Rolling offcolors too punishing

I didn't really expect anyone to respond to this. I just wanted to voice my concerns to the devs. Because, to be clear, I think this is a magnificent game. It's probably in like my top... two. But I didn't really say much of what I wanted to, because I was frustrated and had lost interest in the game. I'm sure I'll play again, I just needed a break. Anyway I should have explained my concerns better in case devs are reading this. I'm going to drop a big, crazy wall of text here, because I suck balls at writing succinctly.


As someone said, I could have used a hybrid chest. I was using a 5L pure STR piece and trying to get RGBBB for cyclone, curse on hit, elemental weakness, enfeeble, and life on hit. The only hybrid piece that would have raised my chances much would be a STR/INT, and the extra energy shield would have actually been worse than useless. It would just disappear the second I cast RF and stay gone until I entered town. It would also raise my righteous fire burn dmg without giving me any additional regen, which is worse than having no ES on the armor at all. I'm in invasion, so I didn't have the option of buying a higher-level STR chest with the right colors. The only decent choice I saw at the time was to pray for four off colors on the really nice STR chest that I picked up and then 5-linked. If I had it to do over again, I would have bought a lower-level pure str chest for easier off-socket rolls and taken a hit on the armor. However, that person was right that I, and a lot of people, discount the possibility of using hybrid pieces. Knowing what I know now about the chances of off-color spawns, I would be less inclined to avoid them in the future.


I normally try to make informed decisions. When I wanted to make a 6L, I researched the odds and found them to be between 1/900 and 1/1000. I calculated that if I were able to get it in less than 1200 tries, I would save money compared to buying the near-perfect 6L Lionseye I wanted. It took 1400. No big deal, I'd make the same choice again. But this time around I only wanted 4 off colors on my chest, and could afford to buy up to 2000 chroms, so I didn't research the odds on off-colors. I figured I had a big enough cushion, and I was okay with using every last chrom. Big mistake.


I don't know exactly what the odds are, but I think we have some halfway reasonable ballparks. I got a single 5-off in my ~2000 chroms, and other people have said that the average is around one 5-off per thousand. That would work out to be exactly a 25% chance for any individual socket to be an off color, so it sounds like a reasonable guess for something the devs would pick (I'm talking about max level gear, chances are higher for lower level gear).


Now the problem here is that to get the 5-off that you want, you only have a 1 in 6 chance. On a 5-off pure red piece, you can have 0 greens, 1 green, 2 greens, 3 greens, 4 greens, and 5 greens, and obviously the number of one color determines the number of the other color if the sockets are off-colors. So now you have a 1 in 6000 chance of getting the piece you want. Unfortunately, if you were going for 5-off on a 6L, you were planning on one socket being a primary color (ditto for trying to get 4-off on a 5L). This means you still have to deal with the 25% chance of getting an additional off-color that would take away the primary color you needed. So you now have to raise your input by 33% to account for having its average value decreased by 25%, in order to maintain the same expected chance:
(1000 - 25% = 750, 750 + 750 * .333 = 1000).


So your chances are now 1 in 8000 to get exactly the 5-off combo you want.

As for getting the correct 6-off, it would be about 1 in 28,000 (1 in 4000 chance of 6-off * 7 possible combos). As a fun fact, it would take 15 and a half hours of clicking on chroms to actually use that many if you used them at a rate of one every 2 seconds.


Now I absolutely agree that more powerful items and builds should cost more, so that your character always has a path to progression. A lot of people say "it should be hard" to craft powerful gear. But then shouldn't we be focusing on 6 links, not on 6 off colors? If my figures were roughly correct about having a 1 in 28,000 chance to get the correct 6 offcolors on something, then it costs over 3 times as much to 6-off something as it does to 6L something (league-dependent). I personally think it should be the opposite.


Now I doubt my numbers are quite accurate, but I don't think they're wildly off either. Maybe I've overestimated it by half. That would mean it's still more expensive to get a correctly-colored 6-off than a 6L. Maybe they're UNDER by half. Then we'd be sitting on a 1 in 56,000 chance. And it really is important to discuss specific numbers, or at least relate it to the difficulty of other crafting, because "it should be hard" can apply to basically any number you can dream up. That's not a recommendation. Try cracking open a game engine that expects a floating point decimal to calculate the chance of an off-socket, and instead typing in "it should be hard."


