Thinking Outside the Box: What if we had Rare Gems instead of sockets?

Just imagine the investment for a pvp set-up :)

Now you just swap in block pen if needed etc and do on he fly swaps to gems to counter oponents, in your proposed system that becomes a 100+exalt investment for a single set-up.

Honestly, socketing/linking etc should have been in a separate tab and should have been character level gated.

Hell it could even function based on RNG with greater odds to get a better result the higher your level is.

I imagine a system where lvl 65 unlocks a 5-link and lvl 91 unlocks a 6-link for example. Maybe still allow fusings to roll so you can obtain a 6-link prior to lvl 91 etc to allow the gamble.

But really, skill diversity and potential should have been gated by character progression to accomplish a "growth" simulation and provide a positive carrot for players to chase irrelevant of RNG factors.

Peace,

-Boem-

"

Boem wrote:

Hell it could even function based on RNG with greater odds to get a better result the higher your level is.

I imagine a system where lvl 65 unlocks a 5-link and lvl 91 unlocks a 6-link for example. Maybe still allow fusings to roll so you can obtain a 6-link prior to lvl 91 etc to allow the gamble.

But really, skill diversity and potential should have been gated by character progression to accomplish a "growth" simulation and provide a positive carrot for players to chase irrelevant of RNG factors.

Peace,

-Boem-

one example of how to do it.
but can't say this is the first time I read something like this proposed in the Forum, which probably means GGG are aware of it too.
they are just too bloody comfortable with a system where one's skill matters so very little in progression of a character, compared to one's ability to flip a coin, and flip an Orb.
a system, where the longer you do it, the more you realize playing it the old-fashioned hack&slash way, is a long road heading absolutely fkn nowhere.

I'm sorry but this is like saying "What if PoE was a different game"

Sockets are like the defining feature of this game, I've never played another game personally with similar mechanics.

HOWEVER!!!!!

I've been very vocal for a long time about wanting a PoE 2 that has different game systems and this is something that could work in a different PoE game.

This could be a good and interesting feature for Permanent Endless Ledge league (eh? eh?) as it's very roguelike, but I do not think it'd benefit the health of the core game. Skill gems need to be predictable to enable predictable character progression in spite of unpredictable itemization.

I do, however, really like the idea of gem affix (I've posted about it w.r.t. vaal orbs and how bland gem corruption is compared to how outrageously exciting it could be), and am in favor of a system of deterministic linking/unlinking that would be balanced by a more effective mana management system and reduced gem magnitude per link (whereby a 6L chest wouldn't immediately be strictly better than a 3+3L or 4+2L).

Putting the affixes on the gems makes it harder/impossible to experiment with different supports. You're taking away a player choice (how to fill the sockets) and giving that decision to RNG. Needless to say, I'm not a fan of the idea.

The slight tweak to transplant the sockets from the item to the gem is still unfavorable because decoupling the sockets from the item transforms a single, complex optimization problem (finding a good item with sockets and affixes) into two, simpler optimization problems (find good with affixes only, and find good gem) which leaves less room for tradeoffs, e.g. give up better sockets for better affixes or vice versa. Decoupling simplifies the tradeoff decisions. Without meaningful tradeoffs the itemization isn't very interesting. Sockets on items are a great way to create that tradeoff and it lasts through the whole game, even future expansions. It's a tradeoff that never goes away even if they add a bunch of new content that totally obsoletes current gear. POE's itemization that couples two orthogonal character build decisions together in a way that makes logical sense is one of GGG's great design innovations IMO. The core is solid, tinkering should be confined to the fringes instead.

However, I agree there is some bad RNG regarding item sockets. If we could set aside economic impact while rebalancing both gear and currency, I think the ideal solution would be to eliminate the concept of linked/unlinked sockets. All sockets on an item would be linked. 6L := 6S. Jewelers would take the roll of Fusings in terms of rarity and market value; this is not a freebie to make 6L easy. Fusings would be eliminated entirely with no replacement. Vendor recipes & exchanges involving any socket-related currency would be changed.

While we're at it, I would also change Chromatics to incentivize experimentation rather than optimization. The way I would do that is to make them work on individual sockets rather than the whole item. Having said that, they would be still be the last piece to perfecting a 6L, so from an economic balance perspective they are the new Fusings. So they can't be ubiquitous, but there should be a middle ground where the demand for balance and the demand for experimentation can meet in the middle.

That's about all I have to say right now.

"

ampdecay wrote:

I'm sorry but this is like saying "What if PoE was a different game"

Sockets are like the defining feature of this game, I've never played another game personally with similar mechanics.

HOWEVER!!!!!

I've been very vocal for a long time about wanting a PoE 2 that has different game systems and this is something that could work in a different PoE game.

Well there's one person who doesn't hate the idea. :3 and yes, this is more of a "PoE 2" idea.

