?" and roll up based on that. But now I'm trying to anticipate the decision to roll to +16. If a piece is going to be 100th in line for that last roll, then I shouldn't even bother with the +8 and just sell it straight away.

I agree. One shouldnt undervalue great epic pieces. They are much easier to get lucky with. I'll take an epic reaction ring with all three % stats any day above the same piece in legendary, which has a 4th bait flat stat which can easily ruin the piece by the time you hit level 12 on it. 

That's really not at all what I meant, Swiftblade. A legendary item with three good substats is 100% worth keeping and rolling to 12, at very least. If you hit the bad stat twice, then go ahead and sell, but if you hit it only once, then you're basically in the same world as if you were using an epic item, except now you have only a 25% chance of getting a bad roll, instead of 33%.

I only keep 5* lego with 4 good subs so can't roll bad.  Speed/Acc/Res/HP-DEF% Speed/CR/CD/xx%.  I'm assuiming with mythic (if I ever start farming Hard mode...) I'll treat 5* like lego allowing for 1 bad sub if the gimmie roll is good.  

These Items seem better than 6* epic with a mystery roll to me.  with bottom items I'll stop at 12 regardless of rolls and oil up...

The silver difference is huge as well, I think?  Anyone have the math on this?

I don't even keep ANY 5 star gear anymore. Just doesn't bring enough compared to my current gear, I feel. All 5 star forge gear, and farmed gear is autosold.

If the chance of upgrading is the same per roll, then 5* costs about 80% of 6*.

(that's the ratio at +16, I also checked at +12 and +8, there 5* is even cheaper, ~70%)