I agree that powerful gear should be expensive. I'm in favor of lowering the chance to roll a 6L. But making it harder to get off colors doesn't make it any harder for the large numbers of players whose particular OP build doesn't require more than three off colors. It just arbitrarily cost-discriminates against some people who want to do different things. I'm not opposed to "hard" at all. I'm opposed to how that "hard" is distributed among the facets of a build, and whose shoulders it falls on. Also, it's worth mentioning that the cost to get off-colors would not simply drop to zero if they changed the odds. Let's say I'm roughly correct about the odds of getting a correctly-colored 6-off. Is anyone gonna be offended if they changed it to an average of 20,000 chromes instead of 28,000? Is that just not enough of a cost to personalize a build? Maybe I'm wrong and the real chance is 1 in 14,000. would it really be a problem if they changed it to 1 in 10,000? How many chromes did you have to spend personalizing your gear?


Also, it IS and will remain hard to get the best gear even if you got to pick your socket colors for free. Few people are using mirrored items, and virtually nobody is crafting them (especially in 4 month leagues). There is literally nobody who is using technically perfect gear, and most people still have tons of room for improvement. Also, this game is about rerolling for a lot of people, not super long-term character investments. What percentage of characters are 95 or higher? It should be taken into consideration. Do we want to design crafting systems around each character being played for 800 hours if the average is only 150? (numbers completely made up). If someone wants to try a 5-off color build, but he probably doesn't want to play it into the 90's, he's heavily discouraged from doing so. We're not talking about mirrored gear and godlike 6 links. I'm talking about using a 5L chest with mediocre gear, and it's still probably going to be crazy expensive to try.


Picture two games: in the first, variety is cost-discriminated against, and power is not. Everyone is OP, and everyone plays the same few builds, because it's too expensive not to. In the second game, power is cost-discriminated against, but there are an enormous variety of builds. Not many people are using overpowered gear, because it's too expensive. Everyone is challenged and has things to work toward. The sky is white with special snowflakes. Which world do we want PoE to be more like?


How do you create the world where players are encouraged to try a large variety of builds, but have the motivational hook of always being able to make significant upgrades to your gear? Which do we make cost more, 6-linking something, or 6-offcoloring something? Well, obviously the 6L contributes more directly to power. What good is a 6 off color if only 3 of the sockets are connected? The 6L will have immense value to many people, because there are tons of builds that need only a few chroms to adjust the 6L perfectly. That transitiveness makes it much harder currency. Even if it's not powerful to you, it's powerful to someone, and they will pay you a lot of currency that you can then spend to make yourself more powerful.


Now, who is going to fully reimburse you for the 8000 chroms you spend trying to get a very specific set of 5 off-colors on a weird piece of armor for your weirdo build that nobody else wants to play? There's a good chance you're screwed if you want to find that particular buyer any time between now and when you quit this game. More likely you'll eventually just decide to take a big hit on the price and dump it so you can move on and play your next build.


I see it as a problem that it's a lot easier to make a 6L than a custom-colored 6-off, because a 6L has a ton of market value and personal power. It's a step in the direction of that game where power is easier to come by than variety. If we want to make the game harder, shouldn't it be more expensive to play an overpowered, cookie-cutter 6L build that requires few off sockets than to play a wonky, untested 5L build that is going to give you a much smaller reimbursement when you move on to the next character and sell your gear?


There's a great deal of value in cost-discriminating against power. That's how you motivate players: by keeping the stuff everybody wants valuable. Is there really as much value in discriminating against variety? Crafting costs should reflect which is more important. When you make a 6L armor hard to create, people will pay whatever they have to to make it, because it's powerful. If you make it cost 20,000 chromes just to try out a build for the sake of variety, very few people will do it. Variety suffers. Even if there are some OP 5-6 off color builds, you still have to consider whether they're any more powerful than spending an equivalent amount of currency on a cookie-cutter OP build that only requires a few chromes to get the right colors, and which allows you to spend those extra 5, 10, or 30 thousand chromes worth of currency on something else.


As somebody said, PoE has a "super awesome skill tree," but some of the potential variety in it cannot or will not be realized due to the high cost of 5-6 off-colors. The armor you choose to wear is a part of your character's customization, and there are a lot of potential builds that cannot combine a certain armor type with a certain skill setup without needing 4-6 off colors. Of course not every build will be feasible, but I think game mechanics should be working to encourage variety, not restrict it. Hybrid armor is one answer, and I've definitely underestimated its value, but there are many builds which are barely able to benefit from the second armor type. That's a huge cost to pay just to have an unusual gem combination, at least when compared to numerous OP builds that don't have to make any similar sacrifice. But I agree with the person who said I should not be so quick to discount hybrid pieces. I guess they're right that GGG wants us to make that choice sometimes. That's reasonable, but I think that it frequently ends up arbitrarily nerfing gem/armor combinations that aren't inherently OP, which leads to reduced variety.