"

GusTheCrocodile wrote:

I'm not, you know, 100% happy with the current state of 5/6 linking. But I think player choice in support gems, along with the passive tree, provides an important balance that keeps the game becoming too focused on random drops, so for me this is kind of throwing the baby out with the bathwater.

My system has more player choice than the current system.

Here's the current system: there is a very light amount of RNG before getting the two supports you want most. Then there is some moderate RNG before locking in #3 on your priority list. Then some rather severe RNG. Then 4th support you want. Then some extreme RNG. Then 5th and final support.

Here's how I image rare gems would go: there is a very light amount of RNG before getting the two supports you want most. Then there is some moderate RNG, during which you might try various third supports, before locking in #3 on your priority list. Then some rather severe RNG, using various 4th supports. Then 4th support you want. Then some extreme RNG trying to get that 5th support you want and using various other 5th supports instead. Then 5th support of choice; one more to go.

"

PolarisOrbit wrote:

Putting the affixes on the gems makes it harder/impossible to experiment with different supports. You're taking away a player choice (how to fill the sockets) and giving that decision to RNG.

In terms of extremely short-term experiments, you've got a bit of a point (although "impossible" is gibberish). But in terms of getting players to try things perhaps they normally wouldn't, as a result of item availability, there would be more experimentation.

"

PolarisOrbit wrote:

The slight tweak to transplant the sockets from the item to the gem is still unfavorable because decoupling the sockets from the item transforms a single, complex optimization problem (finding a good item with sockets and affixes) into two, simpler optimization problems (find good with affixes only, and find good gem) which leaves less room for tradeoffs, e.g. give up better sockets for better affixes or vice versa. Decoupling simplifies the tradeoff decisions. Without meaningful tradeoffs the itemization isn't very interesting. Sockets on items are a great way to create that tradeoff and it lasts through the whole game, even future expansions. It's a tradeoff that never goes away even if they add a bunch of new content that totally obsoletes current gear. POE's itemization that couples two orthogonal character build decisions together in a way that makes logical sense is one of GGG's great design innovations IMO. The core is solid, tinkering should be confined to the fringes instead.

Ah. Another person who enjoys the design of Tabula Rasa. Hey, I do too.

But at the end of the day, this coupling you speak of is a clever idea but not very functional. If it were two well-designed systems then it would be one thing, but its not. Imagine a game where you go through with just white items, using only socketing as a progression mechanic; could you enjoy playing such a game? You would need two mechanics each strong enough to stand on their own before trying to couple then like that.

^Your basically asking players to run around with poorly optimized gem set-ups.

While the few wealthy kids get to shit all over them with the few available optimized ones on the market.

This game is about player progression and character development (alongside the competition) and having freedom in that is important.

Your basically forcing experimentation on players who don't necessarily enjoy it or even want to be bothered with it.

That is to say, sure the first time that arc + cold damage drops it's "interesting" but 3 league's later it's just a nuisance since you already depleted all the options and know what is optimal.

It's similar to what happens in the current situation.

New gem is released

-> experimentation phase begins

- multiple supports are tested and arranged in different set-ups and theory-crafts.

- best ones makes the cut and get published.

- forget all others since they are not efficient in comparison.

You would implement what would inevitably lead to a bad player experience, if not at the start (everything is fun when its new and shinny) then in due time.

Peace,

-Boem-

"

Boem wrote:

You would implement what would inevitably lead to a bad player experience, if not at the start (everything is fun when its new and shinny) then in due time.

Every system would lead to a bad experience eventually. But as I explained to Gus, I think mine wouldn't be a worse experience. That "forced experimentation" wouldn't be a permanent thing, and just like players now eventually 6L, players would eventually get the gem affixes they want most. The difference would be, between 5L and 6L would be the "experimental" 6L, or a 5.5L if you prefer that term. What's worse about that?

There are currently 72 different supports gems.
So for each time you would want to add a mod on your gem, you would have a very low chance to actually gain the one you want. And that is if we don't take level or quality of support into account. Now make this random process five times. Do you swim in orbs ? Even mirror-crafters would find the process horribly tiresome !

Even if we restrict the supports available based on the skill itself, it still leaves a big marge of RNG compared to what we have actually in the game (which is again quite low if you stay at 5L since the masters expansion).

You know the difficulty to find an item with the 4/5 mods that you want in the game (basic 3 res+life+armor for example) ?
Now you had that difficulty on skill gems with even more variety on the mods. Sorry but no.

There are not 72 different support gems for every skill.

Just as a random example, Split Arrow has 44 potential supports, 46 if we're trolly and include LMP and GMP. Cutting some others, like Chance to Ignite, from the pool might be a good idea. I'm guessing about 40 make it.

Caster weapons have 13 available prefixes and 24 suffixes.