I'm taking it on faith that there is nothing inherently imbalanced about offcolor-reliant builds. Partly because the exact same gem combination in a different armor type would no longer be an off-color build. So the argument there is that a certain armor type in conjunction with a certain gem combination is OP. I don't see it. To me that just means that one armor type is the best choice for a specific set of gems and skill points, and that doesn't mean it coincidentaly crossed the line into OP. Don't most builds use the optimal kind of chest armor? I mean some people use uniques that have a suboptimal armor type, but most builds don't spend their points on 200% increased evasion and then wear a pure ES chest.


I guess what bothers me about making it so difficult to get a large number of off colors is that it discriminates against builds fairly randomly. There are tons and tons of wildly OP builds that use between 0 and 3 offcolors. Why go out of the way to cost-discriminate against other builds? Because they might also be OP? If two builds are of roughly equal power, why make one super-difficult to achieve? That just discourages variety, which seems to be the opposite of what this game is supposed to be about. Builds should be cost-discriminated against according to their power. Not just because they "should", but because it preserves a good incentive structure while offering players the most choice. I don't like the idea of having near-unattainable builds unless they are the godlike tip of the mountain that exists as a final destination to keep people motivated for years. How many people are using a good 6 off color item, one with good stats and the correct colors for a specific build? It's practically unheard of. Unless a 6-off build happens to be far and away the most powerful build in the game, I just don't see that as a good thing.


I see that someone was worried that increasing the chance of off colors would increase overall randomness. Maybe other people are worried about that too, so let me address why it's not a problem. Obviously increasing the chance of off colors spawning makes it easier to get any off-color combo, decreasing randomness in those cases. So the only combinations it would hurt involve primary colors. Say you want the hardest primary color combination: all 6 sockets primary. If the odds of an off socket are currently 25% per socket, you have a 17.7% chance. If the chance of off colors were increased to 35%, the odds are only reduced to 7.5%. That's about a 1 in 13 chance to get the "hardest" combination.



I'm sorry that this post has run to Elliot Rodgers/Unabomber lengths. Jesus I am a crazy person. Honestly there's a lot more like to say about the subject, but this has gone too long. Thanks for reading


TLDR variety = good

"

Dryhumper wrote:

As someone said, I could have used a hybrid chest. I was using a 5L pure STR piece and trying to get RGBBB for cyclone, curse on hit, elemental weakness, enfeeble, and life on hit. The only hybrid piece that would have raised my chances much would be a STR/INT, and the extra energy shield would have actually been worse than useless. It would just disappear the second I cast RF and stay gone until I entered town. It would also raise my righteous fire burn dmg without giving me any additional regen, which is worse than having no ES on the armor at all. I'm in invasion, so I didn't have the option of buying a higher-level STR chest with the right colors. The only decent choice I saw at the time was to pray for four off colors on the really nice STR chest that I picked up and then 5-linked. If I had it to do over again, I would have bought a lower-level pure str chest for easier off-socket rolls and taken a hit on the armor. However, that person was right that I, and a lot of people, discount the possibility of using hybrid pieces. Knowing what I know now about the chances of off-color spawns, I would be less inclined to avoid them in the future.

Just to reply on that :

You could have taken a AR/EV chest, making it already easier to get your colors ( you would probably have done it, it didn't take me long to get : as an example ).
If people were less avoiding the "base choice", this would honestly be much much less of a "problem".


Now, upping a little bit the probability of rolling of color sockets .... why not, but it's not a really big deal to me, if you want a weird combination, getting the best base will allow you to get it without too much trouble.
And if the item is a unique ( = no base choice ), well ... that was probably not meant to be used like that then.

I fell it somewhat wrong, that gem colour chances are highly corellated with gear ATTRIBUTE REQUIREMENTS. That is, low-grade armour piece (let's say, from normal) has much higher chances for rolling offcolours, than high-grade one (from merciless). Since endgame players tend to use highest-grade equipment (as it provides the most ES, armour and eva), they often face this issue. I believe it would be better, if offcolour chances were independant on attribute requirements, thus making it easier (and more consistent) to roll colours.

Also, there is another issue. Why do we want to roll so many offcolours at all??? Mainly because gems of "main" colour are too weak or just inexistant. That's a MAJOR flaw in game design. IF you want to make a melee character, you HAVE to use RED support gems (usually - a lot of them). Why? Because all the best melee gems are red (Multistrike, Melee Splash, Melee Physical Damage, Melee Damage on Full life, Weapon Elemental Damage). And if you want to make ranged attack character, you HAVE to use a lot of GREEN support gems, because those are best for ranged (LMP, GMP, Chain, Pierce, Physical Projectile Damage, Faster Attacks). I dont know about blue gems, they seem pretty underrated in my opinion (with only few good options), though some casters, perhaps, can focus on them.

If GGG added a bit more diversity among gems, and set different colours for gems with similar purpose (like 2x melee damage gems and multistrike), the offcolour problem cwould gone, by a certain margin.

"

Fruz wrote:

Spoiler

That would be an example of what he termed "worse than useless"

I understand many/some builds can get away with like, 600 ES chests. Not all, though.

I remember once making a CI Archer. Ever make one of those?

If you don't use a Vaal Reg, you will be gimped.

Since you have no shield, you pretty much need all the ES you can get on your chest. Just the way it is.

Some builds really benefit hugely from an 800+ ES chest, and don't do that well with 400 - 600.

Coloring becomes a massive nightmare, just because you wanted a CI archer. People will tell you that you "don't need the vaal reg" but they are just full of shit, and the people saying that have, ironically, never made those types of builds.

"

Veruski wrote:

I understand many/some builds can get away with like, 600 ES chests. Not all, though.

LOL
Hint : If you are playing a build that cannot efficiently use a 600 ES chest ( already dumb as f**k ) with a 600 ES chest .... the problem does not come from the chest.

Yes I have played CI archer, no I did not have a vaal reg, yes it was good, one more time : the problem here does not come from the chest.
And it was not in Ambush.

And how much ES would have my armor with a vaal reg base ?
798, that means 179 additional ES.

Is 179 ES something that ruins a build at the point that it's unplayable ? Obviously NOT.
( PS : the survivability that I get from the Evasion out of it is probably more important that 179 raw ES )
If you cannot plan a build that does not have to use a 800+ vaal regi, there is a problem, but not coming from the gear.



Seriously, what a bunch of crap your post is there .....

Personally, I don't mind off-colours being hard to roll. However, I do think that it is no fun to have to reroll ALL socket colours over and over until you land the EXACT one combination that you are looking for. It would be nice if you could somehow save individual socket colours that you want to keep along the way while continuing to roll the remaining sockets that are not correct yet.

I could easily imagine the Chromatic Orb being updated to reroll the colours of open sockets. This way, if an item's sockets are empty, all of them will be rerolled upon Chromatic use. Once you place a gem into a socket, only the remaining open sockets will be rerolled upon Chromatic use. So, if you have a socket(s) that are the colours that you are looking for, but not all of the sockets are what you want yet, you can place gems into the sockets that you are satisfied with and continue to roll the remaining ones until you complete your desired combination.

It kind of stinks that, if you want RRRR on a pure dexterity based item, that you have to give up occasional R sockets over and over until you actually land the complete RRRR combo; why not allow each R socket to represent progress along the way instead of being virtually meaningless?

"

Perfect_Black wrote:

I could easily imagine the Chromatic Orb being updated to reroll the colours of open sockets. This way, if an item's sockets are empty, all of them will be rerolled upon Chromatic use. Once you place a gem into a socket, only the remaining open sockets will be rerolled upon Chromatic use. So, if you have a socket(s) that are the colours that you are looking for, but not all of the sockets are what you want yet, you can place gems into the sockets that you are satisfied with and continue to roll the remaining ones until you complete your desired combination.

It would need to ignore the rule : "impossible to get the same color on a socket" then.
And maybe with a color penalty coming with it.

This idea is not bad actually.

"

Fruz wrote:

Yes I have played CI archer, no I did not have a vaal reg, yes it was good

sounds like you were playing in softcore

( Excuse me for having played invasion when it started )

PS : you don't have any credibility when flaming about this kind of thing anyway, not that you have when not doing so tbh.

when i did my CI archer, it was in the hardcore league (back before 4 months existed)

even then, 600 ES on the chest was basically suicide, and we had legacy rainbowstrides back then

the game has arguably gotten more difficult since then, because now we have things like Palace

it's quite ridiculous to come into a thread and tell the OP, after he wrote like a 4,000 word essay describing his issues, that he should have just "sacrificed a few hundred ES on the chest"

totally ridiculous.

you don't know anything about his build, and there is a 100% chance that he understands his build, and what it needs, a lot better than you do.

it's just more of the same Feedback thread trolling. Come into a thread. Tell them their "feedback is incorrect" or "wrong"

Of course, ignoring that it is actually impossible for that to be the case...